View detail for AVR2004: LC

```AVR2004: LC-Balun for AT86RF230
Features
• Balun for AT86RF230 with lumped elements
• Simulation results
• S-Parameter file
Application Note
1 Introduction
In some cases the used balun on the ATAVR®RZ502 Radio Boards must be
replaced by concentrated elements. Therefore the functionality must be given by
the use of capacitors and inductors only. This application note shows how to
transform the differential AT86RF230 RF signal into a single-ended 50 ohm
antenna without using the original balun.
To convert a differential signal to a single-ended type, a balanced to unbalanced
transformation has to be done by using a balun circuit. For this transformation,
hand calculation, simulation, hardware, real measurements and the resulting SParameter file are shown and given.
Figure 1-1. Schematic of a balun structure attached to AT86RF230
L1
CK
RFP
C1
CK
AT86RF230
50 ohms
L2
CK
RFN
C2
GND
Rev. 8113A-AVR-04/08
2 General Considerations
The task to solve is to transform a RF signal from a differential port, which are two
signals relating to each other, into a signal which is related to ground level (a singleended port). Since the common balun structure is a reciprocal element, it is possible
to change input port with output port and still get the same functionality. So if a signal
from a single-ended port is the input signal, the balun structure has to transform it to a
signal on a differential port. Therefore the input signal must be turned into two nonground signals. This can be done by using a low pass branch and a high pass branch
generating a +90° and a -90° signal. Using this as a differential signal fulfills the
desired needs. As an initial step the capacitor and inductor values for the branches
can be calculated following Equation 2-1, as a starting point for balun simulations at
2.45GHz. Consider the capacitors CK from Figure 1-1 as DC blocking elements,
therefore a value of 22pF is enough.
Equation 2-1. Simple balun equation set
Rinner = RIN ROUT = 100 Ω ⋅ 50 Ω ≈ 71Ω
Rinner
= 4.6 nH
2 ⋅π ⋅ f
1
C1 = C 2 =
= 0.92 pF
2 ⋅ π ⋅ f ⋅ Rinner
L1 = L2 =
3 Simulation
For simulation purposes, several software tools do exist. To enable the reader to use
the S-Parameter file from the measurements, the freeware Quite Universal Circuit
Simulator [1] was used here.
The circuit for the simulation consists of a low-pass branch (C_H and L_H) and a
high-pass branch (L_L and C_L) as shown in Figure 3-1 using the calculated values
from Equation 2-1. The single-ended source V1 feeds both branches at their series
input elements. The output signals S_hp for the high-pass branch and S_lp for the
low-pass branch are plotted in Figure 3-1 and shall be seen as two signals shifted by
+/- 90 degrees compared to the input signal S_in.
Since the signals S_hp and S_lp do have a time constant phase difference following
Equation 3-1, these two signals can be combined to the one desired differential
signal.
The balun designer may take a look at the input and output impedance of the circuit,
since it is the goal to interface other RF circuits and provide the best match to them.
For this application note the differential port is the gate to the AT86RF230 pins RF_P
and RF_N, the single ended port is connected to the single-ended antenna. So the
differential impedance must have 100 ohms as the AT86RF230 datasheet states and
the single ended port must have 50 ohms impedance, since a standard antenna
should be used.
2
AVR2004
8113A-AVR-04/08
AVR2004
Figure 3-1. Simulation of the balun principle
Equation 3-1. resulting differential phase
Phase 1 − Phase 2 = Phase differenti al
+ 90 ° − ( −90 °) = +180 °
Figure 3-2. Simulated circuit
3
8113A-AVR-04/08
When the circuit as shown in Figure 3-2 is simulated at 2450MHz, which is almost the
middle frequency of the 2.4GHz ISM-band, the input impedance measured at the
differential port P1 is 100 ohms and the output impedance at port P2 is 50 ohms as
desired and shown in Figure 3-3 and Figure 3-4, inclusive the 180° phase shift
expressed as 0° phase shift due to the differential impedance view at port P1.
Figure 3-3. Simulation result, magnitude and phase of input and output impedance in
[ohm] and [°]
Figure 3-4. Simulation result, return loss of balun ports in [dB]
4
AVR2004
8113A-AVR-04/08
AVR2004
4 Test Device
4.1 Layout
For the circuit from Figure 3-2, a PCB was designed. The layout is shown in Figure 41. The PCB’s material is Rogers 4003 instead of FR4 to extract the S-parameter of
this circuit as exactly as possible. The reason for this is, that Rogers 4003 provides a
more static εr than FR4 does. At the end the S-parameters of this circuit were
recalculated to deembed the traces of the PCB, from the PCB borders up to the pads
of the SMD components.
Figure 4-1. Test PCB
4.2 Measurements
To ensure the measured values rely on the circuitry only, the board was measured
without the SMD components first. Therefore the traces were measured single-ended
from the PCB border when the end of each trace was grounded, to find the electrical
length.
After knowing the influence of each trace to the circuit, the measurements of the
complete structure with populated components were done. Since this may be done
later again with other measurement devices, the procedures were limited to basic 50ohm two-port equipment only. So, no three- or four-port measurement devices were
used. To measure a balun with a two-port network analyzer, the third and unused port
has to be loaded with a 50 ohm match. This ends up in two *.s2p files, but one *.s3p
file is needed. Therefore the popular software MathCAD® was used for crunching the
two *.s2p files to one standard *.s3p file.
Based on that result and the measurements, the traces were deembedded. The
results of this procedure are shown in Figure 4-2, Figure 4-3 and Figure 4-4 as the
desired parameters.
5
8113A-AVR-04/08
Figure 4-2 shows the input return loss at common port in dB. The other ports 2 and 3
are terminated with a 50 ohm match to the ground node, which is equivalent to 100
ohms differential impedance between both of these ports.
Figure 4-2. Input return loss of s11 (common port)
Figure 4-3. Magnitude curves of s21 and s31 (differential ports)
6
AVR2004
8113A-AVR-04/08
AVR2004
Figure 4-4. Phase curves of S21 and S31
The transmission parameter’s magnitudes in dB are shown in Figure 4-3, with S21
marked with circles and S31 with squares. The low-pass and high-pass functions of
the appropriate LC networks and the crossover frequency at band center frequency
(2.45 GHz) can be observed.
Figure 4-4 shows the transmission parameter phase in degrees. The S21’s phase is
marked with circles and S31’s phase with squares. The differential phase of
approximately 180 degrees can be observed at band center frequency (2.45 GHz).
The first measurement showed a shifted cross-over frequency, so slightly changes to
the component values were necessary. This fine-tuning during consecutive
measurements did lead to slightly different values than hand calculation and early
simulations showed. The results are given in Table 4.1. Since these results show
reflection coefficients of -30dB, the developed circuit fulfils the AT86RF230 needs.
Table 4-1. measurement results compared to simulated values
component
Simulated value
Measured value
C1 = C2
0.92 pF
0.82 pF
L1 = L2
4.6 nH
4.3 nH
Best impedance match
2.45 GHz
2.57 GHz
Reflection coefficient at 2.45 GHz
(differential port)
-73.4 dB
-31.2 dB
Reflection coefficient at 2.45 GHz
(single-ended port)
-73.4 dB
-31.5 dB
7
8113A-AVR-04/08
4.3 S-Parameter
The S-parameter of the LC balun for the AT86RF230 comes with this application note
With a circuit, as shown in Figure 4-5, the parameters can be experienced as shown
in Figure 4-6.
Figure 4-5. Simulation circuit to extract parameters
Figure 4-6. Parameters extracted from measured s3p file in [ohm], [°] and [dB]
8
AVR2004
8113A-AVR-04/08
AVR2004
5 Harmonics
Due to its non-linearity behavior, a RF device transmits not only at the main or center
frequency. RF energy is also detectable at integer multiples of the center frequency,
which are called harmonics. A device transmitting at a center frequency fc = 2.45GHz
will also radiate RF energy at 2 x fc = 4.9GHz, 3 x fc = 7.35GHz, and so on. A good
RF transmitter radiates as much of its transmitting power at center frequency as
possible and therefore as less energy as possible at the harmonic frequencies.
International and national regulatory instances like FCC or ETSI limit the power
radiations at all frequency bands. The power level limit from FCC for the harmonic
frequencies 4.9GHz, 7.35GHz and 9.8GHz is -41.2dBm and ETSI limits the radiation
on harmonics to -30dBm.
The used hardware to measure the harmonic power levels was an ATAVRRZ502
Radio Board on STK®501/STK500 combination. The interesting fact here is, that an
AT86RF230 IC was chosen, which provides as much RF output power as possible to
be as close at the regulatory limits as possible or above. The chosen AT86RF230
provided 4dBm output power, which is more than the specified +/-3dBm. For this “too
good” IC it was expected that the harmonics are increased in the same way. Table 51 shows the measured power levels of the used hardware.
Table 5-1. Measurement results, harmonic power level on “too good” AT86RF230
with balun from Wuerth Electronics
frequency
Channel 11
Channel 15
Channel 26
2.45GHz
4.06 dBm
3,99 dBm
3,96 dBm
4.9GHz
-40.78 dBm
-41,94 dBm
-44,68 dBm
7.35GHz
-40.24 dBm
-41,12 dBm
-42,49 dBm
9.8GHz
-47.84 dBm
-47,42 dBm
-47,77 dBm
A second ATAVRRZ502 Radio Board was chosen, providing about the same RF
output power as the Wuerth balun Radio Board, for populating the above developed
LC-Balun instead of the standard balun from Wuerth Electronics. For this case,
inductors from Murata were used (L=4.3nH, 0402, Murata LQG15HN4N3S02) and
capacitors from Phycomp (C=0.82pF, NP0, 0402, Phycomp, SPLR 223886915827,
RS Comp. 616-9357). The circuit from Figure 5-1 was used to measure the harmonic
power. The results are shown in Table 5-2.
9
8113A-AVR-04/08
Figure 5-1. LC-Balun circuit for power measurement at harmonics
Table 5-2. Measurement results of harmonic power level on “too good” AT86RF230
with LC-balun
frequency
Channel 11
Channel 15
Channel 26
2.45GHz
4,22 dBm
4,14 dBm
3,98 dBm
4.9GHz
-36,07 dBm
-35,62 dBm
-34,78 dBm
7.35GHz
-33,54 dBm
-33,81 dBm
-32,73 dBm
9.8GHz
-53,67 dBm
-53,2 dBm
-53,7 dBm
Table 5-3. Harmonics power of LC-balun above FCC
frequency
Channel 11
Channel 15
Channel 26
2.45GHz
n/a
n/a
n/a
4.9GHz
5,13 dB
5,58 dB
6,42 dB
7.35GHz
7,66 dB
7,39 dB
8,47 dB
9.8GHz
0 dB
0 dB
0 dB
From Table 5-1 and Table 5-2 it can be seen, that both baluns radiate more power
than the FCC limits allow, but less enough to fulfill the ETSI limits. It is not surprising
that the Wuerth balun shows this behavior. The Radio Board was designed to operate
slightly below the FCC limits, to provide the maximum output power at the antenna.
This was done taking into account that the harmonics are slightly below FCC limits
also. Now the “too good” AT86RF230 was used, with more output power than the
sold AT86RF230 ICs, to show exactly this behavior.
The LC-Balun consists of a low-pass and a high-pass branch. Here the high-pass
branch operates as it is defined. It passes higher frequencies than the center
frequency it was designed for. This behavior is of course bad for the harmonics. So
as expected, the attenuation of the AT86RF230 harmonics is less than the
attenuation of the Wuerth-Balun.
10
AVR2004
8113A-AVR-04/08
AVR2004
Table 5-4. Harmonics power of Wuerth balun above FCC
frequency
Channel 11
Channel 15
Channel 26
2.45GHz
n/a
n/a
n/a
4.9GHz
0,42 dB
0 dB
0 dB
7.35GHz
0,96 dB
0,08 dB
0 dB
9.8GHz
0 dB
0 dB
0 dB
Table 5-3 and Table 5-4 provide the power levels above the FCC limits. This shows
clearly, that a filter is needed. This filter has to provide at least an attenuation of
5.13dB at 4.9GHz and 7.66dB at 7.35GHz. For the Wuerth balun, this filter will work
too, even if only 1dB would be necessary. Usable filters to solve this problem are the
following types: Cauer, Bessel, Butterworth and Chebychev. A Chebychev filter was
chosen and after dimensioning to a cutoff frequency of 3GHz, it provides attenuation
of -17dB at 2 x fc as shown in Figure 5-3.
Figure 5-3. 3GHz Chebychev filter transmission function (absolute values) in [dB]
Figure 5-4. LC-Balun with 3GHz-Chebychev filter
11
8113A-AVR-04/08
Table 5-5. Harmonics power level of LC-balun with applied filter
frequency
Channel 11
Channel 15
Channel 15
2.45GHz
4,1914 dBm
4,1114 dBm
3,9514 dBm
4.9GHz
-53,07 dBm
-52,62 dBm
-51,78 dBm
7.35GHz
-62,64 dBm
-62,91 dBm
-61,83 dBm
9.8GHz
-90,67 dBm
-90,2 dBm
-90,7 dBm
As it can be seen in Table 5.5, FCC limits are not hurt any longer with applied filter.
ETSI limits were not hurt at all. For ETSI no filter would be needed, but the developed
FCC filter will not disturb the functionality here, so finally one piece of hardware can
be used to pass FCC and ETSI. The final schematic is provided in Figure 5-4.
The input and output impedances have changed a bit, but the matching at 2.45GHz is
still good enough (better than 10 dB as AT86RF230 datasheet states). Figure 5-5
shows the simulated input impedance of the LC-Balun with applied filter, and the
output impedance of the Chebychev filter, seen from a 50 ohms antenna.
Figure 5-5. overall impedance simulation results for LC-Balun with applied filter in
[ohm] and [dB]
12
AVR2004
8113A-AVR-04/08
AVR2004
6 Common Mode Rejection Ratio
The Common Mode Rejection Ratio (CMRR) is a figure of merit, which describes the
performance of a balanced circuit. It is defined as the ratio between the insertion loss
of the differential mode versus the common mode with respect to signal gain following
equation 6.1.
Equation 6-1. CMRR for balun circuit with port 1 as common port
CMRR =
S 1C
S + S 13
= 12
S1D S 12 − S 13
The developed LC balun showed a CMRR following Figure 6.1 and provides a CMRR
based on measurement values of -36dB versus the Wuerth balun, as shown in Figure
6.2, which provides a CMRR of -29dB. So finally the developed LC circuit shows a
better CMRR at the operating frequency band, but it does not help at all to prevent
common mode signals at harmonic frequencies. The Wuerth balun shows a 12dB
higher CMRR 4.9GHz, so finally the LC balun helps in operating space, but not more.
Figure 6-1. LC balun’s CMRR in [dB]
13
8113A-AVR-04/08
Figure 6-2. Wuerth balun’s CMRR in [dB]
7 Conclusion
If the Wuerth balun, which is used on the ATAVRRZ502 Accessory Kit, must be
replaced by a balun circuit consisting of lumped elements, one solution was shown
here and measured in Atmel. The schematic can be found in Figure 5-4. As
components for L1 and L2 Murata LQG15HN4N3S02 should be used and for the
capacitors C1 and C2 Phycomp 223886915827 (RS Component Number: 616-9357)
should be used. The LC balun has the disadvantage, that an additional filter must be
used and the Common Mode Rejection is more worse than using the from Atmel
recommended Wuerth balun as it is shown in the ATAVRRZ502 Accessory Kit.
14
AVR2004
8113A-AVR-04/08
Disclaimer
International
Atmel Corporation
2325 Orchard Parkway
San Jose, CA 95131
USA
Tel: 1(408) 441-0311
Fax: 1(408) 487-2600
Atmel Asia
Room 1219
Chinachem Golden Plaza
East Kowloon
Hong Kong
Tel: (852) 2721-9778
Fax: (852) 2722-1369
Atmel Europe
Le Krebs
8, Rue Jean-Pierre Timbaud
BP 309
78054 Saint-Quentin-enYvelines Cedex
France
Tel: (33) 1-30-60-70-00
Fax: (33) 1-30-60-71-11
Atmel Japan
9F, Tonetsu Shinkawa Bldg.
1-24-8 Shinkawa
Chuo-ku, Tokyo 104-0033
Japan
Tel: (81) 3-3523-3551
Fax: (81) 3-3523-7581
Technical Support
[email protected]
Sales Contact
www.atmel.com/contacts
Product Contact
Web Site
www.atmel.com
Literature Request
www.atmel.com/literature
Disclaimer: The information in this document is provided in connection with Atmel products. No license, express or implied, by estoppel or otherwise, to any
intellectual property right is granted by this document or in connection with the sale of Atmel products. EXCEPT AS SET FORTH IN ATMEL’S TERMS AND
CONDITIONS OF SALE LOCATED ON ATMEL’S WEB SITE, ATMEL ASSUMES NO LIABILITY WHATSOEVER AND DISCLAIMS ANY EXPRESS, IMPLIED
OR STATUTORY WARRANTY RELATING TO ITS PRODUCTS INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTY OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT. IN NO EVENT SHALL ATMEL BE LIABLE FOR ANY DIRECT, INDIRECT,
CONSEQUENTIAL, PUNITIVE, SPECIAL OR INCIDENTAL DAMAGES (INCLUDING, WITHOUT LIMITATION, DAMAGES FOR LOSS OF PROFITS,
BUSINESS INTERRUPTION, OR LOSS OF INFORMATION) ARISING OUT OF THE USE OR INABILITY TO USE THIS DOCUMENT, EVEN IF ATMEL HAS
BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Atmel makes no representations or warranties with respect to the accuracy or completeness of the
contents of this document and reserves the right to make changes to specifications and product descriptions at any time without notice. Atmel does not make any
commitment to update the information contained herein. Unless specifically provided otherwise, Atmel products are not suitable for, and shall not be used in,
automotive applications. Atmel’s products are not intended, authorized, or warranted for use as components in applications intended to support or sustain life.