6G3 power supply question

[billr]

Hello.

I've just built myself a 6G3 clone. it's all working well, i've fitted a master volume control, and put a pot in the feedback loop, just wanted to play with the feedback to see what the overall effect is.

thing is, the power supply.

On the fender diagram, it shows, 333-0-333 into a GZ34, with an output of 375V. i can't see how that works as it's a RC filte, and i would expect 1.4 * the input ac voltage, or thereabouts. And sure enough in my build, i'm getting 450V, (i had a custom wound Xformer, which is about 13V higher on the outupt, but that doesn't explain the difference) with all the corresponding Anode/Plate voltages suitably raised.

I modelled this PSU on PSUD and sure enough, 450 odd volts.

Should i be concerned, anode current seems fine, just that the various voltages are elevated, the PSU caps are close to their tolerance with bothers me a bit.

ANy thoughts?

Lookforward to hearing from someone in case i am missing something.

kind regards

BIll

[Roe]

it depends on the current draw. is it 375v at idle?

[martin manning]

On the fender diagram, it shows, 333-0-333 into a GZ34, with an output of 375V. i can't see how that works as it's a RC filte, and i would expect 1.4 * the input ac voltage, or thereabouts. And sure enough in my build, i'm getting 450V, (i had a custom wound Xformer, which is about 13V higher on the outupt, but that doesn't explain the difference) with all the corresponding Anode/Plate voltages suitably raised.

You are correct, the only way the Fender schematic voltages would make sense is if the 333V PT secondary voltage is at unloaded condition, which would likely mean ~10% drop to ~300V loaded. If you have the heater current for it, a 5U4GB rectifier would bring all the voltages down ~30V.

Another option is to leave the plate and screen voltages where they are, and adjust the 10k feeding the PI node to bring the preamp voltages into line. A 22k 1W should get you close. I would also consider adding screen and control grid stopper resistors to the 6V6 sockets (as in AB763 Deluxe Reverb). That will help keep them happy at the high plate voltage.

[Stevem]

Put a 5V4 in for 15 to 20 volt drop in your V+, or a 5U4 if you want a 30 volt or so drop.

I could never see a need for a master volume in a Fender style amp like this because the character , feel and response of the output section is a intergril part of what this amp is cherished for, and the output section needs to be able to be hit with a certain amount of PI drive signal.

Do you know how feedback used within an amps circuit affects its sound and response?

If this amp is working well and your not playing it with the volume at or above 5 you might not hear much of a difference by twisting that feedback pot around.

[martin manning]

I could never see a need for a master volume in a Fender style amp like this because the character , feel and response of the output section is a intergril part of what this amp is cherished for, and the output section needs to be able to be hit with a certain amount of PI drive signal.

I don't know if I buy that. Fender power amps are generally by-the-book, and therefore pretty clean. Any unique distortion character in this 6G3 may well be in the preamp. It has 220k plate and equivalent 3k cathode resistors on the input stage, 5E3 tone and volume controls, and an off-center 115k/1k5 split-load on the gain/mixer stage.

[Stevem]

It’s just that to me if your not using a amp like this for it’s great near cranked up rock sound ( played above 6 ) which comes by thru way of its rectifier sag and the recovery sustain that stems from that reverse action, then what the heck is the point?

[martin manning]

I don’t think there is much sag in a GZ34 when it’s powering a 2x 6V6 power amp.

[syscokid]

… thing is, the power supply.

I took the band-aid approach to my recent 6G3 clone build regarding higher than comfortable voltages.

Backstory:
I sourced the set of trannies from Modulus Amplification. Website specs for the Modulus Deluxe PT is 355-0-355 Unloaded. However, a month before I ordered the trannies from Modulus, I downloaded the PT’s spec sheet from the site and it stated the voltage as 325-0-325. I assumed that this was the loaded voltages. To my surprise, when I first tested the PT’s voltages without any tubes installed, I measured 375 vac at pin 6 of the rectifier socket. And with all tubes installed, pin 4 & 6 measured 356 and 348 respectively.

At first, I swapped the GZ34 for a 5Y3. B+ went from 437v down to 412v. The plates of the JJ6V6GT’s went from 430 down to 406.

Band-aid approach:
I reinstalled the GZ34 and a couple of weeks ago, I purchased from Granger Amp’s their Amplified Zener Voltage Drop Kit and opted for their 51v drop with a chassis mount MOSFET. B+ dropped 57v down to 380v, and 6V6 plate voltages are now at 372v @ 22 mA’s for about 60% dissipation. All voltages measured at 120v, controlled with a variac.

[billr]

On the fender diagram, it shows, 333-0-333 into a GZ34, with an output of 375V. i can't see how that works as it's a RC filte, and i would expect 1.4 * the input ac voltage, or thereabouts. And sure enough in my build, i'm getting 450V, (i had a custom wound Xformer, which is about 13V higher on the outupt, but that doesn't explain the difference) with all the corresponding Anode/Plate voltages suitably raised.

You are correct, the only way the Fender schematic voltages would make sense is if the 333V PT secondary voltage is at unloaded condition, which would likely mean ~10% drop to ~300V loaded. If you have the heater current for it, a 5U4GB rectifier would bring all the voltages down ~30V.

Another option is to leave the plate and screen voltages where they are, and adjust the 10k feeding the PI node to bring the preamp voltages into line. A 22k 1W should get you close. I would also consider adding screen and control grid stopper resistors to the 6V6 sockets (as in AB763 Deluxe Reverb). That will help keep them happy at the high plate voltage.

thank your for thoughts, i have s5y3gt and a 5u4 kicking around somewhere. I've put a variable bias voltage cct for the output valves, i think i'll tune in the output current a bit better, and change the inter-cap resistors for the preamp sections. I did add grid stop and screen resistors.

[billr]

… thing is, the power supply.

I took the band-aid approach to my recent 6G3 clone build regarding higher than comfortable voltages.

Backstory:
I sourced the set of trannies from Modulus Amplification. Website specs for the Modulus Deluxe PT is 355-0-355 Unloaded. However, a month before I ordered the trannies from Modulus, I downloaded the PT’s spec sheet from the site and it stated the voltage as 325-0-325. I assumed that this was the loaded voltages. To my surprise, when I first tested the PT’s voltages without any tubes installed, I measured 375 vac at pin 6 of the rectifier socket. And with all tubes installed, pin 4 & 6 measured 356 and 348 respectively.

At first, I swapped the GZ34 for a 5Y3. B+ went from 437v down to 412v. The plates of the JJ6V6GT’s went from 430 down to 406.

Band-aid approach:
I reinstalled the GZ34 and a couple of weeks ago, I purchased from Granger Amp’s their Amplified Zener Voltage Drop Kit and opted for their 51v drop with a chassis mount MOSFET. B+ dropped 57v down to 380v, and 6V6 plate voltages are now at 372v @ 22 mA’s for about 60% dissipation. All voltages measured at 120v, controlled with a variac.

Thanks for the idea on the zener drop kit, i'll look into that.

[billr]

It’s just that to me if your not using a amp like this for it’s great near cranked up rock sound ( played above 6 ) which comes by thru way of its rectifier sag and the recovery sustain that stems from that reverse action, then what the heck is the point?

I understand the point you're making. However, it's just a bit of fun, with the variable feedback, rolled off, it's back to standard, and with the master volume at midpoint (ish) that's standard too.

[billr]

Put a 5V4 in for 15 to 20 volt drop in your V+, or a 5U4 if you want a 30 volt or so drop.

I could never see a need for a master volume in a Fender style amp like this because the character , feel and response of the output section is a intergril part of what this amp is cherished for, and the output section needs to be able to be hit with a certain amount of PI drive signal.

Do you know how feedback used within an amps circuit affects its sound and response?

If this amp is working well and your not playing it with the volume at or above 5 you might not hear much of a difference by twisting that feedback pot around.

I put the MV in there for some options, at midpoint, it's roughly spec.

And i do understand what negative feedback does, lowers noise floor, broadens the BW and flattens the overall frequency response, i just wanted to play with it, that's all. i take your point of not hearing much of a difference, i'm not, still it was only another pot/resistor/cap etc. bit of fun.

[billr]

… thing is, the power supply.

I took the band-aid approach to my recent 6G3 clone build regarding higher than comfortable voltages.

Backstory:
I sourced the set of trannies from Modulus Amplification. Website specs for the Modulus Deluxe PT is 355-0-355 Unloaded. However, a month before I ordered the trannies from Modulus, I downloaded the PT’s spec sheet from the site and it stated the voltage as 325-0-325. I assumed that this was the loaded voltages. To my surprise, when I first tested the PT’s voltages without any tubes installed, I measured 375 vac at pin 6 of the rectifier socket. And with all tubes installed, pin 4 & 6 measured 356 and 348 respectively.

At first, I swapped the GZ34 for a 5Y3. B+ went from 437v down to 412v. The plates of the JJ6V6GT’s went from 430 down to 406.

Band-aid approach:
I reinstalled the GZ34 and a couple of weeks ago, I purchased from Granger Amp’s their Amplified Zener Voltage Drop Kit and opted for their 51v drop with a chassis mount MOSFET. B+ dropped 57v down to 380v, and 6V6 plate voltages are now at 372v @ 22 mA’s for about 60% dissipation. All voltages measured at 120v, controlled with a variac.

Hello again, just looked at the Grainger Kit, very reasonable $US23, however, as i am in NZ, the postage brings that up to $65, which seems a bit steep, i may have a go at building one for myself.

[syscokid]

… thing is, the power supply.

I took the band-aid approach to my recent 6G3 clone build regarding higher than comfortable voltages.

Backstory:
I sourced the set of trannies from Modulus Amplification. Website specs for the Modulus Deluxe PT is 355-0-355 Unloaded. However, a month before I ordered the trannies from Modulus, I downloaded the PT’s spec sheet from the site and it stated the voltage as 325-0-325. I assumed that this was the loaded voltages. To my surprise, when I first tested the PT’s voltages without any tubes installed, I measured 375 vac at pin 6 of the rectifier socket. And with all tubes installed, pin 4 & 6 measured 356 and 348 respectively.

At first, I swapped the GZ34 for a 5Y3. B+ went from 437v down to 412v. The plates of the JJ6V6GT’s went from 430 down to 406.

Band-aid approach:
I reinstalled the GZ34 and a couple of weeks ago, I purchased from Granger Amp’s their Amplified Zener Voltage Drop Kit and opted for their 51v drop with a chassis mount MOSFET. B+ dropped 57v down to 380v, and 6V6 plate voltages are now at 372v @ 22 mA’s for about 60% dissipation. All voltages measured at 120v, controlled with a variac.

Hello again, just looked at the Grainger Kit, very reasonable $US23, however, as i am in NZ, the postage brings that up to $65, which seems a bit steep, i may have a go at building one for myself.

I haven’t bothered to read any of the following, but I’m assuming there’s got to be some good info here…
https://ampgarage.com/forum/viewtopic.php?t=27035

[martin manning]

i have s5y3gt and a 5u4 kicking around somewhere. I've put a variable bias voltage cct for the output valves, i think i'll tune in the output current a bit better, and change the inter-cap resistors for the preamp sections. I did add grid stop and screen resistors.

5Y3 will definitely add some sag. Easy to try, and you might like it. Adjustable bias is a good idea for any bias varying tremolo to get the best out of the effect and protect the output tubes.
I am not a fan of MOSFET voltage reducers as they seem prone to failure, but for this low-power amp it may work well. You can assemble the circuit on a terminal strip and give it a try.