AN445: Si4431 RF Performance and ETSI Compliance Test Results

AN445
Si4431 RF P E R F OR MA N C E
AND
E T S I C O MP L I A N C E TE S T R ESULTS
1. Introduction
The purpose of this application note is to provide measurement results and ETSI compliance results for the
Si4431B when operated from 863 to 870 MHz. All tests are performed using an ISM-DK3 kit with a with a 4431-TB1-B-868 (or DKDB1) TX/RX Split RF Test Card. The Wireless Development Suite (WDS) is used to control the
test card. The results can be duplicated by using the same configuration and scripts available on the Silicon Labs
website and referenced in the EZRadioPRO® Quick Start Guide. All settings were used directly from the Excel
Register Calculator worksheet provided on the Silicon Labs web site. For measurement results with different RF
parameters, contact customer support at: https://www.silabs.com/support/pages/contacttechnicalsupport.aspx and
register to submit a technical support request.
2. Relevant Measurements to Comply with ETSI
In this report the rules of the 300220001v020301 ETSI document are applied.
ETSI compliance of the transmitter in the EU pertains to TX frequency error (subclause 7.1), output power
(subclause 7.2 and 7.3), emissions and harmonics (subclause 7.8), and transient power (subclause 7.5). When
operating under the sub-bands (g1, g2, g3, g4, alarm, etc.), the modulation bandwidth limit (subclause 7.7) is also
applied. In narrowband channelized applications, the adjacent channel power (ACP, subclause 7.6) should also be
measured.
In RX mode, the Si4431 is an ETSI Class 2 receiver. For Class 2 compliance, the following measurements must be
checked: blocking (subclause 8.4), sensitivity (subclause 8.1), receiver spurious radiation (subclause 8.6). The
Si4431 does not comply with ETSI Class 1 requirements.
The compliance with each of the above ETSI measurement points are summarized in Table 1.
Table 1. ETSI measurement Point Compliance
EN 300220
Spec Number
ETSI Limit
VDD=3.3 V, Si4431
complies:
VDD=3.3 V, Si4432
complies:
Notes
TX frequency error
(sub-clause 7.1)
For narrowband
channelized systems:
±12 kHz for F≤500 MHz
±20 kHz for F>500 MHz
For other systems:
±100 ppm
At all power levels
At all power levels
Depends on the
crystal selection
+14 dBm (max)
in g1 sub-band
+27 dBm (max)
in g3 sub-band
At all power levels in
g3 and g1 sub-bands
At all power levels
in g3 sub-band
Several steps below
max power in
g1 sub-band
See "4.2. TX Frequency Error (ETSI
300220001v020301,
Subclause 7.1)" on
page 11 for details.
Output power
conducted
(sub-clause 7.2)
See "4.1. Output
Power (Average
Power, Conducted,
ETSI
300220001v020301,
Subclause 7.2)" on
page 9 for details.
Rev. 0.3 4/11
Copyright © 2011 by Silicon Laboratories
AN445
AN445
Table 1. ETSI measurement Point Compliance (Continued)
EN 300220
Spec Number
Effective radiated
power
(sub-clause 7.3)
See "4.3. Effective
Radiated Power
(ETSI
300220001v020301,
Subclause 7.3)" on
page 11 for details.
Transient power
(sub-clause 7.5)
See "4.4. Transient
Power (ETSI
300220001v020301,
Subclause 7.5)" on
page 12 for details.
Adjacent channel
power
(sub-clause 7.6)
See "4.5. Adjacent
Channel Power
(ACP) (ETSI
300220001v020301,
subclause 7.6)" on
page 14 for details.
Modulation
bandwidth
(sub-clause 7.7)
ETSI Limit
VDD=3.3 V, Si4431
complies:
VDD=3.3 V, Si4432
complies:
+14 dBm ERP (max)
in g1 sub-band
+27 dBm ERP (max)
in g3 sub-band
At all power levels
in g3 sub-band
At all power levels
in g1 sub-band,
but limited to
<2.4 dB antenna
gain
At all power levels
in g3 sub-band
Several steps below
max power in
g1 sub-band
Spurious emission limits
(sub-clause 7.8) or less
than 3 dB difference compared to CW operation
At all power levels
At all power levels
200 nW with channel
spacing ≥20 kHz
10 µW with channel
spacing <20 kHz
With default settings
the integrated output
power (16 kHz bandwidth) limited to
+5 dBm when
25 kHz channel
spacing is used.
Integrated output
power (8.5 kHz
bandwidth) limited to
+15 dBm when
12.5 kHz channel
spacing is used.
Applicable to
With default settings
narrowband
the integrated output
power (16 kHz band- channelized systems only.
width) limited to
+5 dBm when
The case of 25
25 kHz channel
and 12.5 kHz
spacing is used.
spacing is
checked.
Integrated output
power (8.5 kHz
With VCO curbandwidth) limited to rent boost and
+15 dBm when
PLL bandwith
12.5 kHz channel
optimization
spacing is used.
some ACP performance
improvement
can be achieved
(see section 4.5
for details)
Special spectrum mask
(refer to Figure 22
in Section 4.6)
At all power levels
The Si4432 comApplicable in the
plies at reduced
sub-bands
power levels
(~14dBm) with
default register settings. By setting the
Ox58, 0x59 and
0x5A register values
to 0x40, 0x80 and
0x81 respectively,
compliance can be
achieved with full
power.
See "4.6. Modulation Bandwidth (ETSI
300220001v020301,
Subclause 7.7)" on
page 22 for details.
2
Rev. 0.3
Notes
The 3 dB difference between
CW and
switched operation is checked.
AN445
Table 1. ETSI measurement Point Compliance (Continued)
EN 300220
Spec Number
Unwanted
emissions in the
spurious domain
(sub-clause 7.8)
ETSI Limit
VDD=3.3 V, Si4431
complies:
VDD=3.3 V, Si4432
complies:
Special spectrum mask
(refer to Figure 26 in Section 4.7)
At all power levels
At all power levels
Sp[dBm] =
10log(BW[kHz]/16)-107
Complies easily,
with large margin
Complies easily,
with large margin
–69 dBm at ±2 MHz offset
–44 dBm at ±10 MHz offset
Complies easily,
with large margin
Complies easily,
with large margin
Spurious emission limits
(subclause 7.8)
Complies easily,
with large margin
Complies easily,
with large margin
Notes
See sections 4.7,
4.8, and 4.9 for
details.
Receiver sensitivity
(sub-clause 8.1)
See "3.1. Sensitivity
(ETSI
300220001v020301,
Subclause 8.1)" on
page 4 for details.
Receiver blocking
(sub-clause 8.4)
See "3.2. Blocking
(ETSI
300220001v020301,
Subclause 8.4)" on
page 4 for details.
Receiver spurious
radiation
(sub-clause 8.6)
Limits valid for
Class 2 receivers.
EZRadioPRO
devices do not
comply with
Class 1 limits
See "3.3. Receiver
Spurious Radiation
(ETSI
300220001v020301,
Subclause 8.6)" on
page 5 for details.
Rev. 0.3
3
AN445
3. RX Measurement Results
3.1. Sensitivity (ETSI 300220001v020301, Subclause 8.1)
The ETSI standard requires –107 dBm sensitivity with a 16 kHz RX bandwidth at 20% PER. If the RX bandwidth is
not 16 kHz, the sensitivity limit is modified according to the following formula:
BW[kHz]
Sp  dBm  = 10 log  ----------------------- – 107
16
Equation 1.
For this test, a much more strict 1% PER requirement is used as this more closely reflects a link that may be
considered usable for most applications. Figure 1 demonstrates the ETSI 20% PER sensitivity limit in red and the
Si4431 1% PER sensitivity over temperature. The sensitivity limit changes for datarates higher than 8 kbps
because the 16 kHz bandwidth can no longer be maintained. A modulation index of H = 1 is used for all cases with
no additional allocation in the IF bandwidth for XTAL tolerance. Even with the more stringent 1% PER
measurements, the Si4431 sensitivity complies with subclause 8.1 with a large margin over the –40 to 85 °C
temperature range.
Rx sensitivity with 1% PER, 3V
1% PER sens. [dBm]
-100
-105
-41
-5
25
55
-110
85
ETSI LIMIT (20% PER)
-115
-120
1
10
100
Data rate [kbps]
Figure 1. Sensitivity vs. Datarate, PER 1%
3.2. Blocking (ETSI 300220001v020301, Subclause 8.4)
Blocking is a measure of the capability to receive a desired modulated signal in the presence of a strong undesired
blocking signal. The unwanted input signal may be present at any frequency other than the spurious responses. All
selectivity results are measured by positioning the input power 3 dB above the ETSI sensitivity limit given in
subclause 8.1.4 (see Equation 1 above) with the primary signal source generator. A second generator with an
unmodulated signal is used as the interferer and combined with the primary signal using a power combiner such as
Mini-Circuits ZFSC-2-4-S+. The second interferer generator is placed at the desired frequency offset and the
power is increased until the PER degrades to 20%. For example, the +2 MHz point in Figure 2 shows –35.4 dBm
spurious signal level. It means that if the spurious level is lower than this the PER will surely lower than 20%.
The utilized data rate and deviation in the measurements were respectively 100 kbps and 50 kHz (H=1). The
appropriate Si4431 RX bandwidth to receive this signal is 208 kHz. The primary generator is set to –92.8 dBm,
which is 3 dB above the ETSI RX sensitivity limit for this bandwidth.
ETSI specifies the blocking limits in absolute values at two points: ±2 and ±10 MHz. The limit for Class 2 receivers
at ±2 MHz is –69 dBm, and at ±10 MHz it is –44 dBm.
4
Rev. 0.3
AN445
The blocking performance of the Si4431 is shown in Figure 2 together with the ETSI limits for class 2 receivers
(red lines). The Si4431 is fully compliant with the ETSI Class 2 receiver blocking requirements with large margin.
The peak at ~ –1.8 MHz offset is the rejection at the receiver's image frequency. This spurious response is not part
of the blocking measurements.
Selectivity over temperature and VDD
Carrier level [dBm]
9
9
9
(76,OLPLW
NESV>G%F@
CW interferer offset [MHz]
Figure 2. 100 kbps Blocking
(Datarate = 100 kpbs, Deviation = 50 kHz, No Allocation in BW for XTAL Tolerance)
3.3. Receiver Spurious Radiation (ETSI 300220001v020301, Subclause 8.6)
If the device under test (DUT) uses an RF connector, then the spurious power delivered to a 50  termination in
RX mode is measured. If the DUT does not use an RF connector and uses an integrated or dedicated antenna,
then the spurious radiation in RX mode is measured.
The measurement frequency range in the first step is between 9 kHz and 4 GHz. If a spurious signal approaches
the limit by less than 10 dB in the 1.5 to 4 GHz range, then the whole procedure needs to be repeated in the 4 to
12.75 GHz range as well.
Figure 3 shows the conducted leakage at the RX connector in RX mode up to 1 GHz. This complies with the ETSI
limit of –57 dBm ERP up to 1 GHz (corresponding to 40.3 dBµV/m generated rms E field strength at 3 m distance).
Figure 4 shows the spurious leakage with a monopole antenna at the RX connector in RX mode from 1 GHz up to
7 GHz. The measurement is provided by Elliot Labs, a certified Electromagnetic Compatibility (EMC) test house.
The measurement of –47 dBm ERP above 1 GHz is ETSI compliant but with less than 10 dB margin therefore,
measurements up to 12.75 GHz are required and shown in Figure 5.
Rev. 0.3
5
AN445
According to Figures 3, 4, and 5, the leakage in RX mode is significantly lower than the ETSI limits.
Figure 3. Receiver Leakage in RX Mode (Conducted up to 1 GHz)
Figure 4. Receiver Leakage in RX Mode, Radiated with Monopole Antenna
(Measured at Elliot Labs)
6
Rev. 0.3
AN445
Figure 5. Receiver Leakage in RX Mode, Radiated with Monopole Antenna
(Measured at Elliot Labs)
3.4. Sensitivity vs. Frequency Offset
The Sensitivity vs. Frequency Offset measurements are referred to as bucket curves due to their shape. This test is
not required by ETSI, but it is of practical importance. The receiver and generator frequencies are calibrated for
zero frequency offset. The signal generator is then offset and the sensitivity measurement is repeated. The
performance for this parameter is directly linked to the programmed receive bandwidth. With AFC disabled, there is
still internal modem compensation, which corrects for offsets up to 0.25 x RX BW, and, with AFC enabled, offsets
can be corrected up to 0.35 x BW. The modem parameters and receiver bandwidths were set using the Excel
Register Calculator worksheet assuming a 20 ppm crystal tolerance on both the transmitter and receiver as shown
below.
Figure 6. Modem Parameters and Receiver Bandwidths
Rev. 0.3
7
AN445
Figure 7 demonstrates the performance for DR = 100 kbps, Dev = 50 kHz, and a BW of 208.4 kHz. For the AFC
disabled (modem compensation only) case, the frequency offset compensation can be calculated as 0.25 x
208.4 = 52.1 kHz, which may be observed in Figure 7. For the AFC enabled case, the frequency offset
compensation can be calculated as 0.35 x 208.4 = 73 kHz. If more frequency offset compensation is desired, the
RX bandwidth should be increased to a value greater than that shown in the Excel Register Calculator worksheet.
This may easily be done by artificially increasing the value used for XTAL Tolerance.
Sensitivity vs Frequency Offset - DR=100kbps, BW=208.4kHz
-60
Sensitivity (dBm)
-70
-80
A FCOFF
A FCON
-90
-100
-110
-100
-80
-60
-40
- 20
0
20
40
60
80
100
Frequency Offs et (k Hz)
Figure 7. Selectivity vs. Frequency Offset (100 kbps)
Sensitvity vs Frequency Offset - 40kbps, BW=95.3kHz
-60
Sensitivity (dBm)
-70
-80
A FCON
A FCOFF
-90
-100
-110
-50
-40
-30
- 20
-10
0
10
20
30
40
Fre que nc Offse t (kHz)
Figure 8. Selectivity vs. Frequency Offset (40 kbps)
8
Rev. 0.3
50
AN445
4. TX Measurements
Unless otherwise specified, all TX measurements are measured with VDD = 3.3 V, txpow[2:0] = 111,
frequency = 868 MHz using the scripts available on the Silicon Labs website and a TX/RX split test card. All
measurements are done at the output of the low-pass filter at the TX SMA connector. The output power is ~1 dB
higher at the TX pin of the chip before the low pass filter.
4.1. Output Power (Average Power, Conducted, ETSI 300220001v020301, Subclause 7.2)
The TX output spectrum in power state 0F with the proposed TX matching of the 4431-T-B1 B 868 testcard is
shown in Figure 9 at 868 MHz. The power delivered to the antenna is 13.7 dBm in power state 0F.
The ETSI sub/bands and limits are shown in Figure 10. The limit in the g1 sub-band is +14 dBm effective radiated
power (ERP) (corresponding to 16.1 dBm effective isotropic radiated power or EIRP), which allows for an antenna
gain of 2.4 dB in power state 0F.
Figure 9. Output Power at Power State 0F at 3.3 V
Rev. 0.3
9
AN445
27dB m with high speed mo dulatio n,
o therwise 25kHz channels
40
30
20
g (860) band
g1
14dB m
10
0
-10
g2
-20
g3
-36dB m
-30
g4
-40
-50
-54dB m
-60
860
862
864
866
868
870
872
868 - 868.6
869.7 - 870
868.7 - 869.2
869.4 - 869.65
Figure 10. ETSI Sub-Bands and Limits at the 868 MHz Band
The conducted power and total current consumption at different power settings and supply powers are listed in
Table 2.
Table 2. Conducted Power and Total Current Consumption at Different Power Settings and
Supply Powers
Power
TX Output
Total Current
Setting
Fund Power Consumption
Register 6Dh [dBm], 3.6 V
[mA], 3.6 V
TX Output
Fund Power
[dBm], 3.3 V
Total Current
Consumption
[mA], 3.3 V
TX Output
Fund Power
[dBm], 1.8 V
Total Current
Consumption
[mA], 1.8 V
8
–4
14.4
–4.1
14.3
–4.7
13.9
9
–2
14.9
–2.1
14.8
–2.8
14.4
0A
2.6
17.1
2.5
17
1.7
16.4
0B
5.8
19.5
5.7
19.4
4.6
18.6
0C
8
21.8
7.9
21.7
6.2
20.3
0D
11
25.9
10.9
25.7
7.5
22.4
0E
13.2
30.6
12.8
29.8
8.2
24.4
0F
14.4
35.3
13.7
34.1
8.8
26.9
10
Rev. 0.3
AN445
4.2. TX Frequency Error (ETSI 300220001v020301, Subclause 7.1)
The frequency error basically depends on the crystal spreading. A typical value is 20 ppm; so, the ±100 ppm ETSI
limit for non-channelized devices is usually satisfied without any difficulty.
The internal capacitance bank of the Si4431 is used to tune the load capacitance of the applied crystal before
measurement.
In the case of narrowband channelized applications, the required frequency accuracy is more strict and depends
on the channel bandwidth. In this application, a more accurate crystal is necessary.
4.3. Effective Radiated Power (ETSI 300220001v020301, Subclause 7.3)
The effective radiated power at the fundamental frequency depends on the power delivered to the antenna and the
antenna gain. The conducted power at power state 0F was already shown in "4.1. Output Power (Average Power,
Conducted, ETSI 300220001v020301, Subclause 7.2)" on page 9. To comply with the g1 sub-band limit, the
antenna gain should be less than 2.4 dB at 0F power state.
Using the configuration with a monopole (shown in Figure 11) the radiated power maximum at the fundamental
frequency is +12.1 dBm EIRP at power state 0F and at 3.3 V. It is below the ETSI limit by 4 dB (14 dBm ERP or
16.1 dBm EIRP).
Figure 11. Test Card with Monopole Antenna for 868 MHz
Rev. 0.3
11
AN445
4.4. Transient Power (ETSI 300220001v020301, Subclause 7.5)
Transient power is the power falling into adjacent spectrum due to switching the transmitter on and off during
normal operation. The modulation bandwidth limits (see chapter 4.5 i.e. ETSI clause 7.7) shall be fulfilled
irrespective of the transient power limit values.
4.4.1. Measurement Procedure According to ETSI
1. The measured receiver shall use a QUASI-PEAK DETECTOR defined in CISPR 16-1-1.
2. RBW should be 120 kHz.
3. Special narrowband and wideband settings:
a.
For narrowband equipment (channel spacing <25 kHz), the measurement shall be started at 60 kHz offset from the
beginning of the upper adjacent channel.
b.
For wide-band devices, the initial offset shall be 100 kHz from the modulation bandwidth edge.
c.
The measurement shall be repeated in 120 kHz steps, up to 2 MHz offset.
d.
The measurement shall be done on both sides of the carrier.
4. The measurement shall be repeated in 120 kHz steps, up to 2 MHz offset.
5. The measurement shall be done on both sides of the carrier.
6. The measurement shall be done in ZERO-SPAN mode.
7. The transmitter shall be turned on and off at least 5 times
8. The power level should be recorded at least for a period of 60 seconds
9. If the resulting power level is above the spurious limit (see Section 4.5 or subclause 7.8 of the ETSI document),
then the measurement shall be repeated with continuous transmission, and the transient level in Step 6 shall
not exceed this power level by more than 3 dB. This is the so-called Step 2 of this measurement according to
ETSI.
4.4.2. Modified Measurement Results
Slight modifications were done in the measurement method to address the worst-case scenario and to simplify the
measurement.
1. The TX_state signal was multiplexed to GPIO2 of Si4431.
2. This signal was used for triggering the spectrum analyzer (to capture PA transient, which is so fast that it could
not be captured by turning on and off the PA five times as requested in ETSI).
3. The sweep time in ZERO-SPAN mode was set to 5 ms, number of points 1001 (it does not make sense to
measure 60 seconds as requested in ETSI due to the fast transient response).
4. The ZERO-SPAN measurement was repeated in 50 kHz frequency steps from –2 MHz to 2 MHz offset around
the carrier.
5. The carrier was set to 868.3 MHz, Si4431 XTAL was tuned to 0 ppm offset.
6. The same measurement was repeated with the max peak detector as well.
4.4.3. Transient Power Results
1. The quasi-peak detector has 1 ms charge time by definition; therefore, it has a huge suppression (about 12 dB
according to CISPR 16) for narrow pulses like a few microseconds in our case.
2. Because of its slow response, the quasi-peak detector cannot detect VCO pulling in our case. Effectively, there
is no difference in the results between the transient signal and the continuous transmission, by using the quasipeak detector for the measurement (the result was the phase noise curve in both cases (see the magenta
curves in Figure 12), which means the 3 dB criteria of ETSI test Step 2 is always met. The difference between
the continuous and on-off switched PA case can only be seen with the max peak detector (in continuous mode,
the peak detector results are the same as the quasi peak detector, while, with switched PA, the pulling appears
with the peak detector (blue curve)).
3. There is no limitation in channel spacing according to these results.
12
Rev. 0.3
AN445
Transient power on Si4431 B0
20
10
0
dBm
-10
MAXPEAK
QUASI-PEAK
-20
-30
-40
-50
-60
-2000
-1500
-1000
-500
0
500
1000
1500
2000
Offset[kHz]
Figure 12. Transient Power Measurement
With the Quasi-peak detector, both the continuous and on-off measurement gives the same curve (magenta), while
with the Maxpeak detector, the VCO pulling can be observed with PA on-off.
Zoom to V CO pulling region
20
10
0
MAXPEAK
dBm
-10
QUASI-PEAK
-20
-30
-40
-50
-60
0
100
200
300
400
500
Offset [kHz ]
Figure 13. Zoomed VCO Pulling Region of Fig. 12
Rev. 0.3
13
AN445
4.5. Adjacent Channel Power (ACP) (ETSI 300220001v020301, subclause 7.6)
The adjacent channel power (ACP) is defined as the amount of the modulated RF signal power which falls within a
given adjacent channel. ACP measurements are applicable to narrowband systems. Narrowband systems are
used in a non-channelized continuous frequency band with a channel bandwidth of less than or equal to 25 kHz or
used in a channelized frequency band with a channel spacing of less than or equal to 25 kHz.
According to ETSI the narrowband systems are divided into two groups. In the first group, if the channel separation
is smaller than 20 kHz, then the ACP limit is –20 dBm in an integrated bandwidth of 8.5 kHz. In the second group,
if the channel separation is higher or equal to 20 kHz, then the limit is –37 dBm in an integrated bandwidth of
16 kHz. The script used for the ACP measurement is given in Annex A.
4.5.1. ACP Measurements with Channel Spacing Higher or Equal to 20 kHz
In this chapter the ACP measured with a 25 kHz channel spacing is investigated. For this measurement the
integrated bandwidth of the adjacent channel is 16 kHz and the ETSI limit for the ACP is 200 nW (–37 dBm).
Figure 14 illustrates the measured ACP at the 869.65–869.7 MHz alarm band with default VCO current and PLL
bandwidth settings (the corresponding lines are ignored in the script in Annex A) with a 1 kbps data rate and a
5 kHz deviation. In theory, the Si4431 can work with much lower deviation as well, but 5 kHz is a practical choice
due to crystal selection considerations. The integrated power around the carrier is +3.8 dBm in 16 kHz bandwidth
and with average detection. In CW mode and with peak detection the measured power is around +4.4 dBm. With
this power the ACP is 0.6 dB lower than the ETSI limit (–37 dBm) therefore the Si4431 complies with the ETSI ACP
limits with 25 kHz channel spacing up to +4.5–5 dBm fundamental output power with the default VCO and PLL
register settings if no margin is left. With a 3 dB margin, the allowed power is +1.5–2 dBm.
Figure 14. Adjacent Power Measurement, 25 kHz Narrowband Channel Spacing, Default Settings,
1 kbps, 5 kHz Deviation (+4.4 dBm CW Power Measured by Peak Detector)
14
Rev. 0.3
AN445
Figure 15 shows the measured ACP at the 869.65…869.7 MHz alarm band with VCO current boost (7Fh is written
into register 5A). The data rate and dev. are the same as the previous example (1 kbps, 5 kHz). The improvement
on the ACP is approximately 1.7 dB due to the VCO current boost. The allowed fundamental power to comply with
the ETSI ACP limits with 25 kHz channel spacing is approximately +6–6.5 dBm (CW, measured by peak detector)
with VCO current boost if no margin is left. With a 3 dB margin, the allowed power is approximately 3–3.5 dBm.
Figure 15. Adjacent Power Measurement with 25 kHz Narrowband Channel Spacing with VCO
Current Boost (the value of register 5A is 7Fh), 1 kbps, 5 kHz Deviation (+4.4 dBm CW Power
Measured by Peak Detector)
Rev. 0.3
15
AN445
Figure 16 shows the measured ACP at the 869.65–869.7 MHz alarm band with VCO current boost (7Fh in register
5A) and with PLL BW increase (EFh in register 58). The data rate and dev. are the same as before (1 kbps, 5 kHz).
The improvement on the ACP is approximately 3 dB due to PLL bandwidth increase. The VCO current boost and
PLL BW result together in a 5 dB improvement and therefore the maximum fundamental output power which
complies with the ETSI ACP limits with 25 kHz channel spacing is ~ +9–9.5 dBm (CW, measured by peak detector)
in this case if no margin is left. With a 3 dB margin, the allowed power is approximately 6–6.5 dBm.
Figure 16. Adjacent Power Measurement with 25 kHz Narrowband Channel Spacing with VCO
Current Boost (7Fh is in register 5A) and with PLL BW Increase (EFh in register 58), 1 kbps, 5 kHz
Deviation (+4.4 dBm, CW Power Measured by Peak Detector)
16
Rev. 0.3
AN445
4.5.2. Measurements with 12.5 kHz Channel Spacing
In this section, the ACP measured with a 12.5 kHz channel spacing is investigated. In this measurement the
integrated bandwidth at the adjacent channel is 8.5 kHz and the ETSI limit for ACP is 10 µW (–20 dBm).
The measured ACP is shown in Figure 17. The Si4431 is modulated with 2.4 kbps and the applied deviation is
2.5 kHz. The power is set to maximum (0Fh in register 5A). The Si4431 is used in default mode (no VCO boost, no
PLL BW increase). The integrated ACP power in the applied 8.5 kHz bandwidth is 6 dB below the ETSI limit,
therefore with 12.5 kHz channel spacing the Si4431 can be used at maximum power level with high margin.
Figure 17. Adjacent Power Measurement with 12.5 kHz Narrowband Channel with Default
Settings, Max Power State, 2.4 kbps, 2.5 kHz Deviation
Rev. 0.3
17
AN445
4.5.3. Measurements with 12.5 kHz Channel Spacing with 9.6 kbps and 4.8 kHz Deviation
In this section, the ACP is measured with a 12.5 kHz channel spacing and 9.6 kbps is investigated. The deviation is
4.8 kHz. In this measurement, the integrated bandwidth at the adjacent channel is 8.5 kHz and the ETSI limit for
ACP is 10 µW (–20 dBm). The operation frequency here is 868 MHz.
The measured ACP is shown in Figure 22 with default settings (no VCO boost, no PLL BW increase). The power is
set to –1.5 dBm (09h in register 5A in case of the split testcard) to comply with the ACP standard with ~3 dB
margin.
Figure 18. Adjacent Power Measurement with 12.5 kHz Narrowband Channel with Default
Settings and with 9.6 kbps and 4.8 kHz Deviation, –1.5 dBm CW Power Measured by Peak
Detector)
18
Rev. 0.3
AN445
Figure 19 shows the measured ACP with VCO current boost (7Fh in register 5A) and with PLL BW increase (EFh
in register 58) at a power of 3 dBm (state 0 A in case of split testcards). The data rate and deviation are the same
as before (9.6 kbps, 4.8 kHz). With a 3 dB margin, the allowed power is ~0 dBm (CW, measured by peak detector),
so the improvement due to the VCO current and PLL BW boost is ~1.5 dB.
Figure 19. Adjacent Power Measurement with 12.5 kHz Narrowband Channel with VCO and PLL
BW Boost and with 9.6 kbps and 4.8 kHz Deviation, CW Power is +3 dBm Measured by Peak
Detector
Rev. 0.3
19
AN445
4.5.4. Measurements with 12.5 kHz Channel Spacing with 9.6 kbps and 2.4 kHz Deviation
In this section, the ACP is measured with a 12.5 kHz channel spacing and 9.6 kbps is investigated. The deviation is
2.4 kHz. In this measurement, the integrated bandwidth at the adjacent channel is 8.5 kHz, and the ETSI limit for
ACP is 10 µW (–20 dBm). The operation frequency here is 868 MHz.
The measured ACP is shown in Figure 20 with default settings. The power is set to 11.4 dBm (0Dh in register 5A in
case of the split testcard) to comply with the ACP standard. The Si4431 is used in default mode (no VCO boost, no
PLL BW increase). With a 3 dB margin, the allowed power (CW, measured by peak detector) is ~9.5 dBm.
Figure 20. Adjacent Power Measurement with 12.5 kHz Narrowband Channel with Default
Settings and with 9.6 kbps and 2.4 kHz Deviation, CW Power is +11.4 dBm Measured by Peak
Detector
20
Rev. 0.3
AN445
Figure 21 shows the measured ACP with VCO current boost (7Fh in register 5A) and with PLL BW increase (EFh
in register 58) with a power of 13.5 dBm (power state 0E). The data rate and deviation are the same as before
(9.6 kbps, 2.4 kHz). With a 3 dB margin, the allowed power (CW, measured by peak detector) is ~10.5 dBm, so the
improvement due to VCO current and PLL BW boost is 1 dB.
Figure 21. Adjacent Power Measurement with 12.5 kHz Narrowband Channel with VCO and PLL
BW settings and with 9.6 kbps and 2.4 kHz Deviation, CW power is +13.5 dBm Measured by Peak
Detector
Rev. 0.3
21
AN445
4.6. Modulation Bandwidth (ETSI 300220001v020301, Subclause 7.7)
The range of modulation bandwidth includes all associated side bands above the appropriate emissions level
(ETSI 300220001v020301, subclause 7.8) and the frequency error or drift under extreme test conditions. The
frequency drift in extreme test conditions primarily depends on the crystal quality, which is not included in this
report.
The ETSI spectral mask with which the radio must comply at the sub-band edges is demonstrated in Figure 22.
Basically, there are only two limit thresholds, and the bandwidth of integration is varied at the different offset
regions.
This can be observed in the spectral mask measurements as well (see Figures 23 and 25), where the limit lines
mark the –30 and –36 dBm levels. As the RBW is changed according to the ETSI spectrum mask, the measured
phase noise level is changing in the –36 dBm level region.
Figure 22. ETSI Spectral Mask Measurement Limits at the Sub-Band Edges
Figure 23 demonstrates the measured spectral mask in the g3 sub-band (869.4–869.65 MHz) with default register
settings. Since the allowed output power in the g3 sub-band is 500 mW, the power level is set to maximum (power
state 0Fh, where the power in CW mode is ~14 dBm). The applied bit rate is 50 kbps with H = 1. With these
parameters, the spectral mask of 4431 complies with ETSI subclause 7.7 at the most critical G3 sub-band with
minimal margin (~1 dB) due to modulated sidebands products. Unfortunately, due to this, the power cannot be
higher at any other sub-bands as well with default register settings. It also means that with the Si4432 a significant
(~6 dB) power reduction is required.
If higher output power is desired, the Si4432 can be used with a modified register setting. By setting the Ox58,
0x59 and 0x5A register values to 0x40, 0x80 and 0x81, respectively, the sideband spurs practically can be
eliminated. This way the Si4432 is compliant with ETSI spectrum mask restrictions even with max power. Figure 24
shows the spectrum mask test result with a Si4432 having ~19.5 dBm CW power. The Si4432 complies with full
power with the above described register settings.
Figure 25 shows the measured spectrum mask in the 868.6–868.7 MHz alarm band where the maximum allowed
power is +10 dBm. To comply with this limit, the power is set to state 0Dh (Vcc = 3.3 V).
22
Rev. 0.3
AN445
The applied bit rate for this test is 19.6 kbps, H = 1. With these parameters the Si4431 complies with the ETSI
spectral mask limits up to the allowed maximum power limit (+10 dBm) in the 100 kHz wide alarm sub-band.
Figure 23. Spectral Mask Measurement in G3 Sub-Band (869.4–869.65 MHz), 3.3 V
(Power State 0Fh (max), 50 kbps, H = 1)
s
Figure 24. Si4432 Spectrum Mask Test at G3 Sub-Band; Full (19.5 dBm CW) Power, 25k Deviation,
50k Data Rate
Rev. 0.3
23
AN445
Figure 25. Spectral Mask Measurement in 100 kHz Alarm Sub-Band (868.6–868.7 MHz), 3.3 V,
(Power State 0Dh to be below the 10 dBm Fundamental Limit, 19.6 kbps, H = 1)
24
Rev. 0.3
AN445
4.7. Unwanted Emissions in Spurious Domain (ETSI 300220001v020301, Subclause 7.8)
The allowed emissions in ERP in the spurious domain are shown in Figure 26.
A critical restriction is the –54 dBm limit below 862 MHz.
0
-10
-20
-30 dB m. RB W 1M Hz, peak det.
-30
74MHz 118MHz
-36dB m. RB W 100kHz, quasi peak det.
-36dB m. RB W 100kHz,
quasi peak det.
-40
-50
-60
-54dB m. RB W 100kHz, quasi peak det.
47MHz 87.5MHz 174M Hz 230M Hz
10
210
470M Hz
410
862M Hz
610
810
1010
Figure 26. ETSI Spurious Radiation Limits in ERP
Rev. 0.3
25
AN445
4.8. Conducted Harmonics
Figure 27 shows the conducted spectrum with a 100 MHz span around the 868 MHz carrier with full power at 3.6 V
VDD. Below 862 MHz, all spurious emissions are lower than –60 dBm. Above 862 MHz the spurious emissions are
below the ETSI –36 dBm limit.
Figure 28 demonstrates the conducted full spectrum below 1 GHz. There are no spurs above the ETSI –54 dBm
limit.
Above 1 GHz, the main spur components are the harmonics. Figure 29 shows the conducted harmonics above
1 GHz up to 12.75 GHz. This measurement is done with 1 MHz RBW and max hold. No harmonics above the ETSI
limit (–30 dBm) are observed.
Peak detectors were used in all the conducted measurements.
Figure 27. Si4431 Spectrum around the Carrier
26
Rev. 0.3
AN445
Figure 28. Si4431 Spectrum below 1 GHz
02:18:53 Oct 13, 2009
Ref 15 dBm
Peak 1
Log
10
dB/
#Atten 25 dB
2
Dl
-30.0
dBm
3
Marker
2.590000000 GHz
-40.2 dBm
Start 800 MHz
#Res BW 1 MHz
Marker
1
2
3
Mkr3 2.59 GHz
-40.2 dBm
Trace
(1)
(1)
(1)
VBW 1 MHz
Type
Freq
Freq
Freq
X Axis
860 MHz
1.73 GHz
2.59 GHz
Stop 12.75 GHz
Sweep 29.88 ms (401 pts)
Amplitude
14.61 dBm
-39.2 dBm
-40.2 dBm
Figure 29. Conducted Harmonics above 1 GHz (State 0F, 3.6 V)
Rev. 0.3
27
AN445
4.9. Radiated Harmonics
ETSI compliance for radiated harmonics is shown in Figures 30 through 36. These tests were performed by Elliott
Labs an ETSI certified Electromagnetic Compatibility (EMC) test house. The test is performed with a Si4431-4431T-B1_D_868 Test Card (2 layer design for direct-tie) at the 868 MHz frequency band at an output power level of
+13 dBm (power state 07, 3.3 V). The measured harmonic levels up to the 10th harmonic in TX mode are shown
below. According to this, the two-layer design direct-tie board is fully-compliant with ETSI radiated limits up to
12.75 GHz using the monopole antenna shown in Figure 11 on page 11.
EN 300 220-1 Radiated Spurious Emissions
7HVW6SHFLILF'HWDLOV
2EMHFWLYH
7KHREMHFWLYHRIWKLVWHVWVHVVLRQLVWRSHUIRUPHQJLQHHULQJHYDOXDWLRQWHVWLQJRIWKH(87ZLWKUHVSHFWWRWKH
VSHFLILFDWLRQOLVWHGDERYH
'DWHRI7HVW 7HVW(QJLQHHU 'DYLG%DUH
7HVW/RFDWLRQ )UHPRQW&KDPEHU
&RQILJ8VHG &RQILJ&KDQJH 1RQH
(879ROWDJH 9'&
*HQHUDO7HVW&RQILJXUDWLRQ
7KH(87DQGDOOORFDOVXSSRUWHTXLSPHQWZHUHORFDWHGRQWKHWXUQWDEOHIRUUDGLDWHGVSXULRXVHPLVVLRQVWHVWLQJ
)RUUDGLDWHGHPLVVLRQVWHVWLQJWKHPHDVXUHPHQWDQWHQQDZDVORFDWHGPHWHUVIURPWKH(87
$PELHQW&RQGLWLRQV
7HPSHUDWXUH
5HO+XPLGLW\
6XPPDU\RI5HVXOWV
5XQ
0RGH
&KDQQHO
3RZHU
6HWWLQJ
0HDVXUHG
3RZHU
&:7;
0+]
0D[
ƒ&
7HVW3HUIRUPHG
/LPLW
5HVXOW0DUJLQ
5DGLDWHG(PLVVLRQV
7UDQVPLWWHU+DUPRQLFV
(1
G%—9P#
0+]G%
0RGLILFDWLRQV0DGH'XULQJ7HVWLQJ
1RPRGLILFDWLRQVZHUHPDGHWRWKH(87GXULQJWHVWLQJ
'HYLDWLRQV)URP7KH6WDQGDUG
1RGHYLDWLRQVZHUHPDGHIURPWKHUHTXLUHPHQWVRIWKHVWDQGDUG
Figure 30. EN300 220-1 Radiated Spurious Emissions Test Results (Page 1 of 7)
28
Rev. 0.3
AN445
5XQ5DGLDWHG6SXULRXV(PLVVLRQV7UDQVPLW0RGH0+]
6LB%617&:7;0+]10DWFK9'' 97;3:5/D\HU%RDUGZR6KLHOG
Figure 31. EN300 220-1 Radiated Spurious Emissions Test Results (Page 2 of 7)
Rev. 0.3
29
AN445
)UHTXHQF\
0+]
1RWH
/HYHO
G%μ9P
3RO
YK
+
9
9
9
(1
/LPLW
0DUJLQ
'HWHFWRU
3N43$YJ
3HDN
3HDN
3HDN
3HDN
$]LPXWK
GHJUHHV
+HLJKW
PHWHUV
&RPPHQWV
7KHILHOGVWUHQJWKOLPLWLQWKHWDEOHVDERYHZDVFDOFXODWHGIURPWKHHUSHLUSOLPLWGHWDLOHGLQWKHVWDQGDUGXVLQJWKHIUHHVSDFH
SURSDJDWLRQHTXDWLRQ( ¥3*G7KLVOLPLWLVFRQVHUYDWLYHLWGRHVQRWFRQVLGHUWKHSUHVHQFHRIWKHJURXQGSODQHDQG
IRUHUSOLPLWVWKHGLSROHJDLQG%LKDVQRWEHHQLQFOXGHG7KHHUSRUHLUSIRUDOOVLJQDOVZLWKOHVVWKDQG%RIPDUJLQ
UHODWLYHWRWKLVILHOGVWUHQJWKOLPLWLVGHWHUPLQHGXVLQJVXEVWLWXWLRQPHDVXUHPHQWV
Figure 32. EN300 220-1 Radiated Spurious Emissions Test Results (Page 3 of 7)
EN 300 220-1 Radiated Spurious Emissions
7HVW6SHFLILF'HWDLOV
2EMHFWLYH
7KHREMHFWLYHRIWKLVWHVWVHVVLRQLVWRSHUIRUPHQJLQHHULQJHYDOXDWLRQWHVWLQJRIWKH(87ZLWKUHVSHFWWRWKH
VSHFLILFDWLRQOLVWHGDERYH
'DWHRI7HVW 7HVW(QJLQHHU 'DYLG%DUH
7HVW/RFDWLRQ )UHPRQW&KDPEHU
&RQILJ8VHG &RQILJ&KDQJH 1RQH
(879ROWDJH 9'&
*HQHUDO7HVW&RQILJXUDWLRQ
7KH(87DQGDOOORFDOVXSSRUWHTXLSPHQWZHUHORFDWHGRQWKHWXUQWDEOHIRUUDGLDWHGVSXULRXVHPLVVLRQVWHVWLQJ
)RUUDGLDWHGHPLVVLRQVWHVWLQJWKHPHDVXUHPHQWDQWHQQDZDVORFDWHGPHWHUVIURPWKH(87
$PELHQW&RQGLWLRQV
7HPSHUDWXUH
5HO+XPLGLW\
6XPPDU\RI5HVXOWV
5XQ
0RGH
&KDQQHO
3RZHU
6HWWLQJ
0HDVXUHG
3RZHU
&:7;
0+]
0D[
5;
0+]
0D[
ƒ&
7HVW3HUIRUPHG
5DGLDWHG(PLVVLRQV
7UDQVPLW0RGH
5DGLDWHG(PLVVLRQV
5HFHLYH0RGH
/LPLW
(1
(1
5HVXOW0DUJLQ
G%—9P#
0+]G%
G%—9P#
0+]G%
0RGLILFDWLRQV0DGH'XULQJ7HVWLQJ
1RPRGLILFDWLRQVZHUHPDGHWRWKH(87GXULQJWHVWLQJ
'HYLDWLRQV)URP7KH6WDQGDUG
1RGHYLDWLRQVZHUHPDGHIURPWKHUHTXLUHPHQWVRIWKHVWDQGDUG
Figure 33. EN300 220-1 Radiated Spurious Emissions Test Results (Page 4 of 7)
30
Rev. 0.3
AN445
5XQ5DGLDWHG6SXULRXV(PLVVLRQV7UDQVPLW0RGH0+]
6LB%617&:7;0+]10DWFK9'' 97;3:5/D\HU%RDUGZR6KLHOG
Figure 34. EN300 220-1 Radiated Spurious Emissions Test Results (Page 5 of 7)
Rev. 0.3
31
AN445
)UHTXHQF\
0+]
1RWH
/HYHO
G%μ9P
3RO
YK
9
+
9
+
9
9
+
(1
/LPLW
0DUJLQ
'HWHFWRU
3N43$YJ
3HDN
3HDN
3HDN
3HDN
3HDN
3HDN
3HDN
$]LPXWK
GHJUHHV
+HLJKW
PHWHUV
&RPPHQWV
0D[LPL]HG
7KHILHOGVWUHQJWKOLPLWLQWKHWDEOHVDERYHZDVFDOFXODWHGIURPWKHHUSHLUSOLPLWGHWDLOHGLQWKHVWDQGDUGXVLQJWKHIUHHVSDFH
SURSDJDWLRQHTXDWLRQ( ¥3*G7KLVOLPLWLVFRQVHUYDWLYHLWGRHVQRWFRQVLGHUWKHSUHVHQFHRIWKHJURXQGSODQHDQG
IRUHUSOLPLWVWKHGLSROHJDLQG%LKDVQRWEHHQLQFOXGHG7KHHUSRUHLUSIRUDOOVLJQDOVZLWKOHVVWKDQG%RIPDUJLQ
UHODWLYHWRWKLVILHOGVWUHQJWKOLPLWLVGHWHUPLQHGXVLQJVXEVWLWXWLRQPHDVXUHPHQWV
5XQ5DGLDWHG6SXULRXV(PLVVLRQV5HFHLYH0RGH0+]
6LB%617&:7;0+]10DWFK9'' 97;3:5/D\HU%RDUGZR6KLHOG
Figure 35. EN300 220-1 Radiated Spurious Emissions Test Results (Page 6 of 7)
32
Rev. 0.3
AN445
)UHTXHQF\
0+]
1RWH
/HYHO
G%μ9P
3RO
YK
9
9
(1
/LPLW
0DUJLQ
'HWHFWRU
3N43$YJ
3HDN
3HDN
$]LPXWK
GHJUHHV
+HLJKW
PHWHUV
&RPPHQWV
7KHILHOGVWUHQJWKOLPLWLQWKHWDEOHVDERYHZDVFDOFXODWHGIURPWKHHUSHLUSOLPLWGHWDLOHGLQWKHVWDQGDUGXVLQJWKHIUHHVSDFH
SURSDJDWLRQHTXDWLRQ( ¥3*G7KLVOLPLWLVFRQVHUYDWLYHLWGRHVQRWFRQVLGHUWKHSUHVHQFHRIWKHJURXQGSODQHDQG
IRUHUSOLPLWVWKHGLSROHJDLQG%LKDVQRWEHHQLQFOXGHG7KHHUSRUHLUSIRUDOOVLJQDOVZLWKOHVVWKDQG%RIPDUJLQ
UHODWLYHWRWKLVILHOGVWUHQJWKOLPLWLVGHWHUPLQHGXVLQJVXEVWLWXWLRQPHDVXUHPHQWV
Figure 36. EN300 220-1 Radiated Spurious Emissions Test Results (Page 7 of 7)
Rev. 0.3
33
AN445
APPENDIX
Si4431 B1 script used in the ACP measurements ("4.5. Adjacent Channel Power (ACP) (ETSI 300220001v020301,
subclause 7.6)" on page 14).
#BATCHNAME 868M
s2 8780
# Crystal Oscillator Load capacitance
S2 896c
#VCO current boost
S2 da7f
#PLL BW boost
S2 D8EF
#TX freq set
S2 F573
S2 F678
S2 F7F0
# TX data rate: 1kbps
S2 EE08
S2 EF31
S2 F02C
#TXdeviation: 5KHz
S2 F208
S2 F100
# PN9 GFS
S2 F133
#TXon
s2 8708
#TXpower state
s2 ed0a
34
Rev. 0.3
AN445
DOCUMENT CHANGE LIST
Revision 0.1 to Revision 0.2
Updated "3.2. Blocking (ETSI 300220001v020301,
Subclause 8.4)" on page 4.
 Added "4.5.3. Measurements with 12.5 kHz Channel
Spacing with 9.6 kbps and 4.8 kHz Deviation" on
page 18.
 Added "4.5.4. Measurements with 12.5 kHz Channel
Spacing with 9.6 kbps and 2.4 kHz Deviation" on
page 20.

Revision 0.2 to Revision 0.3
Updated Table 1 on page 1.
Updated Figure 9 on page 9.
 Added spectrum mask measurements with
optimized register settings to "4.6. Modulation
Bandwidth (ETSI 300220001v020301, Subclause
7.7)" on page 22.
 Added Figure 24 on page 23.


Rev. 0.3
35
Simplicity Studio
One-click access to MCU tools,
documentation, software, source
code libraries & more. Available
for Windows, Mac and Linux!
www.silabs.com/simplicity
MCU Portfolio
www.silabs.com/mcu
SW/HW
www.silabs.com/simplicity
Quality
www.silabs.com/quality
Support and Community
community.silabs.com
Disclaimer
Silicon Laboratories intends to provide customers with the latest, accurate, and in-depth documentation of all peripherals and modules available for system and software implementers
using or intending to use the Silicon Laboratories products. Characterization data, available modules and peripherals, memory sizes and memory addresses refer to each specific
device, and "Typical" parameters provided can and do vary in different applications. Application examples described herein are for illustrative purposes only. Silicon Laboratories
reserves the right to make changes without further notice and limitation to product information, specifications, and descriptions herein, and does not give warranties as to the accuracy
or completeness of the included information. Silicon Laboratories shall have no liability for the consequences of use of the information supplied herein. This document does not imply
or express copyright licenses granted hereunder to design or fabricate any integrated circuits. The products must not be used within any Life Support System without the specific
written consent of Silicon Laboratories. A "Life Support System" is any product or system intended to support or sustain life and/or health, which, if it fails, can be reasonably expected
to result in significant personal injury or death. Silicon Laboratories products are generally not intended for military applications. Silicon Laboratories products shall under no
circumstances be used in weapons of mass destruction including (but not limited to) nuclear, biological or chemical weapons, or missiles capable of delivering such weapons.
Trademark Information
Silicon Laboratories Inc., Silicon Laboratories, Silicon Labs, SiLabs and the Silicon Labs logo, CMEMS®, EFM, EFM32, EFR, Energy Micro, Energy Micro logo and combinations
thereof, "the world’s most energy friendly microcontrollers", Ember®, EZLink®, EZMac®, EZRadio®, EZRadioPRO®, DSPLL®, ISOmodem ®, Precision32®, ProSLIC®, SiPHY®,
USBXpress® and others are trademarks or registered trademarks of Silicon Laboratories Inc. ARM, CORTEX, Cortex-M3 and THUMB are trademarks or registered trademarks of
ARM Holdings. Keil is a registered trademark of ARM Limited. All other products or brand names mentioned herein are trademarks of their respective holders.
Silicon Laboratories Inc.
400 West Cesar Chavez
Austin, TX 78701
USA
http://www.silabs.com
Similar pages