ETC UCC3305

用 UC C3 30S 作均控制恶的 HID 却电子镇优忌
藏黯紫黑潜
用 UCC3305 作为控制器的 HID 灯电子镇流器
摘
山东省临沂市电子工业公司
毛兴武刘维志
山东省临沂市阳光科技有限公司
薛彦状
要:国产f{lD灯电子镇况器尚未进入实用化、 H fl)灯启动、"卢共振"抑制和可靠性设计等情多关
键技术亟待解决 J 本文介绍了 UC C3 305 型 HID 灯控制器的功能特，虫，重点介绍了 1月其作为控制嚣的
I1lD灯电子镇 LfL 器实际电 J各
关键词: 1111) 订
控制 ~S lC <:3 305
全辰1 卤化物灯镇流器
H;
一、引言
3. 输出功率恒定‘大小 líJ j)l;J ;
高强度放电 (11 11) HJ 主要包 i，~高压的 (11 1'5) 灯和l
」
金属卤化物士 r (简称金 I句灯)寺，是 I~ rìíJ )\;;川比较 f
泛的新型节能也 ;UI卓， HID 灯的触发 fr~ z;}J ( l!fJ }，\火) 1t!
f巨幅值往往达数 F 伏以上，
{I 高!切!电视{J飞 I t!怕出 F
4. 内置分频 2揣~， 使1 史 脚!肉问岛却
IJ J 归
1 5 (0
叭()兀
UT川、) 'j川i阳 16
创(0
叭()兀
:
1 呻俨
i川r1丁r、
的输出频 2率字为 PWM 振 7荡克器!衍顿纣烦!韦[的
l向(J 11 川 2 ， 从 rM IIJ 1功
iI'. HID 灯发生声频共振;
S. 启动 tij 稳态 1 付一士 r It! ìfrt 之 lt 11) j).';J:
会出现放 1 1.!， 1 t!弧斗、稳 (H [l "l h Jl 11r(")乃至，);12，弧，使用
环境条件十分恶劣，故用普 j国荧Jt土]电子镇流器去
点燃 HID 灯片:不适合。
进入 20 !lI:纪 90 年代中期，国内许多高科技公 rJj
或大专院校足科研机构，开始致力 FHID 灯电于销
;武器的研究、 Ç4~、 L. 业科技发 JR Jýf I~l 111 j 丰j (19961997n将 "HII 川、 J 1 包 f 锁 ifri (，~ j F 发"夕 IJ }j [1 j主项目 )j'
州'1 科瑞公司悍 j 一 1996 J， j二 6 Jj 11 111] ú: 北点公斤招标，
出资 1000 万元人民币，购买 HPS 灯电 F 镇 j武器技术
r.业产权，当时尤
家投标， ll 能尤 J)J ](1] j且
也今/)
6 ，过压、过流、 1、11 白皮卅路等故障保护，捉内
系统的安全 N:;
7. 工作电压低，电源电流小，峙但输址:驱动 [I.!， i在
达 lAo
UCC330S 为控制手!J!W 功 HID 灯提供所必甫的女
，~i) 功能，是世界 i可款 Hm 灯 I t! (-镇 j武器仁fH I C,
三、用 UCC330S 作为控制器的 3SW DC
金卤灯电子镇流器
用 UCC330S 作为控制嚣的 3S \V 1)( 尘 1M 灯也 f
--rL ，国产 HID 灯 i扛 f 镇流 ~~i 井水 i丘入实川阶段。尔文
介绍的用 UC C3 305 作为控制嚣的 HID 灯 I t! J气镇 1武器
电路，虽其负载 Jjj J，ç 只有一 35\飞/ , (1 1. fJ)为设计 1 /11)灯 I t!
F 镇流器提供 r
个币:要的 11、路，
二、 UCC330S HIO 灯控制器简介
Li C C3 30S 采川 2X JJt~J 川 01(: J.J 在‘ l人J iiM/i j:}0 úlll 'i- J 1
慎 1m 器电路白[l图 2 所不) 35 \V DC 金 l.zï 灯f1 10SBMI
5Y l. VANIA 制造，灯的点火屯庄市 J 500\ ,\c
i支金卤灯镇流器 [)C二输入 I I.!， Lk-;也因为 9 - 18V ,
DC 输入电压经 R9 施 Jm 到机 I J阳J ， 1/" IJ却 J. (r(J 最品 'li
J I~
沟 8V( 并被 6.8V 的卉纳\悦 ?13币 1亿人胞 Jm 到 i在!阳
(I(J It! ifrt: 11~
r
1肌气 ) U 1( UC C3 305 )的大多挝 JJJíì8‘必须
!奸后。 虽然}亥 J，('-Jl IC j是 Jl}己包IJ (L :'-J ITIW I) E) 公 ü] 古
{U]tll 18 (V B0051、) fiûi JJIl
90 年代中期推出的产 il-II ，但、可lI;t H 约用户提供 f 伺
UI JJ却 17( PUMPOUT) 输出频率为振旅器频率 1/2 的
机的 JNJH 扫 i 扑作一剧， j(IÎÚ1E 的山Jl J }f发，是近!树I， j 二发
力-坡，出幅在 Vcc 与地之 r~L AC 方波脉 ìljl !jV功 υ3JF
生的事情。
关，.Ìl且过口和开关管 (INS819) 月 Jl d.i IOV ，施 JJU 全IJ
UCC330S 币 Jj' IC 包含-个 I I.!， i在刑脉宽训 1M 器、
!阳.18"
灯功率调整 2丹、灯 iiIAJ支补住址件和H出阵保护 Il~ 附匀，
其主要特点如l'f:
I AV
个 10V 的山 J J;. 才能实 f见，
川脚 21 (05C) 相脚 1 O( ISET) 可 j 也之间的外出
屯在
CI4 和1
R15 ，
设定打fç 局出颇丰( "..般 JJ
1.固定顿午 L 作与 l 包 1M Jt'! 1宁;j川;
100kllz)o UI R却 10(PWMOUT) 输 !W I' \VM Ivk j1 jJ , IWi)J
2. 既可驱动直流 (DC) 也 11J !lt<: z.JJ 交流 (A C) I J[ D
J)J 宅斤关。 1 ，通过 L2 升 !E 有1
IJI
1] C4 常流陈肢，
".
ι
飞「
4矗阻碍捆'
{也手元 B<非启用 h∞Of在 3 月总第 7 期
8町 124
苦
囚
iH
11
VOUT I
111
SENSE 气
R
141NOTON
131FLT
To
Current 唱-l 11BYPASS
S川
LOAD! 5
ISENSE 气
71LPOWER
囱
1
〉片
t
6!WARMUPV
WARMUPC 13
.
仨」
口口
V
RJV
Li miter
Gain=1.33
-一------------一-
图
- -一- - -- - - - -- - -一-------------------UCC3805
HID 灯控制器内部结构框图
在 SEPIC 型变换器输出端，产生一个 75 - 90V 的 DC
比较重要的环节。
电压给金卤灯供电。在灯点火之前，镇流器输出可视
Ll中的贮能在 Ql 开通时储存，在机关断时
为开路，此时镇流器提供一个约 600V 的开路电压将
释放传输。设 L2 的臣数比为 n ，在灯启动时最坏情况
灯启动 c 在灯点火过程中，随灯辉光放电、弧光放电
下的输出电压 Vo(max) =500V , Q1CIR Fl 310) 漏极承
和灯管击穿，灯管两端的电压急剧下降，尔后随灯管
受的最高电压 Vos(max) = 100V , V1N(maX) = 16v ,
内气压和温度升高，在约 1505 后，电压逐渐达到
n 可由下式求比:
\1"，(m，飞)
Ql 源极电阻 R2 为电流检测元件 ， R2 上的检测
信号输至 Ul 的 23 脚，以履行峰值电流限制功能。
R20 为电流检测电阻，用作监控负载电流和功率。镇
流器输出电压经 R21 、 R22 和 R25 分压输入到 Ul 脚
11 ，用作监控灯电压。脚 2 、脚 3 外接电容 C13 、 C12
及 Ul 内部相关电路用作监控灯温。
在图 2 所示的应用电路中，磁性元件的设计是
= V1N(mad
ν 内，、
+斗一 m8 l1i
(l
I
n
、、且.，，，，
75 - 90V町的稳态值，灯电流约 450mA 。
因此，
机。 (mø.x500V
n='=z6( 臣)
V05 ( 阳
， na，墨川) 一 V1N'1咱叭仙(11川
E
I与
灯电压范围为 6ωO-'1lOV ，帖)(汇
c沾'稳态工作时开关 占
I~一!厅
i
空比由下式计算:
VoU = VIN
• n一旦一­
川""}一 D
(2)
11
用 UCC3305 作殆栓倍1 .a..伪 HID JIl"电子镇忧思
C2
LAMP
iT
R6
270k
OS
ivv
01
川
C2 1
川
I--
,F
R7
100
图 2
在 Vo(max)
= I lO v 、
V， ~(max)
35W DC 金卤灯电子镇流器屯路
=9V 下，最大占空
比可根据公式 (2) 计算:
10 =60 臣，磁心气隙长度 0.56mm o
η…
VO(m.， 1
110
一一--一=一一一---一~=2.04
1 一 Dm制
VI:\(m;nl • n
9x6
Dm;王x
选取 Np = 10 臣，次级绕组iYs = n . Np = 6 x
p
四、用 UCC3305 作为控制器的 35W AC
金卤灯电子镇流器
= 0.67
利用 V川 max) = 16V 和 Vo(min)=60V 及 n=6
和公式 (2) ，可求出最小占空比 Dmin = 0.38 0
对于 j:" = 100kHz 的开关频率
用 UCC3305 作为控制器的 35W:\C 金卤灯电子
镇流器电路如图 3 所示 O
to~(max) =
i京电子镇流器的负载是 35W AC 金卤灯，由飞手IJ
6. 7 阳， lo 'i (min) =3.8 问。在 VI~ = 9V 和 PO = 36W 下
浦照明部制造 O 为避免声频共振， AC 金卤灯工作频
的变换器效率 η= 859毛，平均输入电流则为 4.7A 。若
率必须低于 1kHz ，但又不能低于 100Hz ，以防止灯闪
设输入电流纹波峰→峰值/''i (p _1' 1 = 2A ，那么 ， L1 的
烁。方波驱动电流是比较理想的，原因是方波电流可
电感值为，
使灯电流波峰系数(即灯电流峰值与均方根值之比
值)接近于 1 ，有利于延长灯的使用寿命，且可使灯功
L扣= 川川(，川丁飞"丁飞
V'N( t1
川『?:工兀;:川儿
; ;:;l
率恒定，光输出稳定，消除电流过零时弧光重新点火
利用 EIOO 一 1山
8 磁心丰莉和
T啊lφO. 佣
9 8m
川
川川
11川川1ll川
11 l 的导线 3
列
O 臣
I:!吁泪pP
可
iJ c
若设 L2 初级电感峰值电流为 3A ，那么初级绕组
电感值则为:
0.379) ，磁通密度 B =0. 1T ， 磁心有效截面积 AE=
89x lO- Ó 1l1 2 ，又 L =20 μH ，初级绕组峰值电流 /p =
3A ，初级团数可由下式求出:
12
2O u. H x 3A
入端 (TB1 )的 DC 电压范围是 9 - 16V ，经 LI 、 Q1 、
换器升压，再通过一个 DC/AC 变换器，得到→个 AC
选用 RM IOPA250 - 3F3 磁心(面积乘积 A p =
L . /"
A B
图 3 所示的 35W AC 金卤灯电子镇流器电路，输
L2 、 D1 、 C4 及 Ul 内相关电路组成的 DC/DC 升压变
9•6 . 7
L2p =一 3 一= 20( 州)
~=忡
E '
而出现闪烁的可能性。
电压作为金卤灯的供电电源。
U1 的振荡器频率为 100kHz ，经分频器该频率
除以 512 ，在脚 15 和脚 16 输出 195Hz 的低颇，作为
U2 、 U30R21 1O)高压高速 MO$i栅极驱动器的逻辑
输入控制信号。 IR2110 脚 1 O( H'N) 和脚 12( L"，，) 为逻
~=7 臣
(3)
4发应用脚
〈也手元jS<斗启用 }20∞年 3 月总第 7 l1ll
R31
C31
R2 1
出5k
问，J，
3.3k
11
R l1
5.1k
5%
C25
C19
O.Olll F
图 3
用 UCC3305 作为控制器的 35W
辑输入，脚 7( Ho) 与脚 1 ( Lo) 分别为高端和低端输
'
AC 金齿灯电 F 镇流器电路
变电流。
出，并且 Ho 、 Lo 与 HI~ ， LIN 是同相的。 U2 、 u3 的输出
H lD灯的启动必须借助于点火器(亦称启动
电压驱动。 2 - Q5 四只 MOSFET 组成的桥式逆变
器 )0 H lD灯的点火脉冲电压通常在 4 - 30kV 之间，
器。
Q5 采用摩托罗拉公司生产的 MTP8N50E ，
冷启动灯的点火电压低一-些，而热启动灯的启火电压
e 额定电压 Vnc 为 500( 电流容量为 6A) ，以承受镇流器
则较高。传统的点火器采用升压变压器，大多是利用
450 VIl C 的开路电压。金卤灯连接于 TB2 的③、⑤端之
火花隙 (Spark gap) 元件的快速击穿特性(小于 50ns)
间。当 Q2 、 Q5 导通而 Q3 、 Q4 截止时，电流流向是
在变压器初级产生触发脉冲 ο
Q2→③→金卤灯4⑤→ Q5 →地;当 Q4 、 Q3 导通而
功器，采用品!时管或带负阻特性的硅双向开关
Q2 、 Q5 截止时，电流流向为 Q4→⑤→金卤灯→③→
(SIDAC) 等器件作为触发元件 a 图 4. 为典型的金卤灯
Q3→地。因此，通过金卤灯的电流是低频 (195Hz) 交
触发电路 O 与升压变压 ~i 初级串 t亮的 K2401F1 (或
Q2
-
目前有很多 HID 灯启
K2501 F I)是 SIDAC ú 当施加于 K2401F1 的电压达到
H.V
BALLAST
STEP.UP
TRANSFORMER
其击穿值后，迅速导通，压降仅约1. 5V 。在 K2401F1
导通瞬间，产生一个脉冲，经变压器升压，使灯点火启
动。一旦灯被点燃，灯上的电压降低于 K2401 日的击
穿电压之下，从而使 K2401Fl 阻断。
METAL
HALl OE
LAMP
因 3 示出的金卤灯镇流器电路，在灯点火之前，
输出电压约 450VAC c :在灯点火后历经约 150 秒，灯
进入稳定工作状态。此时灯电压压.约 8ω0-9
列
OVAC ，灯电
流为 450mA ，最高效率 η
孙"川川'
(福特电力电子公司是 Unilro
口
r刷叫)(叫
o巾
di怆
e 中国区代理商之一，绎销
图 4
典型的金属卤化物灯触发电路
UCC3305 专用 I比
C. 联系电 t话舌: 0519 - 6684942 6688597)
13
「气
υ
nυ
η气U
ηδ
…
jC
丰
葛亮东J教:电灯在其校剖器
U
uC 协
·新特器件应用... .~"， \川、 ic
--:-9 一
R
高亮度放电灯及其控制器口 CC3305
;王学明车励熊光宁
摘要:，:白炽灯和曰:尤，灯的亮度和技丰不理怒，本文介绍丁效率高、亮度强的高亮度放
电灯(r.IIPt~1民 1日内灯、高压钩灯、水银灯、金属卤化物灯和新一代尤管射频灯的结构
E
二和特点￥剖J达 1 号卡桥镇流器电路的:工作原理，讨论丁专为高亮度放电灯设计的控制器
;UCC3305 的工作原理、内部结构，并给出丁 UCC3305 在 HID 中的应用电路。
关键词 t声，完皮放电4灯光管射频止了点火起弧纳灯水银灯金属卤化物灯
UCC33g5
:‘主
li、高亮度放电炉右。)': i 三::
结构与水银灯相似，只是灯管的大小和电极
的材料不同。由铝、银z 鸽或不锈钢等导体制
电灯崩理电能转化'成可见光，能的器件，但
成的金属支撑框与放电灯底座间接相连，这
是电灯并不能把所有:的电能全部转化成1可见
样可以使放电灯的结构更加稳定， I 同时又可
光能，其中一部分将变成非可见光能忍如紫外
飞将各个鸽电极连接起来毡，水银灯有雨种电极，
线汪红外线 L 热能等江阜炽灯的效率约为吐 1% ，
二种是启动电极，一种是王作电极 c 启动且极
雨蒙出的红外线品时肖耗能量的毛~ Yo" 热能
的距离很近，边样:可加速惰性句体氧的电
占 ~Q%) ~'日i光灯的效卒稍高，为 23:%、红外线
离。启动电极与石英晶钵或巨移喜晶铝放电管管v相
古 36}~ 曳热能占.4; 1%幸};\ t'
连 4 在多晶铝放电管‘中啦F
芹军生弧光般电
':'l 我们知道月，几光的波长为坷。?、
电使电能转化3为
fg 电f磁幅射;;因此水，绿灯也称
Z70~lIF 可紫安h线随波长为 18p'~穹3，~Qqrn;:: 红外
为弧光管 Q '.在其啪的 7个启动电极主串联一
线的波长为:~1.8.俯帘00伊RY!:红处线和紫外线
只电阻可以限制启动‘电流~~一在这种双电极结
都是不可见光:Þ' - 坦是紫外线ïlJ 以通过磷把其
构的放电灯中，一旦主电极 C工作电极?进入
中一部分转换成可见光。工 if)~二'乏
正常王作状悉与 ì启动电极的电阻比k王作电极
之 7 飞为了提高灯真的效率 p…科技大员又发明
71J
了高亮度放电打耳目D， High ， Inderi~ity ':Dis~
JL11框
外壳
charge) .、目前常见的 HIO 灯有:低压倒灯
釜光罩
(LPS> 、高压铀灯 (HPS) 、水银灯 (MV) 和金
弧光放电内壳
属卤化物灯 CMH) 灯等。
任何科放电灯都需在灯管内添充少量
惰性气体以便芹生电离，常见的气体为氧气
和fffi.凭， i H1D:坷的电极常用销和水银为材
料，在金属卤化物灯中则使用映化锅讪在水
银灯的内壁上大多涂有磷粉，时便把紫外线
转化成可见指点~:\:'
H .:}-飞儿}~
， ;.~j.l~ .:~;l
J亨、水银放电灯的内部结构如嚣1 所示 r 高
压纳灯的结构如国 12 所示?气金属卤他物好的
'一，
4
宅、矗
~，
A
一 10 一
《国外电子元器件))1 996 年第 8 期
1996 年 8 月
之前，灯丝电流持续半秒钟左右扩瞬间居勤
，←安全? 乒. ::;:时两制叫入高吨，可在瞬间内激发
吁
弧光放电内烫'
门世电离。伽
…冷阴极日:光灯中，只有两根灯丝与灯管相
连二所以必须采用瞬间启动方法。采用预热启
帆
动和快速启动时，灯管的阴极必须与加热电阻
支撑框
丝相连，困此称为热阴极。在正常工佯状态下，
电极
通常在两个电阻丝间串联一个电容，便很小的
电流连续流过。电阻丝中电流必须加以控制，
4
准确定时，以免损坏电阻丝，造成故障。
通常日光灯管的内壁都涂了一层磷，这
样可将二分之一→的紫外线转化为可见光。荧
光是通过吸收而产生的可见电磁光'-只要热
图 2 高压制灯的结构
源存在，就、可持续发光。在所有的放电灯中，
都或多或少地应用了荧光原理。
大得多，就不再起作用。
0
放电灯内壳的材料必须能够承受 1300 K
1. 2 水银灯
高温，且不受灯内的水银、卤、销等金属的腐
'最初，水银灯广泛用于工业及商业室外
蚀 L 水银灯的内壳材料采用可熔性珊酸盐'层
照明， .这是因'为，它的效率高于白炽灯、且输出
的二氧化硅苏打石灰玻璃或石英。金属卤化
功率可高达 400W. 此外安装及更换费用都比
物灯的内壳材料采用陶瓷，飞锅灯则使用专门
较低。高压水银灯的效率约为 17% ，每瓦功率
-研制的烧结多晶铝 ('PCÄ1 材料。飞
，可转换为 50 流明 L 但缺点是可见光中蓝光过
3 波长在 '200'----i5 000nm 1之间的光可穿过
强 L 目前，水银灯'已多被金属卤灯和高压纳灯
铝或内壳材料-而其它波长的光则被内壳吸
所替代.，金属卤灯的光色质量较好，通常用手
收，转化为热 J 造成功耗 J
室内照明豆高压纳灯通常用于室外照明。
...-;"\
t~ 放电灯的外壳，通常由透明度极高的砂
1. 3 金属卤灯
酸棚玻璃制成。在玻璃的内壁涂上磷，可以
金属卤灯是由水银灯派生出的，在光色
改善灯光的颜色，同时可将紫外线转化为可
质量和效率方面都比水银灯有所提高噜而且
见光，所以玻璃的内壁呈毛玻璃状。i 水银灯
光的颜色比·标准水银灯白得多，接近于自炽
的内壳和外壳之间充有氮气，这样可使内壳
灯，同时效率较高。光电转换率约为 75"'110
隔热，即使弧光温度很高，外壳也不会过热。
流明/瓦。
下面简单介绍常见发光源的工作原理和
在放电灯中加上卤盐(锢、i 轮、腆化锅、
特点。
锐)是为了在汽化中使金属原子充分集中。这
1. 1 日光灯
样与永银灯相比，在红色区和黄色区内光谱
日光灯是一种低压水银灯，灯内充有氧
气。有三种基本的启动方法:预热启动，快速
色线将增加。
f."4 低压鞠灯
启动和瞬间启动。预热启动是最古老的方
最早的低压放电锅灯在欧溯户泛用于室
法，启动时，首先将鸽丝(电阻丝)加热一秒钟
外照萌斗它的最文优点是效率高:，光电转换率
左右，以激发离子。 1 快速启动时，在弧光产生
可达 200 流明/瓦。它的最大缺，点是:发出的
高亮度放电灯及其控制器 UCC3305
1·A
品呻
光是强烈的黄光，使人无法分辨颜色。
个步骤?以便使设计出的;镇流器更合理。
第 1 步:点火
1. 5 高压纳灯
人，-1 :;:~了，三
高压锦灯效率可高达 31% ，光电转换率
灯中填充的氧气和惊气必须从起始的非
为 102 二:-129 流明/瓦、所谓"高压"只是相对
导电状态激励到导电状态卢 HID 灯才能点
而言，这类放电灯内部压力为每立方英寸
亮，这&个过程称为点火也在:放d 唔灯的两三个电极
5~10 磅 (5.-.;.:10PSD. 还略低于正常气压。
由于高压锅灯弧光管的直径很小，所以
不必用启动电极点火;高压纳灯需要用高压
高频脉冲起弧。起弧后在加热过程中就会发
i
z
国压击穿~
州
电流击穿
v}11千土负植区
?才正阻区
二:\:
f..一一叫
出比低压纳灯更典型的单色黄光。高压销灯
叫
中的惰性气体为惊气。由于氯气击穿电压较
lÒ 2，
可
高+所以点火电路必须提供较高的电压。、，
10
1.'6 光管射频灯-
"
飞、
,.....
、}
pes" ， ~RÞ LÌghtin'g)
二光管 ι 射频灯(" i>ì'
牛
恒功率二JJ，\
1
100
L..
是:一种新型光源，它:采用了微波技术。
2: 45MHz'，的微波可从硫 L 氢及其它气体的原
子中产生一种自由电子和原子核的等离子
固 3 放电灯特性 :iJL丁一飞气 I(A)
t之间加较高电压即可完成点火过程。?
体'0' 与其他高强度放电〈自ID) 灯类似，这些电
:\第 2 步r 盖格区 (Geiger:RègÌòω:
子改变能级时释放光子，发出可见光，但不同
~. ，-盖格区也称早期辉电状态，三当两极之间
的是这种灯不需要电极。目前，在华盛顿国
的电压逐渐升高时，两极间3产生断续小电
家宇宙博物馆中使用的就是大型的光"管"
流 il:泣种J情况相当于在阻扰较高的电容两极
。19忧"Pi肘'儿
板'之间加人.一定的间断性电压-从而在电'容
1::，，~ ì冲这种光"管"是在磁极g产生的电场中旋转
中产生间断的漏电流。但若由手电琉密度较
的玻璃球;在小型先"管"中1 光电转换率霄
小，不能达到所要求的放电状态，所以还必须
达到 '200 流明/瓦，光管的居动和再点束时
增加两电极间的屯压增加电流 tfr:、生放电。
间为 10 秒。由于灯申没有电极和磷 J所以寿
命由镇流器决定。 A
2 、放电灯的启动过程
3
\
第 3 步 z 汤森法放电(离子雪崩') :'
汤森德放电开始时两极间有电流流过，
但是弧光放电仍然是间断的。此时电容两端电
压略有上升，电流就会大大增加，就象老式的
放电灯的工作可以分成 3 个阶段:
人工预热的日光灯管那样 J 大约→秒钟后，灯
①电极释放出自由电子，电子在外加电
管中产生意巧的挥卖?但民再究坏严暗得多。
场中加速。
②自由电子的动能转化为气体原子的激
发能。
③激发能转化为光
第 4 步:电流:岳穿 l' 孤兑‘‘
从电流击穿这一时刻起，瓢光可完全自
持 j 此后弧光放电电流为连续电辘二电流击穿
时，在阳电极处每个正离子都至少~生一个
、~ ~~图 3 为放电灯的特性曲线?放电坷的启
电子-4 因此能够自己维持放屯。，在这ι阶段
动过程就是通过1点央产吏弧'光并使之稳定的
中，放电灯两端电压很稳定;但放电电流可能
过程手下面详细阐述从点火到正常宝作的各
增加几个数量级，在设计镇流器时，必须考虑
一 12 一
《国外电子元器件))1 996 年第 '8 期-1 996 年 8 月
'
这个特点 d 从汤森德放电到电流击穿的、转换
称为辉光一弧光转换。
第 '5. 步 LZ 电压击穿
夺、
.r 认
ι 随着电流不断增加，最终放电灯两端电
压将急剧下降，这种现象称为电压击穿，应当
d
B
扯
l
号、
注意，吗弧光灯的正常工作电压低于起始(电
离)电压。
第 6 步:亚辉尤放电
电压击穿后，放电灯开始亚辉光放电。
国 4
半桥缺 iJÍi器电路
桥式整流器中的 DJ~P4 将交流输九电
此时放电灯呈现负阻特征，随着放电灯电流
压变为脉动直流电压，经电容 Cl 滤波后变
的逐渐增加
成比较平稳的直流电压。电容 C1 的值要足
时放电灯可等效为背对背的两只齐纳二极管
够大，以保证电压基本不变斗若将电容 εl 变
与一个电阻串联，齐纳二极管的电压大约等
成两个电容串联，就可构成半桥电路运半桥电
于放电灯的工作电压。当电压升高到可以击
路的两只开关管交替'导通;，-、等通间隔约为工
穿稳压管时，才会产生弧光。镇流器必须能
作周期的→半。开关管的门极驱动利用自激
适应这一现象 0'
振荡技术，由一个工作频率大约为 40kH~ 的
在放电灯的启动过程中，应避免进木异
自饱和门极驱动变压器来完成;一，
川
常辉光放电状态。?如果放电坷的功率超过额
驱动， HID 灯需把低阻'抗电压源变成高
定值 ν 放电灯就进 A 异常辉先放电状态
阻抗电流源，这通过串联十个电感即可实
sAbriòtm a) GlÞ~， Discharge) ~此肘电流较
现a。当:电流在任一方向使变压器饱和时，1]极
大-放电灯呈现正电阻，电压随电流升高而牙
可区动变压器就改变极性，毛从而使峰值电流受
高ι 放;直灯的今温度急剧升高气这种状态会大
到限制。这是一种原始的电流控制和电流限
大缩短放电灯:的表命，因此要极力防止出现
制方法与但在寸定的输出咣卒豆揄〈电压和谐
这种技态。」吗 4
波分量容许范围内，它，可s助理常工作与采用这
·当电源发生故障或有意中断电灯工作
种技术时，"电灯电流波形为三角波 ~::C?!在启
后，再重新开灯时，、需要热点火。在热状态
动期间，在开始充电至电压稳定(此时电压等
下，; HID 旱期辉光放电所需电压比冷状态
予电源电压的一半)的过程中?还为电流提供
高，因此镇流器必须能够产生足够的电压，使
一条通道，给灯丝加热。
HID 灯重新点火，否则，就要等到放电灯冷
半桥电路适用于大多数 HID ~了 J 包括水
却后才可重新点火。:川
银灯和高压锅灯。这类灯和日光灯最主要的
3 、 HID 灯镇流器设计考虑
区别是点火特性不同。与日光灯相比，
"
主镇流器的功率级应采用双极型器件，驱
动电压'要适:当，输出功率要恒定，并能准确地
HID
灯所需的弧光点火电压要高得多，了一些镇流
器通过串联电容和电感组域的谐振回路也可
达到这个要求。
控制电，灯电流，.0' 了目前最常见的是采用自激式
上述的自激振荡技术可能出现很多问
半桥变频器飞互作原理如图 4 所示。应当说
题，，"1 其中最重要的是它直接影响放电灯的功
明'，.:为了减小电极一触点的金属沉积， 'HID 灯
率二从而影响照明和放电灯的寿命。特别是饱
必须由交流电供电。，
和门极驱动变压器铁芯材料磁导率的容限通
.
'··在
高亮度放电灯及其控制器 UCC3305
qJ
常为士 25% ，这将会改变产生饱和所需的电
强度放电灯要求的控制和保护功能。该控制
流，从而影响对峰值电流和放电灯功率的控
器主要用于自动直流灯的控制，但也可用于
制。应当注意，如果输出端出现过载或短路，
交流供电的高强度放电灯。广
UCC3305 的内部结构如图 5 所示，其主
开关频率和功耗就会大大增加。因为大多数
灯具都可以更换灯泡，所以许多镇流器的高
要特点下:
(1')输出功率恒定，大小可调。
压端可能对用户造成伤害，因此，安全也是个
8 可调起动 f 加热电流
重要问题。如果在开灯电路中或在输出端加
电预置最低和最高保护电压'
入过压检测电路乎就可大大减小事故概率。
水银灯和高压纳灯的控制还存在一些其
他问题。这类灯需要几分钟才能达到稳定状
(2) 最大电灯电流和冷/热电灯峰值电
流可调。 t
态和最大密度。一旦加热后，这类灯就不易
(3) 电流型控制。
再点火，只有充分冷却后，使内部蒸汽的压力
(4) 再点火和定时时间可调，并且具有电
大大降低，镇流器才能严生所需的启动电压，
灯温度补偿 o
、J
使电灯再点火。理想的控制电路应适应放电
(5):有过流保护功能，可防止短路画。
灯的内部温度，相应地调整驱动信号或延缓
(6) 无灯或电灯开路时的过压保护二
再点火。此外，还应考虑温度、湿度、使用年
(7)工作频率固定不变。
限所造成的放电灯电压和照明的变化。
(8) 主输出采用大电流 MOSFET 推拉
电路。
5 、 UCC3305 高强度放电灯控制器
(9) 两路输出的占空比均为 50% ，可为
电灯提供双极性驱动信号。
UCC3305 高珉度放电灯控制器具有高
(0) 带有闭锁故障指
电飘r--
刁亏。
→《【， ι
--…-"...-------一一…------------←----------，
'-'>--
(11)工作电压为 5V.
'
吁'
工作电流很小。
•
(2) 内有 5V 电荷泵。
(3) 为了适用于低频
(200Hz) 交流供电，内部设
有分频器。
UCC3305 控制器通过
调整 PWM 占空比，向电
灯输出恒定功率，同时还具
有许多保护功能，可根据所
用的电灯适当调整各种参
数，快速启动和热态再点火
都可以按应用要求设置。该
IC 的单端驱动可与许多变
换器电路连接 H 如反激式、
图 5 UCC3305 高强度~放电灯控制器、
?黑
正激式、 SEPIC 、升压或降
《国外电子元器件))1 996 年第 8 期
-14 一
1996 年 8 月
压变换器等。否obT 和，Qom 输出端的王作频
功率状态下，电灯电流和电压都被控制电路
率为振荡器频率的 1/512 ，这样可减小大多
IC 限制在一定范围内。
数高强度放电灯中的声音共振 4 这些输出端
6 、结束语 7
的电流较小，可与高压驱动 IC 连接，也可与
许多开关型电源的设计原理也适用于照
半桥或全桥电灯功率级组成网络。
UCC3305 组成的高强度放电灯控制电
明工业。高强度放电灯的电子镇〈下转 P35)
路如图 6 所示。启动时，电灯的电压、电流和
卢噜宁'户'嘈甲
功率波形如图 7 月『示。由图可见，从 h 到 t 1
Vl_
ωWI每格
之间在点火后电灯的电流限制在 ZA ， 电灯的
~
电压逐渐升高。应当注意，电灯的功率也逐
J/r
乍、"
、/
"'--
、刷、
30WI每格
渐上升，直至功率最大时-电灯进入稳定工
--"--
作状态。然后，在 tl 到 t，2之间，电灯的电流和
"--
1"唱格
、、
功率开始下降， ，而电灯的电压开始升高，直
到 t 2 ，电灯才开始稳定工作。从 t 2 开始调整
、、
问悔、
..L_J
•• ...L.晶晶..-
2
10
电灯的功率，使其适应因温度、寿命和产品
,
h.
..
图 7 起动时，电灯的电压、电流和功率波形
结构的差别而引起的电压的变化。在这种恒
、，
,
忡-- 性问
1
t :,
~二量
句句'.宇
:
ζ 飞·→..句..
‘
叫
-
-"
ι
.‘‘
，、飞，呵巧"电-
J
‘
'
臼
俨占­
t>"
飞
、
ι
la
也
、
?、L、
~
RSL 1
L-_
__..-0→ _J ←一
卜
I 一一 _-J 一→一 J...: 一一二→一
PWM ISENSE LOAD LPOWER
OUT IN ISENSE
FB
COMP
VOUT '
SENSE
BAT
,
NOTON ' - - t
DIV
PAUSE
~→一
QOUT
QOUT:
24
,
f ‘
't
1'"
工
--
、 l ，"
-I-l~nT 二i;LLL」
lcsl 工飞。庄r工叩
，'，'、
RIN
;‘, ,
,::
~
、品由、
民飞飞噜
: t '.\' ";.} 主
飞飞、飞
\
4、飞 4
、
，飞、
、
一口
i
眩.
曹
.''1
1
.1
\斐! ，..吨! 2~
/ l' jγ.
l'
吨
~ ~ ，，~
''\飞
4
图 6 UCC3305 典型应用电路巳 ι
..
~
IS0130 高隔离度低价格隔离放大器
-35 一
争，
•
、'，1 、
、
、
‘飞。
‘
J
10kO
In-
In+
1N4 148' 4
工 10~ F
IN4148
'
‘
6.80
2N54Q 1 x 2
:止，已-图 4.150130 用于模拟信号测量
e
近隔离放大器。隔离运算放大器'的输入输出
移， 3 脚应接 GND l.输入 2 脚的电压变化范
端的最小距离应以减小寄生电容为原则进行
围不应超过 100mV 。
布置，这将使放六器保持最佳的隔离性能。
参考文献+
地线不要布置于器件的下面。在选择隔离电
1.((数据采集系统中的放大器如 j闺秀兰
源或稳压器时要注意;也源电压的波动会影
编机械工业出版社
响 IS0130 的位能，要使用稳压性能好的电
源装置。，为宁改善线性和减小非线性温度漂
2. ({世界新型集成电路大拿扎裁τ分册
咨询编号;二 960809
•
吨'哇'哇'中喝'中々哇'中中司，+唱'唱'中唔'中哇'哇'中哇，哇'中非非非哇'中中非非非非非非非非非非非吨'吨'哇'非非非非哇'中++非非非非非非非
(上接 P14)
流器综合了普通技术和功率变换技术。把传
咨询编号 :960803
统的脉宽调节系统加以修改，再加上适当的
编者注:本文由深圳日嘉电子有限公司特约
保护电路，就可控制电灯功率。
刊登，该公司各有详细应用资料和样品，有兴
在使用 HID 1]才要注意以下问题:
趣者可直接与以丁地址联系:深圳市振华路
(1)若外壳破损、大多数放电灯都会产生
苏发大厦 306 栋 805 室.
强烈的 uv 辐射，对皮肤和眼睛极为有害。
(2) 灯内温度通常在 1300K 以上，应使
用灯罩以免灯泡爆裂，造成伤害。
邮编 :518031
电话 :0755-3230748 ， 3230658
传真 :0755-3237394
联系人:王海滨
..
DN-65
Design Note
Considerations in Powering BiCMOS ICs
by Jack Palczynski
Bipolar linear integrated circuits have been with
us for years in the form of PWM and PFC controllers, supervisory circuits and others. Since
these devices have traditionally been built using
relatively high voltage (35V) Bipolar processes,
powering considerations were typically never a
concern. In addition, many of these ICs contained high current protection zeners to keep
higher voltages from damaging the device. With
a number of new BiCMOS ICs now replacing traditional Bipolars, more consideration needs to be
given to powering these low voltage controllers.
This Design Note will provide more details regarding the device specifications and help simplify the powering and use of these energy saving devices.
The most prolific of PWM ICs are those of the
UC3842 family which are easy to use and to
power. The VCC supply can be as high as 34V
and an on-chip zener can sink up to 30mA of
current. It consumes significantly more power in
comparison to the replacement series of Unitrode UCC3802 BiCMOS PWMs. Requiring only
about 10% of the current used by the UC3842,
these BiCMOS parts are a logical replacement
for previous UC3842 based designs. The tradeoff
is that the maximum voltage is 13.5V. With these
specifications in mind, several power methodologies will be described.
Upon first review of the UCC3802 data sheet,
several seemingly contradictory specifications
could be noted. The UVLO start threshold has a
range of 11.5V to 13.5V, while the protection
zener voltage can vary from 12.0V to 15.0V.
However, the absolute maximum supply voltage
of the IC is specified at 12.0V. This absolute
maximum is defined as the lowest possible zener
voltage when driven from a low impedance (voltage) source. Note, however, that the zener volt-
6/95
age is always higher than the UVLO start voltage. These two parameters track each other and
the chip is tested to guarantee that the zener
voltage will never be below that of the start voltage.
For low cost, off-line applications, these newly introduced control ICs offer savings in overall cost
and power in comparison to their predecessors.
For example, the UCC3802 PWM consumes only
1mA (approximate) for full operation − which can
eliminate the need for a bias supply in many instances. To this 1mA, add estimates for the other
currents consumed by the control circuitry (gate
drive, slope compensation, etc.). Working backwards, calculate the resistor value for the circuit
of Figure 1 to deliver this current from the rectified, filtered line voltage. Typically, the AC input
voltage will range from 85VRMS to 264VRMS, or
about a 3:1 ratio. The corresponding DC input
will vary accordingly from 124VDC to 374VDC.
Calculate R1 to provide supply current at the
lowest input voltage. 3mA minimum supply current is used for this example.
124VDC − 12VDC
= 37.3kΩ (use 36k).
3mA
At high line, this current will increase to
374VDC − 12VDC
= 10.1mA
36k
and dissipate more power. The resistor will guarantee that the ICs zener clamp will not be subjected to overcurrent since it shunts only 7.1mA,
the other 3mA are consumed for operation. This
configuration is generally referred to as a "current source" power supply. Before the IC starts, it
only draws 100µA and C1 is charged with nearly
the full supply current. Once C1 reaches the
UVLO start threshold of about 13.5V, the
Design Note
DN-65
UDG-95123
UDG-95124
Figure 1. Off-line Current Source Power
Figure 2. Adding a Bootstrap Supply
UC3802 starts and then uses 1mA for itself, and
an additional 2mA (this example) to power the
gate drive and other functions.
ply has started. Resistor R2 is placed in series
with the bias supply coming from the transformer
in order to again limit the zener current to 10mA.
This circuit will result in a higher efficiency than
that of the first - at the cost of additional components and a bias winding on the transformer.
During start up, C1 again charges through R1
until the turn on threshold of the IC is reached. If
R1 is very large, then the UVLO hysteresis of the
UCC3802 allows the controller to continue running until it reaches its lower threshold. During
this time, the bias supply starts supplying current
and should take over as the primary IC power
supply before the lower UVLO threshold is
reached.
A drawback to the above circuit is that a good
percentage of power is dissipated in R1. If the input voltage range is very wide, or if a high frequency is used or a FET with high gate charge is
used, it may not be possible to guarantee that
the zener current is limited to 30mA at high line
while still being able to provide enough current to
run at low line. To create a more efficient input
supply, a few alternatives are demonstrated. Figure 2 shows a typical bias supply with modifications for the BiCMOS IC. R1 limits current while
the IC is in standby mode so that zener current is
not exceeded as in the last example. In this case,
the resistor may be made much larger since it
will not be the major power source once the sup-
A third solution is to use the Unitrode UCC3889
bias supply control circuitry as seen in Figure 3.
This patented control technology will provide a
UDG-95125
Figure 3. UCC3889 Powers Other Circuits
2
Design Note
DN-65
fully regulated supply from a high input voltage
without the use of transformers. As with the last
circuit, a series resistor from an 18V input to Vcc
will limit current. With the UCC3889 IC, an input
Power Factor Correction IC or other primary side
PWM circuit may also be easily powered.
in designing new, more efficient switching power
supplies.
References:
[1] W. Andreycak, “UCC3800/1/2/3/4/5 BiCMOS
Current Mode Control ICs” Unitrode Applications Note U-132.
The UCC3802 and the entire family of Unitrode
BiCMOS integrated circuits can provide added
efficiency, higher speed, FET totempole output
drivers (which eliminate the need for protection
Schottky diodes) and other added features. With
some care, input power can easily be designed
to meet the ICs power requirements, and assist
[2] J. Palczynski, “UCC3806 BiCMOS Current
Mode Control IC” Unitrode Applications Note
U-144
[3] W. Andreycak, “Elegantly Simple Off-Line Bias
Supply for Very Low Power Applications”
Unitrode Applications Note U-149
UNITRODE CORPORATION
7 CONTINENTAL BLVD. • MERRIMACK, NH 03054
TEL. (603) 424-2410 • FAX (603) 424-3460
3
IMPORTANT NOTICE
Texas Instruments and its subsidiaries (TI) reserve the right to make changes to their products or to discontinue
any product or service without notice, and advise customers to obtain the latest version of relevant information
to verify, before placing orders, that information being relied on is current and complete. All products are sold
subject to the terms and conditions of sale supplied at the time of order acknowledgement, including those
pertaining to warranty, patent infringement, and limitation of liability.
TI warrants performance of its semiconductor products to the specifications applicable at the time of sale in
accordance with TI’s standard warranty. Testing and other quality control techniques are utilized to the extent
TI deems necessary to support this warranty. Specific testing of all parameters of each device is not necessarily
performed, except those mandated by government requirements.
CERTAIN APPLICATIONS USING SEMICONDUCTOR PRODUCTS MAY INVOLVE POTENTIAL RISKS OF
DEATH, PERSONAL INJURY, OR SEVERE PROPERTY OR ENVIRONMENTAL DAMAGE (“CRITICAL
APPLICATIONS”). TI SEMICONDUCTOR PRODUCTS ARE NOT DESIGNED, AUTHORIZED, OR
WARRANTED TO BE SUITABLE FOR USE IN LIFE-SUPPORT DEVICES OR SYSTEMS OR OTHER
CRITICAL APPLICATIONS. INCLUSION OF TI PRODUCTS IN SUCH APPLICATIONS IS UNDERSTOOD TO
BE FULLY AT THE CUSTOMER’S RISK.
In order to minimize risks associated with the customer’s applications, adequate design and operating
safeguards must be provided by the customer to minimize inherent or procedural hazards.
TI assumes no liability for applications assistance or customer product design. TI does not warrant or represent
that any license, either express or implied, is granted under any patent right, copyright, mask work right, or other
intellectual property right of TI covering or relating to any combination, machine, or process in which such
semiconductor products or services might be or are used. TI’s publication of information regarding any third
party’s products or services does not constitute TI’s approval, warranty or endorsement thereof.
Copyright  1999, Texas Instruments Incorporated
APPLICATION NOTE U-161
Ron Fiorello
Unitrode Corporation
POWERING A 35W DC METAL HALIDE
HIGH INTENSITY DISCHARGE (HID)
LAMP USING THE UCC3305 HID
LAMP CONTROLLER
APPLICATION NOTE U-161
UNITRODE CORPORATION
POWERING A 35W DC METAL HALIDE HIGH INTENSITY DISCHARGE (HID) LAMP
USING THE UCC3305 HID LAMP CONTROLLER
by Ron Fiorello
Unitrode Corporation
ABSTRACT
High Intensity Discharge (HID) metal halide lamps are being used in more and more applications where
lamp color, long life and efficiency are important. From automotive and industrial lighting to theatrical and
stage lighting, HID promises to be the light of the future. HID lamps offer many advantages over many
other types of discharge lamps because of their luminous efficiency (their ability to convert electrical power
to visible light) and the color of the light output is closer to an ideal source (the sun) then other types of
discharge lamps i.e.; low pressure sodium, high pressure sodium etc.
The purpose of this application note is to demonstrate the use of the UCC3305 HID lamp controller IC.
Information is presented on a design example to help the user better understand all of the controllers many
features.
INTRODUCTION
The following section specifies typical design
requirements necessary of an HID ballast which
would be powering a DC headlamp in an automotive
application. The headlamp used in this application is
a 35W DC metal halide lamp manufactured by
OSRAM/SYLVANIA.
Input Voltage Requirements - 9 to 16VDC
Startup Requirements - Must run/startup down to
6VDC
Protection/Fault Monitor- Protection against input
overvoltage, output open circuit and output short
circuit.
Power Regulation - Regulate power to the lamp
within +10% over a lamp voltage variation of 60 to
100VDC.
Lamp Ignition Voltage - Provide an open circuit
voltage of greater then 500VDC at start-up in order
to ignite the lamp.
Efficiency - greater than 85%.
Cold Start - The light output on initial start-up must
be within a window as specified by SAE J2009.
Hot Restrike - The ballast must be able to properly light the lamp when hot without a cool down period.
The load presented to the ballast by the lamp is
non-linear. Before ignition occurs, the lamp draws
very little current from the ballast. The ballast sees
essentially an open circuit on its output at start-up.
The open circuit voltage feeds an ignitor circuit
(internal to the lamp) which steps-up the voltage in
order to provide the approximately 20kV ignition
voltage necessary for the lamp. Upon ignition,
metals and gases inside the lamp are ionized causing the lamp voltage to collapse. During ionization,
the lamp will require significant current from the
ballast to properly establish and maintain the arc
discharge. During this time, the current into the
lamp must also be controlled to protect the lamp
electrodes [1].
The initial start-up power into the lamp is higher
then its steady state value. This is necessary in
order to get the light output up to 75% of its steadystate value within 2 seconds, which is a requirement for an automotive application as specified
under SAE J2009. The lamp voltage right after the
glow to arc transition varies from lamp to lamp but
is usually between 20 and 40VDC. As the lamp
warms up and the internal pressure inside the arc
tube increases, the voltage begins to rise and will
gradually reach a steady-state value of between 60
and 110VDC after 150 seconds. This depends on
the age of the lamp. A typical steady-state voltage
of this type of lamp is between 75 and 90VDC.
U-161
APPLICATION NOTE
UDG-96239
Figure 1. Power Regulation Loop
Optimal converter topology
Below is a summation of the different functional
blocks of the UCC3305 and their major electrical
characteristics;
The optimal converter topology for this application
would meet the following requirements;
VCC/OV Protection/VREF/VBOOST Block
1) Output voltage that is capable of being
higher then input voltage.
2) Low input current ripple for reduced input
filter requirements
3) High efficiency
4) Minimal number of magnetic components
5) Minimal number of power semiconductors
There are a few candidate topologies which meet
some of the above requirements. The best choice
for this particular application is the SEPIC converter
which meets all of the above requirements for a
35W lamp. The schematic of this circuit is shown in
Figure 2 [2].
VCC Maximum Voltage - 8 Volts
Must bypass with 0.1µF to 1.0µF Ceramic
Monolithic Capacitor as close to the IC as possible
OV Threshold - Internal Comparator with
reference voltage tied to internal 5V. OV threshold
adjustable with external resistor divider
VREF:
5.0V Trimmed Bandgap Reference
Must bypass with 0.1µF Low ESR Capacitor as
close to the IC as possible
VBOOST Max Voltage - 12 Volts
Supplies drive for output drive stage
Must bypass with 0.1µF to 1µF Ceramic
Monolithic Capacitor as close to IC as possible
The UCC3305 HID controller
The features of the UCC3305 HID controller are
outlined below:
Output Drive Stage:
•
•
•
•
•
•
•
•
•
OV input protection
Output fault protection/timing
Power regulation vs. lamp voltage
Lamp start-up/cool down simulation
Current-mode control
Fixed frequency operation
DC or AC lamp drive capability
High current drive capability
On board charge pump to provide gate drive
down to 6VDC
• Adjustable start-up to steady-state current
ratio
PWMOUT:
1.0A Peak current drive capability
Q and Q not outputs
Outputs to drive external bridge via external
MOSFET drivers Output frequency is fs/512
At lamp start, outputs are disabled via RC from
NOT-ON and DIV. PAUSE
2
U-161
APPLICATION NOTE
Oscillator:
OSC:
Sawtooth Oscillator with Programmable
Frequency DMAX from 0% to 100% possible
With RSET = 150k, Fs ~ 22xe-6/COSC
Maximum operating frequency is 300kHz
I1 =
VREF − KV • VO
R2
I2 =
KV • VO + KI • IO
R1
where KV and KI are the proportionality constants
for voltage and current respectively; since
Load Power and Main Error Amplifiers:
LOADSENSE, LPOWER, COMP AND FB
I1 = I2
The LOADSENSE amplifier, the main error amplifier
and its external associated resistors and capacitors
will determine where the peak of the power curve
occurs as well as the shape of the frequency
response of the ballast. Below is an analysis of this
operational block based on the 35W DC lamp in an
attempt to show how the power curve of the ballast
is determined for this particular application.
VREF − KV • VO
=
R2
KV • VO + KI • IO
R1
rearranging the above equation and solving for IO,
KI • IO = (VREF − KV • VO) •
IO = [(VREF − KV • VO) •
From the simplified schematic of this loop shown in
Figure 2 below, the power curve equation is determined as follows;
R1
− KV • VO
R2
R1
1
− KV • Vo] •
R2
KI
since
PO = VO • IO
Power curve equation
substituting the expression found for IO into the
power equation,
From the simplified schematic of the power regulation
loop, shown in Figure1, the currents I1 and I2 can be
found as follows;
UDG-96240
Figure 2. 35W DC HID Ballast Schematic
3
U-161
APPLICATION NOTE
UDG-96242
The expression for PO1 is valid for lamp voltages from 60 to 105V. The expression for PO2 is valid for lamp
voltages above 105V (This is due to the limiter block inside the UCC3305). Above approximately 105V, the
lamp will be driven in a constant current mode which results in the straight line for power vs. lamp voltage
as shown. The output power can be regulated with a variation of less than ±10% from a nominal of 34.5W
with a lamp voltage variation from 60 to 110V.
As the lamp ages, the voltage will normally increase due to the lamp electrode material erosion over time.
Therefore it is beneficial to limit the current available to the lamp above a certain voltage.
Figure 3. Calculated Power Curve vs. Lamp Voltage of UCC3305 Controlled 35W Ballast Powering DC Metal Halide
Osram/Sylvania Lamp
PO =
REQ = 5.078 • K
VO
R1
R1
• [VREF •
− KV • VO • (1 +
)]
KI
R2
R2
KV ≈ 0.0032
where
REQ =
RA • RB
RA + RB
KI = 0.75
KI is equal to the current sense resistor value
from block diagram of UCC3305, where 1/120 is
the voltage divider attenuation ratio. REQ is the
parallel combination of the 100k and the 5.35k
internal resistors.
KV ≈
Substituting the values found for the constants KV
and KI and the actual resistor values used in the
circuit into the power equation, the power curve can
be plotted for a range of lamp voltages as shown in
the figure below.
REQ
1
•
120 REQ+ 7.85k
KV ≈ 0.0032
RA = 100k
R1 = 4.7k
RB = 5.35k
4
U-161
APPLICATION NOTE
KI = 0.75
By the addition of an external resistor from the ADJ.
pin to ground, this ratio can be programmed. At the
instant of start-up, the output of the limiter is at
zero volts since the WARMUPC capacitor has
not charged. Because of this, the inverting input
to amplifier A1 is at ground with 20µA • RADJ
volts on its non-inverting input. As an example, if
RADJ = 150k, then the voltage at the non-inverting
input is 3V.
R2 = 16k
VO1 = 60, 65..110
VREF = 2.5
VO2 = 110, 115..120
PVo(1) =
VO1
R1
•[(
) • VREF − KV • VO1
KI
R2
•(
PVo(2) =
V
VO(−) = −( (−) ) • 10k
83k
V
VO(−) = −( (−) ) • 10k + V(+)
83k
where
R1
+ 1)],
R2
VO2
R1
•[(
KI
R2
VO(+) = the contribution of the non-inverting
input to VO of A1
VO(−) = the contribution of the inverting input
to VO of A1
V(+) = the voltage at the non-inverting input
of A1
) • VREF − 0.322
• ( R1 + 1)]
R2
Current sense comparators/amplifiers
VO = V(+) • (
The INPUT ISENSE comparitor/amplifier inside the
UCC3305 provides cycle by cycle current control
as in a typical peak current-mode controller. An
added feature allows the user to program the startup to steady-state current ratio of this current. This
allows the ballast to provide increased power to
the lamp at start-up in order to get the lamp light
output up to its steady-state level as quick as possible. The simplified schematic of this section is
shown in Figure 4 below.
10k
10k
+ 1) − V(−) • (
)
83k
83k
on start-up,
V(−) = 0
so;
VO = 3.36 Volts
ISENSEIN
-
-
A1
ADJ
R
WARM-UP C
UDG-96241
Figure 4. Current Sense / Limit Block
5
U-161
APPLICATION NOTE
The current which flows thru the feedback resistor
of the load sense amplifier is;
The current sense threshold is then;
VS =
3.36
10
ILS =
VS = 0.336 Volts
ILS = 430µΑ
This translates to a peak switch current at start-up of;
IP =
Assuming that the current which flows thru the
feedback resistor also flows thru the inverting input
resistor, the voltage across the output current
sense resistor is;
3.36
0.02
IP = 16.8 Amps
VCS = 0.30 − (430µA • 5.1k)
This current threshold will gradually decrease as
the WARMUPC capacitor charges up to 10V.
VCS = 1.89V
The limiter limits the inverting input of A1 at steadystate to 5V. The current limit at this point is then;
(the negative sign of the voltage is ignored since
this is defined as positive current)
VO = 3.36 − 0.6
IL =
VO = 2.76 Volts
IP =
1.89
0.83
IL = 2.3 Amps
0.276
0.02
This defines the maximum startup current which
flows into the lamp at ignition. The current into the
lamp will decay exponentially due to the voltage
charging characteristic of the WARMUPC and
SLOPEC capacitors. The current will decay to a
steady-state value of approximately 450mA after a
period of time given by the time constant of the
internal 50 Meg resistor and the capacitor placed
from the SLOPEC pin to ground. In this example,
the time to steady-state is 150 seconds from:
IP = 13.8 Amps
The switch current in the SEPIC converter is a
combination of the input inductor current plus
the reflected load current. At steady-state, 9 VIN,
the peak-to-peak current thru the output rectifier
is approximately 1A. This reflects back to the
primary as 6A into the switch plus 3.1 Amps from
the inductor. The total current thru Q1 is then
9.1 Amps.
t = 50 exp(6) • CSLOPEC
Output current limit on start-up
The capacitors used for the SLOPEC and
WARMUPC functions must have low leakage characteristics since they are charged from nanoamp
current sources internal to the IC. Any significant
amount of leakage current caused by these components will have an effect on the output power regulation characteristic of the ballast.
From Figure 2 the start-up current limit into the
lamp can be determined. On start-up, the
WARMUPV pin, which is a buffered version of
WARMUPC, is at ground.Therefore, the two 27k
resistors are in parallel resulting in an equivalent
resistance of 13.5k. The current that flows from FB
to ground is then;
I=
5.46 − 0.30
12k
Slope compensation resistor
2.5
13.5k
Slope compensation in the UCC3305 is provided by
the addition of an external resistor in series with the
INPUT ISENSE pin. This resistor adds a portion of
the oscillator ramp into the current sense signal to
provide the necessary slope compensation for duty
cycles exceeding 50%. The amount of slope
compensation that is needed is dependent on the
topology used as well as the inductor values chosen.
In the SEPIC converter, both input and output inductors need to be considered when determining how
much slope compensation is necessary.
I1 = 185µΑ
The voltage at the output of the load sense amplifier is then;
VLS = 185µΑ • 16k + 2.5V
VLS = 5.46V
6
U-161
APPLICATION NOTE
The current sense comparator compares the current
sense signal to the output of the error amplifier to
determine the duty cycle of the power switch [3].
VI, the voltage at the current sense resistor can be
determined as follows
From the simplified schematic of the current sense
circuit;
2
10 • exp(−6)
where S is defined as the slope of the oscillator ramp.
Substituting the following values into the equation for
RP will allow us to determine the value of the slope
compensation resistor.
S=
N
VI = RI • ( S ) • ( IOAV + M2 • t)
NP
+ RI (IIN +
M1
• tOFF)
2
M2 = 178571
NS = 60
M1 = 178571
NP = 10
RI = 0.01
S = 181818
M1
NS
• RI + N • RI • M2
2
P
RP =
2
where;
NS = number of turns of secondary winding
of L2
NP = number of turns of primary windings
of L2
RI = current sense resistor
LOAV = secondary average output current
M2 = down slope of secondary current thru L2
M1 = down slope of primary current thru L1
IIN = average input current
tOFF = off time of switch
RP = 0.064
R4 = 50k
R3 is the slope compensation resistor and R4 is the
internal 50kΩ resistor;
The above equation can be rewritten as follows;
Ns
VI = RI • ( N ) • (LOAV + M2 • (T − tON))
RP =
P
Solving for R3;
M1
• (T − tON)]
2
The first term is the contribution of the output
current thru L2 (output inductor) to the current
sense resistor. The second term is the contribution
of the input current thru L1 (input inductor) to the
current sense resistor
+ RI [IIN +
R3 = 0.064 •
R4
1 − 0.064
R3 = 3.419k
Therefore, the slope compensation resistor chosen
must be greater than 3.42k in order for stable converter operation at duty cycles which exceed 50%.
This signal is set equal to the voltage at the output
of the error amplifier. This results in the following
equation after rearranging terms;
Frequency response of the power regulation
loop
N
RI ( S ) • LOAV = VEA + tON
NP
The frequency response of the ballast is determined by analysis of the power regulation loop.
Since it is really the output current that is being regulated and not the output voltage (any change in
output voltage is attenuated by 1/120), the analysis
can be simplified by modeling the power stage as a
voltage controlled current source with some
transconductance gain GM. This assumption is
valid for a loop cross over frequency below resonance of the power stage which is approximately
10kHz.
M
N
• ( 1 • RI + RI s • M2 − m)
2
NP
− IIN −
R3
R3 + R4
M1
• T • RI − M2 • T
2
In order to eliminate the possibility of subharmonic
oscillations, the term which multiplies tON should
be set equal to zero eliminating any dependency on
duty cycle.
M1
NS
2 • RI + RI • NP • M2 = M
7
U-161
APPLICATION NOTE
The transconductance gain, GM, can be found as
follows;
GM =
phase margin with these component values is
greater than 20dB and 60 degrees, respectively.
∆IO
∆VE
CIRCUIT EXAMPLE:
A 9 to 16VDC input SEPIC converter powering a
35W OSRAM/SYLVANIA DC lamp was built and
tested. Data on efficiency, power curve and various
oscillagrams of current and voltages in the
power/control circuit were taken and are discussed.
The output current is converted to a voltage by the
output current sense resistor. The gain of the power
stage is then;
GP = RIs
∆IO
∆VE
Magnetics Design
or;
L1
GP = RIs GM
The input inductor L1, is designed based on the
same criteria as a boost inductor. Energy is stored in
L1 during the on time of Q1 and transferred during
the off time. L1 is designed to operate in the continuous mode with low current ripple. At 9 VIN with 36W
of output power and a converter efficiency of 85%,
the average input current is 4.7 Amps. If a total peakto-peak ripple current of 2 Amps is assumed the
inductance of L1 can be found. But before we can
calculate the inductance required, the minimum and
maximum duty cycle must be found so that the
maximum and minimum on time of Q1 can be
determined. The SEPIC converter has a DC transfer
function of;
D
;
VO = VIN • n
1−D
The turns ratio, n can be found from the maximum
acceptable voltage stress on Q1. The stress on
Q1 is the sum of the capacitor voltage plus the
reflected secondary voltage. The capacitor voltage
is essentially equal to VIN, so;
VO
VDS = VIN +
;
n
where; Rls = the load sense resistance (0.75Ω)
∆IO = load current change (500mA)
∆VE = error amp voltage change (5V)
GP = −22.5dB
The loop response must now be tailored for good
power regulation (high DC gain) and adequate
phase and gain margin at the loop crossover
frequency. The gain of the LOAD SENSE amplifier
is restricted due to the fact that gain of this stage
effects the power curve characteristic as shown in
above analysis of the power curve equation.
The LOADSENSE amplifier should be set up as an
integrator so that it can filter out switching frequency
noise from the control loop. The pole frequency was
chosen to be at 1kHz to give good rejection of the
switching frequency noise. This results in a capacitor value of 0.01µF. The low frequency gain of this
amplifier is set to 7.5dB. The combination of this
gain and the power stage gain results in −15dB of
low frequency gain with a pole at 1kHz.
The worst case output voltage on startup of the
lamp is restricted to 500V, since this voltage will be
reflected back to the drain of Q1. The turns ratio
must be chosen so that the drain voltage never
exceeds its maximum rating. A IRF1310 was
chosen for this application in part because
of its 45mΩ on resistance and VDS = 100V.
Calculating the turns ratio at VIN = 16V; n is then
found to be 5.8. A turns ratio of 6 is used.
The response can now be tailored with the main
error amplifier. A zero must be added in the amplifier response at some mid-band frequency so that
the DC gain for the overall loop is as high as possible. The high frequency gain of this amplifier must
be well below 0dB to ensure adequate gain and
phase margin for the open loop gain. Since the
16kΩ, resistor has been determined from the
power curve characteristic desired, only the feedback resistor value can be chosen. If this resistor is
chosen so that the high frequency gain is to be less
then −20dB for good gain margin, or the feedback
resistor value of 1kΩ, the capacitor value can then
be determined. If a zero frequency of 3.4kHz this
assumed, this will give an adequate low frequency
gain boost. From this, the value of the capacitor can
now be determined to be 0.047µF. The gain and
The maximum duty cycle can now be determined
from the DC transfer function. To find the maximum
duty cycle, the worst case steady-state lamp voltage
is used of 110VDC at VIN = 9V. Lamp voltages
between 60 and 110V will be within the power
regulation range of the ballast. Lamp voltages
outside of this range will be operated in the
constant current mode. Therefore;
8
U-161
APPLICATION NOTE
DMAX = 0.67
The minimum duty cycle is determined using the
minimum steady-state lamp voltage of 60VDC and
VIN = 16V.
An RM10PA250-3F3 core was used which has an
AP of 0.379. The number of primary turns is then;
L • IP
20µH • 3A
=
= 7T
AE • B 89 • e-6 • 0.1T
(10T is used since this will easily fit in one layer
with the desired core and wire gauge chosen)
NP =
DMIN = 0.38
For a switching frequency of 100kHz, tON MAX =
6.7µS, tON MIN = 3.8µS
This ferrite core must be gapped since it stores
energy. It is desired that the total gap be placed in
the center leg. The gap is calculated from;
The inductance based on tON MAX at VIN MIN can
be calculated;
9 • 6.7µS
= 30µH.
2
L1 consists of 30T of 19AWG wound on a
Micrometals E100 −18 core.
LP =
L1=
0.89 • 100
= 0.56mm = 0.02 in
0.020mH
L2
The secondary turns can be calculated from the
turns ratio as 60T. The core used for L2 has a
center leg gap of 0.022 in. Multifilar wire is used for
both the primary and secondary turns to minimize
the copper losses. The winding sequence used was;
primary-secondary-primary-secondary-primary.
The voltage across the primary winding of L2 when
Q1 is on, is for all practical purposes, equal to the
input voltage (neglecting voltage ripple on the
capacitor) since the series capacitor is switched
across the primary. The inductance of the primary
winding is chosen based on the peak current
desired (it is desired that the inductor current is
continuous). The peak current chosen is based on
a tradeoff between the voltage stress on Q1 and
minimal number of turns to minimize the leakage
inductance which in turn means reducing the number of layers of windings. If the peak current thru L2
is restricted to 3.0A, the primary inductance can
then calculated as;
Performance data
Performance data on the ballast is presented in the
following curves showing efficiency and the measured power curve. Oscillograms of Q1 voltage and
current are also given as well as startup characteristics of the lamp voltage and current. The maximum efficiency achieved was 86.2% at a lamp voltage of 100V. The efficiency decrease after this
point is due to an increase in output power which
occurs at lamp voltages above 100 to 105V. The
lamp cold start voltage and current waveforms are
shown with a time base of 50mS and 1Sec. As can
be seen, the ballast output voltage is 600V before
lamp ignition. Once the lamp ignites, the voltage
collapses and the lamp current increases to 2A.
Eventually, the lamp voltage begins to increase
and the current decreases. They will arrive to their
steady-state values of 80 to 90VDC and 450mA
respectively after approximately 150 seconds.
9 • 6.7µS
= 20µH.
3
The inductance of L1 and L2 could have been set
equal to each other. This would have made both
inductors “easy” to integrate on the same core. This
was not attempted here because of leakage inductance concerns between the primary and secondary windings of L2.
L2 =
The number of turns for L2 can now be determined
based on the particular core geometry chosen. The
area product (AP) required is found from;
AP =
L • IP.• IRMS
= 0.362 cm2
KF • J • BM
This is based on the following parameters;
B = 0.1T
J = 450A/cm
L = 20µH
K = 0.4
uO • uR • NP2 • AE
= 12.56 • 10-7
L
IP = 4.7A
IRMS = 5.2A
9
U-161
APPLICATION NOTE
UDG-96243
UDG-96244
Figure 5. Efficiency and Power Curve of 35W HID Ballast
10
U-161
APPLICATION NOTE
Figure 6. Ballast Output Voltage and Current
Figure 7. Ballast Output Voltage and Current
11
U-161
APPLICATION NOTE
Figure 8. MOSFET (Q1) Gate and Drain Voltage at Steady State
Figure 9. MOSFET (Q1) Drain Voltage and Current at Steady State
12
U-161
APPLICATION NOTE
Figure 10. Output Rectifier (D1) Current at Steady State
Figure 11. Ballast Hot Restrike Voltage and Current
13
U-161
APPLICATION NOTE
35W HID BALLAST PARTS LIST
REF DES
RI
R2
R3, R5
R4
R6
R7
R9
R11
R12
R13
R14
R15
R16
R17,R18
R19,R25,R32,R8
R20
R21
R22,R23
R24
R26,R27
R30
R31
C33
C1
C2,C3,C26
C4
C8,C11
C6,C7
C9
C10
C12,C13
C14
C15
C16
C17,C18,C19
C5,C24
C25
C30
C31
C32
Z1
Q2,Q3
Q1
D1
HS2,3,4,5
U1
HS1
L1
L2
PART DESCRIPTION
4.7Ω 1/4 W CC
0.02Ω 1W
1k 1/4W CC
3.3k 1/4W CC
270k 1/4W CC
100k 1/4W CC
180Ω 1/2W CC
5.1k 1/4W CC
12.5k 1/4W CC
16.1k 1/4W CC
1k 1/4W CC
150k 1/4W CC
250k 1/4W CC
27k 1/4W CC
10k 1/4W CC
0.75Ω 3W CC
565k 1/4W CC
282k 1/4W CC
560Ω 1/2W CC
100k 1/4W CC
18Ω 3W CC
330Ω 3W CC
DIGIKEY NUMBER
10QBK-ND
1KQBK-ND
4KQBK-ND
270KQBK-ND
100KQBK-ND
220HBK-ND
5.1KQBK-ND
15KQBK-ND
16KQBK-ND
1KQBK-ND
150KQBK-ND
250KQBK-ND
27KQBK-ND
10KQBK-ND
VC3D.75-ND
562KXBK-ND
280KXBK-ND
560HBK-ND
100KQBK-ND
VC3D18-ND
VC3D330-ND
10µF/100V POLY FILM
1µF/50V METALLIZED FILM
470µF/50V ALUM ELEC
0.47µF/630V POLY FILM
0.47µF/50V CERAMIC
4.7µF/250V ALUM ELEC
470pF/25V CERAMIC
10µF/35V ALUM ELEC
1µF METALIZED FILM, NISSEI #R68105K63B
150pF/50V CERAMIC
0.056µF/25V CERAMIC
47µF/25V ALUM ELEC
0.01µF/50V CERAMIC
0.1µF/50V CERAMIC
1000pF/50V CERAMIC
100µF/25V ALUM ELEC
180pF/1kV CERAMIC DISK
1000pF/100V CERAMIC
1N5235B, 6.8V ZENER
2N3904, 40V, 0.200mA TRANISTOR
IRF1310, 100V, 0.027Ω
MUR860, 600V, 8A FST REC
THERMALLOY#7128D, HS FOR Q2,Q3,Q4
UCC3305JP
THERMALLOY #6398-P2,HS FOR Q1
E100-8 MICROMETALS
CORE-30T #18AWG
35µH
RM10PA250-3F3 PHILIPS
10T PRIMARY LITZ(2X10X,1)
60T SECONDARY LITZ(1X15X,1)
WINDING SEQUENCE
(PRIM-10T, SEC-30T, PRIM-10T,SEC-30T, PRIM-10T)
14
EF1106-ND
P4675-ND
P1248-ND
EF4225-ND
P4671-ND
P6187-ND
P4808-ND
P1227-ND
P4804-ND
P1240-ND
P1220-ND
P4513-ND
P4525-ND
P4812-ND
P1221-ND
P4119-ND
P4036-ND
1N5235BCT-ND
NEWARK#IRF1310
NEWARK#MUR860
NEWARK#95F715
QTYPER
1
1
2
1
1
1
1
1
1
1
1
1
1
2
4
1
1
2
1
2
1
1
1
1
3
1
2
2
1
1
2
1
1
1
3
2
1
1
1
1
1
2
1
1
4
1
1
1
U-161
APPLICATION NOTE
CONCLUSION
REFERENCES
The performance data presented of a typical
UCC3305 HID lamp controller application, demonstrated it to be a excellent means of controlling a
35W DC metal halide HID lamp. The power regulation and efficiency achieved using the SEPIC
converter topology proved it to be a good alternative to other conventional circuit topologies for an
automotive lighting application. The many protection and control features of the UCC3305 simplify
the task of the ballast designer considerably,
making it an economically feasable choice for AC
as well as DC HID lamp applications.
[1] Waymouth: ELECTRIC DISCHARGE LAMPS
M.I.T. PRESS, Cambridge Mass.
[2] Lloyd Dixon, “High Power Factor Preregular
Using Sepic Converter”, Unitrode Power Supply
Seminar SEM1100.
[3] Abraham Pressman: SWITCHING POWER
SUPPLY DESIGN, Mc Graw-Hill, Inc.
UNITRODE CORPORATION
7 CONTINENTAL BLVD. • MERRIMACK, NH 03054
TEL. 603-424-2410 • FAX 603-424-3460
15
IMPORTANT NOTICE
Texas Instruments and its subsidiaries (TI) reserve the right to make changes to their products or to discontinue
any product or service without notice, and advise customers to obtain the latest version of relevant information
to verify, before placing orders, that information being relied on is current and complete. All products are sold
subject to the terms and conditions of sale supplied at the time of order acknowledgement, including those
pertaining to warranty, patent infringement, and limitation of liability.
TI warrants performance of its semiconductor products to the specifications applicable at the time of sale in
accordance with TI’s standard warranty. Testing and other quality control techniques are utilized to the extent
TI deems necessary to support this warranty. Specific testing of all parameters of each device is not necessarily
performed, except those mandated by government requirements.
CERTAIN APPLICATIONS USING SEMICONDUCTOR PRODUCTS MAY INVOLVE POTENTIAL RISKS OF
DEATH, PERSONAL INJURY, OR SEVERE PROPERTY OR ENVIRONMENTAL DAMAGE (“CRITICAL
APPLICATIONS”). TI SEMICONDUCTOR PRODUCTS ARE NOT DESIGNED, AUTHORIZED, OR
WARRANTED TO BE SUITABLE FOR USE IN LIFE-SUPPORT DEVICES OR SYSTEMS OR OTHER
CRITICAL APPLICATIONS. INCLUSION OF TI PRODUCTS IN SUCH APPLICATIONS IS UNDERSTOOD TO
BE FULLY AT THE CUSTOMER’S RISK.
In order to minimize risks associated with the customer’s applications, adequate design and operating
safeguards must be provided by the customer to minimize inherent or procedural hazards.
TI assumes no liability for applications assistance or customer product design. TI does not warrant or represent
that any license, either express or implied, is granted under any patent right, copyright, mask work right, or other
intellectual property right of TI covering or relating to any combination, machine, or process in which such
semiconductor products or services might be or are used. TI’s publication of information regarding any third
party’s products or services does not constitute TI’s approval, warranty or endorsement thereof.
Copyright  1999, Texas Instruments Incorporated
DN-72
Design Note
Lamp Ignitor Circuit
by Ron Fiorello
Both Fluorescent and HID lamps are becoming
increasingly more popular due to their luminous
efficiency and quality of the light output. These
types of lamps, although they have many advantages over incandescent lamps, have more demanding starting requirements. This design note
briefly describes a few different types of possible
ignitor circuits and is not meant to be a complete
description of all the possible circuit configurations.
Parallel Ignitor Circuit
The circuit in Figure 2 is a simple parallel ignitor
circuit for a gas discharge lamp. The trigger cir-
Fluorescent Lamp Ignitor Circuit
The ignitor circuit shown in Figure 1 is typically
used to start conventional Fluorescent lamps
powered by an electronic ballast. Capacitor C
UDG-96225
Figure 1.
charges upon application of the input voltage
causing the trigger circuit to close switch S2
(where S1 and S2 are typically the power
switches in a half bridge converter). This allows
the series resonant circuit consisting of L1 and
C1 to provide the high voltage (typically 500V) to
ignite the lamp. Once the lamp has been lit, the
trigger circuit is disabled. The trigger element is
usually some type of semiconductor switch such
as a DIAC.
UDG-96226
Figure 2.
cuit, which is usually part of the lamp, will repetitively trigger in order to generate the necessary
voltage pulses to ignite the lamp. The pulses are
due to the storage of energy in the inductor when
the switch S1 turns on. The obvious advantage of
this circuit is its simplicity since it consists of only
two elements, an inductor and switch. Although a
simple bimetal switch could be used, the disadvantage of the circuit is that the switch would
need to be located in or near the lamp.
Series Ignitor Circuit for HID Lamp
The circuit in Figure 3 is typically used to ignite
gas discharge lamps. The trigger circuit drives
UDG-96227
Figure 3.
1/97
Design Note
DN-72
the primary of a pulse transformer inducing a
high voltage on the lamp electrodes. The main
advantage of this series ignitor circuit is that the
ballast is not exposed to the high voltage transients generated from the ignitor. The main disadvantage, which is true for most ignitor circuits,
is that it should be located as close as possible
to the lamp in order to minimize the parasitic effects of wiring. This effect could have significant
impact on the rise time of the voltage pulse delivered to the lamp, increasing the chances of a no
light condition. Methods for reducing the interwinding capacitance of the transformer must be
used in order to reduce this parasitic element.
The necessary voltages required to ignite HID
lamps vary from lamp to lamp and manufacturer
to manufacturer, but are usually between 5kV
and 25kV for short arc lamp and 20kV to 50kV
for long arc lamps.
the tips of the electrodes of the lamp. Once the
arc discharge has occurred, the ignitor circuit is
rendered inactive because the lamp impedance
will drop drastically and the capacitor voltage will
never reach the threshold level of the switch.
The circuit in Figure 4 shows a HID ballast output stage driving a full bridge power stage for an
AC lamp. Upon application of power to the output
The turns ratio from primary to secondary of T1
can be found based on the ignition voltage of the
lamp. If the pulse transformer were ideal then all
of the energy stored in the capacitor would be
transferred to the lamp electrodes. We know,
however, that this is not the case. Depending on
the winding method used for the transformer,
there can be significant energy lost to the interwinding capacitance. Because of this, it is necessary to take precautions to minimize this
element. Usually a turns ratio 25% to 50% higher
then that calculated is required to generate the
necessary voltage on the secondary since it will
not be possible to eliminate this capacitance entirely.
Assume, for instance, that the threshold voltage
of S1 is 400V. C1 will be allowed to charge up to
400V before an ignition pulse occurs. This assumes that the open circuit output voltage of the
ballast is limited to 600V before ignition of the
lamp. The energy stored in the capacitor is
(1/2)CV2. Assuming a lossless switch for S1, all
energy stored in C1 is then dumped into the
transformer primary winding inductance. The primary inductance required to support this energy
can be determined using the above assumption
from:
(1⁄2)
UCG-96228
Figure 4.
• C1V2 = (1⁄2) • LPI2
Care must be taken in choosing the secondary
ignitor inductance since it is in series with the
lamp and is being excited with an AC voltage.
Too large an inductance will reduce the excitation
voltage to the lamp. Luckily the excitation frequency of this voltage is typically between 200Hz
and 1000Hz so the AC impedance can be kept to
a minimum with a fairly large inductance value.
The secondary inductance also provides some
beneficial filtering of the current seen by the
lamp. This filtering helps reduce the chance of
acoustic resonances being excited in the arc
tube by the switching frequency current ripple of
stage of the ballast, before the lamp is lit, Q1 and
Q4 are turned on and held on until such time that
lamp ignition has occurred. It is sometimes necessary to hold Q1 and Q4 on even after ignition
occurs until the arc discharge is fully established.
During this time, C1 is allowed to charge up to
predetermined level set by the threshold voltage
of the switch S1. S1 must be capable of switching significant current in a short period of time
(typically hundreds of amps in a fraction of a microsecond). This high current capability is necessary to get the arc discharge to form properly on
2
Design Note
DN-72
References
the ballast. These resonances can cause problems with the lamp optics as well as lead to destruction of the lamp if left unchecked. Because
of this, AC HID lamps are typically driven with a
low frequency square wave current.
UNITRODE CORPORATION
7 CONTINENTAL BLVD. • MERRIMACK, NH 03054
TEL. (603) 424-2410 • FAX (603) 424-3460
3
[1]
Waymouth "Electrical Discharge Lamps"
MIT Press
[2]
Murdoch "Illumination Engineering from
Edison’s Lamp to the Laser" Visions
Communication
application
INFO
available
UCC2305
UCC3305
HID Lamp Controller
FEATURES
DESCRIPTION
•
Regulates Lamp Power
•
Compensates For Lamp
Temperature
•
Fixed Frequency Operation
•
Current Mode Control
The UCC3305 integrates all of the functions required to control and drive one
HID lamp. The UCC3305 is tailored to the demanding, fast turn-on requirements
of automobile headlamps, but is also applicable to all other lighting applications
where HID lamps are selected. HID lamps are ideal for any lighting applications
that can benefit from very high efficiency, blue-white light color, small physical
lamp size, and very long life.
•
Overcurrent Protected
•
Overvoltage Shutdown
•
Open and Short Protected
•
High Current FET Drive Output
•
Operates Over Wide Battery
Voltage Range: 5V to 18V
The UCC3305 contains a complete current mode pulse width modulator, a lamp
power regulator, lamp temperature compensation, and total fault protection.
Lamp temperature compensation is critical for automobile headlamps, because
without compensation, light output varies dramatically from a cold lamp to one
that is fully warmed up.
The UCC2305 is tested for full performance with ambient temperature
from –40°C to +105°C while the UCC3305 is tested with ambient temperature
from 0°C to +70°C. The UCC3305 is available in a 28 pin small-outline, surface
mount plastic package (SOIC).
BLOCK DIAGRAM
UDG-94091-1
SLUS297B - SEPTEMBER 1995 - REVISED APRIL 2004
UCC2305
UCC3305
CONNECTION DIAGRAM
ABSOLUTE MAXIMUM RATINGS
VCC Supply Voltage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.0V
BOOST Supply Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . 12.0V
PWMOUT Current, Peak . . . . . . . . . . . . . . . . . . . . . . . . . . . . ±1.0A
PWMOUT Energy, Capacitive Load . . . . . . . . . . . . . . . . . 5.0µJ
Input Voltage, Any Input. . . . . . . . . . . . . . . . . . –0.3V to +10.0V
Output Current, QOUT, QOUT, FLT . . . . . . . . . . . . . . ±10.0mA
Output Current, 5VREF, LPOWER, COMP . . . . . . . . . ±10.0mA
ISET Current . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . –1.0mA
Storage Temperature . . . . . . . . . . . . . . . . . . . −65°C to +150°C
Junction Temperature . . . . . . . . . . . . . . . . . . . −55°C to +150°C
Lead Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +300°C
PDIP-28 or SOIC-28 (Top View)
N or DW Package
ELECTRICAL CHARACTERISTICS Unless otherwise stated, VCC = 6.6V, ISET = 100kΩ to GND, ADJ = 100kΩ to GND,
OSC = 200pF to GND, BAT = 4V, LOADISENSE connected to LPOWER, VOUTSENSE = 0.666V, BOOST = 10.5V, COMP
connected to FB through a 100kΩ resistor, –40°C<TA<+105°C for the UCC2305, 0°C<TA<+70°C for the UCC3305, and TA=TJ.
PARAMETER
TEST CONDITIONS
MIN.
TYP.
MAX. UNITS
Overall Section
VCC Supply Current
0.1
BOOST Supply Current
BOOST Threshold to PUMP Stop
9.1
1.0
mA
3.0
5.0
mA
9.6
10.2
V
BOOST Threshold to PUMP Start
9.2
9.7
10.3
V
BOOST Threshold to PWMOUT
4.7
5.4
6.1
V
BAT Threshold to PWMOUT Stop
4.7
5.0
5.3
V
BAT Threshold to PWMOUT Start
4.15
4.8
5.0
V
1
µA
Battery Section
BAT Input Current
BAT = 4V
–1
Oscillator & Divider Section
OSC Frequency
80
100
120
kHz
−70
−50
−40
µA
1.1
1.5
1.9
V
0.8
1.2
1.6
V
–8
–5
–1
µA
5VREF Voltage
4.85
5.0
5.1
V
ISET Voltage
4.8
4.8
5.2
V
FB Voltage
2.4
2.5
2.6
V
FB Input Current
–1
0
1
µA
OSC Pull-Up Current
OSC = 1.5V
DIVPAUSE Threshold to Pause
DIVPAUSE Threshold to Divide
DIVPAUSE Input Current
0V < DIVPAUSE < 6V
Reference Section
Error Amplifier Section
FB Sink Current
VOUTSENSE = 4V, FB = 4V
0.3
1.5
FB Release Delay
VOUTSENSE Step from 4V to 1V
15
30
43
COMP Source Current
FB = 2V, COMP = 4V
–3.0
–0.2
COMP Sink Current
FB = 3V, COMP = 1V
2
0.2
1.0
mA
ms
mA
mA
UCC2305
UCC3305
ELECTRICAL CHARACTERISTICS (cont.) Unless otherwise stated, VCC = 6.6V, ISET = 100kΩ to GND, ADJ = 100kΩ
to GND, OSC = 200pF to GND, BAT = 4V, LOADISENSE connected to LPOWER, VOUTSENSE = 0.666V, BOOST = 10.5V,
COMP connected to FB through a 100kΩ resistor, –40°C<TA<+105°C for the UCC2305, 0°C<TA<+70°C for the UCC3305, and
TA=TJ.
PARAMETER
TEST CONDITIONS
MIN.
TYP.
MAX. UNITS
–2.5
–0.1
2.5
µA
–8.0
–0.4
mA
Load Power Amplifier Section
LOADISENSE Input Current
LPOWER Source Current
LPOWER = 0V
LPOWER Sink Current
LPOWER = 1V
0.4
1.3
LPOWER Voltage
VOUTSENSE = 0.0V
0.32
0.40
0.48
V
VOUTSENSE = 0.45V
0.32
0.40
0.48
V
VOUTSENSE = 0.65V
0.41
0.46
0.51
V
VOUTSENSE = 0.88V
0.43
0.51
0.59
V
VOUTSENSE = 2.0V
0.43
0.51
0.59
V
VOUTSENSE = 0.7V, SLOPEC = 0V
0.29
0.34
0.41
V
COMP = 5V, WARMUPC = 0V
0.16
0.21
0.28
V
COMP = 5V, WARMUPC = 10V
0.10
0.19
0.27
V
COMP = 1V, WARMUPC = 0V
0.07
0.10
0.2
V
OSC = 0V
–15
–5
–2
µA
OSC = 2V
–80
–40
–15
µA
VOUTSENSE Threshold to PWMOUT
4.2
5.0
5.2
V
VOUTSENSE Threshold to FB
1.7
1.9
2.1
V
0.035
0.083
0.140
V
1
µA
mA
Input Current Sense Section
ISENSEIN Threshold
ISENSEIN Bias Current
VOUTSENSE Section
VOUTSENSE Threshold to NOTON
VOUTSENSE Input Current
–1
OUTPUTS SECTION
PWMOUT High Voltage
IPWMOUT = –100mA
PWMOUT Low Voltage
IPWMOUT = 100mA
PUMPOUT High Voltage
IPUMPOUT = –10mA
PUMPOUT Low Voltage
IPUMPOUT = 10mA
1.0
1.8
V
PUMPOUT Frequency
BOOST = 9.5V
35
50
60
kHz
5.0
6.3
NOTON High Voltage
INOTON = –1mA
INOTON = 1mA
QOUT, QOUT High Voltage
IQOUT = –1mA or IQOUT = –1mA
QOUT, QOUT Low Voltage
IQOUT = 1mA or IQOUT = 1mA
FLT Low Voltage
IFLT = 1mA
5.3
5.8
0.1
QOUT, QOUT Frequency
IFLT = –1mA
10.0
0.3
NOTON Low Voltage
FLT High Voltage
9.15
V
0.5
V
V
V
0.3
V
0.1
0.45
V
150
200
250
Hz
6.0
6.3
0.1
0.3
V
5.0
6.3
V
V
Timing Capacitor Section
FLTC Discharge Current
FLTC = 2.5V
35
60
100
nA
FLTC Charge Current
FLTC = 2.5V
–430
–300
–220
nA
4.65
4.9
5.1
V
SLOPEC = 0.5V
–165
–90
–60
nA
SLOPEC = 2.2
–105
–60
–40
nA
SLOPEC = 4.2
–50
–30
–10
nA
ISLOPEC = –125nA
1.3
1.5
1.7
V
ISLOPEC = –50nA
2.8
3.0
3.2
V
FLTC Threshold to FAULT
SLOPEC Charge Current
SLOPEC Voltage
3
UCC2305
UCC3305
ELECTRICAL CHARACTERISTICS (cont.) Unless otherwise stated, VCC = 6.6V, ISET = 100kΩ to GND, ADJ = 100kΩ
to GND, OSC = 200pF to GND, BAT = 4V, LOADISENSE connected to LPOWER, VOUTSENSE = 0.666V, BOOST = 10.5V,
COMP connected to FB through a 100kΩ resistor, –40°C<TA<+105°C for the UCC2305, 0°C<TA<+70°C for the UCC3305, and
TA=TJ.
PARAMETER
TEST CONDITIONS
MIN.
TYP.
MAX. UNITS
40
100
200
nA
Timing Capacitor Section (cont.)
SLOPEC Discharge Current
SLOPEC = 2.2V, VCC = 0V, BOOST = 0V,
BYPASS = 8V
WARMUPC Charge Current
WARMUPC = 0V
–525
–375
–275
nA
WARMUPC = 2V
–525
–375
–300
nA
WARMUPC = 6V
–200
–120
–75
nA
WARMUPC Voltage, Charging
IWARMUPC = –250nA
3.39
3.8
4.1
V
WARMUPC Discharge Current
WARMUPC = 5V, VCC = 0V, BOOST = 0V,
BYPASS = 8V
23
50
126
nA
WARMUPC = 1V, VCC = 0V, BOOST = 0V,
BYPASS = 8V
5
10
34
nA
WARMUPC Voltage, Discharging
IWARMUPC = 25nA, VCC = 0V, BOOST = 0V,
BYPASS = 8V
1.5
1.9
2.3
V
ADJ Bias Current
VADJ = 0V
−38
−20
−12
µA
WARMUPV Voltage
WARMUPC = 1V
0.05
0.125
0.25
V
WARMUPC = 2V
0.09
1.00
1.5
V
WARMUPC = 3V
2.3
2.48
2.66
V
WARMUPC = 5V
4.5
4.8
5.25
V
5.25
WARMUPC = 10V
4.5
4.8
BYPASS Voltage
VCC = 0V
8.8
9.6
BYPASS Current
VCC = 0V, BOOST = 0V, BYPASS = 8V
2.5
V
V
7
µA
PIN DESCRIPTIONS
When the HID lamp is turned off, power to the lamp and
the controller is removed, leaving these two critical capacitors charged to specific voltages. Also, with power
off, the lamp will cool down at a controlled rate. It is essential that the two capacitors discharge at a similarly
controlled rate so that if the lamp is restarted before the
lamp is fully cooled, the controller will have an estimate
of new lamp temperature, and can again command the
correct power for the lamp.
5VREF: Circuitry in the UCC3305 uses the internal 5V
reference to set currents and thresholds. This reference
can also be used for other functions if required.
ADJ: The ratio of cold lamp peak current to warmed-up
lamp peak current is controlled by the voltage on ADJ. To
select this voltage, connect a resistor from ADJ to GND.
BAT: This input is used to detect excessively high input
voltage and shut down the IC if the input exceeds a predetermined level. Connect BAT to a voltage divider
across the input supply. The UCC3305 shuts down when
this input voltage exceeds 5V. To protect the IC in the
event of very high or negative inputs, keep divider impedance higher than 10k.
Power to control the discharge of these capacitors comes
from energy stored in a large capacitor connected to BYPASS. The value of the capacitor required can be estimated assuming a maximum BYPASS current of 5µA, a
discharge time of 60s, and a maximum allowable droop
of 5V by:
BOOST: Although the UCC3305 is powered from the
VCC input, most functions of the device operate from a
supply voltage of approximately 10V connected to
BOOST. This 10V supply can be generated by a voltage
doubler using PUMPOUT as an AC signal and external
diodes as switches.
BYPASS: The UCC3305 compensates for lamp temperature changes by changing the voltage on the SLOPEC
and WARMUPC capacitors. These voltages rise as the
lamp warms up. An internal calculation determines what
power should be applied to the lamp.
C = I•
4
60s
∆t
= 5 µA •
= 60 µF
5V
∆V
COMP: Differences between commanded lamp power
and desired lamp power are amplified by an error amplifier. This amplifier senses the difference between the
voltage at FB and 2.5V, and drives COMP with an amplified error voltage. A capacitor is normally connected from
COMP to FB to compensate the overall feedback loop so
that the system will be stable.
UCC2305
UCC3305
PIN DESCRIPTIONS (cont.)
to normal operation is accomplished by the rise of the
WARMUPC capacitor.
DIVPAUSE: The QOUT and QOUT outputs can be used
to switch lamp polarity in an AC ballast. It is important to
stop polarity switching when the lamp is being lit, so that
the arc across the electrodes can form in the correct
place. Pulling high on DIVPAUSE stops the internal divider which generates the QOUT and QOUT signals, and
thereby freezes the QOUT and QOUT signals.
Current mode control has an advantage over voltage
mode control in that a current mode loop is easier to
compensate. Current mode control has a disadvantage
compared to voltage mode control in that the loop can
enter into chaotic oscillations at high duty cycles. These
chaotic oscillations can be prevented using slope compensation. The UCC3305 contains internal slope compensation in the form of a current proportional to OSC
voltage on ISENSEIN. This current combined with an external resistor from ISENSEIN to the current sense resistor creates a voltage drop proportional to OSC voltage,
which gives slope compensation.
To stop the divider when the lamp is being lit and start after the lamp has lit, connect a resistor from NOTON to
DIVPAUSE and a capacitor from DIVPAUSE to GND.
FLTC: The voltage on VOUTSENSE is proportional to
lamp voltage. If that voltage is too high or too low, the
lamp is either open, shorted, or not yet running. During
normal operation, there is a capacitor connected to
FLTC, and this capacitor is discharged to 0V by a current
source inside the UCC3305.
ISET: Many functions inside the UCC3305 require precise currents to give well controlled performance. These
controlled currents are programmed by a resistor from
ISET to GND. A resistor of 100k programs the IC to normal operating current. Lower resistor values increase the
internal currents. Some of the functions which are influenced by this resistor are WARMUPC charging and discharging, SLOPEC charging and discharging, FLTC
charging and discharging, and error amplifier bandwidth
The UCC3305 monitors the voltage on VOUTSENSE
and compares it to an internal 83mV lower threshold and
a 2V upper threshold. If the voltage is outside this window, then the IC will pull up on FLTC with a current of approximately 250nA. If the fault remains long enough to
charge the external FLTC capacitor over 5V, the controller declares a catastrophic fault and shuts the IC down.
The IC will stay shut down until power is removed from
BOOST.
LOADISENSE: Just as ISENSEIN is normally connected
to a current sense resistor which monitors battery current, LOADISENSE is normally connected to a resistor
which monitors lamp current. Lamp current is then regulated by the controller such that the correct lamp power is
supplied at every lamp temperature, in conjunction with
the lamp voltage sensed by VOUTSENSE.
If the fault clears before the FLTC capacitor reaches 5V,
the capacitor discharges down to 0V. This discharge current is approximately 50nA, representing a five times longer discharge rate than charge rate.
FLT: If the voltage on the FLTC pin exceeds 5V, indicating a severe fault, then a latch in the UCC3305 sets and
PWM drive is halted. In addition, the FLT output goes
high to VCC, indicating a serious system fault.
LPOWER: LOADISENSE directly drives one input of an
op amp in the UCC3305. This amplifier amplifies the difference between the desired load current and the actual
load current, and generates an output signal on
LPOWER which feeds the error amplifier.
FB: Differences between commanded lamp power and
desired lamp power are amplified by an error amplifier.
This amplifier senses the difference between the voltage
at FB and 2.5V, and drives COMP with an amplified error
voltage.
GND: Ground for all functions is through this pin.
NOTON: While the lamp is in a fault condition, such as
excessively high or low lamp voltage, NOTON is pulled
high to VCC, indicating that the arc is not yet correct.
When the voltage on VOUTSENSE is within the 83mV to
2V window, NOTON is pulled low.
ISENSEIN: The power regulating algorithm in the
UCC3305 HID Controller computes a function of lamp
current and lamp voltage and commands the appropriate
battery current to keep lamp power constant. This appropriate battery current is sensed by a connection from
I-SENSEIN to a current sense resistor. This current
sensed pulse width modulation scheme is often referred
to as current mode control.
OSC: The fixed frequency PWM in the UCC3305 operates at the frequency programmed by the OSC pin.
Typically, a a 200pF capacitor from OSC to GND programs the PWM frequency at 100kHz. In addition, this
programs the charge pump at 50kHz and the QOUT and
QOUT signals at 192Hz. The actual oscillator frequency
is a function of both the capacitor from OSC to GND and
the resistor from ISET to GND.
In addition to this current regulation, the UCC3305 contains peak input current limiting. This limiting is set to
0.2V across the ISENSEIN resistor during normal operation and 0.4V during starting. The transition from starting
PUMPOUT: Although the UCC3305 is powered from the
VCC input, most functions of the device operate from a
supply voltage of approximately 10V connected to
BOOST. In normal operation, this 10V supply is gener5
UCC2305
UCC3305
PIN DESCRIPTIONS (cont.)
ated by a voltage doubler using the PUMPOUT pin as
an AC signal and external diodes as switches.
PUMP-OUT is a square wave which swings from VCC
to GND at half of the OSC frequency.
VCC: VCC is the main supply input to the UCC3305.
Many functions in the UCC3305 are powered by VCC,
while others are powered by BOOST. VCC should be
clamped to 6.8V by an external zener diode and kept as
close to 6.8V as practical with a low value resistor to the
input supply.
PWMOUT: The output of the pulse width modulator is a
command signal to a power MOSFET switch. This signal appears on PWMOUT. In normal systems,
PWM-OUT can be directly connected to the gate of an
N-channel power MOSFET such as the IRF540. If the
lead between the UCC3305 and the MOSFET is longer
than a few cm, a 10 ohm resistor from PWMOUT to
gate may be required to dampen overshoot and undershoot.
VOUT-SENSE: The VOUTSENSE input is used to
sense lamp voltage, commonly through a 120:1 voltage
divider. For a normal, running HID lamp, the voltage
across the lamp is between 60V and 110V. It takes
higher than 300V to break down the lamp, and it is desirable to limit the voltage on the starter input to 600V
maximum. A lamp voltage less than 10V is indicative of
a shorted lamp.
The UCC3305 regulates lamp power by commanding
the correct lamp current for a given lamp voltage. In addition, a comparator in the UCC3305 terminates a PWM
cycle if VOUTSENSE reaches 5V, corresponding to
600V on the lamp. This regulates lamp voltage at 600V
when the lamp is not lit. Comparators in the UCC3305
also compare VOUTSENSE to 83mV corresponding to
10V lamp voltage and 2V, corresponding to a 240V
lamp voltage. When the VOUTSENSE voltage is outside this window, the lamp is either not lit, shorted, or
open.
QOUT: The UCC3305 is immediately configured for DC
HID lamps. To operate with AC HID lamps, it is necessary to add a power H-bridge which will toggle lamp
voltage. A practical switching frequency for this toggle
function is the OSC frequency divided by 512, or 192Hz
for a 100kHz oscillator.
The QOUT pin is a logic output which toggles at the
OSC frequency divided by 512, 180 degrees out of
phase with the QOUT pin.
QOUT: The QOUT pin is a logic output which toggles at
the OSC frequency divided by 512, 180 degrees out of
phase with the QOUT pin.
WARMUPC: In addition to the capacitor from SLOPEC
to GND, lamp temperature is estimated by the voltage
on a capacitor from WARMUPC to GND. This capacitor
is charged by a 200nA current source to 4.2V and by a
100nA current source from 4.2V to 10V when the lamp
is on, and discharged by 39nA current sink to 2.5V and
11nA current sink to GND when the lamp is off.
SLOPEC: To track lamp warm-up and cool down, two
capacitors connected to the UCC3305 charge and discharge. One is connected to SLOPEC. The other is connected to WARMUPC. The capacitor connected to
SLOPEC charges up to 5V with a rate controlled by the
resistor from ISET to GND. With a nominal 100k ISET
resistor the charging current into SLOPEC is equivalent
to the current from a 50Meg resistor to 5V.
WARMUPV: The voltage on WARMUPC is used to
modulate the signal fed to the error amplifier through
FB. However, the impedance on WARMUPC is too high
to be directly used. The UCC3305 contains a buffer amplifier which buffers the voltage on WARMUPC and processes it to WARMUPV, making a signal appropriate for
driving FB.
When power is removed from VCC, SLOPEC discharges at a constant current, nominally 100nA.
APPLICATIONS INFORMATION
Full Bridge Output Stage
ing is derived from the PWM oscillator. It is desirable to
switch lamp polarity when running, but switching lamp
polarity can interfere with clean starting. The UCC3305
has a logic output called NOTON which is high when the
lamp is not running (Not On) and low when the lamp is
running. This output is connected to the DIVPAUSE input
so that the low frequency switching stops until the lamp
is fully lit.
The output of the flyback converter is directed to the AC
lamp through a full bridge inverter. The full bridge is
switched at a low frequency (typically 195Hz), so that the
average lamp voltage is zero. The low frequency switch-
The UCC3305 HID Controller IC has two low frequency
outputs, QOUT and QOUT. These outputs are capable of
driving low-side MOSFETs directly at 195Hz, but
high-side MOSFETs require a level-shifted drive. This
Typical Application
This circuit shows the UCC3305 HID Lamp Controller IC
in a flyback converter. The output of the converter is regulated at constant power, so that lamp intensity is relatively constant regardless of small lamp manufacturing
variations.
6
UCC2305
UCC3305
APPLICATIONS INFORMATION (cont.)
can be as simple as a high voltage transistor and a resistor pull-up, combined with the correct choice of phases.
regulated 10V supply on the BOOST output. This 10V
supply drives all other functions on the UCC3305.
Regulated Lamp Input Power Gives
Constant Intensity
Protection From Over Voltage
The most significant stresses in an automotive environment are the overvoltage conditions which can occur during load dump and double-battery jump start. At these
times, the voltage into the ballast can go so high that
even the most overdesigned power stage will be damaged. The UCC3305 is inherently immune to damage
from this when operated with a zener regulated supply. In
addition, the UCC3305 will protect the ballast components by shutting down the PWM in the presence of excessive voltage on the BAT input.
The LPOWER output of the UCC3305 is a voltage
roughly proportional to lamp input power. The UCC3305
regulates constant lamp power over a wide range of lamp
voltages. The range of lamp voltages which produce constant lamp power is set by the limiting amplifier on
VOUTSENSE.
For inputs to VOUTSENSE below 0.5V, such as would
occur with a shorted lamp, the loop regulates constant
load current. For inputs to VOUTSENSE greater than
0.82V, as might occur with a lamp that is open or not yet
lit, the loop also regulates constant load current, but at a
lower current than for a shorted lamp. In between those
two voltages, the amplifier driving the LPOWER pin will
sum the load current and load voltage and produce a signal roughly proportional to load power. The summing amplifier approximates power well enough to hold power
within ±10% over a factor of two in lamp voltage.
This typical application shows a voltage divider consisting of a 270k resistor and a 100k resistor driving the BAT
input. The threshold of the BAT input is approximately 5V,
so this divider sets the shutdown voltage at approximately 18.5V.
Programming the UCC3305
All circuitry on the UCC3305 HID Lamp Controller is operated from a bias current set by the resistor from ISET
to ground. For best operation, this resistor (RSET) should
be between 75k and 150k.
The UCC3305 HID Controller contains a current mode
PWM similar to the industry standard UC3842 and
UCC3802 circuits. This controller uses a high gain op
amp to regulate the output of the LPOWER circuit. This
op amp drives a high speed PWM comparator, which
compares converter input current to the output of the op
amp and uses this signal to set duty cycle.
Oscillator Frequency
The UCC3305 HID Lamp Controller PWM oscillator is set
by the resistor from ISET to ground and by the capacitor
from OSC to ground. Oscillator frequency can be estimated by the equation:
Slope Compensation
In addition to a complete current mode PWM, the
UCC3305 HID Controller contains internal slope compensation, a valuable function which improves current
loop stability for high duty cycles. Slope compensation is
accomplished with an on-chip current ramp and an
off-chip resistor RSL. Larger values of RSL give more
slope compensation and a more stable feedback loop.
FOSC =
2
RSET • COSC
For operation at 100kHz, RSET should be 100k and
COSC should be 200pF.
The PWM oscillator also determines the low frequency
lamp switching rate for AC lamps. The exact lamp switching rate is the PWM frequency divided by 512.
Powering The UCC3305
Lamp Temperature Compensation
Conventional power MOSFETs require at least 8V of
gate drive to ensure high efficiency and low on resistance. Despite this requirement, the UCC3305 HID Controller can be used to build a ballast that will drive power
MOSFETs well with input supplies as low as 5V! The
UCC3305 does this using a charge pump.
Automobile headlights must come up to full intensity very
quickly, but HID lamps require many minutes to stabilize.
The UCC3305 HID Controller contains sophisticated internal circuitry to anticipate lamp temperature and also to
compensate for lamp temperature.
The circuits anticipate lamp temperature by monitoring
charge on capacitors which charge when the lamp is on
and discharge when the lamp is off. The UCC3305 HID
Controller compensates for lamp temperature by driving
the lamp with a higher lamp power when the lamp is cold
and reducing the power to a normal operating level when
the lamp is warmed up. The capacitors which set these
In this typical application, power for the UCC3305 HID
Controller IC is derived from a 6.8V zener supply. This
zener regulated supply gives the application overvoltage
protection, reverse battery protection, low parts count,
and low cost. The output of the 6.8V zener supply drives
the VCC pin of the UCC3305. VCC is the input to the
UCC3305 charge pump. The charge pump generates a
7
UCC2305
UCC3305
APPLICATIONS INFORMATION (cont.)
time constants are external film capacitors CS and CW,
and are connected to SLOPEC and WARMUPC. CS and
CW are critical capacitors and must be selected to
match the time-temperature relationship of the lamp.
When power is removed from the ballast, CS and CW
must discharge at a controlled rate. The discharge currents are programmed by current sources on the
UCC3305 HID Controller. These current sources are
powered by the power supply connected to BYPASS. In
a typical application, a non-critical electrolytic capacitor
from BYPASS to ground stores energy when the ballast
is on and uses this energy to control the discharge rate
when the ballast is off.
In addition to changing the power regulation point, the
WARMUPC capacitor voltage also changes the short circuit lamp current. The ratio of cold short circuit current to
warmed-up short circuit current is set by the resistor
from ADJ to ground.
FLYBACK HID BALLAST
UDG-94092-1
8
PACKAGE OPTION ADDENDUM
www.ti.com
7-Oct-2009
PACKAGING INFORMATION
Orderable Device
Status (1)
Package
Type
Package
Drawing
Pins Package Eco Plan (2)
Qty
UCC2305DW
ACTIVE
SOIC
DW
28
20
Green (RoHS &
no Sb/Br)
CU NIPDAU
Level-2-260C-1 YEAR
UCC2305DWG4
ACTIVE
SOIC
DW
28
20
Green (RoHS &
no Sb/Br)
CU NIPDAU
Level-2-260C-1 YEAR
UCC2305N
ACTIVE
PDIP
N
28
13
Green (RoHS &
no Sb/Br)
CU NIPDAU
N / A for Pkg Type
UCC2305NG4
ACTIVE
PDIP
N
28
13
Green (RoHS &
no Sb/Br)
CU NIPDAU
N / A for Pkg Type
UCC3305DW
ACTIVE
SOIC
DW
28
20
Green (RoHS &
no Sb/Br)
CU NIPDAU
Level-2-260C-1 YEAR
UCC3305DWG4
ACTIVE
SOIC
DW
28
20
Green (RoHS &
no Sb/Br)
CU NIPDAU
Level-2-260C-1 YEAR
UCC3305N
ACTIVE
PDIP
N
28
13
Green (RoHS &
no Sb/Br)
CU NIPDAU
N / A for Pkg Type
UCC3305NG4
ACTIVE
PDIP
N
28
13
Green (RoHS &
no Sb/Br)
CU NIPDAU
N / A for Pkg Type
Lead/Ball Finish
MSL Peak Temp (3)
(1)
The marketing status values are defined as follows:
ACTIVE: Product device recommended for new designs.
LIFEBUY: TI has announced that the device will be discontinued, and a lifetime-buy period is in effect.
NRND: Not recommended for new designs. Device is in production to support existing customers, but TI does not recommend using this part in
a new design.
PREVIEW: Device has been announced but is not in production. Samples may or may not be available.
OBSOLETE: TI has discontinued the production of the device.
(2)
Eco Plan - The planned eco-friendly classification: Pb-Free (RoHS), Pb-Free (RoHS Exempt), or Green (RoHS & no Sb/Br) - please check
http://www.ti.com/productcontent for the latest availability information and additional product content details.
TBD: The Pb-Free/Green conversion plan has not been defined.
Pb-Free (RoHS): TI's terms "Lead-Free" or "Pb-Free" mean semiconductor products that are compatible with the current RoHS requirements
for all 6 substances, including the requirement that lead not exceed 0.1% by weight in homogeneous materials. Where designed to be soldered
at high temperatures, TI Pb-Free products are suitable for use in specified lead-free processes.
Pb-Free (RoHS Exempt): This component has a RoHS exemption for either 1) lead-based flip-chip solder bumps used between the die and
package, or 2) lead-based die adhesive used between the die and leadframe. The component is otherwise considered Pb-Free (RoHS
compatible) as defined above.
Green (RoHS & no Sb/Br): TI defines "Green" to mean Pb-Free (RoHS compatible), and free of Bromine (Br) and Antimony (Sb) based flame
retardants (Br or Sb do not exceed 0.1% by weight in homogeneous material)
(3)
MSL, Peak Temp. -- The Moisture Sensitivity Level rating according to the JEDEC industry standard classifications, and peak solder
temperature.
Important Information and Disclaimer:The information provided on this page represents TI's knowledge and belief as of the date that it is
provided. TI bases its knowledge and belief on information provided by third parties, and makes no representation or warranty as to the
accuracy of such information. Efforts are underway to better integrate information from third parties. TI has taken and continues to take
reasonable steps to provide representative and accurate information but may not have conducted destructive testing or chemical analysis on
incoming materials and chemicals. TI and TI suppliers consider certain information to be proprietary, and thus CAS numbers and other limited
information may not be available for release.
In no event shall TI's liability arising out of such information exceed the total purchase price of the TI part(s) at issue in this document sold by TI
to Customer on an annual basis.
Addendum-Page 1
MECHANICAL DATA
MPDI008 – OCTOBER 1994
N (R-PDIP-T**)
PLASTIC DUAL-IN-LINE PACKAGE
24 PIN SHOWN
A
24
13
0.560 (14,22)
0.520 (13,21)
1
12
0.060 (1,52) TYP
0.200 (5,08) MAX
0.610 (15,49)
0.590 (14,99)
0.020 (0,51) MIN
Seating Plane
0.100 (2,54)
0.021 (0,53)
0.015 (0,38)
0.125 (3,18) MIN
0.010 (0,25) M
PINS **
0°– 15°
0.010 (0,25) NOM
24
28
32
40
48
52
A MAX
1.270
(32,26)
1.450
(36,83)
1.650
(41,91)
2.090
(53,09)
2.450
(62,23)
2.650
(67,31)
A MIN
1.230
(31,24)
1.410
(35,81)
1.610
(40,89)
2.040
(51,82)
2.390
(60,71)
2.590
(65,79)
DIM
4040053 / B 04/95
NOTES: A.
B.
C.
D.
All linear dimensions are in inches (millimeters).
This drawing is subject to change without notice.
Falls within JEDEC MS-011
Falls within JEDEC MS-015 (32 pin only)
POST OFFICE BOX 655303
• DALLAS, TEXAS 75265
IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, modifications, enhancements, improvements,
and other changes to its products and services at any time and to discontinue any product or service without notice. Customers should
obtain the latest relevant information before placing orders and should verify that such information is current and complete. All products are
sold subject to TI’s terms and conditions of sale supplied at the time of order acknowledgment.
TI warrants performance of its hardware products to the specifications applicable at the time of sale in accordance with TI’s standard
warranty. Testing and other quality control techniques are used to the extent TI deems necessary to support this warranty. Except where
mandated by government requirements, testing of all parameters of each product is not necessarily performed.
TI assumes no liability for applications assistance or customer product design. Customers are responsible for their products and
applications using TI components. To minimize the risks associated with customer products and applications, customers should provide
adequate design and operating safeguards.
TI does not warrant or represent that any license, either express or implied, is granted under any TI patent right, copyright, mask work right,
or other TI intellectual property right relating to any combination, machine, or process in which TI products or services are used. Information
published by TI regarding third-party products or services does not constitute a license from TI to use such products or services or a
warranty or endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual
property of the third party, or a license from TI under the patents or other intellectual property of TI.
Reproduction of TI information in TI data books or data sheets is permissible only if reproduction is without alteration and is accompanied
by all associated warranties, conditions, limitations, and notices. Reproduction of this information with alteration is an unfair and deceptive
business practice. TI is not responsible or liable for such altered documentation. Information of third parties may be subject to additional
restrictions.
Resale of TI products or services with statements different from or beyond the parameters stated by TI for that product or service voids all
express and any implied warranties for the associated TI product or service and is an unfair and deceptive business practice. TI is not
responsible or liable for any such statements.
TI products are not authorized for use in safety-critical applications (such as life support) where a failure of the TI product would reasonably
be expected to cause severe personal injury or death, unless officers of the parties have executed an agreement specifically governing
such use. Buyers represent that they have all necessary expertise in the safety and regulatory ramifications of their applications, and
acknowledge and agree that they are solely responsible for all legal, regulatory and safety-related requirements concerning their products
and any use of TI products in such safety-critical applications, notwithstanding any applications-related information or support that may be
provided by TI. Further, Buyers must fully indemnify TI and its representatives against any damages arising out of the use of TI products in
such safety-critical applications.
TI products are neither designed nor intended for use in military/aerospace applications or environments unless the TI products are
specifically designated by TI as military-grade or "enhanced plastic." Only products designated by TI as military-grade meet military
specifications. Buyers acknowledge and agree that any such use of TI products which TI has not designated as military-grade is solely at
the Buyer's risk, and that they are solely responsible for compliance with all legal and regulatory requirements in connection with such use.
TI products are neither designed nor intended for use in automotive applications or environments unless the specific TI products are
designated by TI as compliant with ISO/TS 16949 requirements. Buyers acknowledge and agree that, if they use any non-designated
products in automotive applications, TI will not be responsible for any failure to meet such requirements.
Following are URLs where you can obtain information on other Texas Instruments products and application solutions:
Products
Applications
Audio
www.ti.com/audio
Communications and Telecom www.ti.com/communications
Amplifiers
amplifier.ti.com
Computers and Peripherals
www.ti.com/computers
Data Converters
dataconverter.ti.com
Consumer Electronics
www.ti.com/consumer-apps
DLP® Products
www.dlp.com
Energy and Lighting
www.ti.com/energy
DSP
dsp.ti.com
Industrial
www.ti.com/industrial
Clocks and Timers
www.ti.com/clocks
Medical
www.ti.com/medical
Interface
interface.ti.com
Security
www.ti.com/security
Logic
logic.ti.com
Space, Avionics and Defense
www.ti.com/space-avionics-defense
Power Mgmt
power.ti.com
Transportation and
Automotive
www.ti.com/automotive
Microcontrollers
microcontroller.ti.com
Video and Imaging
www.ti.com/video
RFID
www.ti-rfid.com
Wireless
www.ti.com/wireless-apps
RF/IF and ZigBee® Solutions
www.ti.com/lprf
TI E2E Community Home Page
e2e.ti.com
Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265
Copyright © 2011, Texas Instruments Incorporated
APPLICATION NOTE U-162
Ron Fiorello
Unitrode Corporation
DRIVING A 35W AC METAL HALIDE
HIGH INTENSITY DISCHARGE
LAMP WITH THE UCC3305 HID
LAMP CONTROLLER
7/97
U-162
APPLICATION NOTE
DRIVING A 35W AC METAL HALIDE HIGH INTENSITY DISCHARGE LAMP WITH
THE UCC3305 HID LAMP CONTROLLER
by Ron Fiorello
Unitrode Corporation
ABSTRACT
This application note expands upon Unitrode application note U-161 which describes the use of the
UCC3305 HID lamp controller in a Metal Halide DC lamp application. This application note extends that
topic to include a 35W Metal Halide AC lamp application.
When using these types of lamps it is beneficial to provide an improved means of driving the lamp in an
efficient and cost effective manner. Electronic ballasts offer definite advantages in size, weight, efficiency
and performance over their magnetic counterparts. Electronic ballasts can also provide additional
protection features which would not easily be possible with a magnetic ballast.
Information is presented in a design example to help the user better understand the controller’s many
features.
INTRODUCTION
The UCC3305 controller integrates most of the
features necessary to control one AC HID lamp. It
is tailored to the demanding, instant start requirements of automotive headlamps, but is also applicable to other lighting applications such as emergency or theatrical lighting where HID lamps are
being used.
The basic features of the UCC3305 HID controller
are outlined below;
•
•
•
•
•
•
•
•
•
OV input protection
Output fault protection/timing
Power regulation vs. lamp voltage
Lamp startup/cool down simulation
Current-mode control
Fixed frequency operation
DC or AC lamp drive capability
High current drive capability
Adjustable startup to steady-state lamp
current ratio
A summation of the different functional blocks
of the UCC3305 and their major electrical characteristics was included in U-161 and will not be
repeated here.
7/97
The following section specifies typical design
requirements necessary of an HID ballast that
would be powering an AC lamp in a 12V automotive battery application. The lamp used in this application is a 35 AC metal halide lamp manufactured
by Philips Lighting.
Input ballast voltage requirements - 9 to 16 Vdc
Protection/Fault Monitor/ I/O Controls - Protection
against input overvoltage, output open circuit and
output short circuit
Power regulation - Regulate power to the lamp
within ±5% over a lamp voltage variation of 60 to
100 Vdc.
Lamp ignition voltage - Provide an open circuit
voltage of 400 to 500 Vdc at startup in order to provide the necessary voltage to the ignition circuit to
ignite the lamp.
Efficiency - Greater than 85%.
Hot restrike - The ballast must be able to properly
light the lamp when hot without a cool down period.
Converter topology
The DC-DC converter topology chosen for this
application, as in the DC lamp case is the SEPIC.
This topology was chosen because it is a single
ended topology and the power switch can be easily
driven. Also, the output voltage can be higher or
lower than that of the input. In this application the
output is always higher than the input but varies
widely from the time before the arc lamp is lit to
when the arc has been established. This power
stage converts the battery voltage to a level which
is suitable for the lamp. This voltage is then
converted to an AC voltage by an output H-bridge
which feeds the lamp. All of the power regulation
and protection is provided on the DC output.
– The Hbridge drivers are driven from the Q and Q outputs
of the UC3305. The circuit is shown in the schematic at the end of this application note.
1) An output H-bridge is not required.
2) The associated H-Bridge drivers are not
needed.
3) The ignitor circuit design is simpler.
In addition to the reasons given above there is also
an issue with acoustic arc resonance of the lamp.
This problem arises because the arc tube can be
thought of as an acoustic chamber which has certain unique properties. Given the right excitation
frequency (or the wrong one in this case), a resonance occurs which causes the arc discharge to
become unstable and move around within the arc
tube. This unstable condition can result in the arc
coming in contact with the arc tube wall, resulting in
the destruction of the lamp. As a minimum, this
condition results in problems with the optics of the
lamp. Each lamp has its own resonant spectrum
characteristics which will vary with manufacturing
tolerances. A particular lamp may have several resonant frequencies of which the ballast designer
must be aware and stay away from so that problems can be avoided. Information regarding the resonant frequencies of a particular lamp are typically
not given as part of the design specification. The
ballast designer must find this information on his
own by trial and error or specifically asking the
lamp manufacturer to characterize the lamp over
frequency.
POWER REGULATION LOOP
An analysis of the power regulation loop, including
the derivation of the power curve equation based
on the 35W DC lamp was performed in [1] and will
not be repeated here. This analysis holds for the
AC lamp as well and should be reviewed for a complete understanding of the controller. A graph of the
power curve is repeated below.
Kv: = 0.0032
R2: = 16k R1: = 4.7k
Ki: = 0.83
Vo1: = 60,65..110
This discussion appears to provide a compelling
reason to select a DC lamp. However DC lamps are
not without drawbacks. In some cases the position
in which the lamp can be operated is restricted i.e.;
the electrodes must be vertical or horizontal
depending on the lamp, limiting the applications.
Also, in a DC lamp the cathode electrode must be
designed larger then the anode due to a gradual
migration of the cathode material over the operating life of the lamp. In the AC lamp, each electrode
functions as both a cathode and anode so that
material migration is of little concern. There are
also issues relating to the symmetry of the arc discharge and optics which favor AC over a DC lamp.
Vref: = 2.4
Vo2: = 110,115..120
PVo1: =
Vo1
•
Ki
• Vref − Kv • Vo1 •
[( R1
R2 )
+1
( R1
R2 )]
PVo2: =
Vo2
•
Ki
• Vref − 0.322 •
[( R1
R2 )
+1
( R1
R2 )]
As a result of these issues and others, the debate
continues on the advantages of DC vs. AC lamp
drives. In AC lamp applications, most ballast
designers have settled on drive frequencies below
1kHz to avoid the problems associated with
acoustic resonance but above 100Hz to avoid lamp
flicker problems. Other designs use a frequency
modulation technique to avoid acoustic resonance
from occurring. This technique appears to have
some merit but can still result in problems in some
instances.
DC VS AC LAMP DRIVES
There has been much discussion regarding the
benefits of DC vs. AC lamps over the years. The
main advantage of a DC lamp is obvious to the ballast designer but not necessarily to the designer of
the lamp. Driving the lamp with a DC source is easier and cheaper from a ballast standpoint for the following reasons:
2
UDG-96242
The expression for PO1 is valid for lamp voltages from 60 to 105V. The expression for PO2 is valid for lamp
voltages above 105V (This is due to the limiter block inside the UCC3305). Above approximately 105V, the
lamp will be driven in a constant current mode which results in the straight line for power vs. lamp voltage
as shown. The output power can be regulated with a variation of less than ±5% from a nominal of 34.5W
with a lamp voltage variation from 60V to 110V.
As the lamp ages, its terminal voltage will normally increase due to electrode material erosion. This erosion can occur for a variety of reasons which are described in [2]. Because of this, it is beneficial to operate the lamp with a constant current above a certain voltage. This will insure adequate current reducing the
possibility of the arc being extinguished . Above a steady-state lamp voltage of 240 Vac, the ballast will terminate operation. This type of voltage increase is typical indication of an end of life condition and that the
lamp is close to failure.
Figure 1. Calculated Power Curve vs. Lamp Voltage of UCC3305 Controlled 35W Ballast
Powering AC Metal Halide Philips Lamp
The data suggests that a square wave drive current
is required for long term lamp reliability. This results
in constant power being delivered to the lamp and
therefore a constant light output as well as a crest
factor (ratio of peak to RMS current) close to 1. As
mentioned above, this also eliminates the possibility of flicker since the lamp current transitions quickly through zero without the arc needing to be
reignited at each zero crossing.
added feature allows the user to program the ratio
of startup to steady-state current. This feature provides increased power to the lamp at startup in
order to get the light output up to its steady-state
level as quick as possible. The simplified schematic of this section is shown in Figure 2.
The start to run current ratio is set by by an external resistor placed from the ADJ pin to ground. A
detailed analysis of this feature as well as how the
output current limit on startup operates is provided
in [1].
Current sense comparators/amplifiers
The INPUT ISENSE comparator/amplifier inside
the UCC3305 provides cycle by cycle current control as in a typical peak current-mode converter. An
3
UDG-97118
Figure 2. - ??????????
to steady-state operation. The ability of the controller to predict this optimum starting characteristic
improves the reliability of the lamp and system performance. If the lamp has been started and is
turned off for any reason, it is beneficial to restart
the lamp in a manner that would prolong lamp life
and reduce the cool down period between lamp
starts. Typically, HID lamp ballasts in use today
have no provision for hot restrike and instant start.
To facilitate this, the UCC3305 has the capability to
program the discharge of the SLOPEC and
WARMUPC capacitors in a manner that simulates
the cool down characteristics of the lamp.
SLOPEC AND WARMUPC
The ADJ pin sets the maximum startup current
which flows into the lamp at ignition. The current
into the lamp will decay in an exponential fashion
due to the voltage charging characteristic of the
WARMUPC and SLOPEC capacitors. The current
will decay to a steady-state value of approximately
450mA RMS after a period of time given by the
time constant of an internal 50 Meg resistor and the
capacitor placed from the SLOPEC pin to ground.
In this example, the time required to reach steadystate current is 150 seconds from:
A capacitor on the BYPASS pin provides the necessary energy to the internal current sources when
the power to the IC has been disrupted. This will
determine the discharge time of the SLOPEC and
WARMUPC capacitors. The value of this capacitor
should be chosen based on the current of 5µA and
a discharge time (cool down time) desired based
on the characteristics of the lamp used.
t: = 50 • exp(6) • CSLOPEC
The WARMUPV capacitor has a shorter time constant and is used to tailor the first few seconds of
the run-up characteristic for the lamp. As can be
seen from the block diagram in the data sheet, this
capacitor has no effect on the operation of the ballast controller after it has been charged beyond 5V
due to the built in limit function.
As an example, for a discharge time of 120 seconds and a maximum allowable droop of 5V;
Another function of both the WARMUPC and
SLOPEC capacitors is to program the hot-restrike
characteristics of the ballast. The voltage on these
capacitors also provides information to the controller on the temperature state of the lamp. This
allows the ballast to predict how the lamp should be
started i.e.; whether or not the lamp needs to go
through the warm-up phase or just proceed directly
l•t
120S
= 5µA
= 120µF
V
5V
If the lamp is operating for some period of time and
is then turned off and instantly restarted, its power
will instantly reach that level which occured before
the lamp arc was extinguished. This occurs after
the initial 10 to 20mS power burst which is required
C=
4
to get the arc to form properly on the tips of the
electrodes. As a result, the lamp can be instantly
started under any condition using a properly
designed ignitor.
high frequency properties since the ignition pulse
must have a very fast rise time, usually on the order
of less than a micro-second. The interwinding
capacitance of the transformer must be minimized
since this element, in conjunction with the secondary inductance of the transformer, will act as a
low pass filter. All of this, coupled with the fact that
the secondary may need to generate tens of kilovolts for hot re-strike, puts enormous burden on the
transformer design. The requirement to minimize
the interwinding capacitance requires as low a
number of turns as possible. This obviously conflicts with the high ignition voltage required.
Because of this, the open circuit voltage of the ballast should be as high as possible so that the number of turns on the secondary can be reduced.
The capacitors used for the SLOPEC and
WARMUPC functions must have low leakage characteristics since they are charged from nanoamp
current sources internal to the IC. Any significant
amount of leakage current caused by these components will have an effect on the output power regulation characteristics of the ballast.
Lamp ignition requirements
The ballast design is usually considered separately from that of the ignitor circuit although certain
decisions must be made to help simplify the ignitor
design. Below is a summary of some of the choices which must be made in designing the ignitor.
Typically a triggering device of some type is used in
the ignitor. This triggering device must be able to
handle a very large di/dt since hundreds of amps of
current could be required to get the arc established
within the first couple of microseconds. Currently,
there aren't many semiconductor switches which
meet this requirement. Because of this, most ignitors use spark-gaps as the transformer primary
triggering element. These devices are capable of
extremely fast breakdown times (typically < 50nS)
and can switch hundreds of amps. Once the arc
has been established, the lamp terminal voltage
collapses and the ignitor firing circuit is no longer in
the circuit. The only remaining element is the ignitor secondary inductance which, in a series ignitor,
is in series with the lamp.
In designing the ballast, the open circuit output voltage is usually made as high as possible. This
reduces the burden on the ignitor transformer
design. Typical ignition voltages vary for different
types of arc lamps but usually are in the range of
4kV to 30kV depending on the internal pressure of
the lamp, temperatures and electrode spacing. HID
lamps usually require a lower starting voltage to
ignite a cold lamp and a much higher starting voltage for a hot lamp. This increase in ignition voltage
for a hot lamp is due to an increase in pressure
within the arc tube. Again, this depends on the
lamp design. Some lamps are high pressure lamps
and take approximately the same ignition voltage
hot or cold (usually very high, 30kV) to start. The
ignition pulse must be high enough in voltage to
break down the lamp and ionize the gases and
metals inside the arc tube. Then, within the first
couple of milli-seconds, enough current must be
available to properly establish the arc and maintain
the discharge. During this period the electrodes are
still cold and must be heated quickly to cause thermonic emission of electrons between cathode and
anode. Once this process occurs, the arc is established and has to be maintained. In an AC lamp, the
electrodes act as both anode and cathode due to
the AC drive current. When the lamp is being ignited, it is sometimes recommended that the switching of the AC bridge be paused. This is so that the
ignition process is not interrupted by an AC drive
current which could result in the arc being extinguished. It hasn't yet been definitively proven that
this facilitates ignition of the lamp.
Slope compensation resistor
Slope compensation in the UCC3305 is provided
by the addition of an external resistor in series with
the INPUT Isense pin. This resistor adds a portion
of the oscillator ramp into the current sense signal
to provide the necessary slope compensation for
duty cycles exceeding 50%. The amount of slope
compensation that is needed is dependent on the
topology used, as well as the inductor values chosen. In the SEPIC converter, both input and output
inductors need to be considered when determining
how much slope compensation is necessary. The
analysis of this function was performed in [1] and
will not be repeated here.
Frequency response of the power
regulation loop
The frequency response of the ballast is determined by analysis of the power regulation loop in
the same manner outlined in [1]. It is the output current that is being regulated since the lamp really
The ignitor transformer design is not a trivial matter.
The material chosen for the core must have good
5
regulates the output voltage. A simplified analysis
can be performed by modeling the power stage as
a voltage controlled current source with some
transconductance gain, Gm.
.01µF. The low frequency gain of this amplifier is set
to 7.5dB. The combination of this gain and the
power stage gain results in -15dB of low frequency
gain with a pole at 1kHz.
The transconductance gain, Gm;
Gm = ∆Io
∆Ve
The final loop response is tailored with the main
error amplifier. A zero is added in the amplifier
response at mid-band frequency so that the DC
gain for the overall loop is as high as possible. The
high frequency gain of this amplifier must be well
below 0dB to ensure adequate gain and phase
margin for the open loop gain. The 16Ω resistor has
been determined based on the power curve characteristic desired leaving only the feedback resistor
value to be chosen. If this resistor is chosen so that
the high frequency gain is less than -20dB for good
gain margin, feedback resistor value of 1Ω, the
capacitor value can then be determined. A zero frequency of 3.4kHz is chosen to give adequate low
frequency gain boost. The resulting value of series
capacitor is 0.047µF. The gain and phase margin
with these component values is greater than 20dB
and 50 degrees respectively.
The output current is converted to a voltage by the
output current sense resistor. The gain of the power
stage, Gp is;
Gp= Rls Gm
where; Rls = the load sense resistance (0.75 ohms)
Gp= -22.5dB
The loop response is now tailored for good power
regulation by achieving as high a DC gain as possible. The gain of the LOAD SENSE amplifier is
restricted since the gain of this stage effects the
power curve characteristic as shown in [1].
The LOADSENSE amplifier is set up as an integrator to filter out switching frequency noise from the
control loop. The pole frequency was chosen to be
at 1Khz to give good rejection of the switching frequency noise. This results in a capacitor value of
A PSPICE simulation of the control loop is shown in
the figure below indicating more than adequate
gain and phase margin.
Figure 3. - ??????????
6
CIRCUIT EXAMPLE
UCC3305 does not provide sufficient energy to
drive the BOOST pin and driver. The voltage drop
of the PUMPOUT pin, when used as a charge
pump, is two to three volts which results in insufficient voltage to Vboost for driving the gate of Q1.
A 9 to 16 Vdc input SEPIC converter powering a
35W PHILIPS AC lamp was built and tested. Data
on power curve and various oscillagrams of current
and voltages in the power/control circuit were taken
and are discussed.
Performance data
The magnetics design for this converter is exactly
the same as in the DC lamp case outline in [1] and
will not be repeated here.
Performance data on the ballast is presented in the
following curves showing efficiency and the measured power curve. Oscillograms of Q1 voltage and
current are also given as well as startup characteristics of the lamp voltage and current. The maximum efficiency achieved was 85.8% at a lamp voltage of 100V. The open circuit voltage of the lamp
cannot be seen due to the aliasing of the oscilloscope, but the ballast output voltage is approximately 450Vac before lamp ignition. Once the lamp
ignites, the voltage collapses and the lamp current
increases to 2Aac. Eventually, the lamp voltage
begins to increase and the current decreases. They
will arrive to their steady-state values of 80 to
90Vac and 450mAac respectively after approximately 150 seconds.
The additional circuitry required for the AC lamp
consist of the H-bridge and drivers. The H-bridge
MOSFET must be able to withstand the open circuit
voltage of the ballast which is set to approximately
450Vdc. MTP6N60 devices from Motorola were
used which have a Vds rating of 600V at 6 Amps
continuous. The MOSFET drivers used are 2lRF2110 each of which are capable of driving the
four H-Bridge MOSFETS.
An external boost converter is used to convert the
6.8 Vcc voltage to the 10V boost voltage required
for the gate drive of Q1. This circuit is necessary
since the internal charge pump circuit in the
UDG-96244
Figure 4. - ??????????
7
Figure 5. - ??????????
Figure 6. - ??????????
8
Figure 8. - ??????????
Figure 9. - ??????????
9
CONCLUSION
REFERENCES
The performance data presented of a typical
UCC3305 HID lamp controller application, demonstrated it to be a excellent means of controlling an
AC metal halide HID lamp. The power regulation
and efficiency achieved using the SEPIC converter
topology proved it to be a good alternative to other
conventional circuit topologies for this application.
The many protection and control features of the
UCC3305 simplify the task of the ballast designer
considerably, making it an economically feasible
choice for AC as well as DC HID lamp applications.
[1]
“Powering a 35W MH Lamp HID Lamp
using the UCC3305 Unitrode Application
Note U-161”
[2]
Waymouth: :Electric Discharge Lamps M.I.T.
Press” Cambridge MA.
[3]
Loyd Dixon “High Power Factor Preregulator
using Sepic Converter” Unitrode Power
Supply Design Seminar SEM1100.
PARTS LIST
35W HID BALLAST PARTS LIST
REF DES
RI
R3
R4
R5
R6
R7
R8
R9
R10
R11
R12
R13
R14
R15
R16
R17,R18
R19,R25,R32
R20
R21
R22,R23
R24
R26,R27
R28,R29
R30
R31
R33
C33
C1
C2,C3,C26
C4
C5,C8,C11,
C27,C28,C29
C6,C7
C9
C10
C12,C13
C14
C15
C16
PART DESCRIPTION
10Ω 1/4 W CC
1k 1/4W CC
4k 1/4W CC
240Ω 1/4W CC
270k 1/4W CC
100k 1/4W CC
1.68k 1/4W CC
220Ω 1/2W CC
120Ω 1/2W CC
5.1k 1/4W CC
15k 1/4W CC
16.1k 1/4W CC
1k 1/4W CC
150k 1/4W CC
250k 1/4W CC
27k 1/4W CC
10k 1/4W CC
0.75Ω 3W CC
565k 1/4W CC
282k 1/4W CC
560Ω 1/2W CC
100k 1/4W CC
130k 1/4W CC
18Ω 3W CC
330Ω 3W CC
10Ω 1/4W CC
10µF/100V POLY FILM
1µF/50V METALLIZED FILM
470µF/50V ALUM ELEC
2.2µF/400V POLY FILM
0.47µF/50V CERAMIC
DIGIKEY NUMBER
10QBK-ND
1KQBK-ND
4KQBK-ND
240QBK-ND
270KQBK-ND
100KQBK-ND
1.6KQBK-ND
220HBK-ND
120H-ND
5.1KQBK-ND
15KQBK-ND
16KQBK-ND
1KQBK-ND
150KQBK-ND
250KQBK-ND
27KQBK-ND
10KQBK-ND
VC3D.75-ND
562KXBK-ND
280KXBK-ND
560HBK-ND
100KQBK-ND
130KQBK-ND
VC3D18-ND
VC3D330-ND
10QBK-ND
EF1106-ND
P4675-ND
P1248-ND
EF4225-ND
P4671-ND
4.7µF/250V ALUM ELEC
470pF/25V CERAMIC
10µF/35V ALUM ELEC
1µF METALIZED FILM, ITWPAK #105K050RA4
220pF/50V CERAMIC
4.7µF/50V ALUM ELEC
47µF/25V ALUM ELEC
P6187-ND
P4808-ND
P1227-ND
NISSEI #R68105K63B
P4804-ND
P1240-ND
P1220-ND
10
QTYPER
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
3
1
1
2
1
2
2
1
1
1
1
1
3
1
6
2
1
1
2
1
1
1
PARTS LIST (con’t)
35W HID BALLAST PARTS LIST
REF DES
C17,C18,C19
C20,C22,C24
C21,C23
C25
C30
C31
C32
Z1
Z2,Z3
U2
R2A,B
Q2,Q3
Q1
Q2,Q3,Q4,Q5
D1
D2,D3
U3,U4
HS2,3,4,5
TB1,TB2
U1
HS1
L1
L2
PART DESCRIPTION
0.01µF/50V CERAMIC
0.1µF/50V CERAMIC
1µF/35V TANTALUM
1000pF/50V CERAMIC
100µF/25V ALUM ELEC
180pF/1kV CERAMIC DISK
1000pF/100V CERAMIC
1N5235B, 6.8V ZENER
1N4761A, 75V ZENER
LM317L, ADJ LINEAR REG, TO-220
0.01Ω 3W CC
2N3904, 40V, 0.200mA TRANISTOR
IRF1310, 100V, 0.027Ω
MOTOROLA MTP8N50E,500V TO-220
MOT MUR860, 600V, 8A FST REC
MOT MUR160, 600V, 1A FST REC
IRF2110, MOSFET DRIVER
THERMALLOY#7128D, HS FOR Q2,Q3,Q4
TERMINAL BLOCK
UCC3305JP
THERMALLOY #6398-P2,HS FOR Q1
E100-18 MICROMETALS
CORE-30T #18AWG
30µH
RM10PA250-3F3 PHILIPS
10T PRIMARY LITZ (2X10X,1)
60T SECONDARY LITZ(1X15X,1)
WINDING SEQUENCE
(PRIM-10T, SEC-30T, PRIM-10T)
11
DIGIKEY NUMBER
P4513-ND
P4525-ND
P2059-ND
P4812-ND
P1221-ND
P4119-ND
P4036-ND
1N5235BCT-ND
1N4761ACT-ND
9244B-ND
NEWARK#IRF1310
NEWARK #MTP8N50E
NEWARK#MUR860
NEWARK #MUR160
NEWARK #IRF2110
NEWARK#95F715
44F4435
QTYPER
3
3
2
1
1
1
1
1
2
1
2
2
1
4
1
2
2
4
1
1
1
APPLICATION DIAGRAM
UNITRODE CORPORATION
7 CONTINENTAL BLVD. • MERRIMACK, NH 03054
TEL. 603-424-2410 • FAX 603-424-3460
12