FX Loop Advice for Hot Cat with VVR Build

[bcmatt]

I quickly added a couple 1K resistors. It dropped the preamp voltages slightly less than I expected it would. Displayed in orange.

Garter Hot Cat Preamp Voltages.png

Once I actually wire in the tube for the fx loop circuit (and the "B+D1" node), I imagine that should help pull things down a bit more... Then I can remeasure everything.

[Colossal]

Thanks! I couldn't recall if 360R was your chosen value or left over from the earlier schematic. I might expect your voltage to be around 30-32V if drawing 88-ish mA per tube.

[bcmatt]

Thanks! I couldn't recall if 360R was your chosen value or left over from the earlier schematic. I might expect your voltage to be around 30-32V if drawing 88-ish mA per tube.

Yes, I seem to recall it was about that, but I will remeasure later today and let you know.

But, looking at my previous posts on the build thread, before switching to the garter circuit, I did sample 360R Cathode resistors and got:
"With 360 ohms cathode resistors, it seems to give 404V Plates and 25V and 26V cathode voltages. That seems to put it in the range of 100% to 104%."

Once all 4 resistors arrived and I switched to the Garter circuit I posted that it was hitting 103% dissipation but didn't post the plate or cathode voltages. I just noted the wall voltage was about 123.5VAC that day.

[bcmatt]

So, I measured the current bias. I was wondering if somehow splitting the dropping strings would affect anything.

The EL34 plate Voltages were at 408VDC.
The Cathode voltages were 25V (measured from where the two 360R resisters meet as I was instructed to do; I know it's confusing.) If I measured from the actual cathode to ground it would be 50V but then the cathode resistor would be considered 720R.

So the Plate to Cathode Voltage is considered 383V (408V-25V). Anyways, the tube bias calculator says it's biased at 100.4% dissipation.

[Colossal]

Thank you, Matt

[bcmatt]

I got the loop finished off a bit before going into work late today. I haven't had a chance to try it with any FX yet though. Just made sure it worked and took some voltages. My preamp voltages barely came down with it added, so right before I left, I moved the scaling split to be after the sag resistors. We'll see if that makes much difference. I'll test more tonight with actual fx and see if I can settle on where to leave the two trimpots set. The are both around the middle right now. The recovery from the loop is just set with a pre-set voltage divider of 39K above 220K to ground.
This is the current situation:

Hot Cat Loop.png

I'll take more voltage measurements tonight and play with the dropping resistor values to hopefully get the voltages balanced closer to the schematic voltages...
What are considered safe voltages for 12ax7s?

[bcmatt]

The voltages seem pretty reasonable now. All a bit more than the schematic but I think acceptable:

20250709_182536.jpg

What do you think?

[bcmatt]

Gave this amp a play and I am quite liking it with a loop. I had sort of boycotted FX loops for a few years, but it definitely has its advantages. I backed off both the trimpots before the send a little bit. It's hard to know if I have them dialed just right or not. We'll see if I notice any difficulties at the current trimpot settings.
This is turning into a pretty versatile amp!

20250709_200358.jpg

[pdf64]

The voltages seem pretty reasonable now. All a bit more than the schematic but I think acceptable:
20250709_182536.jpg
What do you think?

The h-k insulation on the cathode follower may be somewhat stressed, I suggest to try to decrease the preamp supply voltages a bit more, eg increase the value of those 1k droppers, perhaps double them or try 4k7, try to get B+4 to 300V.
Also DC elevation of the heater supply seems good practice.

[bcmatt]

The h-k insulation on the cathode follower may be somewhat stressed, I suggest to try to decrease the preamp supply voltages a bit more, eg increase the value of those 1k droppers, perhaps double them or try 4k7, try to get B+4 to 300V.
Also DC elevation of the heater supply seems good practice.

I would like to understand both of those ideas better.

Regarding the B+4 voltage: the idea of 328v on that one plate seems scary. But how does the 193v cathode voltage make it safer? Is this where the DC elevated heaters are required to make the tube feel like it's only like 135v?

From a valve wizard article I just read:
https://www.valvewizard.co.uk/heater.html
It suggests I would maybe switch the ground reference on my artificial heater centre tap to instead be connected to a power tube cathode. Because of the garter system that should be 50v above ground instead of the 25v the power tube feels.. but the VVR would be reducing that all the time so I guess that's a no-go. Also, I don't know how the garter connection to the grids would also ruin things.

What would be an appropriate amount of elevation realistically? Is the s suggested 30-60v really enough for this case? Is the somewhere specifically you would get the DC feed from? I've never done this before.

[martin manning]

Vh-k ratings for tubes range from 100-200V, so ideally you would want to bring it down to 100V. You can add a divider to get 50-60V with a cap to ground (from the center) off any one of the supply nodes. There are limits to Vh-k positive too (check data sheets), so you wouldn't want to make the elevation voltage too high. Maybe a combination of lowering the plate voltage and elevating the heater voltage is the best idea, I think preamps sound best with something less than 200V on the plates.

[lonote]

Is the somewhere specifically you would get the DC feed from? I've never done this before.

I have been using this little add-on (red components) from Sluckey for both heater elevation & draining the reservoir caps after shutdown. Guessing you would (probably?) install this on the non-VVR side of your design, which would negate the main cap drain feature.

Sluckey heater elevate.jpg

[pdf64]

... Regarding the B+4 voltage: the idea of 328v on that one plate seems scary. But how does the 193v cathode voltage make it safer? Is this where the DC elevated heaters are required to make the tube feel like it's only like 135v?

There's an anode voltage limit of 300V, but that assumes the cathode is at 0V. Whereas the cathode here is up near 200V, so the anode (to cathode) voltage limit isn't remotely stressed.

But there's also a heater to cathode voltage limit. Because the cathode is tubular, its heater is inside the cathode tube, and the insulation between them is by necessity very thin, has a limit of <200V, that's DC + signal peaks.
So if there's a cathode follower with a high cathode voltage, the heater circuit voltage really should be elevated, so as to reduce the voltage differential between heater and cathode.

Is the somewhere specifically you would get the DC feed from?

You're right, it needs to be unaffected by the VVR.
I suggest taking a 220k 47k potential divider from the B+4, that will help to pull its voltage down a bit, which is all to the good.
The 47k needs bypassing, eg 10 or 22uF.
See limiting values p5
https://frank.pocnet.net/sheets/010/e/ECC83.pdf

[bcmatt]

Thanks so much for chiming in everyone!

I didn't have any 2W 47K resistors so I experimented with a 68K. Is this excessive? It gave me 72VDC elevation for the simulated heater ground. I can do some parallel or series resisters to get a different value if desired.
Here are my new voltages in green with the 4k7 dropping resistor as well:

Voltages with Elevated Heaters.png

[bcmatt]

I added a parallel 220K Resistor to bring the heater elevation down to +58V:

Voltages with Elevated Heaters.png