Clipping/Distorting issue with new AC15 Build

[Evergreen]

Hello All,

I'm new to the forum here, but I've been building tube amps off and on for several years, and have recently been asked to build an AC15 variant for a local studio.

The amp is essentially the Hoffman AC30 layout with a single channel and the power section halved (similar to the Stout, I'm using the Heyboer 18w trannies.) Only other real mod is a TMB stack and Master volume instead of the cut control. Voltages seem reasonable, a little on the low side, but I am using a 5y3 currently.

What I'm experiencing is a nasty clipping/farting sound present on all notes. If I lower the input signal (i.e. roll back the volume on the guitar or the preamp volume) there is no clipping, but once the input gain reaches a certain amount its almost constant. I've since removed the MV from the circuit just for testing, but it did only clip at at the upper end of the MV. With the MV out of the circuit, its basically constant.

Here's everything I've done so far in troubleshooting: Shielded all input/tonestack wiring, changed the PI components to the Stout config, with the smaller caps and higher resistors (.022uf/470k... though it might be a grid block issue), different speaker, different guitar, two sets of tubes, swapping OT Primaries

Swapping the OT primaries did have some effect, made it less of a squeal and more of a nasty odd harmonic clipping. Like a fuzz pedal cranked to 10. Any advice would be appreciated, I can have some chassis pics up this afternoon.

Cheers!

[Evergreen]

Here's a chassis pic, before shielding the front end wiring. Should give an idea of what i'm working with though

[sluckey]

I see you have grid stopper resistors mounted on pins 1 and 2 of the EL84s. That's a potential problem depending on the manufacturer of your tubes. Not the first time this has come up. Some tubes are manufactured with pins 1 and 2 internally connected together. Others are not connected together. I have several different EL84s. Some have the connection, others don't.

Not saying this is your problem but it could be. If your tubes have pin 1 and 2 connected together then your 1.5K grid stopper is shorted out. I suggest you mount the 1.5K without using pin 1. Then you will never have this potential problem.

To see if your tubes have pins 1 and 2 connected simply measure resistance between pins 1 and 2 on the socket. If you measure 1.5K (or whatever value grid stopper you used) this is not your problem. But if you measure zero ohms this could very well be your problem.

[Evergreen]

I'm measuring the value of the resistor across pin 1 and pin 2. also raised the value to 8k2 for kicks and no real change

[Evergreen]

after some extensive chopsticking, I'm thinking it may be oscillation due to the OT primaries  being close to the choke input and the B+ going to the OT. moving the primaries around I'm getting a lot of squeal and white noise. If I position those wires a certain way the clipping is about 20% of what it was. maybe I should try shielding the ot primaries somehow?

thanks

[Evergreen]

So I ended up swapping out the OT, and that seemed to resolve things almost entirely. With the original OT, I noticed that if I touched the primary side of the transformer, it would oscillate and squeal like crazy. Guess I'll be calling Heyboer on that one.

Anyway, the amp is sounding pretty darn good, only issues at this point is that I'm getting very little preamp headroom (crunchy at about 10 o'clock on the volume) and some pretty serious motorboating and phasey sounds when both volume and master volume are about about 2 o'clock.

any thoughts?

thanks

[Evergreen]

The 40/20/20/20 cap can should be adequate filtration for this amp, right? Still having trouble finding the cause of the motor boating and sub octave ghost note at high volume.

[Willabe]

only issues at this point is that I'm getting very little preamp headroom (crunchy at about 10 o'clock on the volume) and some pretty serious motorboating and phasey sounds when both volume and master volume are about about 2 o'clock.

You might need to use shielded wire for the grids.

[Evergreen]

here's a more recent photo. I've shielded... nearly all wiring haha. still have a nasty clipping that is proportional to input signal. occasionally ghost notes still as well, motorboating has stopped

Update: the unwanted clipping is much less when using 50k grid resistors on the el84s. is that really a permanent solution?

[VMS]

What type of a master volume did you use?

And how did you do this part of the circuit?

[thelonious]

Update: the unwanted clipping is much less when using 50k grid resistors on the el84s. is that really a permanent solution?

If you mean that you raised the value of the 1.5K grid stopper resistors to 50K, I don't see why it couldn't be permanent. While that is higher than the stock output tube grid stoppers you'll find in many amps, it's certainly not abnormal for people to use values that high to prevent blocking distortion and parasitic oscillations. If you're using 470K grid leak resistors, I don't think it will cause any problems. The main things you need to be concerned with are max grid resistance and high frequency attenuation. You're not exceeding the first, and you'd need to go higher than 50K to hear the second. It might be a band-aid on an oscillation problem, but it's an effective band-aid.

[Evergreen]

What type of a master volume did you use?

And how did you do this part of the circuit?

Crossline MV instead of cut control. just removed the cap.

That part of the circuit, I've got a single 220k resistor inline before the .047 cap going to the next 12AX7 grid. Do I need to run that 220k resistor to ground?

[Ed_Chambley]

Do your filaments have a center tap?  Cannot see in the photos, but do you have pin 4 and 5 connected together on your 12Ax7 tubes?  Not sure if these are causing your troubles, I just cannot see a Center Tap.

[VMS]

 Do I need to run that 220k resistor to ground?

You could try to approximate the effect of the missing channel with 330k resistor to ground. This could give you little more headroom.

[Evergreen]

330k to ground instead of the 220k inline? I'll definitely try it.

and yes... heater winding CT grounded to transformer bolt. pins 4/5 connected on the preamp tubes.

[VMS]

330k to ground instead of the 220k inline? I'll definitely try it.
 

no, try like this and you probably could start the R19 with 220k and if you lose too much signal increase the value.