Stand Alone Reverb - ground loop busted

[tubeswell]

I have recently had a customer bring in Kendrick Reverb Unit (6G15 Clone) that has a persistent hum, like you have described.

For determining whether its a ground loop, I'd test it on a range of amps, as well as making up a short non-grounding extension lead that I could plug the reverb unit into (when it is plugged into the other amp). It should be safe enough to operate in this way (assuming the ground return path goes through the cable shield of the reverb-unit output/amp input signal cable).

There is the Fender '63 standalone RI ground return path isolation set-up you could try in the reverb unit (back-to-back 6A diodes in parallel with a 10R (or was that 100R?) FP resistor and a high-voltage .1uF cap - all between the reverb unit's ground return path and the chassis). I haven't tried this yet, because the amp I mainly use my reverb unit with seems to have a ground potential that almost perfectly matches the reverb unit's, and there is very little hum (altho' there is some).

Another possibility I wondered about a while back, was whether you could use a hum balance pot canceller (like a heater hum balance pot canceller) with its own earth connection between a split in the signal cable shield of the cable connecting the two amps, but nobody around here thinks that would work, so I didn't bother trying it. (I was imagining that the pot could be used to equalise the chassis ground potentials)

If its not ground loop hum, it could be other hum (bad internal grounding, bad tube, etc)

[loudthud]

I talked with GW about this subject and he advocates using lifting the power line ground on the Reverb unit. Since he's one of the CT to transformer bolt guys, the reverb is suspect. I think Tubeswell makes a good suggestion to try the unit with several other amps. In a previous thread a reverb unit was found to create hum with one amp but quiet on another. The back to back diode trick is another thing I like to see tried.

[Sir Cuitous]

I am reviving this thread since there may still be some interest and I did say I wasn't done with my experiment. Here are the results.

Since my last post I re-wired the grounds per an R.G. Keen spec that Tubeswell sent to me. I originally followed the Hoffman diagram and the result was fairly quiet when the power line ground was lifted. Using the R.G. Keen diagram the result is very quiet.

With that as baseline I re-wired back to stock 6G15 and got the following:

1)Stock 6G15 --> Power Line Ground Lifted --> Very quiet/Best tone
2)Stock 6G15 --> Power Grounded/Ground Buster Box --> Fairly quiet/Very thin weak tone
3)With Paralllel CF at Output --> Power Grounded/Ground Buster Box --> Very Quiet/Good Tone

Then I duplicated the resistor/cap circuit I found in the Ground Buster Box and tried the isolation transformer I origianlly used.

4)With Paralllel CF at Output --> Plus isolation transformer --> return of weird hum/tone

Number 4 was predictable since that was the result of my original build. Lastly I pulled the transformer out of the Ground Buster Box and tried it. Sounds good just like result number 3.

My conclusions are: the cathode follower is necessary if want any chance to get an isolation transformer to work and.... the choice of transformer is very important. Not all audio isolation transformers will work even if they are 10K:10K impedance. However, I'm still too much of a novice to tell you why one transformer works and the other doesn't.

[tubeswell]

Thanks for that handy feedback Sir Cuitous. So you are saying the R.G. system is the quietest then in your experience? (I may well now re-wire my 6G15 ground to the R.G. system to see how much of an improvement it is :-)

[Sir Cuitous]

It does seem quieter, but it's a somewhat of a pain to keep the jacks isolated from the chassis. I could not find isolated RCA type jacks for the tank connections, so I used wire terminated connectors and ran the wires through the chassis. You can probably just cut off a typical audio patch cable and use that. Anyway, I'm very happy with the outcome and I may try the R.G. Keen grounding on my next amp - thanks for suggesting it. I'm certainly glad to get the transformer isolation to actually work.

[tubeswell]

it's a somewhat of a pain to keep the jacks isolated from the chassis. I could not find isolated RCA type jacks for the tank connections, so I used wire terminated connectors and ran the wires through the chassis

You probably don't need to use isolated RCA jacks, because each jack sleeve is connected to an independent cable shield and there is no possibility of a loop through the pan (if you have an Accutronics pan). So I'd say running the RCA jack sleeves straight to the chassis is okay, provided that you don't hook the tranny secondary ground connection up to the sleeve for the 'to Pan' RCA jack.

[Enzo]

Chances are on many reverb pans you encounter, the jack on one end is insulated. Drill out the rivets and take them off dead pans.


There are insulated jacks for sale though. I'll find the Switchcraft number in a minute - same one they use on the pans. Here is a typical dual jack from MCM:

http://www.mcmelectronics.com/product/27-955

From www.tubesandmore.com, single insulated jacks. Stock S-H268R example



And you can always make your own. Make the panel hole larger to clear. Mount a piece of plastic or phenolic to cover the hole, then drill a hole in that for the RCA jack.

[tubeswell]

Hi Sir Cuitous

If you can't find those insulated RCA jacks (of the exact type Enzo showed in his last post) and you still want some, PM me with your snail-mail. (But I really don't think they are necessary, even with the R.G. system).

[Sir Cuitous]

Yeah, they're probably not necessary, but I followed the Keen schema to the letter (single connection to chassis). If anything I'm good at following instructions. Must be all those years of schoolin'.

I get a lot of stuff from AES, but I seem to have overlooked that item. Thanks Enzo. It did cross my mind to make my own on a piece of phenolic, but just cutting off an old adapter cable worked to test it. The back panel is cut for panel mount jacks so to make it right I believe I will get the insulated jacks.

[SatchmoeddieII]

What active isolation device would work? I am wondering if a transconductance type like Anderton used in a project might work. I have about a dozen stand alone reverb schematics kicking around, and some are solid state too. Matching the impedance to the circuit should work better than just tossing in an Iso trans. I have a grounded 8 amp 120 I use around here, and a 1amp 2 wire in 2 wire out antique I use when calibrating Hickock 539s, or KS 15750s. Would using an EMI/RFI IDEC socket with filters to the main power in help? I got one surplus made by Schurter, and it shut up a notoriously noisy old solid state Ge transistor amp I like to use for playing a Hofner 63 through. No more hum, nor Mexican radio stations straying into the system. I have nothing against Mexicans or their radio stations, and the old KJJJ910 was just as anoying, and they played country music. That amp picks up everything SETI should buy it from me.

[Sir Cuitous]

I ripped the transformer out of a commonly available, passive isolation box. The box was about $35 and it has two channels - so two transformers. I believe just matching impedance does not break the ground loop caused by cabling the reverb to the amp. This issue is not bad grounding or bad design, it is a physical loop wired from earth to earth through the patch cable. Though part of my post was about improving the grounding in the unit, it was never very bad to begin with and was not the main issue in this post. The issue was how to effectively use an isolation transformer to solve a known problem. However, to use the iso transformer, I did need to lower the output impedance of the reverb unit. The isolation is working very nicely with a number of amps. I couldn't be happier with the outcome.