LEM LAX100-NP

Current Transducer LAX SERIES
IPN = 16 - 100 A
Ref: LAX 100-NP
For the electronic measurement of currents: DC, AC, pulsed .... , with a galvanic isolation
between the primary circuit (high power) and the secondary circuit (electronic circuit).
Maxi. 16.5 mm
Features
Applications
• Closed loop (compensated) current transducer using the
•
•
•
•
•
•
•
•
•
•
hall effect
Printed circuit board mounting
Isolated plastic case recognized according to UL 94-V0
Multirange with a single device: 16, 25, 33, 50 and 100 A rms
3 independent primary jumpers.
AC variable speed drives and servo motor drives
Static converters for DC motor drives
Battery supplied applications
Uninterruptible Power Supplies (UPS)
Switched Mode Power Supplies (SMPS)
Power supplies for welding applications.
Advantages
Standards
•
•
•
•
•
•
•
•
•
• EN 50178
• UL508 - UR marking
• IEC 61010-1-safety.
Excellent accuracy
Very good linearity
Low temperature drift
Optimized response time
Wide frequency bandwidth
No insertion losses
High immunity to external interference
Current overload capability
Height less than 16.5 mm for a simplified integration with
power modules
• Low primary inductance.
Application Domain
• Industrial.
Page 1/12
070213/7
LEM reserves the right to carry out modifications on its transducers, in order to improve them, without prior notice.
www.lem.com
LAX series 16-100 A
Electrical data
At TA = 25°C, VC = ± 15 V and RM = 50 Ω, NP = 1 turn, high speed PCB design (see page 11), unless otherwise noted.
Parameters with a * in the conditions column apply over the - 40°C.. 85°C ambient temperature range.
Parameter
Symbol Unit
Primary nominal current rms
IPN
At
Primary current, measuring range
IPM
At
Measuring resistance
RM
Ω
Secondary nominal current rms
ISN
mA
Supply voltage
VC
V
Current consumption
IC
mA
Electrical offset current
IOE
µA
Magnetic offset current
Temperature variation of IO
IOM
IOT
µA
G
Primary turns
NP
Sensitivity error
εG
%
εL
%
of
range
Overall accuracy
Reaction time
di/dt accurately followed
Output current noise
Secondary coil resistance
XG
Typ
± 160
*
49
51
*
0
51
*
0
15
*
± 11.4
-150
VC = ± 15 V ± 5%
VC = ± 15 V ± 5%, ± IPM measuring range,
DC primary current
VC = ± 15 V ± 5%, ± IPM measuring range,
AC primary current
VC = ± 12 V ± 5%, ± IPM measuring range,
DC or AC primary current
at 100 At
50
± 15.75
10 + IS
12 + IS
0
+150
31
65
after a cycle to 50 A
41
70
after a cycle to 100 A
100
230
after a cycle to 300 A
120
250
after a cycle to 500 A
-200
200
-330
330
-500
500
mA/At
VC = full range
%
*
- 40°C .. + 85°C, VC = full range
3
-0.47
0.47
± 100 A range
0.1
0.28
± 50 A range
0.06
0.14
± 100 A range
0.11
0.32
± 100 A range, low speed 1 turn PCB design
-0.91
0.91
di/dt
A/µs
> 100
Ω
- 25°C .. + 85°C, VC = full range
1
0.1
µArms
0°C .. + 70°C, VC = full range
0.5
µs
RS
Apply derating according to fig. 2
100
tra
Ino
Conditions
Maxi
µA
Sensitivity
Linearity error
Mini
= IOE +
εG + εL
high and low speed PCB designs,
dIP/dt = 100 A/µs
0.17
0.1 Hz < f < 49 Hz, IP = 0
0.17
51 Hz < f < 1 kHz, IP = 0
0.11
1 kHz < f < 100 kHz, IP = 0
60
Page 2/12
070213/7
LEM reserves the right to carry out modifications on its transducers, in order to improve them, without prior notice.
www.lem.com
LAX series 16-100 A
Electrical data (continued)
Low speed
PCB design
Symbol
Unit
NP
High speed PCB
design
1 turn
0.1
12
tr
µs
2 turns
0.1
0.2
3 turns
0.1
0.2
> 300
12.2
IP = 25 A, -1 dB
> 300
146
IP = 25 A, -3 dB
> 300
60
IP = 12.5 A, -1 dB
> 300
> 300
IP = 12.5 A, -3 dB
> 300
50
IP = 8.3 A, -1 dB
> 300
> 300
IP = 8.3 A, -3 dB
Response time to 90 % of IPN step (typical)
Frequency bandwidth (typical)
Frequency bandwidth (typical)
Frequency bandwidth (typical)
BW
BW
BW
kHz
kHz
kHz
Conditions
dIP/dt = 100 A/µs
1 turn
2 turns
3 turns
Absolute maximum ratings
Prim a ry AC current rm s (3 prim a ry jumpers
in parallel)
Prim a ry continuous direct current (3 primary jumpers
in parallel)
Symbol
Unit
Conditions
IPN
A
100 A up to T A = 70°C.
Linear derating to 70 A at 85°C.
See figure 2
IPN DC
A
100 A up to T A = 50°C.
Linear derating to 50 A at 85°C.
See figure 2
A
above value divided by 3
VC
V
± 20
RM mini
Ω
See m e a s u ring res istance in "Electrical
data" table
°C
100
Prim a ry DC or rm s current (each jumper)
Maxim u m s u p p l y voltage (not operating)
Minimum measuring resistance
Maxim u m b u s b a r temperature (jumper)
Ambient operating tem p e rature
TA
°C
- 40 .. + 85
Ambient s torage temperature
TS
°C
- 40 .. + 90
Stresses above these ratings may cause permanent damage. Exposure to absolute maximum ratings for extended
periods may degrade reliability.
Page 3/12
070213/7
LEM reserves the right to carry out modifications on its transducers, in order to improve them, without prior notice.
www.lem.com
LAX series 16-100 A
Isolation characteristics
Symbol
Unit
Value
Rms voltage for AC isolation test, 50 Hz, 1 min, between
primary and secondary
Vd
kV
3.5
Impulse withstand voltage 1.2/50 µs
Vw
kV
8
Partial discharge extinction voltage rms @ 10pC
Ve
kV
> 1.3
Creepage distance
dCp
mm
8
Clearance distance
dCI
mm
8
Comparative Tracking Index
CTI
V
600
Isolation application example
The transducer can be used according to EN 50178 and IEC 61010-1 standards under following conditions
(for example):
•
•
•
•
•
Rated isolation voltage: 600 V
Reinforced isolation
Over voltage category OV III
Pollution degree PD2
Non-uniform field
The creepage distance and clearance of the transducer mounted on a PCB are greater than 8 mm only if the primary
circuit tracks stay out of the shaded area shown below:
Figure 1: zone not permitted for primary tracks (to guarantee rated creepage and clearance)
Page 4/12
070213/7
LEM reserves the right to carry out modifications on its transducers, in order to improve them, without prior notice.
www.lem.com
LAX series 16-100 A
Typical performance characteristics
Electrical offset drift (uA)
Maximum primary rms
current (At)
120
100
80
60
AC
40
DC
20
600
Maxi
400
Mini
200
0
-200 -50
-25
-400
0
25
50
75
100
VC = ± 11.4 to ±15.75 V
-600
0
-50
-25
0
25
50
75
Ambient temperature (°C)
100
Ambient temperature (°C)
Figure 3: Electrical offset drift
Maximum measuring resistance
(Ohm)
Minimum measuring resistance
(DC current only) (Ohm)
Figure 2: Current derating
60
50
TA = -40 .. 85 °C
40
30
20
10
0
11.5
12.5
13.5
14.5
250
Vc = ±15 V ±5%
Vc = ±12 V ±5%
200
150
100
50
TA = -40 .. 85 °C
0
0
15.5
50
100
150
Measuring range (At)
Maximum power supply Vc (V)
Figure 4: Minimum measuring resistance (DC)
Figure 5: Maximum measuring resistance
0.8
11
Typical consumption current
(mA)
Maximum linearity error
(% of measuring range)
200
Np = 1, Low speed PCB
Np = 1, High speed PCB
0.6
0.4
0.2
TA = 25 °C
0.0
0
50
100
Measuring range (A)
Figure 6: Linearity error
150
10
9
8
TA = 25 °C, IP = 0
7
11
12
13
14
15
16
Supply voltage (± V)
Figure 7: Consumption current
Page 5/12
070213/7
LEM reserves the right to carry out modifications on its transducers, in order to improve them, without prior notice.
www.lem.com
LAX series 16-100 A
Typical performance characteristics (continued)
Linearity error (% of 100 A)
0.05
0
-50
0
50
100
-0.05
-0.1
0.1
0.05
0
-100
-50 -0.05 0
100
-0.1
-0.15
Primary current (A)
Primary current (A)
Figure 8: Typical linearity error for
high speed 1 turn PCB design
Figure 9:
Typical linearity error for
low speed 1 turn PCB design
0
6
20
0
-5
4
0
-1
-10
Gain
-2
-15
Phase
Gain (dB)
1
Phase (°)
Gain (dB)
50
Gain
2
-20
Phase
0
-40
IP = 25 A
-3
-4
100
1000
10000
-20
-2
-25
-4
100000 1000000
-60
-80
10000
100000
1000000
Frequency (Hz)
Figure 11: Typical frequency response
(high speed 1 turn PCB design)
(IP = 1.6 A)
Frequency (Hz)
Figure 10: Typical frequency response
(high speed 1 turn PCB design)
(IP = 25 A)
0
6
20
0
-5
4
0
-1
-10
Gain
-2
-15
Phase
-3
Gain (dB)
1
Phase (°)
Gain (dB)
IP = 1.6 A
2
Gain
100
1000
Phase
-20
-2
-25
-4
10000
100000 1000000
Frequency (Hz)
Figure 12: Typical frequency response
(low speed 1 turn PCB design)
(IP = 25 A)
-20
0
-40
-60
IP = 1.6 A
IP = 25 A
-4
Phase (°)
-100
0.15
Phase (°)
Linearity error (% of 100 A)
0.1
-80
10000
100000
Frequency (Hz)
1000000
Figure 13: Typical frequency response
(low speed 1 turn PCB design)
(IP = 1.6 A)
Page 6/12
070213/7
LEM reserves the right to carry out modifications on its transducers, in order to improve them, without prior notice.
www.lem.com
LAX series 16-100 A
Typical performance characteristics (continued)
Output (10 mA/div)
Input (20 A/div)
200 ns/div
Figure 14: Typical di/dt follow-up
(high speed 1 turn PCB design)
Input (20 A/div)
Input (20 A/div)
Output (10 mA/div)
Output (10 mA/div)
200 ns/div
Figure 15: Typical di/dt follow-up
(low speed 1 turn PCB design)
10 µs/div
Figure 16: Typical di/dt follow-up
(low speed 1 turn PCB design)
Page 7/12
070213/7
LEM reserves the right to carry out modifications on its transducers, in order to improve them, without prior notice.
www.lem.com
LAX series 16-100 A
Performance parameters definition
The schematic used to measure all electrical parameters is:
(C1 = C2 = 100 nF, RM = 50 Ω unless otherwise noted):
+VC
+
IP
M
IS
This model is valid for primary ampere-turns ΘP between
-ΘPmaxi and +ΘPmaxi only.
At zero input current, the model for the offset is reduced to:
C1
RM
IS = IOE + IOT(TA) + IOM(ΘPmaxi)
0V
In which IOM(ΘPmaxi) is the magnetic offset current due to the
maximum input ampere-turns that have been applied to the
transducer.
C2
-
-VC
Sensitivity and linearity
Figure 17 : standard characterization schematics
Ampere-turns and amperes
The LAX transducer is sensitive to the primary current linkage
ΘP (also called ampere-turns).
ΘP = NPIP (At)
With NP the number of primary turn (1, 2 or 3
depending on the connection of the primary jumpers)
Warning : As most LAX user will use it with only one single
primary turn (NP = 1), most of this datasheet is written with
primary currents instead of current linkages. The unit is kept as
ampere-turn (At) to make clear that ampere-turns are meant.
Transducer simplified model
To measure sensitivity and linearity, the primary current (DC) is
cycled from 0 to IP, then to -IP and back to 0 (equally spaced IP/
10 steps).
The sensitivity G is defined as the slope of the linear regression
line for a cycle between ± IPN.
The linearity error εL is the maximum positive or negative
difference between the measured points and the linear regression
line, expressed in % of the maximum measured value.
Magnetic offset
The magnetic offset current IOM is the consequence of a current
on the primary side ("memory effect" of the transducer's ferromagnetic parts). It is included in the linearity figure but can be
measured individually.
It is measured using the following primary current cycle.
IOM depends on the current value IP1.
I
The static model of the transducer at temperature TA is:
IS = G ΘP + error
In which
With :
error = IOE + IOT(TA) +
εΘ
G
+
P
ε (Θ
L
)ΘPmaxi
Pmaxi
OM
=
Is (t1 ) − Is (t 2 )
2
IP (DC)
IP1
ΘP = NPIP : the input ampere-turns (At)
ΘPmaxi
:
IS
TA
IOE
IOT(TA)
:
:
:
:
G
:
εG
:
εL (ΘPmaxi) :
Please read above warning.
the maxi input ampere-turns that have
been applied to the transducer (At)
the secondary current (A)
the ambient temperature (°C)
the electrical offset current (A)
the temperature variation of I at
O
temperature TA (A)
the sensitivity of the transducer (A/At)
the sensitivity error
the linearity error for ΘPmaxi
t2
0A
-IP1
t1
t
t
Ip(3)
Figure 18: current cycle used to measure magnetic and
electrical offset (transducer supplied)
Page 8/12
070213/7
LEM reserves the right to carry out modifications on its transducers, in order to improve them, without prior notice.
www.lem.com
LAX series 16-100 A
Performance parameters definition (continued)
Electrical offset
Response and reaction times
The electrical offset current IOE can either be measured
when the ferro-magnetic parts of the transducer are:
The response time tr and the reaction time tra are shown in the
next figure.
Both depend on the primary current di/dt. They are measured at
nominal ampere-turns.
The "di/dt accurately followed" mentionned in the electrical data
table is defined as the di/dt of the primary current for which the
response time is equal to 1 µs.
• completely demagnetized, which is difficult to realize, or
• in a known magnetization state, like in the current cycle
shown above.
Using the current cycle shown in figure 18, the electrical
offset is:
Is (t1 ) + Is (t 2 )
I
=
OE
2
The temperature variation I OT of the electrical offset current
IOE is the variation of the electrical offset from 25°C to the
considered temperature:
I
OT
(T ) = I OE (T ) − I OE (25°C )
I
1 0 0 %
9 0 %
Ip
Is
tr
1 0 %
tra
Note: the transducer has to be demagnetized prior to the
application of the current cycle (for example with a
demagnetization tunnel).
t
Figure 19: response time tr and reaction time tra
Overall accuracy
The overall accuracy at 25°C XG is the error in the - IPN .. + IPN
range, relative to the rated value IPN.
It includes:
• the electrical offset IOE
• the sensitivity error εG
• the linearity error εL (to IPN)
The magnetic offset is part of the overall accuracy. It is
taken into account in the linearity error figure provided the
transducer has not been magnetized by a current higher
than IPN.
Page 9/12
070213/7
LEM reserves the right to carry out modifications on its transducers, in order to improve them, without prior notice.
www.lem.com
LAX series 16-100 A
Application data
The LAX 100-NP has been designed to be used at nominal currents from 16 to 100 A. The 3 primary jumpers allow the
adaptation of the number of primary turns NP to the application so as to achieve the best compromise between nominal
current, measuring range and secondary current:
Primary
nominal
current rms
IPN (A)
Number of
primary turns
NP
Secondary
nominal
current rms
ISN (mA)
Primary
current,
measuring
range
IPM (A)
Primary
coil
resistance
@ 20°C
RP (µΩ)
Primary
insertion
inductance
LP (nH)
100
1
50
160
90
15
IN
OUT
50
1
25
160
90
15
IN
OUT
33.3
2
33.3
80
400
60
IN
OUT
33.3
3
50
53
800
136
IN
OUT
25
2
25
75
400
60
IN
OUT
16.7
3
25
50
800
136
IN
OUT
Connections (see PCB
layout )
See also the paragraph "performance parameters definition: transducer simplified model" for more details about ampereturns and output current.
High and low speed PCB designs
The PCB design is very important to achieve good linearity and frequency response with the LAX 100-NP.
High speed designs are best for accuracy, high frequency response and low response times.
High speed design
IP
Low speed design
IP
Page 10/12
070213/7
LEM reserves the right to carry out modifications on its transducers, in order to improve them, without prior notice.
www.lem.com
LAX series 16-100 A
High and low speed PCB designs
High speed PCB designs (recommended)
Low speed PCB designs
High speed 1 turn
Low speed 1 turn
High speed 2 turns
Low speed 2 turns
High speed 3 turns
Low speed 3 turns
Page 11/12
070213/7
LEM reserves the right to carry out modifications on its transducers, in order to improve them, without prior notice.
www.lem.com
Dimensions LAX 100-NP (in mm. General linear tolerance ± 0.15 mm)
Assembly on PCB
Safety
• Recommended PCB hole diameter 2 mm (0/+0.1) for primary pins
1 mm (0/+0.1) for secondary pins
• Maximum PCB thickness
2.4 mm
This transducer must be used in electric/electronic equipment with respect
• Solder temperature
maximum 270°C for 15 s (wave
to applicable standards and safety requirements in accordance with the
soldering)
following manufacturer's operating instructions.
• No-clean process only
Remarks
• IS is positive (sourcing) when IP flows in the direction of the arrow
• Mass: 20 g
Caution, risk of electrical shock
When operating the transducer, certain parts of the module can carry
hazardous voltage (eg. primary busbar, power supply).
Ignoring this warning can lead to injury and/or cause serious damage.
This transducer is a built-in device, whose conducting parts must be
inaccessible after installation.
A protective housing or additional shield could be used.
Main supply must be able to be disconnected.
Page 12/12
070213/7
LEM reserves the right to carry out modifications on its transducers, in order to improve them, without prior notice.
www.lem.com