Valve Amp Power Supplies

[simon]

Great Steve

Where PSUD can be useful is investigating the dynamic behavior of LC stages and getting a nice transient response. But it won't tell you what a good one is, you will have to decide that yourself by experiment.

Yep, that's what struck me about Steve's latest adventure. It'll be interesting for me to experiment vicariously

[Paul Barker]

I must admit to liking PSUD, and if I can use it anyone can.

Just remember that accurate info in = accurate info out.

Some things, like capacitor ESR can be difficult to find though.

ive tested my esr’s, Black Gates in my amp are 1.8 ohm per side. microwave 1 uF before choke is 2 ohm. Ive tested my Mullard vintage Mustard caps which the guitar boys like for tone are not useable. They are too high esr and could kill tubes. All differently too high, but I wouldnt put a single one in my amp. Their capacitance is also wrong beyont 10% tolerence so they are junk.Bought NOS. Generally dont buy old caps if you arent equiped to test them. NOS doesnt always make sense.

Likewise paper in oil. Ive had a few failures with vintage paper in oil. I had a .25 2kv cap I was planning to slightly boos the b+ with. Tested it first: very poor condition cant remember the figure but of esar but woefull. Couldnt use it. (safety first). Dont forget the cap before the first inductor has to be rated for the peak ac voltage and the b+ if the choke wasnt in choke input. On a 400v b+ that Ive scoped Ive seen an additional 100v peak for example. The cap after the shoke doesnt see this. This is why I wanted to use a 2kv cap only thing I had with the extra voltage. This is where microwave caps are perfect. 1kv is enough, I wouldnt use a b+ rated cap, no sir! That cap will be youre first failure. Knowledge of what youre messing with matters.
Not to mention my Black Gates are probably 20 years old. Test fine and sound fine right now, but for how long? I dont know. Ill probably live long enough to find out.

Ive already lived long enough to find out efficient speakers, or at least my AER’s and Lowthers dont live forever.

[Nick]

They are too high esr and could kill tubes

How? Why would esr kill tubes? They will normally be used as coupling caps. Also what frequency did you check ESR, that can matter. Not that I think its relevant for small caps anyway as their reactance will be more than the internal resistance.

[Paul Barker]

They are too high esr and could kill tubes

How? Why would esr kill tubes? They will normally be used as coupling caps. Also what frequency did you check ESR, that can matter. Not that I think its relevant for small caps anyway as their reactance will be more than the internal resistance.

These mustard caps wer high esr all differently and capacitance out of tolerance and all differently so they didnt get tried in a circuit to measure for affect on voltage but sure sign they are leaky. I could perhaps make a jig and test for leakage. esr is measured with an esr testr. I didnt actually pay attention to the spec sheet so I dont know the frequency it tests at. Oops. Im not someone who can work out esr off my own back. I am at the mercy of esr tester. But I can compare capacitors and if same cap is different and higher than all the caps ive tested, I can take a hint. But leakage is easy, measure dc current.

[Paul Barker]

There is consistency. Maybe you are asking the wrong engineers. RC stages are fine for low current they work just fine. It's using them in a power stage that becomes a problem.

Oh fekc, why have a missed this all my days! Now I can see a purpose for multi supplies. to set free the power stage. And that is what I was tipping over. Now that means I need to make. separate 150v supply from a different transformer, for the driver stage.
I know in my knower now that should make improvements so worth puting into practice. Thank you, oh boy non so blind as them as can’t see. Ive have never tested aurally for that which means it was ignorance, though you might call it prejudice.

[Paul Barker]

Why I fill my PSUs with bloody capacitance,

Didnt you start that after reading Morgan Blencoe (which I dont haveI so dont know) but that was about when you bought massive cap banks hook line and sinker.

I suspect Barbaric is a casualty of that era in youre growth.

[Cressy Snr]

Why I fill my PSUs with bloody capacitance,

Didnt you start that after reading Morgan Blencoe (which I dont haveI so dont know) but that was about when you bought massive cap banks hook line and sinker.

I suspect Barbaric is a casualty of that era in youre growth.

Yes I started it from reading Blencowe and Harvey Rosenburg. I was out to build a balls-out rocker of an amplifier. What I failed to take into account is that with push-pull, the class of operation is important and that you need to decide on that, then build your power supply accordingly, then BLOODY STICK WITH IT!
The high capacitance supply is associated with class B operation, which is what I wanted and when I’d finished it, with it’s 600V operating point with the KT88s almost in cutoff when idling, it kicked arse big time. I loved it!

Then I started becoming influenced by dogma (crossover distortion class B is crap, you don’t wanna do that blah blah blah) and took it to class A/B. This was a silly decision, because the supply was then just sitting there yawning instead of working its balls off as it was supposed to do. That kind of storage capacity is simply not needed with class A/B operation, especially when the thing is operating mostly at the class A end of its working range.

The Barbarik and it’s monoblock sibling is a classic example of someone not sure of their abilities, being influenced by bollocks on the internet rather than sound engineering principles I’d originally built it to. If I had the Barbarik back in my house, I would cap input it again which would bring the HT back to 600V then bias it just above cutoff so that it would once more be the loud and proud thing it was before. It was a valve Naim when it was like that.

[Nick]

I don't understand why the amp class matters that much (well I do but I think it may be clouding the issue)

The problem is not the C's its the R's

[Paul Barker]

Why I fill my PSUs with bloody capacitance,

Didnt you start that after reading Morgan Blencoe (which I dont haveI so dont know) but that was about when you bought massive cap banks hook line and sinker.

I suspect Barbaric is a casualty of that era in youre growth.

If I had the Barbarik back in my house, I would cap input it again which would bring the HT back to 600V then bias it just above cutoff so that it would once more be the loud and proud thing it was before. It was a valve Naim when it was like that.

Not being funny and no pressure but I don’t recall the price was it the worth of the transformers?

I wouldn’t use it now I’m back building my stuff so if you make a low offer or know the price, it’s a deal!

[Cressy Snr]

Didnt you start that after reading Morgan Blencoe (which I dont haveI so dont know) but that was about when you bought massive cap banks hook line and sinker.

I suspect Barbaric is a casualty of that era in youre growth.

If I had the Barbarik back in my house, I would cap input it again which would bring the HT back to 600V then bias it just above cutoff so that it would once more be the loud and proud thing it was before. It was a valve Naim when it was like that.

Not being funny and no pressure but I don’t recall the price was it the worth of the transformers?

I wouldn’t use it now I’m back building my stuff so if you make a low offer or know the price, it’s a deal!

Yep it was basically the worth of the iron in it.

[Cressy Snr]

I don't understand why the amp class matters that much (well I do but I think it may be clouding the issue)

The problem is not the C's its the R's

I don’t think there were any resistors in the Barbarik PS. It was just a brute force solid state bridge rectifier then CL followed by a big cap bank.

I’m aware now of course, thanks to your input on this thread, that resistors in the power stage supply are not ideal for what I’m doing at present. I’m going to remove them from the 6V6 amp and report later. Just had to walk down to Screwfix and bought a new multimeter. My old one has gone stark raving mad!

[Nick]

I don't understand why the amp class matters that much (well I do but I think it may be clouding the issue)

The problem is not the C's its the R's

I don’t think there were any resistors in the Barbarik PS. It was just a brute force solid state bridge rectifier then CL followed by a big cap bank.

Ok, my mistake, I assumed it was related to what was posted before. Ignore that.

[Cressy Snr]

Hmmm. About 20V difference in sag with the resistor in there.
Dunno enough about interpreting the transient response though, to tell whether it's good or bad.
Perhaps someone could comment on that one.

6v6psunoR.png (92.11 KiB) Viewed 2369 times

6v6psuwithR.png (100.25 KiB) Viewed 2369 times

[Nick]

Well, there is no overshoot.

[Cressy Snr]

Well, there is no overshoot.

OK.
If I'm looking correctly, then the total reaction time to a sudden transient is 350 milliseconds which I would think is pretty mediocre, leading to subjective bass slowness I want a reaction time measured in tens of microseconds. Or am I either not understanding what the sim is saying or being completely unrealistic given the type of supply under simulation. IOW am I seeing the downside of a cap input supply being shown here?

Or on the other hand, could this be where the use of a voltage regulator comes to the rescue, 'cos there'd be no sag on a transient, therefore no reaction time issue and the bass problem would be history, or something along those lines?