18Watt.com

Will someone tell me if I am dreaming again ...
So I was looking at an amp circuit, and it suddenly occurred to me that a tube rectifier is the best surge protector ever devised if you don't want your PSU electrolytics to go die - so, apart from the compression effect, valve rectification has another benefit - and it can't go short circuit without physical intervention or real violence. Now the world is ruled by these new caps that never go sour, is it reasonable to use thermionic rectifiers? Me, I always favour such devices because they cannot reverse, which every other rectifier can do when it fails.
Discuss ... please

dotfret; are you saying that you prefer tube recto to ss? or asking if one is prefered to the other?

sound issues aside, In all the years I have been tinkering/repairing amps the failure rate of ss rec is zero as apposed to one blown tube rec. I don't really think long term reliability of ss rec is a problem in the relatively low power amps we discuss on the forum.

In my experience anyway

dotfret wrote:Me, I always favour such devices because they cannot reverse, which every other rectifier can do when it fails.

As this event has a rate of occurrence approaching zero, it would seem unnecessarily cautious to use it a factor when choosing rectification type.

Valve rectifiers are reliable enough for most peoples needs too, so I would choose using the other factors; whether or not you want a "firm" voltage, size, simplicity etc.

Wow, I guess I should play the lottery...
I repaired my friend's bf bassman, it would hum really loudly and blow a fuse as soon as you turned it on.
I traced the problem to a failed PS diode that was conducting in both directions.
But I agree, you shouldn't choose tube over ss rectification for this reason alone.

I agree with the choose for tone crowd. The failure rate of either tube or SS rectifiers is so low as to not be that much of a concern. Even if a rectifier fails, you are still isolated by the transformer and the result is a loss of voltage, actually making the amp a little safer to us humans (but it can still sting you pretty good, so be carefull). In any even, if you had a really bad rectifier problem the fuse would blow which is a pretty good warning of an issue.

I have also seen a blown diode in a power supply, but I've also seen bad rectifier tubes. And since even a rectifier tube can have a short, I'm not so sure it is a lot safer in all cases.

However, if you're really worried then do what a lot of amp makers do and double or triple up diodes on each side. If one shorts, then the next one in line can act as a fail safe.

Alexo wrote:Wow, I guess I should play the lottery...
I traced the problem to a failed PS diode that was conducting in both directions.

Well I guess that single anecdote proves how unreliable ss diodes are then!

How many ss diodes do you think you use, and have used, in your daily life, in all applications, to date? To the nearest hundred-thousand will do.

I'm not trying to dispute the reliability of ss diodes, just didn't realize what a freak occurence that was at the time.

Sorry, I interpreted your statement as irony!
Thats the Brits for you.

No worries. It does sound pretty sarcastic now that I think about it.

Let me try again -
In my eureka moment it came to me that
1) the PSU components that most often die are the smoothing caps
2) they die more frequently in amps using SS rectifiers
3) there is a voltage surge when switching on (the switching effect), which thermionic rectifiers do not pass on to the smoothing caps, because they have not warmed up at this point
4) SS diodes do pass that surge to the smoothing caps.

Now the above is observationally true for electrolytic capacitors - but does it still hold for the modern types of capacitors? That is what I asked.

Yes, a failure of either class of diode is rare - and there is a rare condition of the dying thermionic diode which can cause the power transformer to run so hot that it fails. But I have decided that, on the whole, I prefer thermionic rectification, because it is kinder to the capacitances. Fizziks?

dotfret wrote:Let me try again -
1) the PSU components that most often die are the smoothing caps
2) they die more frequently in amps using SS rectifiers


Now the above is observationally true for electrolytic capacitors

..... because it is kinder to the capacitances. Fizziks?

Are the above statements actually true?
I'd like to see the "fizziks" and "day-ta" to prove it.

dotfret wrote:Now the above is observationally true for electrolytic capacitors - but does it still hold for the modern types of capacitors? That is what I asked.

I don't know what "modern" capacitors you are referring to. The last 12 amps I built all used Electrolytic caps for filters. I guess I'm behind on the right caps to use...

I don't think the power on surge is going to matter a whole lot on an amp that is designed correctly and played a lot. Plus not all rectifier tubes have a slow turn on, not that the warm up time for a tube vs solid state rectifier even matters in most amps.

With an 18 watt the filters are on the other side of a standby switch from the transformer B+ output line (as with most tube amps). The surge is going to make no difference no matter what rectifiers you are using. They will get the same inrush anyway.

Maybe for a Champ it could make a difference, since there's no standby switch, but not for most amps. And before you cite the Fenders with the Caps on the hot side of the standby switch, they are configured so that the inrush is within their safe operating range and it will not shorten their life.

The best reason to use a tube rectifier is for feel. Even a simulated voltage drop doesn't quite match a tube for the nice soft compression it gives. Otherwise, it really doesn't matter.

From what I understand the corresponding "switch effect" over voltage for a capacitor is a factor of the components of the capacitor itself and therefore would be built into the duty rating of the capacitor so...not an issue.

On the otherhand... the inrush current when switching a capacitor on after a tube rectifier can lead to rectifier failure which is why tube recto's have a spec for the max cap imediately following it.

I have had close rated diodes fail mostly...maybe all... open but to protect against a short two in series makes sense. Modern diodes are more than up to the task so again the choice should be tone related.

-Andy

dotfret wrote: 1) the PSU components that most often die are the smoothing caps
2) they die more frequently in amps using SS rectifiers
3) there is a voltage surge when switching on (the switching effect), which thermionic rectifiers do not pass on to the smoothing caps, because they have not warmed up at this point
4) SS diodes do pass that surge to the smoothing caps.

For that reason, some people recommend a smallish 5W resistor between SS rectifiers and the first filter cap.

Richard, re capacitors - I still use electrolytics, too - by modern, I meant the large poly types which weren't available when I was a lad.
I concluded that a large zener diode, rated about 30V below the caps max value, between B+ and the ground might set my mind to rest - then I thought - but what if the zener shorts?
WTH I'm going for a beer.