Ampeg B22X with multiple problems

[SoulFetish]

The amp sat untouched for the time it took to receive the 6U8 tube and adapter. I installed it, and with nothing plugged in, both volumes down, the machine gun sound happened as soon as the tubes warmed up. I put the 7199 back in, some result. I am including a video link to see if someone may have any insight of what this is, or how to approach it, because I really just don't know what else to do about this.

https://drive.google.com/file/d/1lAn...ew?usp=sharing

okay, that was helpful... in that it was not quite what I was imagining. Ooof... so, I guess I was imagining something the speed of Hendrix's Machine Gun. But it was a good description.
From your description earlier, I was thinking that you may have had a coupling cap leaking DC onto a grid. This causes motor boating, but I've never experienced that in the frequency you have on the clip. Plus, in my experiences, it starts out quiet and builds louder and louder.
I'm going to go back and read through this again to see what you've done and what you haven't checked. But, can you isolate the stage? If you pull preamp tubes starting with V1, and going from there, is there a point before the output tubes it stops?

[Randall]

Be careful with the tube number system with this one. V1 is actually a power tube. I can tell you it does the machine gun sound with all tubes removed except V3, the 7199 PI tube, and the power tubes. Same result with the 6U8A and socket adaptor.

[SoulFetish]

Be careful with the tube number system with this one. V1 is actually a power tube. I can tell you it does the machine gun sound with all tubes removed except V3, the 7199 PI tube, and the power tubes. Same result with the 6U8A and socket adaptor.

Try lifting C19 & C20 one at a time. Or even lift one side of both, and tack in new coupling caps.

[Randall]

How bizarre this all is. I replaced C 19 and C20 with .1uF as that was the closest I could get to .33uF. Turned it on, and no machine gun! "AH", I thought, Soulfetish nailed it. I then put the originals back in one at a time to see if I could narrow it down, but guess what? Still no more machine gun. Neither of them seem to be leaking. I put them in series with my meter from my 350vdc supply, They are both back in circuit now and still no more machine gun. How does simply removing and returning two coupling caps cure this odd sysmptom?

I will mention that one of the .33uF caps is not like all the other flattened blue caps on the board, it is a yellow rectangular cap. I wonder if it was replaced at some point, and possibly a clue to what I am facing?

The current symptom is channels one and two pass signal to a point. It is clean up unttil about halfway, and then it goes into the low, loud howling oscillation as before.

I am just getting my butt kicked on this one.

[nosaj]

How bizarre this all is. I replaced C 19 and C20 with .1uF as that was the closest I could get to .33uF. Turned it on, and no machine gun! "AH", I thought, Soulfetish nailed it. I then put the originals back in one at a time to see if I could narrow it down, but guess what? Still no more machine gun. Neither of them seem to be leaking. I put them in series with my meter from my 350vdc supply, They are both back in circuit now and still no more machine gun. How does simply removing and returning two coupling caps cure this odd sysmptom?

I will mention that one of the .33uF caps is not like all the other flattened blue caps on the board, it is a yellow rectangular cap. I wonder if it was replaced at some point, and possibly a clue to what I am facing?

The current symptom is channels one and two pass signal to a point. It is clean up unttil about halfway, and then it goes into the low, loud howling oscillation as before.

I am just getting my butt kicked on this one.

Solder joint........

nosaj

[SoulFetish]

damn! After reading the preview in the email notification, I though I nailed it too. I was feeling all good about myself.
It’s the worst when you get hung up on a job, and you end up ultimately paying for the repair in bench time just to get it out.

but I suspect solder joints as well.

I’ve seen both of those types of caps as stock parts in old ampegs.

[Randall]

But where to go from here? What an albatross this has become. Why is it freaking out with this loud oscillation when turned up a bit?

[nosaj]

How bizarre this all is. I replaced C 19 and C20 with .1uF as that was the closest I could get to .33uF. Turned it on, and no machine gun! "AH", I thought, Soulfetish nailed it. I then put the originals back in one at a time to see if I could narrow it down, but guess what? Still no more machine gun. Neither of them seem to be leaking. I put them in series with my meter from my 350vdc supply, They are both back in circuit now and still no more machine gun. How does simply removing and returning two coupling caps cure this odd sysmptom?

I will mention that one of the .33uF caps is not like all the other flattened blue caps on the board, it is a yellow rectangular cap. I wonder if it was replaced at some point, and possibly a clue to what I am facing?

The current symptom is channels one and two pass signal to a point. It is clean up unttil about halfway, and then it goes into the low, loud howling oscillation as before.

I am just getting my butt kicked on this one.

Since you can make it happen on command that helps. if it were me I would scope the output of the PI turn it up till it howls see if its showing on scope. If so pull the output tubes and see if it's still howling. This should give you a direction to go preamp PI or Outputs.
nosaj

[Enzo]

And scope the B+ nodes to see if the oscillation is riding them.

[Randall]

OK, so after two days being off, I powered it up with a scope probe on PI pin 1, and machine gun sound is back, with no input. I did see it as a hard square type wave, like + dc to - dc. But I only saw one cycle, half the screen was pos, and half was negative, I didn't have time to adjust the time/div on my scope. I powered off, conected to pin 8 of the PI and this time no machine gun, but I do see a very high freq. oscillation that I can also hear as a buzzing noise, but only with the scope connected. And again, no machine gun after the initial turn on. No oscillation on pin 1.

I get the same results with the output tubes removed.

As for the howling noise, which remember is seperate from the machine gun noise, with an input signal I see no noise on the PI or B+ nodes with output tubes removed.

Here's where it gets tricky, because it has not been consistant. Witht the output tubes in, I can get the howling sound by turning up a volume with no input, and sometimes the machine gun sound if I turn up with an input. On the B+ nodes I don't see a noise, but I do see the supply getting pulled down. I do see a noisy sine wave on pin 1 of the PI when a channel with no input is turned up. Pin 8 is different in that it make a loud noise when probed anyway, but will also make the machine gun sound sometimes, and sometimes the howl when an input is applied, both I can see on the scope.

Seems like the PI is a problem? I cannot imagine what is the problem, I have never seen this before.

If I don't blow this shop speaker I will be amazed.

Here are the 7199 pin voltages:

1 - 320v
2 - 154v
3 - 66v
4 - 0v
5 - 0v
6 - 0v
7 - 0v
8 - 120v (machine guns when probed)
9 - 99v

[Randall]

I'm going to try resoldering and checking values in the PI stage. I have already done the socket. I just noticed since I have the speaker wired just to the yellow 8 ohm wire and chassis ground, the NFB is not connected. This shouldn't cause any of this noise, should it?

edit: Well, I reflowed the whole main board, went completely thru the PI, measuring and replacing anything that was out of spec, especially the 5% ones, which amounted to 5 resistors and a.1uF cap. the 3.9M 5% R25 was over spec, but I didin't have one so I didn't change it. I turned it on, and no machine gun. Good. But when I turned up channel one with no input, it howled. Pin 8 still DOES NOT want to be probed. Right now it goes into machine gun when touched.

I tried swapping the 7027As from side to side to see if there was any interaction differences, there was not. I pulled them and pin 8 still oscillates at a very high frequency, around 8MHz when probed.

I just dont know.

[Randall]

Help! Don't leave me, please. Taking any and all suggestions. I am truly stumped.

[nosaj]

Help! Don't leave me, please. Taking any and all suggestions. I am truly stumped.

We're up on generators now lot of outages in NWFL. do you have a shop amp that you could tack the FLiptops OT in and check it? Or another KNOWN good OT just to tack in for testing purposes?Otherwise I would reconnect the NFB the howling sounds just like positive feedback.

nosaj

[Randall]

I do not currently have an amp to test the OT with. But, I do have a 4ohm 35W Fender Pro Reverb OT I could clip in. And I will try the NFB as well and report.

Thank you.

[Randall]

Well, today is a very good day indeed. The amp is operating normally. I swapped out the OT for a Fender 4 ohm sub, and the amp did not make any of the problem howls or machine gun sounds. So, I thought the new OT was at fault. I hooked it back up, but this time I connected it exactly as the sub OT, with the black ground wire on the speaker jack. I had it connected as the original was before, with the black ground connected to a filter can ground lug. I read .3 ohms between the jack and the can lug, but somehow it made all the difference. With the ground on the jack, the amp works correctly, with it on the can lug, it howls. I don't understand this, but after all I have been thru with this amp, I will take it.

Other points to be made are the NFB wire made no difference either connected or not with either ground scheme. Also, I still see a very high frequency oscillation on pin 8 of the PI tube when probed, but now it doesn't make any noise. Could this be normal, ie is my scope making it oscillate?