Odd AmpView

WD4BBQ
Posts: 6
Joined: Thu Aug 19, 2021 12:21 am

Odd AmpView

Postby WD4BBQ » Thu Apr 28, 2022 9:51 pm

I just set up Pure signal on the 7000DLE MKII.
I am running a Henry 3KA at around 1100W output to an XDC-4SO with and additional 10 dB attenuator.
The following picture is what I got with AmpView using Two-tone. I was expecting more of a football shape.
It seems to be overly wide in the top right area. Is this shape indicative of anything I am doing wrong?

Pure.jpg
Pure.jpg (690.54 KiB) Viewed 1895 times


I appreciate all the help!!

73
WD4BBQ
Walt
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Odd AmpView

Postby w-u-2-o » Thu Apr 28, 2022 11:05 pm

That shape shows that you are absolutely hammering the amp, your drive level is way too high.

At 70% of your selected drive level the Henry becomes fully saturated and can deliver no more power. This is a badly nonlinear condition. It is impressive that PureSignal looks to be dealing with it to a large degree.

I realize that the Henry ought to be good for at least 2KW. It would appear that your amp may be out of adjustment/broken/needs new tubes/etc.

I'm a solid state guy. Maybe a hollow state person will chime in here about what might be going on with your amp. In the meantime definitely don't run it that hard. A little bit of compression is ok, but any more than a little is pointless because PureSignal will fight you you tooth and nail, i.e. the more you turn it up the more PS will turn it down.
User avatar
w9ac
Posts: 290
Joined: Sun Apr 09, 2017 4:01 pm

Re: Odd AmpView

Postby w9ac » Fri Apr 29, 2022 12:00 am

A pair of 3-500Z tubes in a grounded-grid configuration will require much more drive power than comparable ceramic tubes. Still, you should get better linearity with optimal plate loading.

I've attached AmpView screen captures from my Alpha 77Dx grounded-grid amp. The first image displays correction when the amp is underloaded. It looks much like your AmpView display.

The second image shows correction when observing AmpView while adjusting the plate loading control for minimum degree of correction. Both screen captures show correction when the amp is delivering 1500W into a Bird Termaline load.

A few questions: Do you see the same result on all bands? What's the amp's maximum output power while remaining under a safe value of control grid current? How much plate idling current do you see when the amp is keyed but not driven with RF? The manual states that normal idling current is 170 mA to 220 mA in the higher B+ SSB position. This is a quick indication that the amp is biased correctly in the SSB position.

The 3K-A is a straightforward grounded-grid, dual 3-500Z design. When the amp is keyed, a Zener diode in the directly-heated cathode sets bias into class AB1. When unkeyed, it's biased into cutoff.

According to the schematic, there's no EBS so memory effect should be a non-issue. The input section uses a separate Pi network per band. Hopefully they were designed physically close to the cathodes. With a grounded-grid design, it's important for low IMD because the input impedance into the cathode changes over the RF cycle. The trick for a designer is keeping Q high enough for the Pi input network to function as a flywheel, but low enough to present a low input SWR across a given band.

Paul, W9AC
Attachments
Alpha77Dx-1KW-Underloaded.jpg
Alpha77Dx-1KW-Underloaded.jpg (53.41 KiB) Viewed 1876 times
Alpha77Dx-1KW-Optimal Loading.jpg
Alpha77Dx-1KW-Optimal Loading.jpg (51.51 KiB) Viewed 1876 times

Return to “PureSignal Operations (PowerSDR & Thetis)”