Best way to shut off the signal to one of the output tubes?

[ontariomaximus]

Just finished an Express (kelly '90). TONS of fun I tell ya'! What I want to do though is shut off the signal to one of the output tubes for a quarter power option. In a cathode biased amps it's easy, just ground the grid of one of the output tubes. In a fixed bias amp, its a little harder because grounding the grid also grounds the -bias supply. So can I switch a capacitor across the LTP phase inverter's 470 ohm cathode resistor to get the same effect? Any other ideas would be appreciated.
Also what is the most popular feedback resistor 100k or 47k? My estimate is that 100k offers very little feedback. 47k does help lower the noise, but is there a sonic preferance?

[sliberty]

Shutting off power tubes is a popular technique when you have 4 power tubes to work with. You shut off two, and all is well (but quieter). However, when you only have two power tubes, you cannot shut one of them off without seriously bad tone resulting. You would be better off just trying out a pair of 6V6's (which the Express does support). If that is not quiet enough for you, other options include Power Scaling, speaker attenuation (build an AirBrake, maybe even build it into the Express), lower efficiency speakers, etc.

[mlp-mx6]

Kevin O'Connor demonstrates techniques where you do shut down one tube of a pair, resulting in semi-single-ended tone. It is not impossible, I don't think. The quiescent current is still flowing through the other tube, so the OT is in balance. Keep in mind that at low volumes a class AB amp is actually class A at lower volumes.

[mhuss]

It's a simple trick I first saw in Mark Durham's Vajra amp. Just remove the drive signal from one of the tubes (keeping the bias the same). The bias keeps the idle current going through the unused half of the OT to avoid premature magnetic saturation, and both halves of the PI put out a full 360 degree waveform.

--mark

[CaseyJones]

Yeah, and Randall Smith says he thought it up in his swimming pool!

[ontariomaximus]

It's a simple trick I first saw in Mark Durham's Vajra amp. Just remove the drive signal from one of the tubes (keeping the bias the same). The bias keeps the idle current going through the unused half of the OT to avoid premature magnetic saturation, and both halves of the PI put out a full 360 degree waveform.

--mark

Thats what I am trying to accomplish. I suppose the easiest way is just to put a switch between one of the PI's coupling caps and the 220k/1k5 resistors. As I said earlier its easy to remove the drive signal from the one of output tubes, but is trickier with fixed bias. I was just looking for a different way to do it, perhaps by bypassing the PI's cathode resistor.

[Rick]

It's a simple trick I first saw in Mark Durham's Vajra amp. Just remove the drive signal from one of the tubes (keeping the bias the same). The bias keeps the idle current going through the unused half of the OT to avoid premature magnetic saturation, and both halves of the PI put out a full 360 degree waveform.

--mark

Thats what I am trying to accomplish. I suppose the easiest way is just to put a switch between one of the PI's coupling caps and the 220k/1k5 resistors. As I said earlier its easy to remove the drive signal from the one of output tubes, but is trickier with fixed bias. I was just looking for a different way to do it, perhaps by bypassing the PI's cathode resistor.

Funny you should say that, I had inadvertently left a poor solder joint on the cathode connection that goes across the phase inverter once. The amp worked and sounded not too bad, but it had very low power. I finally spotted it and resoldered it and the power went way up back to normal (LOUD) on a 36W TMB.

[dartanion]

Keep in mind that at low volumes a class AB amp is actually class A at lower volumes.

Sorry, but this is just not true. Class of operation is determined at full unclipped output.

[mlp-mx6]

OK. Then let me state it this way - at low volumes, both tubes conduct the full signal top to bottom. Until the signal swing exceeds the bias voltage the tube never cuts off.

[ampdoc1]

mhuss,

The onlly Vajra schematic I can find shows a switch to go from fixed bias to cathode bias. The description sounds pretty simple, but do you have a schematic of the Vajra you mentioned?

Thanks,
ampdoc

[dartanion]

It's not actually a switch, it's the pot paralleled across the gridleak on one power tube. This is effectively half a PPIMV. I have used this on a few different builds and it is really cool. I have this on both my 18watt and 30watt Liverpool type builds.

[ampdoc1]

Thanks Dar,

I had on my "far away" glasses when I was peering at the schematic!

ampdoc

[d95err]

In a cathode biased amps it's easy, just ground the grid of one of the output tubes. In a fixed bias amp, its a little harder because grounding the grid also grounds the -bias supply.

Look at it this way: What the Vajra mod does is essentially to shunt one of the grid bias feed resistors. In a cathode biased amp, the grid bias voltage just happens to be 0V. So, you should be able to do the same in a fixed biased amp.

At least, I can see no reason why it wouldn't work.

[novosibir]

You just may remove one of the 220K bias feed resistors and put/wire a 250K log. pot in this place.

- Output lug connected to the bias supply voltage
- Input lug connected to the coupling cap
- Wiper connected to the output tube's control grid

Watch for a low measuring 250K pot around 220K, there are plenty among the Alpha's.

I've tried it, but to be honest, I don't like the sound of a diminished or fully switched off side of a push-pull power stage! Simply sounds weak IMO.

Larry

[dartanion]

I've only tried it on a couple of Wreck builds, and I do enjoy the tonal variety. I actually like it in full SE mode, but that's my opinion. It is an easy mod to try and remove if you don't like it.