Mismatch transmission losses and effects on maximum power output of SSPA

[webharvest]

This is a weird way of looking at matching losses so I want to confirm I am looking at this properly. Normally you would look at it as reducing RF output power.
I have tuned my solid state power amplifier (SSPA) so I have a pretty good match (1.5:1 worse case). The antenna it is connected to has a 2:1 mismatch. Now this is were it gets interesting; we only made power output measurements with a 2:1 sliding load connected to the output and rotated the angle of the load to get a minimum output power measurement. Let's say that measurement is 30dBm. Someone wants to know what the theoretical maximum power of the SSPA could be based on this minimum measurement because we can't retake the measurement to find the angle at which the maximum power is produced. Keep in mind the output of the SSPA has an isolator so the active devices are fairly well isolated from the load and don't get pulled around during the rotation of the angle.
Here are my thoughts:
1) The 2:1 mismatch has a transmission loss of 0.51dB
2) The 1.5:1 mismatch of the SSPA has a transmission loss of 0.18dB

When we made the first measurement we intentionally rotated the angle of the load to find the worse case match or minimum Pout point. If you connect a 2:1 sliding load again is it correct to say that you could theoretically find a match that would produce a power output of 30dBm + 0.51dB + 0.18dB = 30.69dBm?

Can you explain either way?

Thank you ahead of time for any help you can provide.

 

[VEBill]

Some points for your consideration.

Your mention of sliding loads and isolator, as well as perhaps your choice of forum ('Microwave...'), implies that this is a microwave transmitter. Therefore transmission line losses cannot be ignored - especially when you're accounting for the last 0.01 of each dB. The transmission line loss will be impacted by the VSWR (the standing waves increase the losses on a line as compared to a 'flat' line).

Also, transmission line lengths will have an influence. If the two mismatches are both in the same 'direction', and the transmission line happens to have an electrical length of a multiple of a half wavelength, then perhaps the two mismatches will be more efficiently matched than just one mismatch at one end. There are an infinite number of solutions on the Smith chart that can have wrongs (mismatch), and a transmission line of the 'correct length', make a right.

So the short answer is that the situation is far more complicated than simply adding the two VSWR mismatch losses in a linear fashion. There are other factors at play (a la Smith chart, etc.).

 

[webharvest]

I understand and I should probably add something to make my question a little more clear.
Since the 30dBm is measured with a sliding load and finding the angle when the Pout is at the minimum point if I was to hook up a double stub tuner instead of the sliding 2:1 load could I find a conjugate match that would produce an output that is 30dBm + 0.51dB + 0.18dB= 30.69dBm or would it just be 30.51dBm? I ask because the amplifier source impedance is 1.5:1 and the sliding load is a 2:1.

 

[VEBill]

Your question almost needs something like a 45 minute discussion to explore, even at a high level. Given the limitations of this type of written forum, I'll try to hit some the high points.

As per EE101, your amplifier will (theoretically, and probably practically too) produce maximum power when it is presented with a load that matches its actual Zo. Do you know your amplifier's Zo (as opposed to the 1.5:1 VSWR)?

With RF, you must always keep the Smith chart concepts in mind. The match needs to be presented at the correct location considering the electrical length of the transmission line.

If Zo is not a standard impedance (e.g. 50 ohms), then what do you plan to use as the transmission line and test equipment?

How are you measuring power if the line is not flat? You might want to investigate exactly how your power meter works. Many detect peak voltage and present this as power assuming a 50 ohm flat load.

The 0.18 dB mismatch loss is probably lost forever (the actual RF power was probably never brought into existence for the most likely scenario). But this 0.18 dB is swamped by some of the factors I've already mentioned.

 

[Higgler]

don't forget that VSWR's multiply.
When connecting an amp to antenna, energy will bounce back and forth between the two and with any bandwidth at all, you'll see some real ripples in maximum power out the antenna.
Power transmission Ripple due to VSWR interactions result in ripple spacing across frequency, peak to minimum power(in ghz) will be 8/cable length(inches), assuming teflon cable.

To optimize output power at a certain frequency, change/optimize the transmission line lengths on both sides of the isolatorm. Don't forget teflon cables change thru phase versus temp. 70F to 50F can see significant phase change in a long cable at high freq. ex, 18 ghz, 50 foot cable, 80 degree phase change, from past recollection.