https://ampeg.com/support/files/Schematics/S Series/SVT (1972, 6550 tube)/SVT 1971 6550 Schematics.pdf
I see two problems...
1. The plates will be running even higher than the already high 695VDC due to less tubes pulling the voltage down. No idea how tubes would fair at say 750+ VDC but probably pushing it.
2. The OT's impedance will be mismatched.
The possible problem solutions I see here are:
1. Plate voltage
a. use a separate power transformer for the plate voltage on the power tubes. This will also mean the bias will need to be different. I have no idea what kind of current the grid bias draw is, but I'd think this could be dealt with by divider resistors being adjusted to alter that voltage.
b. maybe an easier or cheaper solution than a. could the original power transformer be used, but voltage dropped somehow? Maybe a solid state relay with PWM.
c. I have no idea here but one other thought... tube rectifier(s). I seriously doubt there's any tube rectifier that could deal with this kind of power... but maybe one other alternative?? EDIT: forgot about the rectifier heater supplies. Possible maybe, but not at all practical.
2. OT Impedance mismatch
a. change the wiring of the speaker cab to get the impedance up to 3x what it originally calls for since we'd have 1/3 of the tubes and roughly 3x impedance on the output transformer's primary side.
Probably a little crazy to consider this, but it'd have a few advantages:
1. Less cost (and abuse) on the power tubes over the life of the amp.
2. Crank an actual SVT without having to be at crazy power and volume levels.
3. Still owning an SVT that could be put back to stock power later if needed.
I see two problems...
1. The plates will be running even higher than the already high 695VDC due to less tubes pulling the voltage down. No idea how tubes would fair at say 750+ VDC but probably pushing it.
2. The OT's impedance will be mismatched.
The possible problem solutions I see here are:
1. Plate voltage
a. use a separate power transformer for the plate voltage on the power tubes. This will also mean the bias will need to be different. I have no idea what kind of current the grid bias draw is, but I'd think this could be dealt with by divider resistors being adjusted to alter that voltage.
b. maybe an easier or cheaper solution than a. could the original power transformer be used, but voltage dropped somehow? Maybe a solid state relay with PWM.
c. I have no idea here but one other thought... tube rectifier(s). I seriously doubt there's any tube rectifier that could deal with this kind of power... but maybe one other alternative?? EDIT: forgot about the rectifier heater supplies. Possible maybe, but not at all practical.
2. OT Impedance mismatch
a. change the wiring of the speaker cab to get the impedance up to 3x what it originally calls for since we'd have 1/3 of the tubes and roughly 3x impedance on the output transformer's primary side.
Probably a little crazy to consider this, but it'd have a few advantages:
1. Less cost (and abuse) on the power tubes over the life of the amp.
2. Crank an actual SVT without having to be at crazy power and volume levels.
3. Still owning an SVT that could be put back to stock power later if needed.
Last edited:
Use 4X the impedance. That will work better than 3X when you consider the higher unloaded B+. 16 ohm speaker wiring on a 4 ohm tap would do it. As would using a 100W 8k OPT, which you can get (Edcor, which would give you back impedance flexibility). This will optimally run one pair of 6550 on 700 ish volts. Either way It wouldn’t be any more or less abusive than running full power off 3 pair, but considering what 6550’s are going for these days you can see the logic behind it. If you ever go back to original operation, you’d want 6 fresh tubes (don’t just put 4 more back in after two have a lot of hours on them).
One of the old GEC KT88 data sheets had an up-to-five-pair design where you change the OPT primary along with the number of pairs. It’s essentially doing the same thing.
One of the old GEC KT88 data sheets had an up-to-five-pair design where you change the OPT primary along with the number of pairs. It’s essentially doing the same thing.