1970s Ampeg B-15N, screeching oscillation after recap

[cobascis]

I just picked up a very clean (entirely untouched inside) Ampeg B-15 from the early to mid 70s. I did minor testing of the amp with all the old electrolytic, but had a recording session in the next few days, so I replaced all the electrolytic and the can cap with parts from FlipTops. I also installed a 3 prong grounded cable (abraded chassis and used a large hi wattage iron to solder ground to chassis right as it enters amp), and disconnected the AC outlet and the polarity switch. I did brief testing after the recap, and the amp sounded good. It wasn't until the session that I learned all was not OK.

Here's what happens:

• When I turn the volume past noon (with an input plugged in) a horrible oscillation/screech comes out of the speaker. The screech is fixed when I turn volume down below noon. At lower levels the amp seems to function normally.
• The entire amp is extremely microphonic, I can tap anywhere on the chassis and it is very loud. I tried all new power and preamp tubes, no change in microphonic aspect
• I tried another CE Can cap to see if it was a faulty cap. No improvement.
• The screen does not occur if I pull out the preamp tubes
• I've checked and reflowed solder on ground connections

What else should I check? Hoping it is something simple that I am overlooking, need to get this thing back in action for another recording session coming up. Happy to post voltages or anything else that might help track this down. Thanks!

Here is a video of the microphonic nature of the amp and what happens when I turn the volume up when an input is connected.

[passaloutre]

Is there a specific preamp tube that kills the oscillation? Got any spares

[cobascis]

Is there a specific preamp tube that kills the oscillation? Got any spares

I swapped out all 3 of the GSL7 preamp tubes with known good tubes and no change to the squeal. I also tried switching the order of the preamp tubes. No dice. Put new power tubes in as well just on the off chance that was a factor, no change.

[passaloutre]

Put them one at a time. Which one kills the oscillation?

[cobascis]

Put them one at a time. Which one kills the oscillation?

Removing V1 kills the oscillation in channel one, removing V2 kills oscillation in channel 2. If V1 and V2 are removed, no oscillation in either channel. If V3 is removed and V1/V2 are left in, no oscillation.

[sluckey]

Did the amp have this issue BEFORE you replaced the caps? If not, I suggest reexamining the cap replacement with close attention to the cap that feeds the preamp tubes. Look for loose connection or poor solder on positive and negative cap terminals.

     https://el34world.com/charts/Schematics/files/Ampeg/Ampeg_B-15N.pdf

[cobascis]

Did the amp have this issue BEFORE you replaced the caps? If not, I suggest reexamining the cap replacement with close attention to the cap that feeds the preamp tubes. Look for loose connection or poor solder on positive and negative cap terminals.

     https://el34world.com/charts/Schematics/files/Ampeg/Ampeg_B-15N.pdf

I believe it did not, unfortunately I didn't spend a ton of time testing the amp before I recapped it.  But I'm fairly certain I turned it all the way up with no issues.

I've triple checked the solder on the can cap, and replaced the can cap already so I'm sure it's not the connection at the can cap itself. But perhaps where it meets the board and powers the preamp section could have a bad solder joint. Looking at the schematic, it is the middle 40uf (middle from right to left) that powers the preamp. I will see where that connections and reflow that solder. Anything else I should try?

[sluckey]

Looking at the schematic, it is the middle 40uf (middle from right to left) that powers the preamp.

No. The 40µF on the left feeds the preamp tubes (V1 and V2).

The issue may not be directly related to the cap can replacement, but rather some collateral damage that occurred during the process

[cobascis]

Looking at the schematic, it is the middle 40uf (middle from right to left) that powers the preamp.

No. The 40µF on the left feeds the preamp tubes (V1 and V2).

The issue may not be directly related to the cap can replacement, but rather some collateral damage that occurred during the process

Got it. Just reflowed all solder joints where the cap is providing power. Also reflowed both preamp tube socket pin 5. Checked to make sure both those resistors were at around 470K, they were. Reflowed cap can ground. No improvement so far. Hmm...

[pdf64]

R37 supplies the preamp HT node. The symptoms align strongly with that node not being adequately decoupled, eg the 40uF cap connected to R37 either being bad or not properly connected.

It's interesting to note that the heater circuit looks to have a negative DC elevation (something would probably be needed to mitigate heater hum leaking into the signal path at the unbypassed input stage cathodes).

The standby is hot switching, its use will pointlessly stress the GZ34.

[cobascis]

R37 supplies the preamp HT node. The symptoms align strongly with that node not being adequately decoupled, eg the 40uF cap connected to R37 either being bad or not properly connected.

It's interesting to note that the heater circuit looks to have a negative DC elevation (something would probably be needed to mitigate heater hum leaking into the signal path at the unbypassed input stage cathodes).

The standby is hot switching, its use will pointlessly stress the GZ34.

Thanks--good to know re: standby, I typically don't use standby on my amps but I will make note not to on this one in particularly.

After work I will again check, R37, R7 and R11, make sure they are measuring correctly and inspect or reflow both solder joints. Since I've tried 2 brand new can caps and done a basic capacitance check on the can cap, and reflowed and inspected the hot and ground, I don't think it could be the connection where the can cap itself connects to the amp. I reflowed where all the can cap wires terminate at the board, but perhaps a solder joint down the line from there was stressed?

Let me know if theres anything else I can physically check on when I am in there tonight. Thanks for the help everyone!

[cobascis]

I just tested resistance, removed old solder and resoldered R37, R38, R7,R11, R19,R15, and resoldered pins 2 and 5 on V1 and V2. I also touched up a few other solder joints across the board that had micro cracks. All the resistors are reasonably within spec, none open or intermittent that I can tell.

Could C3/C5, C10/C8 have anything to do with it? I haven't pulled the board up from the chassis yet, not sure if I need to. All the solder joints are visible from above, but plenty of wires routed below. I realized the only tube I haven't swapped out is the 5AR4. C17 I also replaced during the recap (replacing the large firecracker cap). I have no way to test that other than MM capacitance, which it is on spec for. With "Ultra Low" switched on, neither channel will squeal. But with it off, it will. I suspect this is just due to gain (Ultra Low is quieter usually than standard).

Would grabbing voltages help anyone further diagnose? Thanks again for the help.

[AlNewman]

Did you add any shielded cable, or is it possible you rerouted the OT lines or other signal lines?

The fuse in the top left seems a bit weird, doesn't look like factory really, and at least one of the shielded cables on the bottom right maybe looks like it's bridged?

[cobascis]

Did you add any shielded cable, or is it possible you rerouted the OT lines or other signal lines?

The fuse in the top left seems a bit weird, doesn't look like factory really, and at least one of the shielded cables on the bottom right maybe looks like it's bridged?

I forgot to note--I had to replace the inline fuse. This amp has the standard exterior replaceable fuse, then an inline pigtail fuse. So you're right that it is not factory, but I replaced it in the same way it was originally wired.

I will check on the potential bridging of those shielded cables. I did not add them, or reroute any signal cables to my knowledge. I was just dealing with the can cap, power cable, fuse, and then the other electrolytics on the board. 

[tubeswell]

You say the issue is affected by the amount of volume you dial in. So, if you keep the volume maxed out but reduce the treble, does the issue go away? Have you got a scope handy?

[cobascis]

You say the issue is affected by the amount of volume you dial in. So, if you keep the volume maxed out but reduce the treble, does the issue go away? Have you got a scope handy?

Just tried, turning the treble all the way down does not stop of the issue if the volume is up. I do have a scope handy.

 

Did you add any shielded cable, or is it possible you rerouted the OT lines or other signal lines?

The fuse in the top left seems a bit weird, doesn't look like factory really, and at least one of the shielded cables on the bottom right maybe looks like it's bridged?

I confirmed the shielded cable is not bridged, it looked like it was due to some flux. Cleaned that up and confirmed no continuity.

I checked some voltages, which are a bit high as expected due to my 122V AC wall voltage. That said, things are nearing the voltage rating on the can cap. I'm getting 448V DC on two of 40uf nodes to the right on the schematic, but the node that feeds the preamp tubes that connects to R37, I am getting 520V DC. The cap is only rated for 525V DC. The other PSU cap (C17) I'm getting 530V DC on the hot side (this is a 600V cap, though). This is translating to the following preamp tube voltages:

V1:
Pin 2: 215V DC (should be 175V DC)
Pin 5: 218V DC (should be 165V DC)

V2:
Pin 2: 213V DC (should be 175V DC)
Pin 5: 218V DC (should be 165V DC)

[AlNewman]

What are the voltages at pins 3 and 5 of the 6L6's?

[cobascis]

What are the voltages at pins 3 and 5 of the 6L6's?

Both 6L6s are at around:
Pin 3: 529 V DC (should be 450V)
Pin 5: -52V DC (can't find what it should be on schematic)

529V DC instead of 450 seems like a huge difference... should I try another 5AR4?

[tubeswell]

I do have a scope handy.

Can you post some scope screen shots for the output and the plates of the various preamp tubes?

[cobascis]

I do have a scope handy.

Can you post some scope screen shots for the output and the plates of the various preamp tubes?

Sure, do you want those measurements to be while its shreiking? Or just at idle at a lower volume?

[BrianS]

Did you change any of the grounding points from their original location(s)?  I had an Ampeg do weird stuff after I recapped it and changed one ground location.  Moved it back and the problem disappeared.
I also had to move a seemingly microphonic tube socket off the circuit board of a 1970s Ampeg.  If I remember right, the problem was quite similar to what you are experiencing.  I tried all kinds of things, but the problem seemed to be coming from the socket.  They were unusual circuit board mounted sockets.  I ended up mounting a standard socket to the chassis and running wires up to the circuit board.  This solved the problem.  Never seen anything like it.  Ampegs can be weird.

[cobascis]

Did you change any of the grounding points from their original location(s)?  I had an Ampeg do weird stuff after I recapped it and changed one ground location.  Moved it back and the problem disappeared.
I also had to move a seemingly microphonic tube socket off the circuit board of a 1970s Ampeg.  If I remember right, the problem was quite similar to what you are experiencing.  I tried all kinds of things, but the problem seemed to be coming from the socket.  They were unusual circuit board mounted sockets.  I ended up mounting a standard socket to the chassis and running wires up to the circuit board.  This solved the problem.  Never seen anything like it.  Ampegs can be weird.

The only changes re: grounding were the addition of a grounded power cable, and the removal of the death cap which was connected to ground. The can cap remains isolated from chassis via a washer and grounded via the stock ground scheme. I will double check all grounds.

Here's how the amp looked before I did the work for reference: