AB763 grounding scheme

[Roe]

It's not just the ripple voltage; decoupling of the stages is also important. The last filter is powering no less than six triodes, where two or at most three is usually considered the best practice. It's not quite as bad as it seems, though. Typically only one channel is in use at a time, so the worst case is it supplies four active triodes, and the phase relationship of the reverb recover stage to the others is all over the map. I find there are no issues with the one I built, and in fact the original circuit design and layout, random grounding and all, has been working pretty well for almost 60 years :^)

after rebuilding two, the noise decreased significantly

[martin manning]

after rebuilding two, the noise decreased significantly

Rebuilt how? What changes were made wrt the original?

[Oddvar R]

Yes, please?

[Roe]

after rebuilding two, the noise decreased significantly

Rebuilt how? What changes were made wrt the original?

I rebuilt the grounding and power supply, using merlin as a guide. One filter cap per preamp tube

[Oddvar R]

after rebuilding two, the noise decreased significantly

Rebuilt how? What changes were made wrt the original?

I rebuilt the grounding and power supply, using merlin as a guide. One filter cap per preamp tube

Could you be a bit more specific?

[nworbetan]

I can't help with the specific details as they pertain to rebuilding an AB763, but I'll second Roe's recommendation to read Merlin's grounding advice. He made the Grounding chapter of his his book about preamps downloadable completely free on his website.

http://valvewizard.co.uk/Grounding.html



This is a really easy follow explanation of where the current flows in the ground path, and why micro-managing current flow is important.

[Roe]

Rebuilt how? What changes were made wrt the original?

I rebuilt the grounding and power supply, using merlin as a guide. One filter cap per preamp tube

Could you be a bit more specific?

use multi-star grounding as defined by MERlin the tube wizard. see the excellent post above

[bepone]

moreover on this tip that every node must to have el.capacitor close to the valve which is supplying, it is a good habbit to think about plastic chassis. or how you can make amp in plastic chassis without humm . then if you do so, ground the amp close to input jack in one point and all is done.

[corkhead]

Hi,
I am the original poster, and I wanted to say thanks to all the contributors. The Grounding chapter a couple of posts up is a great find! Lots of excellent information there. My kit has made it to Australia, so it should only be a week or so until I have it in my hot little hands. I like the ideas and info posted. I will make a decision on which way to go with the build, and let you know how I go. Cheers all!

[arjepsen]

Sorry for waking up an old thread, but I thought it'd be better than creating a new copy of this...
I'm planning a build of a DR, and is considering the whole grounding scheme.
Right now, Im leaning towards Martin Mannings way of doing it.
However, if I understand correctly, he has both channel's input jacks grounded straight to the chassis?
Doesnt this itself create a potential ground loop?

I read in Merlins book about isolating the input jacks, and creating a preamp ground star as close the the input jack as possible... so I was wondering:
Why not move the second channel's input jacks over to the first channel's jacks, so all four (or in my case: two - since I see no point in the lo input). This way both channels' input are as close to the ground point as they can be.

What do you guys think?

[martin manning]

I can say that the DR with modified ground scheme that I built is still going strong, and it is still very quiet. Since I imagined the grounding scheme and then successfully built it, I can't say which of the changes had the most effect, and I'm suspicious that one or two might not really be necessary. Overall, eliminating the brass plate and separating the grounds by their supply nodes is probably the most effective aspect. On the inputs, You could isolate the input jacks from the chassis, and run all of the input grounds to a single point, but in the end it worked out fine..

[Roe]

I read in Merlins book about isolating the input jacks, and creating a preamp ground star as close the the input jack as possible... so I was wondering:
Why not move the second channel's input jacks over to the first channel's jacks, so all four (or in my case: two - since I see no point in the lo input). This way both channels' input are as close to the ground point as they can be.

see 15.11: Grounding Multi-Channel Amplifiers, page 15/276: https://valvewizard.co.uk/Grounding.pdf

[martin manning]

That is not a problem in my experience. Both channels on the DR in question have been in use at the same time, one for a guitar and the other for a synth running a sequence. No issues.