I’ve rarely seen capacitors across diodes in a bridge rectifier. Thoughts?

P

pauly

Fractal Fanatic
Hi all,
The below video seems to indicate we should see a lot more capacitors across diodes, however I wonder if part of the noise he’s getting in his tests is related to the wires becoming open within proximity of transformers and mains voltage frequencies. Any thoughts?
Thanks
Pauly

 
The diodes are likely not used in typical amplifiers because the noise is out of the audio range and if it is in range the noise is removed by the filtering in the supply. Caps are cheap but manufacturers won’t add parts to the process if there isn’t a problem to be solved. That’s my best guess anyhow.
 
The filter caps are already across the rectifier output.

Push pull amps are also basically humbucking and cancel out most remaining power supply hum due to their opposing phases on each side of the OT primary.
 
The noise created is right in the audio range. It's at the mains frequency and harmonics thereof. These are "snubber" capacitors and damp the transients that occur when the diodes turn off.

Some of our reference amps have these capacitors. IIRC one of them is a Marshall. It's a good design practice if you want to keep mains noise to a minimum.
 
I remember this being done on some of our AX84 designs back in the day. I can't recall if the use of Fast Recovery Diodes like the UF4007 eliminated the need for the caps or not but it was a minimal expense for small scale production. For Marshall scale, somebody certainly would have nailed down if it was cheaper to have the cap+1n4007 versus a UF4007. @FractalAudio probably knows the answer to whether the UF4007 is sufficient or if a cap is still a good idea.

I regret that I spent much of the heyday of AX84 just trying to keep things "traditional" and straightforward so that laypeople (i.e. me) could understand them. There was some real efforts at innovation along the way that I stifled. Some of those folks obviously knew a heck of a lot. Merlin Blencowe springs to mind but there were others wanting to push the envelope as much as they could.
 
Chris Hurley said:
I remember this being done on some of our AX84 designs back in the day. I can't recall if the use of Fast Recovery Diodes like the UF4007 eliminated the need for the caps or not but it was a minimal expense for small scale production. For Marshall scale, somebody certainly would have nailed down if it was cheaper to have the cap+1n4007 versus a UF4007. @FractalAudio probably knows the answer to whether the UF4007 is sufficient or if a cap is still a good idea.

I regret that I spent much of the heyday of AX84 just trying to keep things "traditional" and straightforward so that laypeople (i.e. me) could understand them. There was some real efforts at innovation along the way that I stifled. Some of those folks obviously knew a heck of a lot. Merlin Blencowe springs to mind but there were others wanting to push the envelope as much as they could.
Click to expand...
With Marshall it was mostly likely an economic decision. FR rectifiers are always more expensive than standard diodes and caps are cheap.
 
mr_fender said:
So are the snubber cabs used in place of larger filter caps or in addition to them?
Click to expand...
They go across the diodes. When the voltage changes polarity the diodes continue to conduct briefly. This is known as reverse recovery time. During that time there is "shoot-through" where all the diodes are conducting. This creates a large current transient which produces noise. The caps dampen the noise.
 
You must log in or register to reply here.
Back
Top Bottom