Did some experiments with putting an isolator in front of the LNA and trying to subtract out its loss. It definitely improved the bumpiness on the Y-factor measurement.

Oscillations are a possibility, when we started taking data the NF was like 10 dB but the amp was singing like crazy. Capacitors and resistors seems to have fixed that.

I will look at the VSWR uncertainty calculation...

I think part of the larger picture is that just maybe the noise figure shown in a MMIC data sheet was measured on-wafer, and the user has to suffer some loss mounting it to a circuit board. For yucks we might but a Qorvo CMP2626 LNA eval board

https://store.qorvo.com/products/detail/qpa2626eval-qorvo/597407/

But it does not appear to have a "through line" to calibrate out the board losses.

Not sure why it would be a Doherty amplifier, it looks like a simple two-stage power amp to me. Two FETs driving four FETs. Could the input and output be through the board? Maybe send a photo of the board flipped over?

That is an interesting board! I found an article where those transistors were used at L-band to make power amplifiers

https://cds.cern.ch/record/180485/files/CM-P00059135.pdf

Ignore this advice if the amplifiers are in coax packages....If you are talking about cascading MMIC chip amplifiers, 60 dB gain is enough so that a little radiation from wirebonds or microstrip can cause positive feedback and then oscillations (You need to have some space between them and preferably channelize the signal path with small enough cross-section to cut off any waveguide mode where gain is available (not a problem when you are at 2 GHz!) In a pinch some RF absorber over the top of the MMICs can do wonders, but I believe that if you put absorber over the first MMIC you might degrade the noise figure. Good luck and let us know how you solved this issue... even better if you supply some photos and text for a Microwaves101 page!

We will be "presenting" at one of the Young Professional events at IMS 2024, I think on Thursday. If anyone wants a Microwaves101/MTTS T-shirt, we will have them there. Stop by the booth any time during the week and use the password "Afghanistan Bananastan" and we might be able to hook you up with some merch if you can't make the YP event.

And there is little chance of asking something anonymously on LinkedIn. But your peers can admire the professional glamour shot that you posted of yourself to comply with your company's social media directives

https://www.linkedin.com/in/steve-huettner-12817a77/

It seems like answers to questions that are posted are routed to just the question asker, and not posted as a comment. That does not seem like the best way to build a community...

Steve

Glad to see you there! It turns out we were too late to organize a livestream. Instead they may allow us to have a booth a Women in Microwaves or Young Professionals event which of course is members only and not what we had in mind. But it looks we still have time to launch an MTT-S Microwaves101 tee-shirt to give away... We will try to get some interest in a live stream for 2025 IMS, however, the only way to get something like that launched for sure it to get more involved in IMS committees, which is not something we are anxious to do. We'll keep you posted!

Mike... that is an excellent article. At first glance I don't see any mention of the amplifiers load-pulling each other in a reactive combiner. I think that could be a significant effect.

I totally forgot about the forum (message board) being wiped clean. It never really came back after that, people moved on. Here we are, the proverbial tree falling in the forest.

Thanks for your thoughts!

Steve

That is a great question! I took a look around, some designs just show a three-port box with labels...

Maybe a picture like this?

https://en.wikipedia.org/wiki/Orthomode_transducer#/media/File:Orthomode-transducer_01.jpg

Sorry I am not a huge help, I'll try to ask an antenna person.

Steve

Thanks

there will likely be some routing between the combiner and the amplifiers, so I consider the phase angle of the failure unknown, and possibly a strong function of frequency. The failure could be a short circuit somewhere back inside the amplifier, or a blown wirebond, or almost anything. For sure, adding the isolation resistor will help the remaining amplifiers see a decent match, with the caveat that the resistor does not blow up...

I will take a look at the reflection coefficients that are seen by constituent amplifiers in a combiner, when the amplifiers have random phase and amplitude errors, and see what that tells me..

Not sure I understand where this comes from

Z= Zo x sqrt(mu_r/epsilon_r)

But I'll try to give it some more thought!

Thanks
Steve

Thanks! Someone should be getting in touch...

You know, I have been looking for an excuse to visit them... I'll try to get over there and ask. Need to be sensitive about anything proprietary of course. Thanks for asking!

Thanks! That is a very active message board, I will peruse it. But I don't want to take on a project like this, it might end up in forgotten pieces as my attention span is pretty short these days. I wonder if Shariar at Signal Path (on Youtube) would do contract work or knows someone that does...

Steve

Thanks, I will add a link to that info. I know who to call when I need help on de-embedding!

Great news on the scikit-RF shirt my old one is in tatters and stained from doing an oil change. Will order a new one.

Steve

Is this for waveguide?

Turns out MSE Supplies in Tucson can thin SiC wafers if anyone else is interested

Thanks

My friend
You have pointed out the weak part of Wilkinsons in SSPAs.... the resistor is going to blow up!
Steve