Williamson amplifier PCB

[Grover Gardner]

[Grover Gardner]

[David McGown]

Grover,

Fantastic results! Congrats!

Where did you get the plastic rings around the holes in the topplate for the 6SN7 and KT66? Those look really handy to dress up the holes cut in the plate.

On the cap across the feedback resistor, how does that compare with what you were getting on the handwired version? I think based on your (or Dave G.) experiments on stability you found that even a 0pF cap worked fine. I think that is what I did on mine. Been too busy of late to put my amp under the scope.

David

[Grover Gardner]

Grover,

Fantastic results! Congrats!

Where did you get the plastic rings around the holes in the topplate for the 6SN7 and KT66? Those look really handy to dress up the holes cut in the plate.

On the cap across the feedback resistor, how does that compare with what you were getting on the handwired version? I think based on your (or Dave G.) experiments on stability you found that even a 0pF cap worked fine. I think that is what I did on mine. Been too busy of late to put my amp under the scope.

David

Those are co-called "2-inch" cable grommets. They actually fit a 1-7/8" hole:

https://www.amazon.com/dp/B0B9R4BL62?re ... tails&th=1

A cheap way to dress up the needed holes for the tube sockets.

The response from the p-to-p version wasn't as even or consistent from tap to tap. That's a remarkable result for all three taps, they usually vary much more than that. There's definitely a cleaner high-frequency response with the PCB, better HF extension calling for a some extra capacitance across the feedback resistor. You can hear it, too, there's more detail and definition in the top end. I noticed it immediately on organ recordings, there's more character to the colorations and voicings in the mid- and high-register voicings. But everything sounds better, more presence and energy.

[David McGown]

Grover,

Interesting. I wonder if true point to point on the tube sockets (and any terminal strips) would improve things over using the tagboards in the builds. That greatly reduces wiring which is likely a major contributing factor. The tagboards are certainly convenient (not as much as PCB). I may rebuild my Musicians Amplifiers point-to-point to see how much that improves them.

David

[Grover Gardner]

Grover,

Interesting. I wonder if true point to point on the tube sockets (and any terminal strips) would improve things over using the tagboards in the builds. That greatly reduces wiring which is likely a major contributing factor. The tagboards are certainly convenient (not as much as PCB). I may rebuild my Musicians Amplifiers point-to-point to see how much that improves them.

David

Maybe, but here's my friend Paul's Heyboer Williamson, recently back to replace the noisy Hammond power trannies. This was point-to-point and, while it's not bad, it's not as good as the PCB version. Same Delrin top-plate with oak base, just different power trannies and choke.

[SoundMods]

Maybe, but here's my friend Paul's Heyboer Williamson, recently back to replace the noisy Hammond power trannies. This was point-to-point and, while it's not bad, it's not as good as the PCB version. Same Delrin top-plate with oak base, just different power trannies and choke.

Yet -- the square wave is cleaner. The ringing is suppressed.

[Grover Gardner]

Maybe, but here's my friend Paul's Heyboer Williamson, recently back to replace the noisy Hammond power trannies. This was point-to-point and, while it's not bad, it's not as good as the PCB version. Same Delrin top-plate with oak base, just different power trannies and choke.

Yet -- the square wave is cleaner. The ringing is suppressed.

You mean the bump at the top of the square wave? That's intentional. I could flatten it with a bit more capacitance if I wanted to. There are different philosophies about the "ideal" 10kHz square wave. I tend to go with the camp that likes a bit of a bump there, keeps the amp lively but doesn't affect stability. What's *after* the bump can create instability, and in this case the smooth and quick recovery is what's important. It's not really ringing, just HF tuning. Here's ringing, a very unstable version of the same amp:

Last edited by Grover Gardner on August 26th, 2023, 11:15 pm, edited 1 time in total.

[Grover Gardner]

The current amp, regardless of the tap configuration, is stable with no load, with a capacitive load only up to .22uF and with a combined resistive and capacitive load of .22uF. It will not break into oscillation. That's pretty damn stable. Here's frequency sweep. The FR is basically flat out to 40kHz, but supersonic anomalies are suppressed to insignificant levels:

[dberning]

You should check for unconditional stability. I go through a series of checks with different capacitor-only loading. It should maintain stability with various capacitors starting with 0.05 uF up to perhaps 2 uF. Expect some ringing which is OK provided it damps out after a few cycles. Do the check at various output voltages, up to clipping. You can do this at 1 kHz, or any convenient frequency.