pure signal w/8000 dle and Alpha 91b

[cys]

I'm trying to get pure signal functioning using Thetis, an 8000 die and an Alpha 91b with 1500W output using the -55 db RF sampler from VK-amps. To date pure signal increases IMD when routing the signal after the Alpha 91b from the RF sampler into the 8000 die pure signal input port. In contrast, pure signal using sampling on the 8000 die alone (ie PS jumper installed), but the 8000 dle output still used to excite the Alpha 91b, leads to a clearly improved signal. I'm judging signal improvements by recording SSB transmissions on remote SDR's.

I've read here that some 8000 dle's have the pure signal ports reversed. Is it advised to try feeding the signal from the external RF sampler into the PS feedback port instead of the PS input port? I don't think this is my problem in that I am seeing "green" coded pure signal feedback.

The other thing that I'd like to ask about is if the issue is likely to be connected with a memory effect associated with the use of an Alpha 91b.

Chris

[cys]

I found this thread that mentions the Alpha 91b uses EBS, here: viewtopic.php?f=20&t=2398&p=2659&hilit=alpha+91b#p2659


It would seem that the way forward is to bypass the EBS on the Alpha 91b. If anyone has done this and has some pointers, that would be appreciated. I'll look at the schematic in the meantime. I know Island Amplifier works on these based on what I've read on the net, so that's another option.

[cys]

I was able to diminish the effects of the EBS on the Alpha 91b by increasing the value of C56, increasing the time constant. This has allowed pure signal to give a better result.

A significant separate problem I'm seeing is that the step attenuator is not outwardly functioning based on what S-ATT is reporting during PS-A use, which is a static value of 31, unchanged. I can't rule out that I've missed some obvious setting. I've seen YouTube videos where S-ATT is changing to an optimal value during transmission using PS-A, so I'm left wondering why this value is static when I do the same.

I'm using a Bird 4273 Thruline WattMeter Sampler that is adjustable to provide the input signal for PS after the Alpha 91b. I also have a VK-Amps -55 dB sampler that I was using before I switched to the adjustable Bird 4273 sampler. The idea was to optimize the sampler dB input and this is when I noticed that S-ATT is invariant despite changes to the sampler dB input. I'm hoping the step attenuator is not damaged in the 8000 dle. Before going down that path I'm going to assume it might be the firmware version in the 8000dle or some side effect of the version of Thetis I'm using (v2.10.3.8). I'm attaching below some images so hopefully someone can note the key misconfiguration that solves all my problems.

Chris

1.png (230.69 KiB) Viewed 2231 times

2.png (189.81 KiB) Viewed 2231 times

3.png (271.73 KiB) Viewed 2231 times

4.png (632.28 KiB) Viewed 2231 times

[cys]

Of course I do the critical troubleshooting control right after sending a help request:

Going to PureSignal with the 8000dle alone by putting the external jumper back in place on the rear of the 8000dle between PS feedback and PS input, and then running pure signal with that configuration leads to S-ATT stepping from 31 to 13 or 14.

I'll try some more troubleshooting with the feedback from the sampler after my linear, the Alpha 91b.

One thing I've seen is that some 8000dle's have been reported to have the PS feedback and PS input ports connected opposite to the labelling. I'm assuming if I see green feedback with PureSignal using the RF sampler feedback that this condition doesn't apply to my 8000dle. I'm kind of afraid to try swapping the feedback input from the Rf sampler to the "wrong" port as a check, though.

Chris

[w-u-2-o]

Not seeing anything wrong. I recommend that the MAP option be unchecked unless it is absolutely necessary for the amp behavior.

It would be safer to use the VK Amps coupler. Any time an adjustable coupler is used there is a danger that damaging levels might be accidentally achieved.

Before going any further, it would probably be best to upgrade that 8000 to Protocol 2 firmware.

Edited to add: you do need to confirm which port is the PS input port. This is relatively easy. You can feed the supposed output port into an external analyzer or receiver to confirm it is producing the output from the internal coupler. Once the internal coupler output port is identified, the other is obviously the input.

[cys]

Not seeing anything wrong. I recommend that the MAP option be unchecked unless it is absolutely necessary for the amp behavior.

It would be safer to use the VK Amps coupler. Any time an adjustable coupler is used there is a danger that damaging levels might be accidentally achieved.

Before going any further, it would probably be best to upgrade that 8000 to Protocol 2 firmware.

Edited to add: you do need to confirm which port is the PS input port. This is relatively easy. You can feed the supposed output port into an external analyzer or receiver to confirm it is producing the output from the internal coupler. Once the internal coupler output port is identified, the other is obviously the input.

Thanks very much for your input and suggestions.

I've gone ahead and updated the firmware to protocol 2.

I also made measurements on the two PS ports. There's signal on both ports, but one has more than the other, so I'll go with the one that has more signal as the PS feedback and use the other port as the PS input.

.