Diode rectifiers

[martin manning]

^+1 Mr. Phil!

I'm just gonna put this right here, skip to about the 3:10 mark.

http://www.youtube.com/watch?v=wkw1sz3QVOE

I'm awful, I know.

He doesn't say whether there is a mechanism to restore the bias and supply voltages when switching from vacuum to SS, and that could be a large part of what he is hearing.

[TUBEDUDE]

Just a note of caution; If using 2 diodes in each leg for a higher PIV, EACH diode needs a bypass cap! Don't use one cap across 2 diodes.

[Ken Moon]

I know this is way out there, but I bought 100 of these "Stealth Dual Diodes" several years ago, and have used them on about 15 amps. They are handy when using a PCB-based power supply.

They are in a TO-220 package, and don't need a heat sink or external snubbers. I only build 20W amps these days, but they should work fine for any size amp, as far as I can tell from the specs.

http://media.digikey.com/pdf/Data%20She ... 8120P3.pdf

If anyone wants to try a couple, send me a PM, and I'll pop them in an envelope and send them to you for free

[teemuk]

I'm beginning to rethink my tube purist ideology and go the diode route. As much as I love having that extra tube in there, the voltage drop, PT strain and decreased bass response just don't seem worth it. Not to mention saving more than a few bucks on parts.

Personally, since invention of solid-state diodes I haven't seen much point in using vacuum tube diodes.

Obviously you are not too fond of characteristics of high impedance rectifiers (e.g. increased voltage drop, sag, etc.) and even if you were you could still use solid-state diodes and achieve those effects simply by placing some series resistance to diodes.

[teemuk]

Originally I put the snubber caps on them but I later read that they can cause noise on a fast diode, so I clipped them.

Mostly it's not super obvious difference to begin with and when noise is considered the rectifier diode is just one part of a complex scheme.

For example, diode switching noise is dependent on transformer's leakage inductance: When diode becomes reverse biased it will still conduct for a brief period of reverse recovery time, during this time leakage inductance of transformer will store energy, which will subsequently be released in a quick transient when diode eventually turns non-conductive.

There are basically two ways to fight this: One is to use very fast diodes, like schottkys. Another one is to introduce parallel capacitance to diodes. These methods can not be used interchangeably, they depend on what type of diode is used. Fast diodes do not need caps, in fact, the caps would hinder their operation. Slow diodes (like silicon rectifiers) "need" caps. A parallel capacitor will actually slow down the diode even further (this is why you do not want to fit them to circuits with fast diodes) and while the added sluggishness will not prevent the aforementioned phenomenon it happens to decrease amplitude of the resulting transients.

Similarly to power transformer there will be leakage inductance in large capacitor banks. This can be reduced with properly designed snubber. Not any snubber will work effectively (or even at all), the circuit needs to match the characteristics of the power supply.

The characteristics of the transformer are naturally important too. For example, toroidal transformers have much lower leakage inductance.

All in all, all this noise reduction can still be nullified with a poor noding scheme of power supply and common return currents. ...or just by poor built practices overall that result to noisy circuits.

Probably bad ears but I can't tell a difference.

It's really not a difference that is strikingly clean to hear. Basically we are talking about slight amounts of noise and noise-induced distortion in the signal. Since the issue is minor to begin with you will need finer test equipment than ears to measure it properly.

So.... hearing the effects isn't always the most important point. An amp with lower/reduced diode switching noise doesn't, for example:

- cause radio interference to surrounding equipment
- send high frequency interference to mains grid

An important point to realize is that the noise characteristics are largerly design dependent. In one specific design with its own specific characteristics you might hear noise with slow diodes without parallel caps. In another circuit the characteristics may be such that switching transients aren't that big issue in the first place and the parallel caps do not considerably affect at least the audible noise.

[Turret]

Another excellent post by teemuk

Thank you

[Ken Moon]

"...you will need finer test equipment than ears to measure it properly."

Not to argue with teemuk, but to me this marks the border of "I don't care" land.