Ampeg Reverberocket vs hum

I have an early 90's Ampeg R12R Reverberocket combo guitar amp. A really sweet amp, but it has a hum/buzz issue that is making me a little crazy(er).

I got it off ebay and it sounded good, but after a while started to sound really bad, then would go back to sounding OK,.. I found a wire jumper that showed signs of heating and had burned the pcb slightly at one end where it was soldered badly. So I resoldered that and it sounded much better, but had a lot of hum. To the point that it really irritated you to listen to it when you weren't playing the guitar through it.

It does not vary with the clean Volume pot, but on the Gain channel the Master Volume does make it less, or as bad as the clean channel if you turn it up.

I plugged a shorted plug into the Receive input of the Send/Receive jacks and the amp was dead quiet.

From looking at the schematic and the behavior of the two Volume controls I isolated the noise to be coming from the second 12ax7 tube, V2. I used a scope and can see about 5mV of AC noise when the Standby switch is turned on. It goes away when in standby, along with the hum.

I have replaced all the electrolytic caps, and put all new tubes in the amp. I have tried swapping V2 out with the old tube and it made no difference.

I am trying to resolder all the resistors around V2 to see if perhaps I have another bad solder connection, but other than that I don't know what to try next?

24281566339_60989df03c_h.jpg
[/url]rr_2 by capndenny, on Flickr[/IMG]

24567098131_7da903fd64_h.jpg
[/url]rr_1 by capndenny, on Flickr[/IMG]
 
Send/Receive

The amp has a Send/Receive set of jacks that are between the preamp and the power amp sections. By plugging a shorted 1/4" jack into the Receive input, it disconnects the signal from the preamp in the amp and applies zero signal to the power amp. When I did this I had no hum to be heard in the speaker. So I knew the hum was coming from the preamp part of the amplifier. It is J4 on sheet 2. If you don't plug into the receive input, then it just keeps the signal from the preamp connected to the power amp.
 
Send and receive jacks refers to the "effects loop" jacks, or the preamp out and power amp in jacks. By shorting the return jack, we are grounding the signal path at the power amp input. So if that is silent, then the power amp is not the source of the hum.

Here is a pdf file of the schematic, I find that easier to use than the screen shot:
http://bmamps.com/Schematics/ampeg/Ampeg_R-12-R,_R-212-R,_R-50-H_Re-Issue_Schematics.pdf

because the standby switch removes the B+ from the amp, it is normal for any hum or other signal to go away when in standby.

If your filter caps had been bad, then even the power amp stages would have hummed too, so they were not going to be the issue.

5mv of noise WHERE?

So neither gain nor volume affect the hum level, but the master does, yes? Or volume does not, but gain does, plus master too?

Crate/Ampeg amps often rely on a number of ground connections, note schematically there are two grounds up under the input jacks as well as a couple different ground circuits in the amp. SO make sure the grounds through board mount screws are firm, and any other grounding. This affects earlier stages more than later.

None of the grids in V1, V2 are at DC, so you can safely ground them. BY grounding a clip wire then selectively grounding each grid down the signal path, you can shunt any signal to ground. That will help isolate the hum source to a channel.

You can also "ground" the signal path when DC is present if you use a DC blocking cap. I use a 0.047uf/630v cap, because I have a large number of them handy, the value is not critical. But put in series with a grounded clip wire, I can shunt the signal path to ground even when DC is present. Basically in plate circuits.
 
Gain had no effect. Volume also had no effect. Only the Master Volume reduces it, which is how I traced it to the V2 tube.

I ordered new relays, but even that doesn't really explain what I am seeing.

it doesn't really matter if I have a guitar plugged in or not.

If my last filter cap was open, it would affect V1, and then I would have even more hum, and I would think the clean Volume control would have some effect, but it does not.
 
I don't see how relays would cause hum, but I guess we will see.

As to filter caps, I have to disagree. The first couple are there for ripple smoothing, but by the time it gets to the screen node, the hum is gone from the B+. The additional caps farther down the line are there mainly for decoupling the stages. If you didn't have ripple and the second to last filter, you won't have any at the last cap either, even if you remove it. If that last cap were leaky, you'd wind up with crosstalk or instability, not ripple. the volume control would have an effect, but likely on oscillation or other artifacts of coupling. So close, but no cigar, as they say.

I believe you on V2, but look for a moment at R13 over there on the grid side. Note the signal at V2-1 is 5 or 6 volts. That feeds through voltage divider R13/14, which at 3.3M and 470k, is about a 8/1 reduction in signal level. That is what hits the grid, maybe 0.7v of signal, which the tube amplifies about 40 times to get the 33-38v on the print. In other words your 5mv at R17 ( the top end) will be a lot smaller at the grid side.

Did you try grounding the grids of V2 one at a time?

Looking at the drawing, 70-90mv at V2-2 results in 1v or so at the master volume pot. So your 5mv at the control (which end of r17 by the way?) might have only been 0.5mv at C7. And a half millivolt signal can get lost in the clutter on a scope trace sometimes. Doesn't take a lot. grid grounding is a tried and true way of dividing and conquering.
 
Cool and thanks

I will try the shorting thing. I noticed that just touching the B+ with my dmm made a loud pop. I always thought a dmm on volts was pretty much an olen circuit.

I appreciate the help.

I was doing a Pspice sim of a 12ax7 circuit and it said to get 5mV on the output of the tube the B+ on that tube would have to be about 10mV. I can't use the scope on it because it is above the max voltage for the probes. I have so e higher voltage probes on the way.
 
Ok, I finished touching up the solder. Put it all back together and powered it up. Flipped the standby on, and it wasn't too bad. After about 30 seconds though it started to buzz. So back to where it was.

I shorted the signal at R10 to GND and still buzzed.

So I shorted out R14 the input to the second half of V2 and silence! So it's almost like R14 is open or something? But it's not.

The fact that it was ok for a while kind of makes me suspect the blocking cap connected to R13. I know caps will behave that way, so perhaps that is the problem. But where is the noise coming from? I didn't buy the coupling caps, so I can't replace it, unless I can borrow one from work tomorrow?

I tried shorting the other end of R13 the 3.3M, but the buzz was still there. I think that's what I saw, but that makes no sense?

So some progress, which is cool!
 
Last edited:
Your volt meter has an internal impedance, pprobably more like a 1 meg resistor than an open.

When you refer to a resistor, please indicate which end. For example R13. You can say the left and right end (looking at the drawing), or you can say the C9 end or the grid end (or the R14 end). Example: the left or C9 end of R13 will have about 8 times the signal level of the right or grid end of R13. So it matters.

So grounding pin 2 did not kill the hum, but grounding pin 7 did. That points to the noise coming in at the left triode of V2.

If you think C8 is weak, do you have any caps remotely similar? 450v 10uf 40uf 50uf 100uf? If so, clip it parallel to C8. if the cap leads are not handy, clip to the top end of R11 or R15. I could be wrong about that cap, never hurts to find out.

Wait, you ground the right end of R13, and no hum, but the left end and the hum is still there? If that is true, did the hum not diminish at all? if that is the story it sounds more like and open R13 than R14. Power off and discharge, measure R13. Do it both directions. if the resistance is different, then ther is some residual voltage in the circuit.
 
Yep

I am pretty sure the cap end of R13 did not get rid of the hum, but the grid side it did.

I rang out R13, R14, and made sure they connected to each other and that R14 goes to GND. It's as if the filament circuit was somehow bleeding into the grid circuit. But it's not the tube.

I can disconnect the filament supply and make sure there isn't a resistive path between pin 9 and pin 7 somehow. There is a lot of dried flux on the pcb because the sockets are soldered on top of the pcb, the component side.

When I worked as a tech about 30 years ago I remember one board I could not figure out what was up. I finally did an a-b comparison using an ohmmeter between the bad board and a good one. I found a part in the circuit where the resistance was about 100k instead of 200k or something, but the parts were all good. I finally traced it to an ink stamp of a part number that was stamped by hand across two pads on the pcb. I scratched through the ink with my probe and the board started working.

I don't have another amp to compare, but there are 2 other 12ax7 tubes on the board.

It was weird that it actually seemed to be OK for a while, and then started to hum again.
 
Don't just assume the heater is the course of hum just because it is 60Hz. ALL hum ultimately comes from the mains. Did you determine yet if your hum is 60Hz or 120Hz? Same note, octave apart. A scope can tell you.

120Hz is rectified power supply ripple. Weak filter caps give you that, but so does grounding where power supply ground currents share a path with signal grounds. 60Hz is from other sources.
 
yes, and so...

I will figure the frequency out tonight. From listening to it I think it's 120Hz, plus it is really buzzy, not a smooth 60Hz. But I will check tonight. I meant to do that last night but it was late when I got it all back together and got to testing it. Tired + High Voltage = Not a good plan

I need to have some kind of plan to look for something?

I have a 6V kids car battery I could use to test for it being filament induced.

I guess this stage has a pretty high input impedance, so it would be easier for noise to drive the 470K R14 since R13 is 3.3M it won't put up much of a fight. As compared to the previous stage where the gate pull-down is 475K, but it goes directly to the cap. Which tends to explain why grounding the C9 R13 junction doesn't do much. That end is stabilized by the previous stage and C9 which has an impedance of 12K at 60Hz.

I'm just not too sure where to go next.
 
More info ?

I finally got back to work on the Ampeg.

It is 60Hz noise, but there are spikes in the noise waveform, as if it was rectifier noise. But the first spike goes positive, then 8.33ms later there is a negative going spike. That's really weird. The 60Hz noise would be heater noise, and there shouldn't be any rectifiers in the circuit. Perhaps it arcing noise in the filament circuit. Like I have a cold solder joint somewhere, and each time the voltage crosses zero volts the current drops to zero until the voltage rises enough to arc over. Then it arcs and makes good contact until the next half cycle?

I can see it even with the amp in Stand By mode, but of course I don't hear anything. I found a spot where the filament signal runs right next to and then under the resistor R13 and R14 junction, which is right where the noise is getting into the signal path. Here is a picture of it.

24594822572_7c71fdde9d_b.jpg
[/url]ampeg by Dennis Kelley, on Flickr[/IMG]

This isn't the source of the noise, but is probably where it is getting in.

I noticed when I turn the amp on the filament in the V3 tube glows really bright at the bottom of the tube for a few seconds. Then it lights up along with the rest of them. It may not be the problem, but I will try swapping it out, or looking for a place where the heater circuit might be arcing. Including the original issue I found, which I thought I fixed.
 
scope shot

Top trace is the junction of R13 and R14, where the noise first appears. The bottom trace is the heater circuit waveform. The has in the upper waveform may or may not be real. It comes and goes with moving the scope probes. But the 60 Hz signal and the spikes are really there, because I can hear them.

[url=https://flic.kr/p/DtxpyA] scope pic by Dennis Kelley, on Flickr[/URL]

I found some articles from back in the 90's, when this amp was made, that talked about one of the issues with using ss rectifiers being the turn off transients from the slow to turn off rectifiers at the time. Their solution was to add a snubber across the rectifier output, or to use schottky diodes, or to put a .01uF cap across each diode. The modern fix is to use a modern fast turn off rectifier diode, such as a UL4007. I have 4 UL4007 diodes and 4 100V, X7R 0.01uF caps on the way from Digi-Key, so we shall see.

The other possibility is to try to just filter it off the heater circuit waveform?

Or just move the two signals away from each other so the heater waveform doesn't couple into the signal at R13/R14.
 
Heaters in not the only way 60Hz can get into a circuit. ground loops can as well, and so on.

Just my opinion, but when you get into it to the point you are deciding the amp needs to be re-engineered, I think you are off the mark. If it were a board layout problem, then ALL of this model would make the hum, not just yours.

If you have a solid 6v battery or real clean DC supply, then try disconnecting the 6vAC from the circuit and powering the heaters from the DC. That will tell you if the hum is heater related.
 
yep, I agree.

I don't know that they don't hum. I had a R50H, which is the same amp in a head version. I recapped it and I remember it had a bit of hum to it as well. But I decided to sell it because it was almost 3 feet wide and it was just too big. Whether it was better or worse I can't say in terms of hum. I remember thinking it needed something done for the hum issue, but I assumed it might just be a tube.