Motorboating returned, and this time it's a problem

[daz]

remember my problem that really wasn't a problem....motorboating only as the amp powers down after hitting the switch? well, for some reason out of nowhere it has returned, only this time it's constant. the kicker is it only does it with a celestion G12-H100. But if i use my EV it doesn't so apparently the slight difference in resistance in the celestion sets it off. Both are 8 ohms but the EV reads something like 5.6 ohms and the celestion i believe read about 7. (5.6 is normal for EV 8 ohm'rs...i've had several and they all read about there)

So can anyone tell me how i might start troubleshooting this? Oh, and it is greatly affected by my variable NFB pot as would be expected.

EDIT: great...i just tried it again a minute after posting this and now it won't do it. More help....how to i set it off again so i can troubleshoot it?! crazy shite man. I'd like to know which component i can change the value of to put it far out of spec for motorboating. i think this may also be affecting the tone even when it's not doing it.

[Enzo]

How about instead of impedance differences, it might be due to resonant frequencies.

Is this a combo amp? If so, pull the chassis and get it as far from the speakers as possible. Does it still do it?

Bang on the chassis. Does it respond at all? It shouldn't. I'd be thinking about a ground connection that opens intermittently. Like maybe a filter cap.

[Jag]

About five years, or so, ago I chased motorboating in a Bassman for about two weeks. It was a bad bias supply cap so, my money is with Enzo.

[daz]

Well, it's cathode biased, and as for speaker orientation, the speaker it started motorboating with was in a different combo. I have 2 combos. the amp i'm talking about has a EV12l in it. Next to it is a smaller 6v6 combo with the celestion. i plugged the amp into that speaker when it started motorboating. So it was further from the amp than the amp's internal speaker, the EV.
But in any case it may not matter anymore because i have made many changes to the amp in the last week, and tonite i realized some of them while sounding better turned out not to be good in a mix. So i changed a lot of things back. therefore this issue will likely be gone as of now. But i'll post back if not. so crazy how what seems like killer tone sometimes doesn't work so great in a mix. But thats the real challenge isn't it?

[Wakculloch]

That your amp motorboats under certain conditions is an indication that there is instability in the power amp. The conditions that cause it to motorboat are not the problem.
Check the layout of the feedback circuit, how much feedback your using and if presence control components are causing excessive phase shifts.
Grid stoppers on the PI and power valves can certainly help, 1K and 10K respectively.
Other than that it may be physical layout of components.

Hope this helps.

[daz]

Layout seems a likely one. I'll see what i can do, tho it's not doing it at all now. I don't get that.

[loudthud]

See if raising or removing the feedback resistor will kill the motorboating. That would indicate a phase issue in the power amp. If it still motorboats with the feedback removed, then look at lead dress, layout or power supply caps.

[daz]

See if raising or removing the feedback resistor will kill the motorboating. That would indicate a phase issue in the power amp. If it still motorboats with the feedback removed, then look at lead dress, layout or power supply caps.

I have variable NFB via a pot and a 33k resistor, so max NFB is 33k and min is 83k. (50k pot) It DID affect the motorboating as i turned it, but like i said the motorboating is gone now. But i have no dount this phase issue you say it is still exists and is set off when conditions are just right. i just don't know how to look for a phase issue. I have no scope and wouldn't know how even if i did.

[loudthud]

I suggest you try increasing one or both of the caps between the PI and the output tubes.

[Steve Conner]

We've been through this with Daz before. I believe that his tweaking for ultimate tone has led him to values of coupling capacitors that make the power amp unstable, by building up enough phase shift in the NFB loop that the negative feedback becomes positive. He probably got there because the positive feedback causes an increasing bass boost before it finally goes unstable, and maybe he was chasing that "rich tone".

I explained the theory behind this in a previous post, but broadly speaking, it's how the tremolo oscillator works. Every capacitor in the power section contributes to the phase shift, including the PI's input and feedback capacitors.

The cure I suggested was to change the whole power section back to the Marshall standard circuit, but I don't see him doing this.

I also insist that motorboating is NOT caused by layout or lead dress. The time constants associated with stray capacitances and mutual inductances between wires are far too short: they lead to oscillations in the ultrasonic and RF region. The lowest oscillation frequency I've ever got from bad lead dress was about 1kHz.

[daz]

The cure I suggested was to change the whole power section back to the Marshall standard circuit, but I don't see him doing this

Uh, it IS ! It's never had anything but standard .022uf coupling caps (aside from brief trying of different values that came out in minutes) and a .1uf cap for the NFB. except for being cathode biased everything from the PI on back IS pure marshall. Theres a variable NFB pot, but thats no different than a set rsistor, and i usually have it set where a standard value would be. So unless it's because it's cathode biased i don't see anything i can do to change it to "pure marshall". And it sounds much better cathode biased, and since it's not doing it now i'm sure not going to return to fixed.

Oh, and the experimentation with different values of NFB caps was only a temp thing i tried for a few days and it's been .1uf since. Plus it never motorboated when i was using the non standard values.

[Steve Conner]

Well then, if all the other components are Marshall spec, it must be your cathode bypass caps on your power tubes. They will introduce extra phase shift to the loop.

If you're determined to stick with cathode bias, then maybe you can compensate for it by reducing the phase shifts elsewhere. So maybe making the coupling caps to the power tube grids 5 times bigger, or the PI input coupling cap, or something.

[daz]

It's not likely those because i changed to different values and i'm now back to the original values and it's done it both ways. Maybe a clue is that when it did it fully while the amp was on i had plugged into a celestion speaker. I've used it many times so it's not the speaker itself, but the difference in resistance must have set it off. By the way, i also recently changed the cathode reistors and went to a higher value and it did it had done this before and now after. In any case it hasn't done it since, but i haven't been using the celestion either. And if you recall i said it just stopped for no reason....well, that was with the celestion still plugged in, tho i'm back to the EV now. But whatever the reason it hasn't done it since so i'm just gonna forget about it unless it returns.

[octal]

The only other cause of phase shift I can think of is the primary inductance of the output transformer. If that's lower than normal it can cause bass roll off & phase shift.

I do wonder if the unequal values of cathode bypass caps on the output tubes might not have something to do with it. Let me put it this way; if you hear a difference in the tone caused by the unequal values, that probably means that they are rolling off inside the audible pass band and causing phase shift which can cause instability. I'd go for much bigger and equal value caps.

[daz]

I do wonder if the unequal values of cathode bypass caps on the output tubes might not have something to do with it

Yeah, i think thats what steve was saying, and it does make sense. But like i said to him i had put them back to 22uf on both tho, but now that i think about it i may have done that AFTER it motorboated and not before. But it's 22uf/22uf now, so if it doesn't do it again i'll know thats it. On a related note, the reason I can't stop tweaking this thing is that every time i make a change it seems like a number of other things much then be changed and before i know it the thing is worlds different and i end up going back to square one. (that being how it was a few weeks ago before i started phase to of my endless journey) But in doing this i learn a lot of valuable lessons in how various things relate to each other.