best layout practice: coupling caps

[ric]

Hi All, I know grid stoppers should go "downstream", right on the tube socket. How about coupling caps? The signal rides on the DC, but it seems to my untrained hobbyist brain that it might be best to put the cap as far "downstream" as possible. What do you all think, and why. I went back and read through the 8 year history of Theory and Design. I have pages of notes, but basic layout as it pertains to this is not discussed. Thanks for your time. This is the electronics course, focusing on tubes, I thought I'd never find, and couldn't have afforded.

[mhuss]

Grid stoppers are best right on the tube pins to minimize the length of the "antenna" between the stopper and the grid. DC is the opposite of this, and length doesn't matter. So the most convenient place is the best place to put coupling caps.

[ric]

Thanks for the reply mhuss. That leaves my options open.

[Mick Bailey]

A coupling cap can be sensitive to induced fields or vibration, especially in the first stage of a high-gain amp and even more so in a high-gain combo. Some amps have very microphonic early-stage coupling caps, regardless of cap technology. In addition, whilst the DC doesn't matter, any additional cable or PCB track length or sub-optimal routing can pick up AC which gets amplified through the next stage.

[ric]

A coupling cap can be sensitive to induced fields or vibration, especially in the first stage of a high-gain amp and even more so in a high-gain combo. Some amps have very microphonic early-stage coupling caps, regardless of cap technology. In addition, whilst the DC doesn't matter, any additional cable or PCB track length or sub-optimal routing can pick up AC which gets amplified through the next stage.

. Mick, is this a vote for "downstream" coupling cap placement, or more about secure cap mount, with as short of leads as possible?
Thanks for the reply. ric

[Mick Bailey]

A little of each. But let's say you have a typical coupling cap of 0.022uf in a point-to-point build. You may have a couple of inches of connection wire to the next tube in some cases. If there's a chance of that lead picking up 50/60Hz hum, it would be better if the cap was right at the end of the wire where it could act as a high-pass filter to reduce the noise induced into that wire. If it was at the beginning, then the wire would pick up the noise and get passed straight into the next stage.

So my vote is downstream.

[mhuss]

If there's a chance of that lead picking up 50/60Hz hum, it would be better if the cap was right at the end of the wire where it could act as a high-pass filter to reduce the noise induced into that wire.

That would seem to make sense intuitively, but if you do the math(s) given a typical .022uF / 1M inter-stage, the -3db frequency is ~7hz! I would argue that certainly the layout of the entire inter-stage connection should be chosen to minimize hum pickup, but the cap placement (start vs. end) won't affect hum much.

Microphonic caps are another story altogether, and some sort of physical damping applied to the cap itself would be best to treat this scenario.

[Enzo]

We can rationalize scenarios where some small effect might occur, but for me, the cap placement is a non-issue. A zillion Fender amp work perfectly well with the caps all on the eyelet board and wires flown to the sockets. Only the grid stoppers are at the sockets.

[ric]

Mick, Thanks for the reply. Sounds like i'm not completely off base with my thinking on this. mhuss, I see where you are coming from. Really a non issue when one does the math. Ignorance is bliss, and I'm afraid I am not yet up to speed on frequency calculations and blissfully build without them at this point. I am here to learn, however. Thanks. Enzo, I know your stance of: 'A million Fender amps can't be wrong". I respect your opinion, but I want to know "best practice". What's the best way. Also Enzo, I want to thank you for giving me hours of your shop time to answer lists of questions I would bring over. You also recommended Ampage, where I would go to the library and download as much as I could in my one allowed hour. Have a smart phone now, so I found this place. It's a real world class resource, literally.

[Enzo]

Great! You are welcome. I am happy to help people who come to learn.

[Mick Bailey]

That would seem to make sense intuitively, but if you do the math(s) given a typical .022uF / 1M inter-stage, the -3db frequency is ~7hz!

That's right. You'd have to have a much lower effective resistance then 1M for the HPF effect to be of any significance.
I have recent experience with a reverb recovery circuit where the coupling cap location did have an impact on hum level, due to the wiring running close to a toroidal mains transformer cable entry. Moving the cap to the other end of the lead fixed the hum, and in that particular case the maths worked out. Equally I suppose a screened cable may have worked.

[soundmasterg]

Some of those Fenders, especially the SF ones will be right on the edge of oscillating because of poor wiring practices. The fix for those in some cases is to move the coupling caps right to the pots. I've seen this practice elsewhere also, so I vote for downstream myself. I try to do it on all my builds though for repairs I leave well enough alone if the amp works fine.

Greg

[ric]

Some of those Fenders, especially the SF ones will be right on the edge of oscillating because of poor wiring practices. The fix for those in some cases is to move the coupling caps right to the pots. I've seen this practice elsewhere also, so I vote for downstream myself. I try to do it on all my builds though for repairs I leave well enough alone if the amp works fine.

Greg

Greg, thanks for the reply. A sensible approach. Gerald Webber recommends tone caps right on their pots in his books, I have also followed that practice. I didn't bring him up in the OP because I didn't want the thread to turn against his opinion. ( I called him up at the very start of this quest and he was very encouraging and happily chatted me up till he had to pick his child up from school, he was just as cool as Enzo about sharing time and knowledge, asking nothing in return.) So it's looking to me like best practice is to keep coupling caps as far downstream as practical.

[soundmasterg]

Greg, thanks for the reply. A sensible approach. Gerald Webber recommends tone caps right on their pots in his books, I have also followed that practice. I didn't bring him up in the OP because I didn't want the thread to turn against his opinion. ( I called him up at the very start of this quest and he was very encouraging and happily chatted me up till he had to pick his child up from school, he was just as cool as Enzo about sharing time and knowledge, asking nothing in return.) So it's looking to me like best practice is to keep coupling caps as far downstream as practical.

The way I look at it is before the cap it will be part of the plate circuit and after the cap it will be part of the grid circuit. I feel the grid circuit is a lot more sensitive to interference so minimizing that possible antenna by shortening the grid path is best approach. But from what others have said it could be splitting hairs too.....it works for me so I stick with it.

Greg

[Enzo]

Try this. Find a cap from plate to grid in some amp you have. Go right to the cap. Take a small screwdriver, making sure to keep your fingers off the metal, and touch the wire lead of the cap with the driver blade. Touch it on either end of the cap. What difference do you hear?


No, I don't have a secret answer. This is one of those ten second experiments that tells you how much the cap isolates the grid from the plate.