Signal looks a bit different with my Alpha 9500

[JJ4SDR]

Switched from using a SS amp (just to maintain the tube in good working order in the Alpha 9500) to the Alpha. I have the Xtronic RF coupler and also a variable high quality RF attenuator in line to use PS with the Alpha.

This isn't comparing apples to apples necessarily, but when I two-tone and look at the Panadapter during the calibration I see something I'd not seen before (it could well be that I did see this also before, but didn't pay attention to it and attribute what I see to the Alpha ). Also, the skirts on my TX bandwidth aren't as clear-cut (depending on the band/antenna I am on) as they were when I used my RF2K-S SS amp with a built-in ATU. It has a built RF coupler and the RF tap provides a - 55dB attenuated RF sample to the ANAN 7000 DLE MKII with PS working well .

Q1: The RF2K-S has an ATU and I did need it switched on on some bands. It does a great job getting my SWR very close to 1:1.
The Alpha 9500 doesn't have an ATU but it does automatically tune itself (after you tune the different bands segment by segment first) based on the tuned segment settings. This way optimal power transfer is ensured.
Does the increased SWR on some bands (especially the 40m band in this case) perhaps account for "dirtier" skirts?
The Amp View looks good (on all bands I have tested the set-up) so I guess I am good.
Q2: I know that anything below -75 that shows up when seen with DUP ON (I have it always ON) is not going to be heard/seen by anyone.
Despite the above, I am wondering if there is some cross-talk inside the RF coupler (RF sample is extracted from the "Forward" tap on the coupler) between the "Reflected" and "Forward" (given higher SWR on some bands) circuits as I am seeing "dirtier" skirts with this set-up compared to how it looked before, using the RF2K-S and its built-in RF coupler.
Q3: I assume that it doesn't make any sense to add/subtract attenuation to see if I clean up the TX signal even more as visualized on the screen as I see something at about -70? Diminishing returns and nobody would hear or see the "improved" TX signal anyway.
I started off at around -58dB (just to be "sure") which is 3 dBs of extra attenuation although amp is putting out the same power as the SS amp did.
Q4: I am seeing 133 in green. Is there a thread that talks about these numbers and if there is a "sweet-spot" that I would need to shoot for ideally? Perhaps this is an exercise in futility, but I would just want to better understand what I am seeing and why?

73,
Juha
UPDATE (not sure why the graphic looked a bit "funky" , here is one that looks a tad better.)

[w-u-2-o]

This isn't comparing apples to apples necessarily, but when I two-tone and look at the Panadapter during the calibration I see something I'd not seen before (it could well be that I did see this also before, but didn't pay attention to it and attribute what I see to the Alpha ). Also, the skirts on my TX bandwidth aren't as clear-cut (depending on the band/antenna I am on) as they were when I used my RF2K-S SS amp with a built-in ATU.

Of course you are going to see a difference in performance and in the character of the residual IMD products. The two amp's are not the same, don't have the same non-linearities, and therefore any residual IMD products will be distributed differently.

Q1: Does the increased SWR on some bands (especially the 40m band in this case) perhaps account for "dirtier" skirts?

Yes, the output impedance mismatch on the amp will cause more IMD.

Q2: I know that anything below -75 that shows up when seen with DUP ON (I have it always ON) is not going to be heard/seen by anyone.
Despite the above, I am wondering if there is some cross-talk inside the RF coupler (RF sample is extracted from the "Forward" tap on the coupler) between the "Reflected" and "Forward" (given higher SWR on some bands) circuits as I am seeing "dirtier" skirts with this set-up compared to how it looked before, using the RF2K-S and its built-in RF coupler.

The answer is "no". A coupler does not have crosstalk. You can add a 1W, 50 Ohm terminator to the unused reverse port if you are concerned but that should make no difference.

Q3: I assume that it doesn't make any sense to add/subtract attenuation to see if I clean up the TX signal even more as visualized on the screen as I see something at about -70? Diminishing returns and nobody would hear or see the "improved" TX signal anyway.

It's always worth optimizing the feedback level. The short answer is shoot for an S-ATT of 25 or even a bit higher. Long answer here: viewtopic.php?f=20&t=2384

An inexpensive 10dB (1dB steps) rotary step attenuator off of eBay can be very handy if you are constantly changing amp's.

Q4: I am seeing 133 in green. Is there a thread that talks about these numbers and if there is a "sweet-spot" that I would need to shoot for ideally?

I have no idea what that number is because it is unscaled. The S-ATT value the auto-attenuate algorithm comes up with is the best indication IMHO.

Perhaps this is an exercise in futility, but I would just want to better understand what I am seeing and why?

You do have a predilection for over-analyzing things but it's hard to fault a thirst for knowledge

[JJ4SDR]

Scott,

As always, I appreciate your "interventions."

I have a "rotator" type variable RF attenuator that is indeed very handy. Not something that I got off of Ebay. Don't really care if it is cheap but prefer high quality. Presently, I have 3 tube amps and one SS amp and what I do is I bounce among them at frequent intervals ( tube amps) to keep the tubes in good working order.

Is predilection akin to having a proclivity? If the answer is affirmative then, yes, I have a predilection too, LOL!

73,

Juha
NI2M

[JJ4SDR]

Went through the calibration exercise again (adjusted the total RF sample level down to -54dB). I seem to recall that you had written that a -50dB total sample may be the "sweet-spot." Correct me I got that wrong.

The ANAN step attenuator indicates 25dB during TX on all the bands I operate on except on 20 it shows 23dB.

Juha

[w-u-2-o]

As already written above, the long answer is here: viewtopic.php?f=20&t=2384