What are the "critical" traces on a tube preamp PC

[crazytooguy]

Hello all. I'm working on laying out a tube preamp PCB, and I'm wondering which traces on the board are the ones I have to pay special attention to as far as size and separation from other traces to prevent oscillation and crosstalk problems. Any tips or references for guidelines on this? Thanks!!

 

[PRR]

Don't let the output near the input.

 

[Larrchild]

And make the filament traces wide enough to not incur voltage drop. Also, if the filament is AC, and you can route the filament traces away from low signal areas, thats good too.

 

[BradAvenson]

You might think about just running the filaments as twisted wires to pads near your tube sockets.

 

[crazytooguy]

Thanks all. The filament supply is rectified, but I have other tube amps that are not, and twisted pair filament supply sounds like a great idea for reducing noise, especially in one circuit that's pretty high gain. It would be easy to break the PCB traces and solder wire ends in place. Hmm...

 

[clintrubber]

[quote author="crazytooguy"]Thanks all. The filament supply is rectified, but I have other tube amps that are not, and twisted pair filament supply sounds like a great idea for reducing noise, especially in one circuit that's pretty high gain. It would be easy to break the PCB traces and solder wire ends in place. Hmm...[/quote]
The separate & twisted wiring for ff is defin. a good idea. It'll have been done quote a lot of times. At least I've seen this several times in tubed consumer reel-to-reels.

Bye,

Peter

 

[CJ]

Run your grid and plate traces 90 degrees to each other when possible.

 

[clintrubber]

Related, sort of: pay attention to the orientation of film-caps w.r.t. the outer foil. If inclined so you could provide a marking on the PCB.

 

[BradAvenson]

I thought of another thing. Make sure the trace spacing is wide enough for the high voltages.
check out this chart for a guide:
http://www.mitsi.com/PCB/Hints/spacing%20guidelines.htm
I'm not sure where this info came from since they don't site it, but I have seen the same chart in a few locations.

 

[crazytooguy]

OK Thanks!! These are great. I'll be making up a list of these and keeping it with my design schematics.

 

[toffifee]

[quote author="clintrubber"]Related, sort of: pay attention to the orientation of film-caps w.r.t. the outer foil. If inclined so you could provide a marking on the PCB.[/quote]
Could you please elaborate? I´m building a tube mic pre at the moment, so further information would be great. I´m sure this has been covered in another thread, but I did not find it (searched for "orientation"...).
Questions: How do I know which side connects to the outer foil (do I have to cut one cap open?)? And then connect it to what?
At which places does the orientation matter? Only in the signal path, or as well at B+ filtering?
Toff

 

[bcarso]

[quote author="toffifee"]Questions: How do I know which side connects to the outer foil (do I have to cut one cap open?)? Toff[/quote]

If you have a scope, probe, and signal generator you can put a signal on one lead of the cap, and gen and scope ground on the other. Run a healthy voltage at 1 kHz and bring the probe to the center of the cap case. Note the component at 1kHz. Now reverse the cap. The highest voltage on the scope means that config has the outer foil on the generator hot lead.

I thought there might be a convention as to the orientation of the marking on the case and the lead-outer foil connection, but the first two experiments with the same manufacturer's parts were different. Some vendors mark it with a band though, and that banded end is the outer foil.

 

[PRR]

> orientation of film-caps w.r.t. the outer foil.

This rarely matters on audio coupling caps. If the cap is big enough to pass bass, its capacitiance is SO much higher than the external stray capacitance that it really does not matter.

It does matter in things like tone controls. The outside foil should go to the lower-impedance side (usually the source) to help shield the cap. But this will be a minor difference in nearly any audio application. If you NEED shielding, just putting the outside foil to the low-Z side of the cap won't be enough.

It gets more important in radio work, which is why outside foils were marked. Even if the self-sheilding was not important, the stray capacity makes 50% of your production tune a little different from the other 50%: consistently putting the outside foil the same way makes tuning more consistent.

 

[toffifee]

Thank you!
This place is wonderful :thumb: :guinness:
Toff