Capacitors in rectifier

[Leevi]

There is a capacitor connected over each rectifier diode in the Fender's Pro Junior.
I have understood that they are preventing some oscillation there. Since those
caps are not widely used in other amps I'm asking if they really have some practical meaning there?


http://rudn.nodevice.com/preview/big/353/353008-1.jpg


/Leevi

[pompeiisneaks]

Actually, from what I understand, those caps are there (usually 1kv) to help control spikes that could be harmful to the circuit.  They're there for a good reason and should be left.  Quite a bit of newer amp design incorporates those.  I recently worked on a Vox AC100cph that I ended up gutting and rebuilding, and kept that same circuit design to ensure protection. 

~Phil

[Leevi]

I checked also Blues Junior schematic which is very close to Pro Junior.
Those caps are missing there.


http://www.blueguitar.org/new/schem/fender/blues_jr.gif


/Leevi

[drgonzonm]

The caps have been recommended as part of upgrades on some older amps likes the Soundcity 120.

I am checking the web and my records to verify this information. 

[PRR]

If the amp already has them, leave them.

If the amp doesn't have them, probably no reason to add them.

If building "clean sheet", diode caps or not may be more of a religious decision than a practical matter.

For sale of a large number, these caps *may* reduce problems with radio interference testing.

[pompeiisneaks]

Here's the section from the vox I spoke about:

[hesamadman]

There's a thread about this somewhere. I honestly think the caps are most generally used in higher voltage amps. It is to distribute the voltage more evenly across the diodes in series. One thing that was recommended to me was use better diodes. A 1n4007 rated for higher voltages would be good. But in this case I'm not sure why it's necessary. I hadn't looked to far into your amp but a blues Jr. shouldn't be too high of voltage I wouldn't think.

But after looking, it looks like a bridge. Probably same idea applies.

[HotBluePlates]

... Since those caps are not widely used in other amps I'm asking if they really have some practical meaning there? ...

Ugghhh...

PRR gave the best recommendation already.

Sometimes power supply components do more than 1 thing at a time. For example, the two 220kΩ resistors across series filter caps in older Fender amps will both enforce equal voltage distribution among the two caps, as well as provide a bleeder resistance to drain cap voltage when the amp is off.

The caps across diodes similarly perform at least two functions. First, they enforce equal-voltage distribution among the series diodes (you should look at them as capacitive voltage dividers for a.c. voltage), since the caps are equal-value. Second, they may reduce some RF noise at the moment the diodes switch off.

Modern electronic regulations specify how much RF interference a piece of gear may generate. Meanwhile, some vintage amps have prices going through the roof, and lack such gilding (and rarely use the cheapest diode available in batches of 100,000). If you want to spend more for additional parts, by all means...

[kagliostro]

On Merlin's PS Book there is a pair of pages about the reason of the capacitors in parallel with the diodes, reverse recovery and switching noise

He says that the capacitor value to be placed in parallel to the diodes is critical and the use of the wrong value can have not effect at all or make the problem worse

the council to avoid problems of this kind due to a wrong cap value is to use a single capacitor (10nF to 100nF) connected across the wires of the transformer (instead of one in parallel with each diode)

One other, more modern, solution is to use a fast diode (UF4xxx) in series with the output of the rectifier, or use all fast diode for the rectifier

Franco

[PRR]

> Blues Junior schematic ... Those caps are missing there.

No per-diode caps, but does have a 0.1uFd cap across the AC side of the bridge.

> section from the vox I spoke about:

Yes, when you must series-string diodes to get your Voltage Rating, the voltages won't split equally unless the diodes are very-very similar. On older plans you would find both resistors and caps to try to equalize the voltages. We rarely see this done today, rectifiers are better. We can get higher voltages, and two rects from the same tape are very likely to be very nearly "matched". Even so, some design rooms have House Rules which require various safety-parts be added just-in-case. So again it may come down to what you believe.

In that VOX plan, today, DIY quantity, I would use single 1600V rects instead of two rects series each side. However if I was making a million of them I would do my sums to see if cheaper diodes plus caps saved any money.
__________________

> Sometimes ...components do more than 1 thing at a time. ...two 220k Ohm resistors across series filter caps in older Fender amps ...equal voltage distribution ..., as well as provide a bleeder resistance

Yeah; but Leo never put bleeders in his amps. He knew how to work on amps.

[trobbins]

For some reason they have used a beefy 1N5062 diode in the full-bridge, and that diode is certainly not fast recovery.  EMI testing or listening tests could have forced the use of caps as the simplest addition to an already designed pcb - but just speculating.


I like the idea of just using the ubiquitous UF4001 fast recovery diode (series connected when VAC needs it, and paralleled when current margin needs it).  That setup provides a very low off-state capacitance path between the PT secondary (where there is significant diode induced dV/dt happening and mains voltage distortion) and the first filter cap.


Suppressing rectifier induced noise could be further alleviated by placing a tuned snubber (eg. using a quasimodo test technique) across the PT secondary.

[Willabe]

Suppressing rectifier induced noise could be further alleviated by placing a tuned snubber (eg. using a quasimodo test technique) across the PT secondary.

And that test would be?

[trobbins]

http://www.diyaudio.com/forums/power-supplies/ has a few good threads on Mark Johnson's quasimodo test jig, or a diy jig could be used to step a voltage across the PT secondary (other windings effectively shorted to recreate the in-situ conditions) and to observe the resulting transient voltage waveform across the secondary winding with a snubber providing the dominant loop path for leakage inductance energy.  The snubber can be 'tuned' in many situations to give a nice damped transient - compared to the inductive energy path ringing around stray shunt and circuit capacitances. 


I have a diy jig.  For vintage valve diode supplies the transformer series resistances are often enough (plus the softer diode turn off) to make any such noise effect negligible.  But modern transformers, and ss diodes, may get some benefit from a snubber if you are keen to suppress that noise contributor - whether its audible or not is another issue.

[Leevi]

If the amp already has them, leave them. If the amp doesn't have them, probably no reason to add them.

I asked this since I got my clone Pro Junior done and I didn't add those caps.
So based on your answer there is no reason to add them.


/Leevi

[Leevi]

> Blues Junior schematic ... Those caps are missing there.No per-diode caps, but does have a 0.1uFd cap across the AC side of the bridge.

Does this have same effect?


/Leevi

[kagliostro]

As far as I can know yes


Franco

[Leevi]

Noticed that the 0.1uFd cap across the AC side of the bridge will
at least prevent popping when the power is switched off.
/Leevi