Ampeg R15R hum problem

[hasserl]

I've been working on an Ampeg R15R amp for awhile, trying to cure it of a hum problem, which I finally did track down. After all the usual suspects were addressed and corrections made (filter caps, input, grounds, etc.), I did get some relief, but the main problem remained. A semi loud hum that did not change pitch or volume when any of the controls were moved. Other than the hum the amp sounded good and worked normal. In fact, it's a really cool sounding amp, if a little dark. But the reverb and tremelo are really cool, as good or better as any I've heard. When you turn it up and play the sound of the guitar drowns out the hum, but when you quit playing the hum is right there, as obnoxious as could be.

Anyway, like I said, I did finally track down the cause of the hum, but I don't think there is anything wrong with the amp. From all I can tell it is operating as designed, and the hum is just part of the design. But that obviously doesn't sound reasonable, so I want to run this by you guys.

What I found was the hum is being caused by the way the circuit blends the wet reverb signal and the dry signal together. Checkout the schematic below of the stock amp. Go ahead and follow the signal path thru the circuit, do you see where the reverb signal is split off then where the two are joined back together?

http://img508.imageshack.us/img508/4616/ampegr15rschem6ao.jpg

The reverb signal comes off the plate of V2a, thru the .01uf coupling cap to V2b then the reverb tank, then is recovered by V3a and sent to the grid of V3b. But what about the dry signal? The dry signal is taken off the CATHODE of V2a thru a 22k resistor to V3b cathode. The two signals are blended in V3b and the signal then goes from the plate thru the .1uf coupling cap to the PI at V4. I ended up removing the 22k resistor between the two cathodes, and wiring up the reverb ala Fender, with a 470k resistor off the Reverb pot, and a 4.7M reverb blend resistor off the .01 coupling cap from the plate of V2a, and both wet and dry signals are returnd to the grid of V3b. I experimented with various sizes for the reverb blend resistor, settling on the 4.7M as giving the best blend. The amp picked up a huge amount of gain, which due to the dark tone of the amp did not sound too good. But with a few other tweaks to the circuit I rolled off some bottom end and brightened it up a bit and the amp sounds good now.

Heres the circuit as it now stands:
http://img157.imageshack.us/img157/1050/ampegrr15rhschem1mt.jpg

So what do you think. Is there something about the stock design I missed that could be causing the hum that I should have addressed rather than re-engineering it?

[hasserl]

Whoa, I don't know why that second schematic turned out so big, it should have been the same size as the first one. Sorry.

[Enzo]

I can't imagine why the cathode coupling would introduce hum any more than some other way. I suspect the hum was coming from somewhere else, like the reverb return circuit, and the mixer simply amplified it like it would any other signal. I'd ground V3-2 right at the tube socket and see if it killed the hum. Did the reverb control affect the hum level?

For that matter, did you remove the reverb pan from the area?

[hasserl]

As I recall, grounding V3 pin 2 did not silence the hum. I believe the reverb control did have a slight effect on the hum, emphasis on slight. I removed the reverb pan from the area, but of course at that point it was disconnected too. The hum remained the same, reverb pan connected or not.

[dai h.]

not sure if it has anything to do with the hum, but looking at other ampeg schematics that have very similar looking circuits, I noticed that some of the other amps have a separate filter for the input stage, while the one above shows the input stage sharing the filter with the reverb pan driver stage which appears to have more current flowing through (low value cathode R, high wattage cath.R and plate R plus it's purpose to drive the pan impedance). I guess the 22k connects the dry signal output AC current loop voltage from the cathode to the mixer stage cathode and the difference with the other way from the plate side cap is it's out of phase? I assume the grounding is correct.

[hasserl]

Good pickup on that Dai, but if there was a problem with the driver and recovery sharing the filter wouldn't the problem still exist with the circuit connected like I have it now? I haven't changed that, I've only taken the dry signal off the plate and mixed it back with the wet the same way Fender was it with their reverb amps; instead of coupling it via the cathodes.

The amps sounds really good right now, no out of phase issues. I might try adding just a bit of presence, but other than that it is really close. The hum is a non-issue at this point, and I suppose I am going to leave the reverb circuit connected like it is rather than going back to the original circuit and trying to hammer it out. I was getting nowhere real fast that way. I'm not anxious to go back to it.

I'm open to any other ideas though.

TIA
Hasse

[dai h.]

well I was staring at it for a while but I couldn't really figure out a reason why it would hum less and those were just some things I noticed.

a) reverb driver sharing filter with 1st stage
-node is not at end of filter chain (the quietest)
-reverb driver stage looks to have relatively more current
through it

b) non-bypassed cathode Rs for V2a, V3b? (bypassing is favorable in regards to less hum--AC grounds the cathode-- but don't know if it would be a big noticeable problem vs. just a relative not very noticeable increase in hum). (More hum from cathode - heater capacitance?)

c) the difference now looks like non-inverted signal taken off the cathode vs. inverted signal taken off plate

I guess the question I would want to ask is if the noticeable hum was normal with the same exact model amps, or not. If not, then maybe there was something wrong with the configuration before.

[hasserl]

You got me thinking about the phase of the signal. Obviously it is now phased opposite of what it was. But in relation to the wet signal, it should be in phase, right. The wet signal gets inverted once at V2B, then again at V3A, so as it comes back to the grid of V3B it should be in phase with the dry signal taked from the plate of V2A. Am I overlooking something? So then with the original circuit the wet signal should be out of phase, right? That doesn't seem logical, I must be mising something. I suppose I could try taking the signal from the cathode of V2A and direct coupling it with series resistance to the grid of V3B, which should put it back in phase to the original circuit.

[AL30]

You've got MONO input jacks

AL

[hasserl]

Yes, that is correct. Don't most amps? Not trying to be a smartass, I don't understand your point.

[dai h.]

right. Mono jacks, therefore the mechanism for the hum reduction, is....??

[AL30]

Sorry, I phrased that a little harsh. I didn't mean to come across as a know-it-all j@ck@$$. I meant it as a question. :o

I have an old Ampeg Gemini 1 with mono jacks and I was told the jacks were a source of hum in the amp. It was recommended to change them out for other jacks. I haven't done it yet so I don't have anything to compare but it's a quick/cheap try.

Again, sorry about the phrasing - it wasn't intended.

AL

[hasserl]

Okay, that's cool, I thought maybe you were pointing out something obvious that I should have picked up, and I couldn't figure it out. :D

The jacks have been replaced with Switchcraft shorting jacks wired up Fender style with a hi and lo input. Note that the amp does not hum now, with the rewired reverb circuit. I'm just unclear as to why it did previously with the original circuit.

[dai h.]

no offense taken here. :) I was just hoping you had an explanation or something that was overlooked. By "mono jacks" I guess you mean to say check if the jacks are the shorting type (hot shorts to ground when jack is not inserted)?

[AL30]

Hasserl and Dai - Yes, that's exactly what I was referring to. Thank you both for explaining it much better than I could.

Hmmm? Interesting about the reverb circuit. When I picked up my Gemini 1 someone had done some tinkering with the reverb circuit. I didn't spend a whole lot of time with the reverb circuitry but I know I don't like the reverb -which is odd of an Ampeg (they are supposed to be spectacular)- and the gain is waaaay too much for my tastes. Maybe I should revisit that? There is still a hum though.

AL

[Ted]

You got me thinking about the phase of the signal. Obviously it is now phased opposite of what it was. But in relation to the wet signal, it should be in phase, right. The wet signal gets inverted once at V2B, then again at V3A, so as it comes back to the grid of V3B it should be in phase with the dry signal taked from the plate of V2A. Am I overlooking something?

Yes. Its also inverted by V2A.

[Ted]

Although, thinking some more about it, what difference does it make whether the signal is inverted or not? Its being mixed with the wet signal from the reverb tank which is delayed by quite a large and varying amount. I can't see that it the phase of the dry signal makes any significant difference. Or am I missing something?

[hasserl]

I don't know Ted, but it seems to work just fine. My guess is you're right.