Totally agree. Combine a good process and good people and there is nothing to worry about. This was more of a thought experiment gone awry! Thanks Desert Sage.

I did not consider TDR, thanks madengr!

Note to readers... you need to log in to see the photo!

Steve

and yes, madengr... no access to the ports....

That seems like a very good idea! My concern is for a combiner where they are assembled with eentsie chip resistors, thinking that the solder process could short them out or leave them open...

I think I can answer my own question... the amplifiers should see a good match with a reactive combiner. This is apparent when the amplifier is analyzed in even mode where both ports are excited equally.

It turns out that when you "mix" S-parameters with high-side LO, you need to take the conjugate of S21 and S12, and then group delay comes out as expected. I did not make this up, it is detailed in a paper by Dr. Dylan Williams...I'll post some updates on this shortly....

Thanks to Frauke for pointing this out. Here is the reference:

D. F. Williams, F. Ndagijimana, K. A. Remley, J. A. Dunsmore and S. Hubert, "Scattering-parameter models and representations for microwave mixers," in IEEE Transactions on Microwave Theory and Techniques, vol. 53, no. 1, pp. 314-321, Jan. 2005,

I know that plating plastic is possible, and some companies do it well. But the devil is in the details. Not just in the plating but in holding tolerances on the plastic part. I am considering adding an entry on this on our career killer page... good luck!

Steve

You might try Microsemi

https://www.microsemi.com/product-directory/rf-microwave-a-millimeter-wave/3592-rf-discretes

Their web site states:

"Microsemi also can offer a significant number of existing configurations to minimize your NRE and provides many customers with microwave components no longer available from their original suppliers. Microsemi's extensive library of products and designs gives us the ability to respond quickly with solutions to meet your needs, quickly and cost effectively."

Good luck!
Steve

Uli's list has circular waveguide dimensions:

https://www.microwaves101.com/encyclopedias/rectangular-waveguide-dimensions

I will forward this message post to him to see if he has any advice.

Steve

I think the best consideration is to stay away from minimum widths that would be hard to resolve in manufacturing. Maybe one way to think about this is that the RF currents will be crowded to the edges of the strips no matter what the width. That's all I have..

Steve

Andrea

all good points! Any discussion of very high impedance transmission line should start with the caveat that for all practical purposes you cannot do much better than a few hundred ohms (for example, wirebond high above a ground plane) before the media falls apart and is useless.

Thanks for the discussion!
I am going to answer my own question after thinking about this some more... impedance is defined as SQRT(L'/C').

if a microstrip is infinitely separated from its groundplane, the capacitance would be zero, and it would have infinite impedance! and infinitely useless.

This came about from playing with the wire-over-ground calculator, and observing what happens when the wire is a mile above ground.

https://www.microwaves101.com/calculators/874-wire-over-ground-calculator

This just in.... we have some data on a self-biased amplifier where the current drops when it is saturated. I guess this is yet another good reason NOT to use the self-biased trick on a power amp or driver amp... as the power will be limited by the shift in quiescent point.

Jim... we will put this into the monthly email blast. Hopefully get someone's attention!
Steve

That is too hard of a question for me! Can you simulate it in HFSS? Also, if there was a way to make "air microstrip" I think we could call it TEM at last. Maybe we could use magnetic levitation to hold up the conductor.

Steve

OK, I think I get it. The reason I asked is because of an "Excel mixer" I am playing with. Here is a page on it, it is still a mess

https://www.microwaves101.com/encyclopedias/excel-s-parameter-mixer

when I "mix" and amplifier's S-parameters to another frequency using high-side LO and look at group delay it is negative. But that is not a physical thing, just a figment of a slightly bogus way to simulate a receiver, right?

Thanks

yes this is possible. You need to calibrate the equipment without the LNA, then insert it, and then measure the difference in output signal.

Be careful to NOT saturate the LNA. I recommend that the input signal be 10 dB (or more) below the input one-dB compression. Good question, and good luck!

Steve

You need to make the lines 1/4 wave long at center frequency
You should plot S11, S22, S23 and S24. The rest of the parameters will be repeats of these due to symmetry. You can make a goal of -20 dB at band center for all of them.
The widths of the lines should correspond to 71 ohms. ADS should be able to calculate the line widths and quarterwave lengths for you.
That should get you started

This seems like a homework assignment...

Steve

II downloaded the Grieg and Engelmann paper, the formula is not in it. I also grabbed a few others from the 1950s, but no luck. Grieg and Engelmann were ITT employees, and microstrip was an ITT technology. The authors of the ITT reference data might have had access to memos that we will never find... I might do a little more digging.

Found it in ITT's reference data for radio engineers, 1983, sixth edition, page 24-25.
The formula came from D.D. Grieg and H.F. Engelmann, "Microstrip - New Transmission Technique for the Kilo-Megacycle Range, proceedings for IRE, Vol. 40, Dec. 1952. There are other reference, let me know if they are needed,

I will download that paper later to check it out.

Cool stuff, I like microwave history.

Steve