Trying to avoid ground loop hum

[Chuck H]

I'm doing the grounds for speaker driven reverb amp. The speaker of a "host" amp drives the tank and then the tank output is re-amplified. To avoid ground loops I am separating the grounds between the host amp and the re-amp up to the re-amp input, but I'm not sure where to ground the re-amp input 1M bias resistor. I think it should be grounded with the re-amp circuit because it's this 0V that is relative to the +V. But it does mean the tank output has one ground and the re-amp input 0V reference is on a different ground at the input??? Not sure on this one.

TIA

Chuck

[Wilder Amplification]

I'm doing the grounds for speaker driven reverb amp. The speaker of a "host" amp drives the tank and then the tank output is re-amplified. To avoid ground loops I am separating the grounds between the host amp and the re-amp up to the re-amp input, but I'm not sure where to ground the re-amp input 1M bias resistor. I think it should be grounded with the re-amp circuit because it's this 0V that is relative to the +V. But it does mean the tank output has one ground and the re-amp input 0V reference is on a different ground at the input??? Not sure on this one.

TIA

Chuck

I say just lift the shield on the return cable that feeds the slave amp (lift it on the can end).

[Chuck H]

Good call. Tried it too, but...

Finished the build and tested it as it was (with the output shield redundantly grounded) and it hummed. Disconnected the redundant shield ground and it still hummed. Which I expected because it's not that redundant as the RCA's on the tank aren't isolated from 0V. So I alligator clipped the isolated ground buss for the host amp to the re-amp ground. No more hum.?. (well, a tiny bit but I've never met a reverb circuit that didn't hum a tiny bit. Maybe it's the tank circuit ground scheme.)

So ground loop wasn't an issue in this case. But why would the circuit hum more with the isolated grounds??? Since I built the host amp also, I know that the outputs negative is connected to 0V on that amp. Hmmm?

Thanks Jon

Chuck

[dai h.]

is there a schematic to look at? (I figure you have re-grounded it correctly somehow since it works the way you want it to.)

[Chuck H]

Thanks for being willing to take a look Dai. I don't have a schem for the reverb amp. I just sort of designed and built it out of my head. But I will do a brief explaination. This is a seperate combo amp to generate reverb only. The reverb tank is powered by plugging another amps speaker output into this "reverb amp" and then via parallel jacks, into the host amps speaker or cabinet.

So I had it set up where the input circuit for the reverb tank (which is adjustable for amps from 5 to 100 watts) used isolated jacks and all grounds for the adjustment, panel meter and tank overload protection were isolated to the host amps 0V. I then treated the actual reverb, from it's input grid foreward, as it's own circuit and grounded it with it's chassis as I always do. I thought this dual ground topology would be insurance against ground loops, but it hummed. So I wired the isolated buss (which is wired to the speaker jack negative terminal, which is grounded to the host amps chassis) to the reverb amps chassis as well. Guess what? No hum AND no ground loops. It works, I just don't know why. Or why it didn't work the first way.

Chuck

[imaradiostar]

Chuck, what tank are you using?

I've had problems with tanks having the input and output coils grounded or ungrounded- depending on the circuit it makes for big trouble.

I was also thinking a cheap-o DI box transformer might be useful- could provide isolation and even signal drop or gain depending on how you connect it.

I suppose if it's quiet now then there is no reason to mess with it!

How does it sound? I thought this was a great idea when you mentioned it previously. Frankly I'd love to see something like this with an integrate delay too but that makes things a lot more complex- you'd need another amp to drive the tank...etc, etc.

jamie

[dai h.]

sorry Chuck, kind of lost me--a bit complicated for my 90-pound weakling of a brain to follow without a diagram. Also, I'm not quite sure what the internal construction of a tank is (which is something I should learn) which might make a difference (in knowing where the grounds should go). However, having said that my guess is that you grounded what was supposed to be grounded, and it was floating before when it shouldn't have been (but I could be wrong! lol).

Anyway, sounds interesting. Didn't Holland(?) amps have something like that?

[Steve Conner]

Surely this is why reverb tanks are available with a choice of grounded or isolated input and output circuits. Choose the appropriate one for your application and it'll break the ground loop.

The input coil should be grounded to the driver circuit. The output coil should be grounded to the same spot as the cathode of the recovery amp. The casing can be grounded to either, but not both.

[Chuck H]

Surely this is why reverb tanks are available with a choice of grounded or isolated input and output circuits. Choose the appropriate one for your application and it'll break the ground loop.

The input coil should be grounded to the driver circuit. The output coil should be grounded to the same spot as the cathode of the recovery amp. The casing can be grounded to either, but not both.

Right, I made the mistake of ordering a "9" tank that is otherwise the same as the standard Fender type tank. That means no isolated jacks. As I outlined above, I did try to iso the whole tank circuit to the driver amp without success. Since it's working now I'm not worried about it. But If the user doesn't like the "9" tank (see below) I'll switch it for a "4" with iso jacks and retry it with your suggestion.

Jamie, It sounds VERY big and wet. Maybe too wet. Could be the "9" tank (6 springs). Some guys think these sound TOO big and now I can hear why. The effect is very impressive and clear though compared to an on board reverb. Now my customer can plug in his boutiquish amps (that achieve their distortion with power tube overdrive) and the reverb won't become a distorted mess. I built this for him because he loves power tube distortion, but he also loves reverb. The two don't get along well in the same amp as most of us know.

Dai, I'm away from my home computer for work right now. I'll draw up a schem and post it here when I get back to it... Or maybe sooner If I find time on my hands. I can do it with this poot but it's a slower process.

Thanks guys

Chuck

[Chuck H]

OK
I did some additional experiments and found that if I grounded the input grid of the reverb amp directly to the host amp buss I got my old ground loop hum back. Duh, I know... But bear with, it's part of the process. So I unclipped that and pulled the jack from the host amp out of the reverb amp and the small amount of hum I have now went away. OK, so it's obvious now that I actually grounded the ground loop hum when I grounded the iso buss in the reverb amp. So I just took my Dremel and drilled the rivets on the output jack of the reverb tank to isolate it.

As Steve Conner described, I should be able to lose the last bit of hum by re-isolating the host amp and keeping the tank output coil 0V on the reverb amp chassis. I'll report back in a bit when I refab the tank output and get it hooked up.

You can tell Steve does electronics for a living and see's this sort of thing ad nauseum. Things I have to hunt for seem obvious to him. And he gets it by reading these half baked posts. Steve, you should work in some kind of tech support. I think you could do well at it with far lower risk of electrocution

Chuck

[dai h.]

yeah what does he think he is? Some kind of engineer?

[Chuck H]

That did it. Woo Hoo

No trace of hum. Thank's to all for helping me deliver a better product to my customer.

I'll post a schem when I have it. It could be a fun project for the guys who want power tube OD and reverb. Or if you just want to add reverb to any non reverb amp without modification.

Chuck

[Chuck H]

schematic

If anyone is interested...

A couple of concessions. Use a true full range speaker. Right now this amp has a Emi legend 1058 and it sounds a bit boxy. But a single 1058 in a combo would sound boxy for a guitar amp. Not sure what I was thinking there. This is an effect, not the dry tone. I think a true full range speaker would liven it up even more. Likewise with the OT. It's a small budget OT and it lacks bottom end. If I build another one of these I'll upgrade the OT.

The split plate load feeding the PI and both V1 cathodes bypassed is a design contigency. I built it this way so I could control the gain up or down as I wasn't sure what I was in for. You could probably get a cleaner tone by removing the V1b cathode bypass cap and increase signal at the split plate load. I may yet make all these changes before I'm done. A very worthwhile and satisfying project so far.

Chuck

[imaradiostar]

Neat, thanks for sharing! Looks like a lot of gain is available- where do you usually have the volume?

I built what amounts to a stereo blackface champ (6l6 or 6v6 switchable) with the thought that a main amp could drive a pair of verb tanks and this little stereo amp could drive a pair of 10's on either side of the main cabinet for verb. From your experience would it be underpowered in this application?

Depending on how much volume is needed and how compact you want to make it I bet you could make it pretty tiny- LND150 mosfet into a 12ax7 LTPI into EL84's or even jfet into 12dw7 gain stage/split load, cutting the tube count back. Or even jfet into a pair of 6bm8's...dangit, I need to stop!

How much does the reverb amp's tone contribute to the sound of the verb- as in how crucial is the design of the verb amp?

Sorry for all the questions, I think this is an interesting and underutilized idea!

jamie

[Chuck H]

Neat, thanks for sharing! Looks like a lot of gain is available- where do you usually have the volume?

I set up the split plate load so the power tubes are barely flattening the waveform at full tilt when the sensitivity is adjusted correctly. This isn't supposed to be a distorting amp after all.

Using my 20 watter I like the volume of the reverb amp around two and a half to three. You can get full saturated reverb (ala BF amp with the reverb dimed) with the volume set to five in conjunction with this 20 watt amp.

Keep in mind that I designed this to be useful with amps from five to one hundred watts.

I built what amounts to a stereo blackface champ (6l6 or 6v6 switchable) with the thought that a main amp could drive a pair of verb tanks and this little stereo amp could drive a pair of 10's on either side of the main cabinet for verb. From your experience would it be underpowered in this application?

Depends... What size amp are you pairing it with?

How much does the reverb amp's tone contribute to the sound of the verb- as in how crucial is the design of the verb amp?

In my case I tried to keep it really straight foreward. As in, not too complicated and not too cavalier either. Since this is basically a "clean" amp (with generous leway for actual distortion figure) I kept it simple but added a NFB loop and balanced the gain stages. And as I said earlier, I've realized that this is an "effect" amp and would benefit from a higher fidelity speaker (and OT). Basically, it should be a "clean" amp capable of reproducing the effect. Not limited to the same considerations as the main tone amp, and even having different (but not stringent) requirements, IMHO.

As for "gain available", as I said, I wasn't sure what to expect. This amp needs to amplify a signal of only three to seven mV. So you can see where extrapolating gain I wanted to build leeway into the circuit.

As a side note... I did plug my guitar into this thing using an RCA to phone adaptor. The results were predictable. It sounded as if I'd plugged my guitar into an old phonograph amp. A little muddy and boxy but interesting in a vintage sort of way.

Chuck