Quote:
did a few quick measurements of a GALI-51 eval board this evening (when biased at 65mA) to look at how the OIP3 changes when the GALI-51 is driven quite hard.
I got the results I predicted back in post #82 where the OIP3 degrades at very high output levels. It looks like the 35dBm figure only applies when the output level of each test tone from the GALI-51 is below about +3dBm. If each output tone seen on the analyser is at +10dBm the OIP3 degrades to about +32dBm and with GALI-51 output tone levels each of +13dBm the OIP3 degrades to about +25dBm.
My spectrum analyser is quite old now but it does have a built in OIP3 measurement function so it helps speed up the whole process. I used 60dB internal attenuation for the highest output levels and I also checked the result using an external attenuator. The distortion performance below is definitely that of the GALI-51 and not the analyser. This may explain why you only saw an OIP3 of +30dBm with some of your tests. You may have been driving it quite hard.
All of the tone levels are measured at the output of the GALI-51 as seen on the spectrum analyser.
|
Thank you so much taking time to do this
. Your results look very close to what I would expect.
What are the frequencies of your two fundamentals and their frequency separation? How sensitive are your measurements to the frequency separation (if you have control over it)? As I wrote before I could only get it working by trying different frequency separation of the two fundamental frequencies.
The use of external attenuators is a good way to check if the results are not coming from the built-in signal generator of your spectrum analyzer. Also the intermodulation junks could come from the power combiner itself if there is not adequate isolation between ports.
I thought I drove the Mar-1+ and Gali-51+ both with about -35dbm (can't 100% remember if this was the case)two-tone measurements at the input of the MiMIC preceded by a 10db attenuator, power combiner and the 20 db attenuators to the outputs of the signal generators. Also I was not sure really why the result for the Mar-1+ is ok but Gali-51+ is abnormal using both -35dbm drive levels. I am taking the TinySA results with a pinch of salt, having a bit of fun really.