WAD 300B PP Rectifier

[Richard]

Hi Tony,

I was on hols so have come to this a bit late, but yes, I've tried 4 types here,

Mullard GZ37
Mullard GZ34
Russian 5U3 (coke bottle 5U4)
National 5U4GB (Philips USA straight tube)

All work fine though the direct heated ones will turn on HT more quickly.

All sound quite different. GZ37 was the best imo and I immediately ordered a couple of spares. 5U3 was pretty good. I didn't like GZ34 somehow harsh sounding, and 5U4GB was very dark and I felt something was missing.

All very subjective I know but do try a few types and you'll hear the differences quite clearly.

It may be a construction thing and impedance/voltage drop as Ian says seems possible too.

Rich

[Ianm2]

try the chinese we type 274 if you can get em, if they will tolerate the conditions, too, said to be a deriviative or the other way around from 5u4, so may not be to taste

ahh, my memories coming back a little, certainly 5z3/4 chinese types are variants on 5u4

also try pototo masher, but wasn't to my taste in s.e. amp

and also 5v4 i think it is, that's going a long way back, its quite a hi volts rugged one i am thinking of that's not used much....maybe 5v3, can't quite recall....

[Tony Moore]

Surely though it must be possible to create a solid state equivalent to the Gz37 for the function of start-up delay plus the voltage drop? It may even be possible to make it sound _better_ than a Gz37? :p (heresy I know) I mean, we already have semiconductor rectification after all!

I bought a spare GZ37 too but I'd much rather there was a cheap SS alternative, plus it would save a few amps drawn off the power tx also which has got to be a good thing?

Cheers,
Tony

[Ianm2]

well problem is voltage drop is caused by resistance, which dissipates heat ,but that easy, only thing is, adds to the output impedance of the power supply, as viewed by the signal circuitry, which ideally should be zero at all freqs.

that's why the more adventourous over on audio talk start implementing active psus' aka regulated ones, to get the output res. down, and regulation up, its the logical follow on, after one goes from el84 to el34, to 300b, to big transmitters, back to lill 45s, then hey presto what's left....the psu to play with.....

delay can use timer or capacitor, again, in principle, simple

why not simply remover the valve rect., put a switch in the HT cct, fire up heaters first, then add hi volts, say 30 secs later, and check all is ok, perhaps rebias, add series resistance to get operating points the same

and even heresy, remove chokes and replace with equivalent resistors, but that's a pet theory, EASILY replaceable, 2 wires each ( AS CHOKES ARE GENERALLY NOT NEEDED IN PP APPLICATIONS!!!!!)

[Richard]

Originally Posted by Tony Moore

Surely though it must be possible to create a solid state equivalent to the Gz37 for the function of start-up delay plus the voltage drop? It may even be possible to make it sound _better_ than a Gz37? :p (heresy I know) I mean, we already have semiconductor rectification after all!

I bought a spare GZ37 too but I'd much rather there was a cheap SS alternative, plus it would save a few amps drawn off the power tx also which has got to be a good thing?

Cheers,
Tony

Well, let's have a look at what this ps actually is.

Yes, it will slow-start the HT but I think it is more than that, and that is why we can hear such a diference with different rect valves.

Look at the supply after the solid state bridge and it is a valve ps.

The solid state diodes and caps before it rectify and smooth the ac fairly well (it will still have a couple of volts ripple), and take a lot of the work off the following GZ37. All the supply though then passes through the GZ37 which is wired here as a half wave rectifier.

It is then choke smoothed and fed to the 22uF reservoir caps. (These are more than enough to supply the amp as can be testified by many old designs such as the 30W p-p 5-20 which used 8uF in this position.)