swopping a 5Y3 to a GZ34

[Domocaster]

Hi everyone,
Question for all you clever tech-heads;
Amp Model; Gibson GA 5T 1960 model; tube line up;
1x 12AX7, 1x 6BM, 1x 5Y3.
In "Inside Tube Amps" by D. Torres, he states one can change a 5Y3 for a GZ34, rectifier tube.
Is this so?
The reason I'm asking is this; The amp sounds a lot better with the following filter caps; 40uf/500v; 20uf/500v, 10uf/500v.
As opposed to the 20uf/500v, 10uf/500v, 10uf/500v, that are specified on the schematic.
In the above book, it is stated that 5Y3's cannot have more than a 20uf/500v filter cap in the first filter posistion.
So, to sumarise; do I need to change the rectifier tube to a GZ34 if I wish to use the larger caps?

[eugenedunn]

I'm not sure about your particular amp, but going from a 5Y3 all the way to a GZ34/5AR4 type rectifier tube seems like a bit of a stretch. I am thinking this because the 5Y3 drops voltage going through it by 50-60 volts and the GZ34 series only drops voltage around 17 volts or so.

If you were to just slap in a GZ34, it seems that too much voltage might be running through your power tubes.....you would have to compensate by replacing some resistors to get the bias in the correct range..... that's my initial impression.

I think you could sub another rectifier for the 5Y3, however...... maybe a 5V4 wouldn't be so far off. You would really have to check your power tubes to see if they get overtaxed by any change in rectifier though.

[muchxs]

Originally Posted by eugenedunn

If you were to just slap in a GZ34, it seems that too much voltage might be running through your power tubes.....you would have to compensate by replacing some resistors to get the bias in the correct range..... that's my initial impression.

The GA-5 is cathode biased. That means just plug the 5AR4 in there and SMILE. Especially if you're swapping to oversize filter caps at the same time. It's a standard Gerald Weber mod for small amps, do a cap job, increase the size of at least the first filter, use a 5AR4 rectifier.

DISCLAIMER: Kids, don't try this on a stock '60s or '70s Champ. You might think the GA-5 and the Champ are similar enough and that it will work. The 5AR4 draws 2 amps on the filament, the 5Y3 draws 2 amps on the filament. In practice the plate voltage in a Champ is high enough. A Champ will blow fuses w/ a 5AR4. I'm talking a stock Champ with a stock transformer. Clones w/ aftermarket iron will take it.

DISCLAIMER, PART "B": ALWAYS measure voltage at the first filter cap when swapping rectifiers. It's easy to exceed the breakdown voltage of the capacitors at which point they break down! Imagine that!

DISCLAIMER, PART "C": Amp tech wannabes, make sure you know what you're doing before messing around inside an amp. There's over 500 volts available inside a GA-5 plus live house current. Lots of opportunities there to learn things the hard way.

If you have test equipment available to measure capacitance and leakage at full working voltage you'd be shocked (no pun intended!) to learn how many filter caps had their best years 40 years ago.

[Domocaster]

Hi, thanks very much for that info, muchxs, I will proceed, with due caution.
One other thing you may be able to help with, is it possible to measure Filter Caps or coupling caps with a digital multimeter, whilst there not attached to a circiuit?
I can do it with resistors, but can't figure how to with caps.
Many Thanks, Dom.

[TNO]

I've re-built a bunch of vintage Gibbies and 30-20-10 on the filters works really well. A 5y3 can handle 30uf easily. Bumping up that first filter from 20 to 30 makes an incredible difference in how much bass the amp can handle. Nichicons are available in 30uf or you can parallel a 20 and a 10. Going over 10 on the preamp filters makes for a less lively feel.

[ThermionicScott]

Originally Posted by TNO

I've re-built a bunch of vintage Gibbies and 30-20-10 on the filters works really well. A 5y3 can handle 30uf easily. Bumping up that first filter from 20 to 30 makes an incredible difference in how much bass the amp can handle. Nichicons are available in 30uf or you can parallel a 20 and a 10. Going over 10 on the preamp filters makes for a less lively feel.

FWIW, the 5F2A circuit calls for 32 uF as the input capacitance for its 5Y3. I wonder, though, if the fact that these smaller amps don't have standby switches (and thus don't have instantaneous current spikes) is how they can get away with it...

- Scott

[bradpdx]

GZ34/5AR4s drive 40mfd caps all the time - that is the first stage of the power supply in the Deluxe Reverb, Super, Pro, etc. (actually old Fenders used 2x16uF = 32mfd, but most of these are replaced with 2x20mfd = 40mfd as the amp is serviced). That is not a practical problem.

The practical problem is the plate voltage, which will rise by some amount when compared with the 5Y3. How much? Maybe 20 volts or so.

Before doing this, check the B+ voltage. If it is already high - more than 375VDC, don't bother. If it is low (say, 325VDC) then the GZ34 might make an interesting change.

Increasing the plate voltage this way will give you a bit more headroom, but not a lot. If the amp is already running a high B+, then it might stress the output tube too much and cause it to redplate - but probably not.

The main thing is that with single-ended class-A amps, there is no dynamic "sag" in the power supply with any rectifier tube. So the advantages of the lower internal resistance of the GZ34 are not realized as they would be in a push-pull amp like a Deluxe, etc.

I wouldn't bother with this change, really.

[teleamp]

Originally Posted by Scott S

FWIW, the 5F2A circuit calls for 32 uF as the input capacitance for its 5Y3. I wonder, though, if the fact that these smaller amps don't have standby switches (and thus don't have instantaneous current spikes) is how they can get away with it...

- Scott

My understanding of why a standby isn't used is, that the rectifier tube doesn't pass B+ voltage until its heater element is at working temperature.

[marshman]

This question (or maybe the answer) throws a wrench into my mental works...I guess I was always under the impression that all the filter caps were basically combined, in principal, into one large 'tub' of capacitance, and though it wouldn't matter if an amp was built with 6x10uf or 2x30uf or 1x50uf+1x10uf...apparently that is NOT the case?

[bradpdx]

Originally Posted by marshman

This question (or maybe the answer) throws a wrench into my mental works...I guess I was always under the impression that all the filter caps were basically combined, in principal, into one large 'tub' of capacitance, and though it wouldn't matter if an amp was built with 6x10uf or 2x30uf or 1x50uf+1x10uf...apparently that is NOT the case?

It matters when the capacitors are separated by elements (resistors/chokes) in order to create isolated supply points for different stages of the amp at different voltages.

Otherwise, paralleling caps together just results in one big cap.

[ojaverde]

Can you swap the other way and replace a GZ34 with a 5Y3 in a blackface deluxe? Will that result in more sag?

[eugenedunn]

Can you swap the other way and replace a GZ34 with a 5Y3 in a blackface deluxe? Will that result in more sag?

Generally yes...... however because of the larger amount of voltage drop through that 5Y3GT, it is possible your power tubes might not get the amount of volts they need to be driven optimally.... but that varies with the type of amp and where your power tubes were at to begin with....

[ThermionicScott]

Originally Posted by ojaverde

Can you swap the other way and replace a GZ34 with a 5Y3 in a blackface deluxe? Will that result in more sag?

Yep, and with the current draw of two 6V6s plus a handful of preamp tubes, it's a pretty safe swap in this case. You'll note that the idle dissipation will go way down, so you may consider readjusting the bias to get the best tone.

- Scott

P.S. With a non-adjustible fixed-bias amp, you'd be at the mercy of your resistor values. Here's what some rectifier tube swaps did in my 5E7 clone, with 123 V at the wall. Note that since I'm using true 19W 6L6s, the percentage of idle dissipation will be in terms of that.

5Y3: 387 plate volts (53% idle dissipation)
5R4: 395 volts (61%)
5U4: 405 volts (75%)
5AR4: 425 volts (100%)

Not sure if that was relevant or interesting, but it was for me.

[marshman]

So, Scott, does this mean: For any given healthy amp, the difference in voltage stepping down from a 5AR4 to a 5Y3 will be about 10%? Or the reverse, the step up will about 10%?

I'm guessing that going 'down' (eg-from 5AR4 to 5Y3) will only be detrimental to your tone, but going up could seriously damage something...sound about right?

[bradpdx]

Originally Posted by ojaverde

Can you swap the other way and replace a GZ34 with a 5Y3 in a blackface deluxe? Will that result in more sag?

Yes, and it's a pretty sad sound at that. This type of power supply limiting results in an amp with less headroom and punch, not simply lower wattage. It is certainly harmless to try, but you will need to increase bias current (which will further increase sag).

IMHO this is a lousy way to run a fixed bias amp - just an unexciting result. That's why Fender used the GZ34.

[ThermionicScott]

Originally Posted by marshman

So, Scott, does this mean: For any given healthy amp, the difference in voltage stepping down from a 5AR4 to a 5Y3 will be about 10%? Or the reverse, the step up will about 10%?

Probably varies a lot based on the tube complement. I don't yet possess the fleet of amps I'd need to chart out trends properly.

I'm guessing that going 'down' (eg-from 5AR4 to 5Y3) will only be detrimental to your tone, but going up could seriously damage something...sound about right?

I think that's mostly right, but one other thing to consider is that using too wimpy of a rectifier tube can cause it to arc when you hit the Standby switch (my 5E7 does this with the 5Y3). Supposedly that inrush of current is bad for the tube and filter capacitors, so it's probably not a good practice.

- Scott