Tag Archives: PowerDividers

How to Measure Mixer IP3 and Identify Potential Error Sources

Mixer linearity is continuously and permanently a critical problem faced in RF system design. The nonlinear action of all physically realizable RF mixers propagates throughout signal chains generating undesired, unfilterable output harmonics, multitone intermodulation, and nonrecoverable nonlinear signal distortion. For example, nonlinear mixing action can cause undesired output harmonics (i.e., spurs) such as the 2f_{RF} \times 2f_{LO} or the 2f_{RF} \times f_{LO} spur instead of the desired f_{RF} \times f_{LO} converted signal. In multitone applications, such as data transmission and radar tracking, mixing heavily exacerbates the problem of spectral purity by not only introducing a second set of unwanted, unfilterable output harmonics but by also introducing multitone intermodulation distortion (IMD). IP3, or the 3rd order intercept point (TOI), is the figure of merit by which industry judges the linearity of all active, power-consuming RF components and their ability to maintain core linear assumptions about circuits. The mixer is no exception.

A power calibrated vector network analyzer and an external driving synthesizer is the standard tool to do IP3 measurements. While VNAs have their own errors, such as limited RF input power and poor 2nd input harmonic suppressions, it is the most convenient and time efficient way to measure IP3 over broadband frequency sweeps. Spectrum analyzers, although highly susceptible to user error, provide a secondary way of doing comparable IP3 measurements. Both measurement on a spectrum analyzer and a VNA should agree. This document serves as a tool for debugging, developing, and utilizing the spectrum analyzer, it’s supporting synthesizers, and other RF paraphernalia for mixer IP3 measurements.

IP3 and IMD Explained

The most important thing to understand about the IP3 specification is that, like a mixer’s spurious suppression, it is a measure of your in-band spur-free dynamic range, and how linear your output signal is. IP3 is not a directly measured mixer parameter like conversion loss, return loss, or isolation. IP3 is a figure of merit of device linearity based on the relationship between a 2-tone analytic signal and its resulting self-intermodulation terms.

To calculate the input referred IP3 (IIP3) we use:

IIP3 (dBm) = P_{Input} + \dfrac{P_{Fundamental}-P_{IM3}}{2}

To calculate the output referred IP3 (OIP3) we use:

OIP3 (dBm) = P_{Fundamental} - \dfrac{P_{Fundamental} - P_{IM3}}{2}

OIP3 (dBm) = IIP3 + Conversion Loss

For the case where a mixer is used as a downconverter with a low side LO, P_{Input} is the average power of the analytic 2-tone RF input signal. P_{Fund} is the average power of the down converted 2-tone RF input to the frequencies f_{RF_1}-f_{LO} and f_{RF_2}-f_{LO} . P_{IMD3} is the average power of the spurious output that appears at (2f_{RF_1}-f_{RF_2})-f_{LO} and (2f_{RF_2}-f_{RF_1})-f_{LO}. P_{IMD3} is a mixing product of the 2-tone RF input and mixer generated second order intermodulation products to the (2f_{RF_1}-f_{RF_2})-f_{LO} and (2f_{RF_2}-f_{RF_1})-f_{LO} frequencies. For an upconverter, the principal is the same.

Depiction of where the 2nd and 3rd order IMD products fall in a mixer output spectrum based on a 2-tone RF input signal. Frequency plan based on a mixer used in a down conversion of [math]f_1 and f_2[/math] using a highside LO.

1. Depiction of where the 2nd and 3rd order IMD products fall in a mixer output spectrum based on a 2-tone RF input signal. Frequency plan based on a mixer used in a down conversion of f_1 and f_2 using a highside LO.

IIP3 is input referred IP3 and OIP3 is output referred. This can be measured in either an up-conversion or down-conversion and the result of which should be roughly equivalent for the same or similar frequency plan (RF, LO, and IF frequency bands and power levels). To visualize what OIP3 means, see figure 2 for what could typically be seen on a spectrum analyzer for a downconverted 2-tone signal.

2. Example output spectrum of an IP3 measurement. Output will include unwanted harmonics such as those close to the carrier.

2. Example output spectrum of an IP3 measurement. Output will include unwanted harmonics such as those close to the carrier.

First, there is the need to clarify the difference between IMD products and spurious products of a mixer. IMD tones are generated from multiple, close input fundamental frequencies, f_{RF_1}  or f_1 and f_{RF_2} or f_2. They are not generated from harmonic mixing of the RF/IF signal and the LO signal. An example IMD product is the (2f_{RF_1}-f_{RF_2}) - f_LO harmonic. Spurious products, such as the 1f_{LO}-3f_{RF} spur, are the result of the unwanted harmonic mixing of the RF/IF signal and the LO signal.

Similarly, for a mixer’s IP2, you’re not interested in measuring the power of the f_{LO} - 2f_{RF} spur, you’re interested in the 1f_{LO}-(f_{RF_1}-f_{RF_2}) or 1f_{LO}-(f_{RF_1}+f_{RF_2}) spur.

3. DUT test frequency plan for a 2-tone RF input. Mixer tested as a down converter from 6GHz RF to 300MHz IF.

3. DUT test frequency plan for a 2-tone RF input. Mixer tested as a down converter from 6GHz RF to 300MHz IF.

Testing Part I: Quick and Dirty

The frequency plan above is the frequency plan of the running example of this document. The example measurement in this document will be done as a down conversion with the MM1-0312HS (unless otherwise specified) from 6GHz +/- 0.5MHz RF to 299MHz +/-0.5MHz IF with a low side fixed 5.701GHz LO. For convention, RF refers to the high frequency small signal input that will be mixed down to the low frequency IF output. The RF signal is input into port 1, LO into port 3, and IF is output from port 2 (configuration B).

4. Mixer input/output signal configuration. RF signal is input into port 1, LO signal into port 3, and the IF signal is pulled from port 2.

4. Mixer input/output signal configuration. RF signal is input into port 1, LO signal into port 3, and the IF signal is pulled from port 2.

First, we generate a combined 2-tone input signal with 2 synthesizers and any power divider (i.e., power combiner); for instance, the PD-0010 resistive power divider.

f_1 is the 6.0005GHz tone and f_2 is the 5.9995GHz tone. Both tones have +0dBm power, verified on both a filtered power meter and our Rohde and Schwarz FSW85 spectrum analyzer. It’s important to understand we’re trying to only test our mixer so we must always power calibrate to the plane of the mixer’s RF port. To avoid measurement error, we must calibrate to only the fundamental tone. Using an unfiltered power meter to do this measurement is slightly erroneous because power meters measure the total output power of a signal. This includes all harmonics in addition to the fundamental tone power.

5. Quick and dirty mixer IP3 measurement. Measurement may or may not be accurate with this setup.

5. Quick and dirty mixer IP3 measurement. Measurement may or may not be accurate with this setup.

We repeat the power calibration step for the LO source set at 5.701GHz, +15dBm. It is extremely important that sufficient LO power is used to fully turn on a mixer or else risk unpredictable results. Mixers are heavily LO power dependent, if LO power is varied, the IP3 measurement result can and will vary widely. If insufficient LO power is delivered, the conversion loss suffers heavily and will provide erroneous IP3 results.

6. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 5.

6. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 5.

After calibrating, the mixer is connected as shown in figure 6. When we calculate the IP3 value, we typically average both down converted tones and both IMD spurs. So, remembering that the data displayed on the screen of the spectrum analyzer is the mixer’s output, we can first calculate the mixer’s OIP3 and then back calculate the mixer’s IIP3 because we know the RF input power.

OIP3 = \frac{(f_{Out_1}+f_{Out_2})}{2}+\frac{\frac{(f_{Out_1}+f_{Out_2})}{2}+\frac{(f_{IMD_1}+f_{IMD_2})}{2}}{2}    latex OIP3 =\frac{(-8.55+-8.28)}{2}+\frac{\frac{(-8.55+-8.28)}{2}-\frac{(-54.02+-53.93)}{2}}{2}=+14.4dBm$

IIP3 = OIP3 - Conversion Loss = 14.4-\frac{(-8.55+-8.28)}{2}= +23.9 dBm

For a “quick and dirty” measurement the result is not bad. It relatively matches the value displayed on the datasheet. This was a straight forward measurement where we did not consider non-idealities in the measurement, and implicitly trusted our test equipment to report the actual IP3 value. As a rule of thumb, if the mixer’s IP3 is below +25dBm, this is a sufficient setup with modern test equipment. To provide more accurate data however, we must account for the non-idealities.

Error Source #1

To account for all of the non-idealities in the measure we look mainly at the small signal RF input for the down conversion measurement. In figure 8, we see that the “quick and dirty” measurement’s spectrum is fairly clean by casual standards; the IMD3 spurs are not stronger than -80dBm. This establishes part of the IP3 measurement floor. This floor isn’t good enough for the most accurate mixer IP3 results. To get the best results, we need to decrease the IMD3 tone power of the input. Getting an ideal result of no input intermodulation products is not possible, although they can be greatly suppressed.

 

Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 5.

7. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 5.

The 2nd harmonic inputs at 2f_1 and 2f_1 are very low power (or should be). The 2f_1 and 2f_2 will mix down to (2f_{RF_ 1} - f_{RF_2})- f_{LO} and (2f_{RF_2} - f_{RF_1}) - f_{LO} and should be accounted for; a passive diode mixer will respond to any signal of any power. If an RF amplifier is used, the 2nd harmonic should be filtered off.

8. Diplexers added after the signal sources from figure 4. Lowpass port of the diplexer passes the signal. Input [math]2f_1 and 2f_2[/math] terms terminated in highpass port with a 50Ω termination.

8. Diplexers added after the signal sources from figure 4. Lowpass port of the diplexer passes the signal. Input 2f_1 and 2f_2 terms terminated in highpass port with a 50Ω termination.

Lowpass filter the 2nd harmonic of f_1 and f_2to eliminate a low power source of intermodulation within the mixer. This is done with the lowpass port of an applicable diplexer. Unwanted high frequency signals will not be reflected but instead be sunk into the highpass port and its 50Ω termination.

9. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 8.

9. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 8.

There is a marginal gain in the 2-tone RF input IP3 as a result of the lowpass filters. Not surprisingly this results in no change in the mixer’s IP3 measurement.

10. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 8.

10. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 8.

Error Source #2

Another major, but less obvious, source of error is the cross talk between the two RF input synthesizers. All synthesizers have a phase locked loop (PLL) to lock the output frequency to an internal (or external) reference frequency.

When a reverse signal is present, either from a mismatched load reflecting power back, or from cross-talk introduced from an external signal source, the phase detector in the PLL will respond. Even if an RF amplifier is at the output of the PLL the amplifier reverse isolation is finite and will leak power into the phase detector.

11. Example phased locked loop showing a possible reverse signal path.

11. Example phased locked loop showing a possible reverse signal path.

The feedback path of the PLL will account for anymismatched reflection. If instead, a 2nd RF input synthesizer were to be the reverse signal, then the phase detector block of the PLL will produce its own harmonics from unwanted intermodulation due to the non-linear action of the phase detector (i.e., mixer).

To reduce the unwanted reverse signal, we can do 2 things. First, we provide isolation through the power divider used to minimize the cross talk between the 2 RF input synthesizers. Second, we can provide additional isolation on each synthesizer arm through the use of attenuators or isolators.

12. RF input diagram using power divider that provides isolation.

12. RF input diagram using power divider that provides isolation.

If instead of the PD-0010 resistive power divider we were to use a power divider that provides isolation, like the PD-0R413 Wilkinson power divider, the cross-talk should be reduced. Power traveling from one output port of the power divider to the other will be attenuated by the power divider’s isolation value.

Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 12 using the PD-0R413 power divider.

13. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 12 using the PD-0R413 power divider.

Swapping in the PD-0R413 Wilkinson power divider reduces the 2 tone RF input’s IMD3 terms by 4dB. Using an even higher isolation power divider like the PBR-0006 should provide a slightly better result.

Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 12 using the PBR-0006 power divider.

14. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 12 using the PBR-0006 power divider.

Incrementally increasing the power divider isolation marginally improves the measured output IP3 of the 2-tone RF input signal. While the 3rd order tones decrease, the higher order intermodulation products may change.

Mixer IP3 measurement with the PD-0R413:

IF output spectrum of the MM1-0312HS with the block diagram setup from figure 12 using the PD-0R413 power divider.

15. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 12 using the PD-0R413 power divider.

Mixer IP3 measurement with the PBR-0006:

IF output spectrum of the MM1-0312HS with the block diagram setup from figure 12 using the PBR-0006 power divider.

16. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 12 using the PBR-0006 power divider.

The mixer’s measured IP3 is effectively the same. So we continue the example using the PBR-0006 because of the higher isolations. To improve the isolation 10dB attenuators are added to the circuit as seen in figure 18. The lowpass filters used in front of the synthesizers are DPX-9516s. They are reflective filters near the crossover point of the diplexer. Outside of the DC-9.5GHz lowpass filter passband, high frequency signals will reflect back towards their source.

17. 10dB attenuators added between the diplexer and power divider to improve the [math]f_1 to f_2[/math] isolation by 20dB. Possible signal paths shown in blue and red.

17. 10dB attenuators added between the diplexer and power divider to improve the [math]f_1 to f_2[/math] isolation by 20dB. Possible signal paths shown in blue and red.

What is of interest for the reflective filters is the device under test (DUT) mixer’s finite LO-RF isolation and the mixer’s internal harmonics generated from the fundamental LO. The LO signal can, does, and will leak through from the LO port to the RF port and output into the RF input synthesizers. By adding 10dB attenuators between the lowpass filters and power divider, the LO fundamental tone and LO harmonic spurs’ leak through is reduced by 10dB. Any LO harmonic reflections will be reduced by 20dB before remixing in the mixer.

e.g., For a typical Marki GaAs MMIC double balanced mixer, the LO to RF isolation will be approximately 40dB. So an input of a+15dBm fundamental into the LO port will give -25dBm at the RF port that could remix if reflected. If the 2nd LO harmonic is strong enough, that can also remix. Unless the LO frequency is low, the higher order harmonics will be rejected (reflected) by the RF balun of the mixer.

Error Source #3

Because the LO harmonics can be an issue, along with any passive intermodulation products, the power divider’s combined output should be filtered.

18. RF input block diagram from figure 18 with a diplexer presented to the mixer.

18. RF input block diagram from figure 18 with a diplexer presented to the mixer.

Instead of using another a filter in front of the mixer’s RF port, a diplexer (not duplexer) is a good choice. A diplexer is wideband enough to provide a path for the 2nd and 3rd order LO to RF harmonics coming out of the RF port to be terminated in a 50Ω load. The diplexer used must be carefully chosen to avoid the LO harmonics from reflecting back into the mixer. Otherwise, the power divider is a better interface to present for impedance matching considerations. This provides a 1dB improvement in the IMD3 spur power that is input into the mixer.

19. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 20. 20dB internal attenuation in the spectrum analyzer.

19. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 18. 20dB internal attenuation in the spectrum analyzer.

There is no change to the measurement of the mixer’s IP3 from the addition of the 3rd diplexer in front of the mixer’s RF port.

20. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 18.

20. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 18.

Error Source #4

All spectrum analyzers have a front-end mixer prior to their IF envelope detector circuitry; that mixer limits the dynamic range of the system. To make sure that the spectrum analyzer’s mixer does not generate its own harmonic content and disturb the measurement, the power into the spectrum analyzer must be attenuated to prevent overloading the receiver. Secondarily, we want to suppress the front-end mixer from re-mixing higher order products.

If we check against the spectrum analyzer datasheet, our Rohde and Schwarz FSW85 has a typical IIP3 of +30dBm with a minimum of +22dBm for an input frequency of 100MHz to 1GHz. For our input powers (-8dBm on both fundamental IF tones) into the spectrum analyzer that floor is approximately -70dBm if we back calculate with our standard IP3 formula. So the IMD3 spur can be down ~-70dBm (with no internal attenuation) before the spectrum analyzer will hit its dynamic range limit. This is the second floor of the measurement that can fundamentally limit the measured dynamic range of the DUT. You must have a better part than the DUT within the test equipment to measure the DUT.

As a precaution, we can filter the input of the spectrum analyzer to prevent high order harmonics from mixing within the spectrum analyzer.

20. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 18.

21.  IF output spectrum of the MM1-0312HS with the block diagram setup from figure 18.

Using a DPX-0R5 to filter the mixer IF output, we see that the above measurement was not impacted or changed when we correct the OIP3 value for the approximately 3.5dB insertion loss of the filter. IIP3 will not change because the RF input to the mixer has not changed. OIP3 will because it is an output referred value.

Next, we can attenuate the mixer IF output by 20dB to prevent overloading the spectrum analyzer’s front-end mixer and improve the impedance match. 20dB is used arbitrarily and a significant enough attenuation value should otherwise be used.

21. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 18 with a DPX-0R5 lowpass filtering the IF output.

22. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 18 with a DPX-0R5 lowpass filtering the IF output and a 20dB attenuator.

If we add 20dB back to the above OIP3 measurement, we see no negative impact to the measurement; just an additional hoop in the IP3 calculation. So we can disregard this source of error unless the measurement calls for it. An example of when this could be necessary would be when the DUT (device under test) mixer’s IF output has a high power (>+5dBm) or when the spectrum analyzer’s internal attenuation is insufficient.

Error Source #5

From a mixer use standpoint, the argument in Error Source #4 makes sense. To prove it, what if we remove all attenuation into the spectrum analyzer?

22. Measurement from figure 23 repeated with 0dB internal attenuation on the spectrum analyzer

23. Measurement from figure 21 repeated with 0dB internal attenuation on the spectrum analyzer.

The remeasured OIP3 is +15dBm. To interpret the result, do we say that the measurement improved or is this an erroneous result? We can verify this by removing the DUT. If we compare measurements of the OIP3 of figure 21 with and without internal attenuation turned on in the spectrum analyzer, we see that the measurement significantly degrades.

23. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 20. 20dB internal attenuation in the spectrum analyzer. Identical to figure 21

24. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 20. 20dB internal attenuation in the spectrum analyzer. Identical to figure 19.

23. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 19. 0dB internal attenuation in the spectrum analyzer.

25. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 19. 0dB internal attenuation in the spectrum analyzer.

Because the measurement of the 2-tone RF input without attenuation is noticeably different than the measurement without attenuation, it’s very likely the spectrum analyzer’s front end mixer was overloaded and gave an erroneous result when no attenuation is applied. This is the same case for the MM1-0312HS and a measurement without attenuation protecting front-end mixer of the spectrum analyzer should be disregarded.

There should always be some attenuation present to protect the front end mixer of the spectrum analyzer from being overloaded. Typically, 10-20dB attenuation is sufficient. In addition, it is always good practice for the instrument to be run in a as low a resolution bandwidth (RBW) as possible. The reduced resolution bandwidth allows for a slower, more precise measurement to be taken. This reduces the measurement bandwidth of the analog to digital converter within the spectrum analyzer. A resolution bandwidth of 10Hz will measure a tone with 10Hz instantaneous bandwidth versus a resolution bandwidth of 300KHz which will attempt to measure a tone with up to 300KHz instantaneous bandwidth. Using a slower resolution bandwidth avoids measuring the side of the tone aswell as improving the noise floor of the measurement.

Generally, the resolution bandwidth will automatically adjust and lower itself when the frequency span of the instrument is set. It can also be manually set for most modern test equipment. Refer to the operator manual for guidance.

Testing Part 2: Final Test

Final block diagram of the recommended mixer IP3 test setup on a spectrum analyzer. Optional blocks are highlighted.

26. Final block diagram of the recommended mixer IP3 test setup on a spectrum analyzer. Optional blocks are highlighted.

The block diagram is the final test setup in figure 29. Its measured MM1-0312HS IP3 data is below in figure 29. For the measurement in figure 30, the LO filter was omitted. Practically, the high order spurs that mix off the 2nd LO input harmonic that fall on top of the (2f_1 - f_2) - f_{LO} and (2f_2 - f_1) - f_{LO} intermodulation spurs should be at a much lower power than the intermodulation products.

Like the results from section Error Source #4, the LO filter is a test condition dependent block. In particular, it is undesirable to filter the 3rd harmonic if you’re driving a classic, hybrid T3 style mixer with a square wave LO; less so if you’re using a newer MMIC T3. The reason behind this is, because for a T3 style mixer, the 3rd and 5th harmonics of a square wave LO remixes and biases the diodes within the mixers in such a way that the mixer nonlinearity is improved over a normal sine wave LO. For a double balanced mixer, the LO filter may or may not be necessary. Typically, it is not because the LO’s second harmonic can be pushed out of band of the mixer.

If we retest the MM1-0312HS with the block diagram from figure 28, we should see an improvement compared to the block diagram from figure 5.

27. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 28.

27. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 26.

Compared to the quick and dirty measurement setup and mixer IP3 data:

28. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 5. Identical to plot to figure 6.

28. IF output spectrum of the MM1-0312HS with the block diagram setup from figure 5. Identical to plot to figure 6.

There was an improvement of 0.5dBm in the measurement of the mixer’s IP3. This is only slightly more in-line with the VNA measurement. This 0.5dBm change could practically be only attributed to measurement uncertainty. The measurement uncertainty in our Rohde and Schwarz FSW85 is 0.3dB for this measurement frequency. Practically speaking, the IP3 measurement between both test setups are virtually identical. So confidence in the accuracy of the measure can be confirmed.

However, the real improvement is in the IP3 measurement of the 2 tone RF input.

29. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 5. Identical to figure 7.

29. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 5. Identical to figure 7.

Final Test Setup:

30. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 20. Identical to figure 21.

30. Spectrum of the 2-tone RF input into the MM1-0312HS with the block diagram from figure 18. Identical to figure 19.

What this means is that we should be able measure very high IP3 values because the measurement is closer to ideal conditions where the input intermodulation products are not present and mixing down to the  (2f_{RF_1} - f_{RF_2}) - f_{LO} and (2f_{RF_2} - f_{RF_1}) - f_{LO} frequencies.

Choosing the MT3-0113LCQG,and instead of a +15dBm LO, we use a higher power +25dBm square wave LO and repeat the comparison. The higher LO power should give a high IP3 measurement and push the measurement system to its dynamic range limit. The test is: Will the improvement to the 2-tone RF input IP3 allow for high mixer IP3 measurements?

Using the block diagram from figure 5, we compare the quick and dirty mixer IP3, taken with the newly specified +25dBm square wave LO:

31. IF output spectrum of the MT3-0113LCQG using the block diagram from figure 5 with a +25dBm square wave LO.

31. IF output spectrum of the MT3-0113LCQG using the block diagram from figure 5 with a +25dBm square wave LO.

and final test setup IP3 measurements:

32. IF output spectrum of the MT3-0113LCQG using the block diagram from figure 28 with a +25dBm square wave LO.

32. IF output spectrum of the MT3-0113LCQG using the block diagram from figure 26 with a +25dBm square wave LO.

There is an improvement of 1.5dB in the measured mixer IP3. This is a 40% improvement in the measured value over the quick and dirty mixer measurement.

If the same test setup is used but the mixer device under test from the MT3-0113LCQG to a T3-03MQP or any other mixer, what can be seen is that the dynamic range of the spectrum analyzer was reached or very close to being reached. While the mixer IP3 measurement has improved, there is no such improvement that can be made without using a better spectrum analyzer.

If we test the H diode version of the mixer which differs from the L diode version of the mixer based on the forward turn on voltage of the mixer’s diodes, the MT3-0113HCQG, we see that the resulting measurement is the same.

33. IF output spectrum of the MT3-0113HCQG using the block diagram from figure 28 with a +25dBm square wave LO.

33. IF output spectrum of the MT3-0113HCQG using the block diagram from figure 26 with a +25dBm square wave LO.

What can be inferred from the H diode’s measurement is that for this IF frequency ~300MHz, the spectrum analyzer can only measure the up to a ~+25dBm OIP3. The spectrum analyzer hit its dynamic range limit. It is an expectation that a higher barrier diode variant of the same mixer (same circuit typology, RF/LO baluns, wiring, etc.) to have better linearity than its lower barrier version. The higher barrier diode version of a mixer should have a an IP3 of 20\log{\frac{V_{F_H}}{V_{F_L}}} dB higher than the low barrier diode mixer; where V_{F_H} is the forward turn on voltage of the high barrier diode and V_{F_L}is the forward turn on voltage of the low barrier diode. This can be verified against the datasheet. For further comparison for the argument of which diode variant should be superior, measurement of identical mixers with 2 diode variants such as the MM1-0626HSM and MM1-0626SSM should be explored.

Both MT3 variants’ corresponding datasheets agree that for the same test conditions on a VNA, the measurement of IP3 at 6GHz RF differs by approximately 3dB; the datasheets can be found here for the H diode version and here for the L diode versions.

For a T3 style mixer, the difference in diode version matters to a lesser extent, a similarly strongly driven L and M diode T3-03, for instance, will typically have a very similar IP3 value. This also holds true for the MT3s. To improve a mixer’s IP3, circuit topology in addition to diode forward turn on voltage matters.

Conclusion

A +5dBm improvement in the 2 tone RF input IP3 is a lot. It’s a 10dBc improvement in the IMD3 spur suppression relative to the IF down converted fundamental frequency. For very specific circumstances, where the front-end mixer of the spectrum analyzer is in a “sweet spot” this improvement can greatly improve an IP3 measurement.

The time and capital consuming test setup for high IP3 measurements can be completely unnecessary. The ability to do the measurement depends highly on the dynamic range of the 3 synthesizers and the spectrum analyzers used. For the measurement provided in this document, it appears the front-end mixer of the spectrum analyzer is the limiting factor stopping the measurement of mixer IP3 values higher than ~+31dBm. This is evidenced by the decreased measurement resolution when spectrum analyzer input attenuation is removed and the disagreement between the measurement of the high barrier diode mixer and with its sister low barrier diode variant.

For a normal, or even good, mixer measuring the IP3 using a quick and dirty setup is not punishing. Results agreeing to within 1.5dB against the published datasheet is very good considering the time and capital costs (approximately $150K USD) of setting up a proper IP3 measurement station. Results are very much acceptable using a bare-bones test setup. In order for any of the solutions in this document to be of use, the dynamic range of the measurement equipment must be higher than what was used to gather the presented data and the DUT mixer must have an extremely high IP3.

The RF/Microwave parts used for the test station can be found here for high isolation power dividers and here for high suppression filters. Look forward for high >+35dBm IIP3 highly linear MMIC T3 mixers covering high frequency bands in the near future. Contact info@markimicrowave.com for more information regarding all products.

6 Ways to Make an N-Way Power Splitter

N Way Branched Resistive Power Divider

One of the unique products that we have at Marki Microwave is our broadband, high isolation 3-way and 4-way power dividers. In this blog post we will answer some common questions we receive, including:

  • How to make a 5 way power divider
  • How to make a 6 way power divider
  • How to make a 7 way power divider
  • How to make an 8 way power divider
  • How to make a 10 way power divider
  • How to make a 12 way power divider
  • How to make a 16 way power divider
  • How to make a 32 way power divider
  • How to make an way power divider

How to Make PAM4, PAM8, and PAM16 Signals

Yesterday I wrote about how it was possible to create a PAM4 signal using a Wilkinson power divider. Our Wilkinson product line also includes more rare 3 and 4 way power dividers, which means that we can combine more than two signals together, making higher order amplitude modulation possible.

To review, here is how to make a four level pulse amplitude modulated (PAM4) signal from a single PRBS generator:

PAM4 setupYou just take a PRBS signal, decorrelate it from itself, attenuate one signal by 6 dB, phase align them, and then recombine them. It’s really pretty easy if you have all the components. The other option is to use two data sources with the amplitudes already set to the correct value, with common clocks to align the signals in time. Here is what you get:

pd0220

So how to make a PAM8 signal? Just add another PRBS signal, but this time it has to be attenuated by another 6 dB. To recap, a PRBS signal, a different (or uncorrelated) PRBS signal with 6 dB lower amplitude, and another different PRBS signal with 12 dB lower amplitude walk into a bar, and out comes a PAM8 signal:

PAM8 setup

As long as the bar is a three way wilkinson power divider, you get this:pam8 8gbps

Not that great. It is a PAM8 eye diagram for sure, but it doesn’t look that nice. This is because the phase decorrelators I was using had a lowpass response, as well as the multiple reflections present in my combination system. So it’s possible to do some crude testing using this setup, but I wouldn’t recommend it for any postdeadline OFC papers.

Personally I don’t like PAM modulations, especially above PAM4. At least with PAM4 you get twice the number of bits, but with PAM8 or god forbid PAM16, you are sacrificing a big chunk of your noise margin (remember the spread on each of the eyes is gaussian) for an improvement of only 50% or 33%, not an order of magnitude or anything like that. I think it is much more promising to use either the phase of the optical carrier itself or an electrical carrier on the signal for doing IQ modulation in interesting ways. Speaking of IQ modulation, we sell some excellent IQ mixers that would be perfect for that. Contact support@markimicrowave.com to learn more.

 

Yes, Wilkinson power dividers also work for combining data

After investigating and concluding that yes, Wilkinson power dividers work for splitting data, the natural question was whether they work for combining data.This is a more complicated question for splitting data. For one thing, the role of isolation was unclear. Also the implications of the return loss increasing at lower frequencies was unclear. Furthermore, I didn’t know when anyone would want to combine two baseband signal together.

Today I found answers to all of these questions. To accomodate increasing datarate requirements, signal integrity engineers are transitioning from on-off-keyed (OOK) to pulse amplitude modulated (PAM4) signals. The difference is that OOK uses two signaling levels (for 1 and 0), while PAM4 has 4 levels (for 11, 10, 01, and 00). Therefore PAM4 can transmit 2 bits per symbol in the same amount of time OOK transmits 1 bit, thereby doubling the datarate.

In order to create a PAM4 signal using OOK test equipment, the technique is to attenuate one signal by 6 dB in power (to create 1/2 the voltage signal) and combine it in phase with another signal using a power combiner! Here’s my chance to test my theories. Would a Wilkinson create an intelligible eye diagram using PAM4, or would it highpass filter it somehow? What would I see?

First off, here is the PAM4 signal using a resistive power divider, the PD-0030, running at 13 GS/s. All eyes are taken with a 2^31 PRBS pattern with significant low frequency content.

pd0040

There is some distortion, but the eyes are open. The amplitude of the middle eye is different from the outer eyes because the amplitudes were not tuned to be perfect, and I’m not sure what the optimal amplitudes (they need to be optimized with respect to the noise in the system). The distortion could come from the return loss in the power divider, the lack of isolation in the power combiner, or the high pass filtering from the cables used to create the signal. I didn’t investigate to find out.

Now here is the output from a Wilkinson, the PD-0220, of the same signal:

pd0220Booyah. Not only is the eye open and legible, it looks better than the PD-0030 eye. It has a higher amplitude due to the lower loss of the Wilkinson, and it might also look better due to the improved isolation reducing some of the reflections. The important thing is that it looks better than the resistive.

Now this is the result of a 2-20 GHz Wilkinson. As I stated, this is a 2^31 PRBS pattern, which at 13 GHz means that most of the spectral power density is below 4 GHz. Maybe the 2-20 GHz Wilkinson was masking the distortion it introduced. So I tried the same experiment with one of our surface mount PD-0434 Wilkinsons in a test fixture:

pd0434As you can see, it still works. The amplitude is the same as the PD-0220, which is higher than the resistive PD-0030. There is some additional distortion, which may be from the lower isolation of the PD-0434 vs the PD-0220, or it may be from the degraded return loss from the surface mount transition.

What if we go really crazy, and put all of the signal outside the ‘band’ of the Wilkinson? Here’s the PD-0434 creating a 4 GS/s PAM4 signal:

pd0434_4gbpsStill looks beautiful. So there you have it. You can use Wilkinsons for almost any task you have in signal integrity. I prefer Wilkinsons because they have 3 dB less loss than resistive power dividers, meaning that you get twice as much power output from them, and they have isolation, which tends to calm down resonances in a system. If you like your resistive power dividers, you can stay with them. When you are trouble shooting or optimizing your experiment or system, it’s probably worth it to look at both.

 

 

 

Suppression vs. Isolation

In making the datasheets for the first Microlithic frequency doubler (MLD-1640), it occurred to us that not enough has been made about the difference between isolation and suppression.

In mixers and amplifiers, some parameters are expressed relative to the input powers, while some are expressed in terms of the output power, with the conversion loss or gain calibrated out. This includes third order intercept point (IP3), which can be expressed as either input IP3 (IIP3), or output IP3 (OIP3). In general it is better to use OIP3 for mixers, since what really affects the dynamic range of a system is the amplitude difference between the output signal and output spur, expressed in dB relative to the output signal or carrier (dBc). This is illustrated in the table below, where the difference between the T3 and competing mixers is even greater when the superior conversion loss of the T3 is considered.

Mixer IIP3 Conversion Loss OIP3
T3-05 33 dBm 6.5 dB 26.5 dBm
Imitator 1 25 dBm 10.7 dB 14.3 dBm
Imitator 2 30 dBm 9 dB 21 dBm

Note that it is better to use IIP3 in amps, for the opposite reason, namely that you want to give the amp credit for it’s gain. So in parts with gain the appropriate measure is IIP3, while in parts with a loss the appropriate spec is OIP3.

When the same logic is applied to spurious products in mixers and multipliers, the input referred value in dB is called isolation, while the output referred value in dBc is called suppression. Suppression is the preferable number to use, because it expresses the important value to the system. The isolation can always be improved by increasing the conversion loss of the mixer or multiplier, but this is obviously undesirable. There are, however, some issues using suppression.

The first comes with mixers. In all mixers we express the spurious output of the LO in terms of isolation, since it is dependent on the input LO power. Since the LO power does not change the conversion loss referenced to the input, this means that the suppression can vary by several dB with different LO drive levels.

The second complication is that the input signal, converted signal, and spurious tone are all at different frequencies. For example, when using a doubler with an 8-20 GHz input range, the output doubled frequency is 16-40 GHz, and the undesired tripled frequency is 24-60 GHz. This means that the isolation curves look like this:

MLD-1640 isolations

 

While the third harmonic suppression looks like this:

MLD-1640 third harmonic suppresssion

 

The curves look slightly different. The suppression is more stretched out, and distorted by the curve of the conversion loss of the doubler. This is the result of the suppression calculation, which can basically be thought of as a 1:1 mapping of the isolation through the conversion loss.

Marki Microwave Passives in a Single Chart

I was thinking about the difference between power dividers, baluns, and couplers, and realized that they could all be thought of as power splitters. The characteristics that make them different are the relationship between the outputs in terms of amplitude, phase, and attenuation between outputs. Here is a brief chart that explains them all:

Passives Overview Chart

Phase Delay vs. Group Delay

When you master phase, you become like a God, capable of performing wonders that mere mortals can only dream of. Wonders like making laser beams (using phase engineered quarter-wave reflectors), communicate tremendous information great distances through thin air (all modern communication formats use both amplitude and phase), and create amazing products (balanced amplifiers, balanced mixers, phased array antennas, Mach Zender modulators, the list never ends).

BUT…phase is the hardest thing to understand in microwaves, RF, and photonics. It is hard to measure, hard to visualize, and makes some very confusing homework problems that kept me in the late night coffeeshops of Champaign-Urbana well past my bedtime.

In this post we will make a dent in the universe of phase understanding by clarifying the difference between phase and group delay, and in the process explain why you can’t match phase with variable line lengths. When you buy a phase shifter, it is sometimes what I would call a real phase shifter, and sometimes what I would refer to as a ‘group delay shifter’. The trombone type variable delay lines (we like the ones from sage) are actually variable time delay elements, and not phase shifters.

A group delay (or time) shift is easy to understand: it is how long the pulse (or wave) takes to arrive at your measuring receiver. Differential delay is therefore the difference in how long it takes for two pulses or waves to arrive. In passive components it is just the distance divided by the speed of light (or whatever your wave is) at your frequency in your material.

Phase is much more difficult. It is the integral of group delay over frequency (plus an offset), or differently the group delay is the derivative of the phase vs. frequency. This is why filters can be used as time delays; the edges of the filter have significant phase variation that leads to significant group delay variations over a narrow bandwidth (this is called Kramers-Kronig relation).

A variable length delay line, therefore, can only change the phase by changing the group delay. But by changing the group delay, you are changing the integral (slope) of the phase vs. frequency. This means that the phase change will be different at different frequencies. This is very different than what you get from a quadrature hybrid coupler, or a balun, where the phase shift is constant across frequencies. The difference is shown below. First is a plot of the phase difference between the two outputs of a BAL-0520 Balun (180°), a QH-0226 quad hybrid (90°), a coupler plus two 37.5° Schiffman phase shifters we developed as a custom (165°), a PD-0220 wilkinson power divider (0°), and a PD-0220 with an extra .570″ adapter on one side (variable).

Phase Differential of various components

As you can see, the phase is flat across the bandwidth of the device for everything except the PD-0220 with the extra delay line (adapter). This has a rapidly changing phase across frequencies. If we take the derivative of this we should get the group delay, but instead I measured the differential group delay with the PNA-X.

Differential Group Delay of Various Components

 

Here you can see that the differential group delay between outputs for each of the devices is 0, except for the power divider with the adapter, which has a flat constant group delay (ignore the big hump, I think that is from the calculation the PNA is doing with the phase flip).

So what is the lesson? You can phase match two outputs using a variable delay line, but only at a single frequency. Otherwise you have to do it with a coupler, a balun, a Schiffman, or some other true variable phase circuit.

Yes, Wilkinson power dividers work for splitting data

The reactive power splitter is like Rodney Dangerfield: it gets no respect. Often people will resort to the primitive, high loss resistive power divider simply because a Wilkinson is specified over a limited band, not down to DC. Don’t get me wrong, resistive power dividers have their place. They are much cheaper, anyone can make them, and they can cover very wide bandwidths when made properly.

The truth, however, is that a Wilkinson will work for one application very well outside of the specified bandwidth. The application it will work for is splitting a signal into two well matched loads. While it is true that you can’t have a reciprocal, lossless, and matched three port device in general, you can have such a device if the output ports satisfy one condition: they must be common mode or differential mode. That means that the signals need to be identical to each other (common mode) or opposite of each other (differential mode).

This happens all the time when someone is trying to split an incoming signal into multiple well matched loads. This is why our baluns work so well for so many people. While they don’t have any isolation, and the return loss from a single ended signal put into one output port is terrible (3 dB intrinsically), they are usually used by people to input data to a chip for testing, or combining the differential outputs of a chip. In this case the return loss looks good all around. In contrast, if they were used to combine identical outputs from a chip, then it would be pure reflection. A balun looking into an in phase power divider just looks like an open circuit.

This brings me to the Wilkinson. Outside of the specified band, the Wilkinson will have no isolation and work just like a reactive power splitter. It will have 3 dB nominal splitting loss, but no additional loss (unlike a resistive power divider). If you are using it to split an incoming sine wave, it will work beautifully, as long as the frequency isn’t very, very low (below tens of MHz). Since the group delay and insertion loss are flat, it can also be used to divide data.

Here is the eye diagram from our PD-0020, a resistive power divider:

pd0020

 

 This is from a 10 Gigabit per second, 2^31 length psuedo-random bit sequence (PRBS) pattern, so there is significant low frequency content. The input eye is saved in the background. As you can see, the output is very clean, but with significant attenuation (1/2 the voltage). This is typical of a resistive power divider.

Next up is the output from a PD-0140. This has a 40 GHz high frequency cutoff, well above what we need to pass 10 Gb/s data, but the 1 GHz low frequency cutoff is high enough that some data will fall beneath this frequency. Here is the output:

pd0140

The eye looks pretty open. There is some small overshoot associated with some group delay variation, but nothing too bad. The inside of the eye is wide open. If the low frequency content was compromised, we would see baseline wander. Since the eyebrows are just as narrow as with the PD-0020 case, we conclude that there is no significant low frequency content degradation.

Fair enough, but maybe the data slipped under the 1 GHz limit. Maybe the part has a conservative spec. Next we try the PD-0218, a Wilkinson power divider with a 2 GHz low end cutoff, well above a good amount of frequency content in a 10 Gb/s signal:
pd0218

Once again, narrow eyebrows, no low frequency content distortion. Once again, there is some wiggle in the eyebrows due to group delay flatness. It is true that the group delay will not be quite as flat in a Wilkinson as a resistive power divider. This is an inevitable result of the fact that a resistive PD is just shorter than a Wilkinson, and the impedance transformation is performed resisitively instead of reactively. The tradeoff is that the voltage out from a Wilkinson is .707 times the input, vs. .5 times the input for a resisitive.

Is there some point at which the low frequency content will eventually catch up to us? Yes there is. At some point the Wilkinson’s reactive impedance match will no longer work, and the input sees a 100 ohm load that it just reflects from, causing significant problems. This frequency is very low compared to the operating frequency of the power divider, though. To illustrate, here is 100 Mb/s data passed through the same PD-0218:
100mhz

At this point you can see some degradation in the eyebrows due to the Wilkinson power divider. So if you are trying to push data from lower than 100 Mb/s through the same system as 40 Gb/s data, then you’ll have to use the resistive.

 

 

How to tell when a spur will matter

When I first entered the world of mixers, customer concerns about spurs were very cryptic. When one spur mattered against another seemed totally arbitrary. Over time I learned that certain spurs matter in certain situations. Here are those situations:

1 LO x 0 IF/0 RF (LO-IF/RF Isolation): As we discuss in the mixer basics primer, this matters all the time, but particularly in an upconversion with a low frequency IF. In this case the LO will need to be filtered from the nearby RF. This is actually a good reason to use a lower frequency IF before final transmission, since that makes the final stage filtering easier. The LO – IF can also be a problem in conversions with a high IF.

1 RF x 0 LO (RF-IF isolation): This isolation, like the LO-IF isolation, is important in conversions with a high IF frequency. For example a DC-6 GHz up/downconversion to 7-13 GHz will need the RF/IF filtered out of the output.

1 LO x n IF: This spur is important in the same situation: upconversion from a low frequency IF. In this case the LO will be on the inside of the LO + IF and LO – IF, and the other side will be the LO + 2 IF or LO – 2 IF, as below:

SAM_0105This is an upconversion of a 20 MHz IF with a 6 GHz LO using an ML1-0220L. You can see that the LO is suppressed due to the RF-IF isolation, and the 1 LO x 2 IF is suppressed due to the balance of the mixer.

2 RF x 2 LO (also m LO x m RF): This is the brother of the 1 LO x n IF, but for downconversions. In a downconversion to a low IF, the 2×2 will show up at double the IF frequency, and requires a low pass filter between the IF and 2IF.

m IF x 0 LO (Harmonic IF Isolations): This is important for medium to high level IF frequencies translated to low to medium level RF frequencies. For example, a 1 GHz IF translated to a 3 GHz RF will have to contend with the 3xIF harmonic, which is probably fairly strong. Odd harmonics, in general, are stronger than even harmonics.

n LO x 0 IF (Harmonic LO Isolations): These are generally only a problem for band conversions, usually for satellite work. If the LO is lower than the IF or RF, then it becomes a problem similar to the harmonic IF isolations, but worse because the LO is stronger. The 2LO can be a problem for high IFs as well.

2LO x n IF: These are an issue when doing a low side upconversion. The LO – 2,3,4 IF will cross the fundamental when the IF increases high enough, as shown below (from our spur calculator):

2LO spur examples

 

n RF – (n-1) LO: These will show up in a typical downconversion with a high side LO:

High Side Downconversion

n LO – (n-1) RF: These, similarly, will show up in a low side LO downconversion:

Low side downconversion spurs

 

Other spurs will show up in many disparate situations, with differing levels, particularly in unusual conversions. One conversion that you haven’t seen mentioned much in this blog entry? The high side upconversion. This is because the high side upconversion is the frequency translation with the fewest spurs:

High Side Upconversion

 

In this case the only interfering spur is the 3rd harmonic of the IF, and that is relatively low level. The high side upconversion causes a reversal of the frequencies, so we recommend a heterodyne system with a high side upconversion to a high IF, filtering, andd then a downcoversion to a lower IF for final processing. The other downside? You need a wideband mixer to use the high frequency LO. A wideband mixer much like the ones sold by Marki….

 

What is the power handling of this device?

We get this question a lot: how much power can part XYZ handle?

Power handling is a difficult topic, because the ways in which a device can fail depend so much on the operating conditions that it is subjected to.

We specify the max power on (for example) the PD-0165 as 1 watt only to be extremely conservative. Here are some use scenarios for the PD-0165 and the power handling I would estimate:

– Ideal use case: 50 ohm matched at all ports, using the device as a power divider. In this case the device is only dissipating the excess insertion loss. Depending on the heat sinking it has attached, it should be able to handle 10s of watts of CW power or more at 43 GHz. At a high enough power the connectors will fail.
– Worst case CW performance: Out of phase reflections at both output ports, or use as a power combiner with two signals that are 180° out of phase. In this case all the power will be dissipated in the isolation resistors, which means that the power is limited to what the resistors can dissipate. This is where the power handling will be limited to about 1 W before the resistors pop.
– Pulsed case: In this situation the power is limited by the voltage breakdown in the device. If the peak power is high enough the voltage will break down the dielectric either in the connectors or the substrate, this isn’t clear. The amount of power it can take depends on the pulse width and hence the peak power.

So the amount of power that you can put through the device depends on how you are using it and how much heat sinking you provide to it.