Blowing Up Phase Inverter Tubes

[Potato Cakes]

Hello, everyone,

Lately, I have been enjoying building and modding tube guitar amps. I enjoy how they sound and I find that I enjoy playing guitar again. More so than when I first picked up a guitar. Recently I have been experiencing the phase inverter tubes blowing up when I start to push the volume. Specifically this occurs in a two different push-pull amp (Spitfire and modded Champ) both which have a master volume control which shorts each output of the phase inverter tube via a 1M ohm potentiometer. I have read differing opinions about the use of master volume controls on tube amps so I'm not sure if this is culprit in these amps. If these are, is there a better way to implement a volume control at the output section?

I've attached schematics of the amps to which I am referring. The modded Champ does not show the master volume control. It is setup the same way as the Spitfire.

Thanks!

Paul

 

[radardoug]

Probably a breakdown between the tube heater and the cathode, which you have sitting at more than 100 volts.

 

[Potato Cakes]

The cathode of the inverter on the Spitfire is 62V. The other cathode voltages are 1.26V and 11.2V. I do not see a cathode voltages on the modded Champ schematic but I could measure it when I open it up again.

I have gone through several different tubes at the inverter position, so I'm less inclined to think it's a tube problem and more of a circuitry issue. Unless all of the various tubes I tried (NOS and new production) were all bad, which for me is possible.

Thanks!

Paul

 

[ccaudle]

Is having a pot capable of putting a short across the outputs of the long-tail pair like that common practice? I know the output impedance of a common cathode stage is pretty high, but as a solid state guy it just looks weird to short directly between the outputs of a diff pair.

 

[Potato Cakes]

That's what I don't know. The methodology I described was what I pulled from a couple of Matchless schematics I found. So maybe instead of shorting the anodes of the of the inverter after the capacitor it would be better to use a dual gang potentiometer to shunt the signal to ground.

I'm pretty new to tube guitar amps so I don't know if master volume controls are a bad idea in general or if there is a better way to do this electrically.

Thanks!

Paul

 

[musipol]

I prefer the dual pot version myself, but I have seen the single used in a number of amps and have not seen a PI tube killed because of it. My 2c

 

[metalockpick]

The cathode of the inverter on the Spitfire is 62V. The other cathode voltages are 1.26V and 11.2V. I do not see a cathode voltages on the modded Champ schematic but I could measure it when I open it up again.

I have gone through several different tubes at the inverter position, so I'm less inclined to think it's a tube problem and more of a circuitry issue. Unless all of the various tubes I tried (NOS and new production) were all bad, which for me is possible.

Thanks!

Paul

1) The Electro Harmonix 12AX7EH will absolutely not work in the champ circuit, due to the cathode breakdown voltage issue mentioned by Radardoug.
2) That EH tube (and some others) don't comport to the 12AX7 specification, but "work" in most circuits, but will often fail in cathode followers or circuits like that modded champ.
3) All tubes eventually, absolutely FAIL. Sometimes, it's shortly after turning the thing on with a new tube. Tubes I see today have a higher fail rate than they used to, and that's without pushing the voltages. Also, most manufacturers test their tubes before they go out. The good ones get gold plated pins, but even some of those can fail out of the box.
4) Spitfire circuit looks fine. If your tubes are failing in that one, smash them (so you or someone else doesn't go through this again) and get better tubes. Or just use them in Tremolo circuits (if the work).

 

[Potato Cakes]

The tubes I had used were new production and NOS, some of which I got from Christian Whitmore from Pro Audio Tubes, who is an awesome resource for all things tube. I also did not have any 12AX7EH tubes to be used in the process. I know that tubes fail, however the process to make them fail at the phase inverter section was repeatable, no matter the type of tube being used causing me to believe the issue is something circuit related.

I think I will switch the master volume controls to the dual gang method. I like this approach over shorting the both outputs of the inverter tube. I have some replacement tubes coming this week. I will install and report back.

Thanks!

Paul

 

[Potato Cakes]

Is having a pot capable of putting a short across the outputs of the long-tail pair like that common practice? I know the output impedance of a common cathode stage is pretty high, but as a solid state guy it just looks weird to short directly between the outputs of a diff pair.

I don't know if it's common but it is how it is done per the Matchless schematics I was using as an example.

 

[pucho812]

1) The Electro Harmonix 12AX7EH will absolutely not work in the champ circuit, due to the cathode breakdown voltage issue mentioned by Radardoug.
2) That EH tube (and some others) don't comport to the 12AX7 specification, but "work" in most circuits, but will often fail in cathode followers or circuits like that modded champ.
3) All tubes eventually, absolutely FAIL. Sometimes, it's shortly after turning the thing on with a new tube. Tubes I see today have a higher fail rate than they used to, and that's without pushing the voltages. Also, most manufacturers test their tubes before they go out. The good ones get gold plated pins, but even some of those can fail out of the box.
4) Spitfire circuit looks fine. If your tubes are failing in that one, smash them (so you or someone else doesn't go through this again) and get better tubes. Or just use them in Tremolo circuits (if the work).

I can attest to higher failure rate. At my former gig we would buy tubes by the 1000 at a time. Although tested by manufacture, we did our own testing. While tubes didn’t outright fail, we rejected a good 20% for being microphonic. The manufacture would replace them without question. I suspect they resold them elsewhere. But that is a guess.

 

[murrayatuptown]

I assume your pots are 'normal' with the wiper at the center terminal.

I have run into odd pots like multi-turn ones that have the wiper elsewhere and Japanese ones (usually dual-gang) that have really puzzling pin patterns I won't even attempt to describe...maybe they work better for PCB layout vs. eyelet/lug wiring...there must be a reason...

 

[radardoug]

Your B+ on the PI is 311 volts. Because of the resistances around the tube, you will have more than 150 volts on the cathode. Get you meter out and measure it. 150 volts will kill the tube. PIs are normally not this stressed.

 

[Potato Cakes]

When I get home next week I will.

I did wind up using the modded Champ for entire recording session after changing out the phase inverter once again. The input volume was kept about halfway and the master volume was disengaged. Sounded really great. I'll still check the voltages and report back.

Thanks!

Paul

 

[Potato Cakes]

In the meantime, I think that it is noteworthy that this schematic has been used many times by other builders and I can't find any information about the cathode voltage on the inverter being a problem.

Thanks!

Paul

 

[Potato Cakes]

Got new (NOS) preamp and phase inverter tubes from Christian at Pro Audio Tubes. Cathode voltage measurements as follows:

V1:
Pin 3 - 1.3V
Pin 8 - 1.3V

V2:
Pin 3 - 1.5V
Pin 8 - 42.5V

V3 & V4:
Pin 8 - 20.4V

I used the modded Champ (Tweedle Dee schematic) today for 6 hours straight for a recording session with no issues. It sounds awesome. I did not crank it to 10 as I used pedals on the front end to drive the input, so I don't know if the phase inverter tubes would give out again. At this point I don't if I want to keep testing tubes to these extremes.

Thanks!

Paul

 

[radardoug]

Your circuit does not have tube numbers by the tubes, so we have no idea what those voltages refer to. But on your circuit, there is 300+volts on the PI. This voltage will divide in thirds between the tube, the plate resistor, and the cathode resistors. This follows the constant current formula, the current is the same in the three parts. So you either have the wrong voltage going in the top, or the stage is not drawing optimal current.

 

[Potato Cakes]

V1/V2 are 12AX7 and V3/V4 are 6V6. I went with a half wave rectifier instead of using a GZ34 or any other tube rectifier because of space restrictions with the chassis. The voltages at the filter caps are close to what is shown on the schematic so it could be the latter of your options. I am fairly new to tube guitar amps, which I have found that quite a bit of decision making is based around subjective tastes of individuals. I know that this design is based off of Dumble modifications for the Fender Champ so if that particular section is not drawing optimal current there is probably a reason.

Also, my initial problem may be related to the above posts related to tube failure. I have another session tomorrow and we'll see how that goes.

Thanks!

Paul