Ampeg B15 oscillations at high volume

[imo]

Hello,
I am fixing a 70's Magnavox B15 for a friend of mine. I replaced the filter caps and some of the tubes. This model uses 3 6l7gt's, 2 6L6's, and a GZ34.
I am getting a slow beating, motorboating oscillation at high volumes with the bass all the way up.

I Switched 1st filter cap from 30U 600v, to 2 20U 600v caps, so now 40U at 600v. I replaced the other filter caps as well with a new cap can wired the same way.

I switched out the diode and cap on the bias supply. I think this, combined with the tubes was contributing to the hum.

I have one new 6L7gt, and tested the others with my emissions tester( all i have) I ended up using the new tube in the phase inverter/driver position, figuring that if one of the others was bad, then i'de have one working channel.
--So it sounds really nice, just wondering about ways to tame the oscillations
Ideas
Ian

------I am noticing that now my voltages are a bit higher.

[Mars Amp Repair]

Ian,
I had one of these recently where the filter can was grounded to the chassis & then the ground wire ran up the front of the turret board & then grounded at the input jack.

It created a ground loop. I ended up cutting the ground wire about half way up the turret board just were it grounded the PI tube section. The hum loop was solved & the amp was quiet as ever.

I don't know if the original wiring might not have grounded the filter can at the can or not as someone else had replaced it. But, dividing the ground path solved the ground hum loop.

Perhaps your motor boating is from a similar issue...glen

[stokes]

I would suggest finding where the oscillation is originating.Since the volume and bass controls affect the osc.it is likely before them in the circuit.I dont have a schem for that amp,but pull the tube that feeds the vol. and tone and see if the oscillation dissapears.Then take it from there.

[imo]

I will check both scenarios. Is it possible that the oscillation could be caused by over voltage? The plate of V1A and B are supposed to be around 135-145. They are closer to 235-240. The rest of the voltages are high, V3, and the 2 power tubes, but within reason. Should i try to lower it? I tried increasing the resistance that goes to the plates (doubling the power resistor from 22k to 44k, but this has a negligible effect on the voltage
I checked the ground. When i disconnected it after the supply caps the buzz increased. That is the last place I can easily cut it. After that it goes to a ground plane on the PCB.
I'm going to try to look for the oscillations and the buzz. Hard to find it with just a load resistor, and quite annoying listening to an oscillating 100 hz signal
Any more ideas are more than welcome
Ian

[Am Peg]

Ian -
What are the chances that you have a) an oscilloscope or b) a DMM with max/min voltage capture?

The reason I'm asking is because I'm chasing a similarly nasty problem in a '63 Ampeg. What I've found is a charge / discharge cycle at about 120hz, with a 10% swing from the 400VDC foundation. That's 40 volts! All it does is thump. All new caps throughout too.

If your B+ is oscillating like that, you'll find it throughout the circuit. You can't track the source of the oscillation when its in the supply. That means its everywhere. It's feeding garbage supply voltage to all of the tube circuits. The voltage oscillations hit every grid and plate, are amplified and pushed on to subsequent sections.

If Glen (from Mars Amps) did indeed resolve an oscillation thumping noise (not just hum) from a ground loop - then I think he's on to something. (Tell me that's so Glen!)

http://www.schematicheaven.com/ampeg..._portaflex.pdf
If this is your amp - pull Valve 3 (the phase inverter), then turn the amp on for a moment. See if the thumping is gone. It's not the tube (valve), its just that you've cut off the signal path to the 6L6's. There's no oscillating noise to amplify.

I'm open to more ideas too!

Les

[Jag]

I had a similar problem a few years ago after recapping a BF Bassman. I wasn't paying attention and had put the bias cap in the wrong way around. Of course, the cap was toast. Check that out, make sure it goes positive to ground.

[Am Peg]

Jag -

Someone was in this amp before me. They put in two bias caps backward. Electrolytics of course. I don't know how the caps survived. I triple checked mine. (used to repair medical electronics).

[imo]

Les,
I do have both (scope and DMM) as well as a nice sig gen. I don't see that type of movement, at least not with the DMM. I have the amp closed for the night, need to do some recording<G>
How are you seeing this 120hz charge discharge on your scope? DOes it happen all the time? Is this audible?
Here is specifically what i have. I have a low hum with volume up and nothing plugged in. It seems loud to me, would definitely bug me if recording, but don't have a similar amp to compare if this is par for the course. It is pretty noticeable.
I also get a motorboating oscillation when i turn the volume up to about 8 with the bass all the way up. Maybe i'm pushing these tubes too hard? I replaced one with a new old stock (in the driver, phase inverter tube), but the other 2 are both old Magnavox tubes. For what its worth neither seems to contribute more or less hum in either V1 or V2 (which are separate channels)
Again, maybe i'm simply pushing the circuit to much. Thats why i was asking about the high voltage.
I would love to look into grounding issues. It would be lovely to quiet the beast down
Ian

btw, bias caps are in the right way

[Enzo]

It is highly unlikely the B+ is oscillating. The amp is oscillating, and each pulse drags on the B+ hard so it sags with each pulse. Yes, that will spread it through the amp, but it wouldn't be the B+ at fault.

A missing ground somewhere could do this. SOme cap somewhere loses its ground and so can charge up to some unwanted voltage until the circuit dumps it. This is the classic relaxation oscillator effect.

[Am Peg]

Enzo - That's exactly the case. The power supply itself is fine. Something else is charging and discharging back into the supply forcing the oscillation.

The thumping starts as soon as the tubes come up to power. If I pull V4, it'll stop, because there's no amplified signal to drive the speaker.

This occurs with the volume at zero. Interestingly, if I change the volume, the frequency of the thumping changes. Adding fuel to the fire I suspect.

And, as mentioned before - I replaced all of the caps. It's hiding. It's in there somewhere...

Here's the schema for reference.
http://www.schematicheaven.com/ampeg...ocket_12rb.pdf

Thanks!
Les

[Am Peg]

IMO -

Yours doesn't start thumping until you get the volume up? (And the bass up too?)

With your scope, I'd be interested to know if you can see a change in your supply voltage when the oscillation starts. If you were to look at the B+ for instance (with care of course ~400VDC) with the volume down, you should see a relatively stable DC voltage. As you raise the volume, it would be interesting to note if the DC level of B+ starts to slowly jump up and down (+ / - 40 volts DC, Freq ~ 1 to 2 Hz).

With what Enzo said about the ground fault - yours would be induced at a higher point. This could be a brittle (cold) solder joint that expands when its heated (higher current passing through it). The expansion literally disconnects the solder from the leg of the wire or device, opening up the circuit at that point. And if that's a bypass cap to ground, well...

The tell-tale would be if the oscillations stops immediately at some volume setting, or after a few moments when its had a chance to cool down - and the solder joint contracts back to contact.

[imo]

I will open it back up to see. The oscillation comes when certain low end notes are hit on the bass, with the volume almost all the way up and the bass up. I am having a hard time triggering the oscillation with a sig gen. It seems to happen, or at least be clearest when playing. It requires a decent level signal to trigger it.
This is where troubleshooting becomes tricky for me. The oscillation becomes clearest triggered by a played note. If i'm looking for a possible bad ground or leaky coupling cap does anyone have any troubleshooting techniques that are good for pinning this down? I will check the B+ around DC and see what i can find
Ian

[Enzo]

When the volume control affects the freq of a motorboating, that means it is some sort of positive feedback. SO either the signal path is compromised, or there is insufficient power suply decoupling. Or a lack of low impedance ground path.

[imo]

I did some more work on it. Still the same. The one thing i did find was that moving the wire from pin 3 of the Output 6l6 around had some effect on the oscillations, lessening it a bit as i moved it around.
I have been checking the ground connections. Every one i have checked seems solid. The power supply has more filtering than on the schematic (40U first cap instead of 30) and all new caps.
I looked on a scope and found the oscillation all the way back to V1, even a bit on the 1st half plate there, but couldn't this be caused by something downstream?
I am baffled every time i deal with positive feedback loops. I just don't know where to look, or what to look for when chasing them down.
Ian

[Enzo]

Once the output yanks so hard on the B+ that it does this, then of course it will be all through the amp - the B+ will be spreading it innocently.

How about moving the 6L6 grid wires - pins 5 - instead of the plate wires. They are the wires that pick up what the tube amplifies.