SSS again

[thyx]

The Peavey PT has a 56V tap for bias. If I run a voltage doubler from that tap I might come close, according to the Duncan PSU designer I'll get 130V.

I CAN get a - voltage from a doubler right?

Close, but no cigar. It has to be at least 150 when using Dumbles SSS circuit, or you might not have enough negative bias at the control grids of the 6550s. When done, you should be able to adjust the bias to read -60 to -75 (or thereabouts) at the cathodes of the drivers. That's the range that will allow you to properly bias the amp for 6550s.

[Bob-I]

The Peavey PT has a 56V tap for bias. If I run a voltage doubler from that tap I might come close, according to the Duncan PSU designer I'll get 130V.

I CAN get a - voltage from a doubler right?

Close, but no cigar. It has to be at least 150 when using Dumbles SSS circuit, or you might not have enough negative bias at the control grids of the 6550s. When done, you should be able to adjust the bias to read -60 to -75 (or thereabouts) at the cathodes of the drivers. That's the range that will allow you to properly bias the amp for 6550s.

No, this amp will use 6-6L6's same as the Peavey so the bias ckt will handle this just fine. The concern is for a direct coupled driver which apparently will require more current than this will supply.

[thyx]

The Peavey PT has a 56V tap for bias. If I run a voltage doubler from that tap I might come close, according to the Duncan PSU designer I'll get 130V.

I CAN get a - voltage from a doubler right?

Close, but no cigar. It has to be at least 150 when using Dumbles SSS circuit, or you might not have enough negative bias at the control grids of the 6550s. When done, you should be able to adjust the bias to read -60 to -75 (or thereabouts) at the cathodes of the drivers. That's the range that will allow you to properly bias the amp for 6550s.

No, this amp will use 6-6L6's same as the Peavey so the bias ckt will handle this just fine. The concern is for a direct coupled driver which apparently will require more current than this will supply.

The SSS uses a directly coupled 12AX7 and drives from the cathode...no problem. This circuit ALSO provides the bias. If you set up the circuit correctly (see the SVT), the max current it will have to draw is less than 2mA (per side) before running into forward biasing the power tubes. But you need the 150 volts if you intend to use a cathode resistor of 100k, as in an SSS. 130 volts won't cut it.

[David Root]

Recently I drew up this section mainly to get the negative bias insertion in to the grid and tail of the driver stage clear in my mind.

Please check it out and critique at will. I am using B- =-175V because I will be using a separate 120 VCT transformer to generate it, so it will come out around -175V because the primary is for 115VAC.

[Bob-I]

Recently I drew up this section mainly to get the negative bias insertion in to the grid and tail of the driver stage clear in my mind.

Please check it out and critique at will. I am using B- =-175V because I will be using a separate 120 VCT transformer to generate it, so it will come out around -175V because the primary is for 115VAC.

Excellent. I never thought of using a separate xformer but why not?

[David Root]

This is a big enough chassis (rj's) 26 x 8.5 x 2.75 to do stuff like that in!

It will be inside the chassis as I already have PT, OT, choke and 6V3 PT for the 12AX7s on top!

[lovetone]

Sorry to jump in, i have used this idea before, you can use a 6.3v filament transformer in reverse and run if from the heater winding. this would give you 110 / 120 ac depending on the mains winding of the transformer.

I have used the method and it works well.

Geoff

[Bob-I]

Sorry to jump in, i have used this idea before, you can use a 6.3v filament transformer in reverse and run if from the heater winding. this would give you 110 / 120 ac depending on the mains winding of the transformer.

I have used the method and it works well.

Geoff

Don't be sorry...

I'm not sure this PT will handle all the heaters as it is. I'm beginning to think I should just go with a cap coupled CF, the bias is just too problematic.

[John_P_WI]

Actually the reversed transformer off of the filament supply should draw very little current. I would not be worried about that. What you will gain is a low impedance supply that is very robust. Well worth it. You can also use a 120 / 12 v transformer reversed very effectively.

[wake010]

The only issue that might occur is an imbalance on the secondary of the transformer due to the draw from the 12BH7.

SVTs used -190V for their rail, but they also have a secondary winding to development that supply.

Gary

I second that .. The Marshall method on some builds can get mushy and develop quite a lag time to recover due to the in-balance.. This can effect tone greatly especially when playing at loud volumes for long periods.. Your bias doesn't recover as quickly so the tubes run hotter(or colder bias however you look at it)..This maybe more of an issue with the current hungry BH... You can speed up the recovery time/constant by going with smaller say .022 grid couplers as well as 100K bias feed resistors instead of the 200k's..Also fast reacting diodes in the bias supply will also help some.. Doublers can also have an effect on the lag time..Just be aware

Tony[/quote]

How about you run a (negative) full wave diode pair off both sides of the main HT winding, ie in parallel with the main positive supply. That would kill the imbalance caused by running the Marshall style bias pull off only one side.