Early breakup on the PI

[Bob-I]

On my 4x6V6 Dumble HRM with D-Lator and reverb, I'm getting what I believe is very early breakup on the PI. I'll need to take more measurements, but total output with 414B+ biased at 70% is 26 watts. I was expecting more like 30, and looking at the various stages it appears that the PI is the first to go.

I switched the reverb and Dlator out and watched the traces on the scope. I see nice symetrical sine waves on each stage of hte clean channel, then after the master volume I see a dramatic drop in voltage swing, then breakup at the PI.

I'm measuring with an Airbrake as I don't have a load resistor big enough to handle this amp without getting very hot.

Here's the schematic.

Discuss

[heisthl]

No grid stoppers on the output tubes? Is that intentional? Those resistors help keep the output tubes from drawing PI current. I use 5K1s on a 6l6 amp. You can also try reducing the size of the input cap and/or the plate caps.

[tubeswell]

What are your PI plate voltages? and what is the Peak 2 Peak voltage at your output tube grids? Don't output tubes go mushy if they aren't getting enough voltage driven into the grids?

[Bob-I]

I have 1.5K grid stoppers, I just forgot to put them on the schem. This is 4 6V6's so I thought 1.5 would be right, should I go higher.

I'll measure the voltages later, at work now.

[Bob-I]

What are your PI plate voltages? and what is the Peak 2 Peak voltage at your output tube grids? Don't output tubes go mushy if they aren't getting enough voltage driven into the grids?

PI plates are at 260V. I'll measure the swing later.

[Bob-I]

I measured the AC voltages. Turns out my calculations were way off, my load is 4 ohms, not 8 . I'm getting about 40V swing at the speaker out, so 40/2*.707=14VRMS. Plug that into this ohms law calculator for simplicity http://ourworld.compuserve.com/homepage ... hmslaw.htm and I've got 49 Watts.

I measured the voltage swing at the power tube grids. I'm seeing just under 80V P-P before clipping. That seems terribly high to me, but I've checked my scope, set at the 2V / division with a 10x probe I see almost 4 divisions.

Does all that make since?

[tubeswell]

80V P-P does seems high for a 6V6. According to Jack Darr, with 315V Plate and 250V screen, a 6V6 needs a 25V P-P input signal. Curious about why did you have 24k/820R in the LTP? I am still learning about LTPs, and my simple understanding is that a LTP is a driver as well as a PI (and you've got several gain stages before that). So maybe you are driving the 6V6 grids too hard? (or maybe not? - I guess it depends on what the plate and screen voltages are). Possibly you could maybe try 10k/470R instead (based on because fender schemes involving fixed bias 6V6PP amps have these values where they have a LTP)?

[Bob-I]

80V P-P does seems high for a 6V6. According to Jack Darr, with 315V Plate and 250V screen, a 6V6 needs a 25V P-P input signal.

Well, this is before clipping, so the 6V6's are handling it. I have 414 on the plates and 411 in the screens.

Curious about why did you have 24k/820R in the LTP?

Because that's what Dumble's use. Remember that this is a 12AX7 LTP, not a 12AT7 like a Fender.

I am still learning about LTPs, and my simple understanding is that a LTP is a driver as well as a PI (and you've got several gain stages before that). So maybe you are driving the 6V6 grids too hard? (or maybe not? - I guess it depends on what the plate and screen voltages are). Possibly you could maybe try 10k/470R instead (based on because fender schemes involving fixed bias 6V6PP amps have these values where they have a LTP)?

I'm learning too which is why I posted. Hopefully some other folks will help out and we'll all learn.

[Fischerman]

Bob, Is it possible you cranked the V-scale trimmer on your scope to get a better look at something and forgot to put it back to the 'calibrated' setting? I've done that before. Just a shot in the dark.

[Bob-I]

Bob, Is it possible you cranked the V-scale trimmer on your scope to get a better look at something and forgot to put it back to the 'calibrated' setting? I've done that before. Just a shot in the dark.

I just checked, no that wasn't it but good thought.

I also double checked my measurements. Turns out the 6V6's are clipping first, not the PI, my bad. The PI seems fine, well balanced and enough voltage swing. The amp's putting out 48 watts so that's good too.

I just somehow expected more clean headroom from this amp, I guess I was wrong.

[heisthl]

reduce the size of the input cap (.02) into the PI - start with .001 and listen to the difference - you will notice a lot more clean headroom at the possible expense of low bass . You can increase from there until you find the correct spot for your amp. This really works - try it. This is one place where a good cap substitution box comes in handy.

[heisthl]

double post

[odourboy]

heisthl - That post belongs in the special book of tweaks - or in Omar's Wiki. Ever thought of writing a book on this stuff?

[heisthl]

heisthl - That post belongs in the special book of tweaks - or in Omar's Wiki. Ever thought of writing a book on this stuff?

Well it works for me but YMMV and lets see what Bob-I says.

[Bob-I]

reduce the size of the input cap (.02) into the PI - start with .001 and listen to the difference - you will notice a lot more clean headroom at the possible expense of low bass . You can increase from there until you find the correct spot for your amp. This really works - try it. This is one place where a good cap substitution box comes in handy.

That's a good thought. Maybe even the power tube couplers too. This amp has a ton of bottom end so this should do the trick.