hum problems - why did this solution work?

[goose42]

Hi, all,
I got called in to eliminate some hum last week, and I'm not really sure how I did it. Be forewarned, the following is like a word problem from a studio maintenance midterm (I wish I'd taken that class, then I might be able to answer the questions!)

The client (friend) has a brand new Korg SV-1 and brand new Acoustic B200 1x15 combo amp. Smart pick; he doubles on keys for one or two songs with his band, and the bass player and pedal steel player both use B200 heads with separate 1x15 cabs. This way, if either of the others go out on the road, it can be the backup.

The house has one properly grounded outlet, ground installed and verified by a licensed electrician. The tenant has connected one big-mama Furman power conditioner to this, and has run three high-grade power strips out into the live/practice room from it. They feed power to the SV-1, the B200 1x15, 2 guitar amps, a bass amp, and a simple PA head. Not the ideal situation, but the point here being that they're all on the same circuit, with protection. Doesn't seem to have any bearing, but the house is about a mile from a low-power AM radio station.

Here are the symptoms:
- Mono unbalanced out from the SV-1 plugged into the B200 yields overpowering 60 Hz hum, with nasty high-frequency fizz in it.
-- Bad on the Active jack (z=?, level = consumer line, -10 dBv), unusable on the Passive jack (hi-Z).
-- Varies with preamp gain and EQ; master volume has little effect.
-- As might be expected, high-frequency component disappears when horn disengaged. The overall volume level of hum drops, too.
-- Turn off SV-1, and the hum is still there. Unplug SV-1 from mains power, and the hum disappears.
-- No hum on the headphone jack or when sent to line inputs on interface.
-- Changing cables has no effect.

- When the amp is plugged into a simple 3-2 prong ground lifter (keyboard still grounded), the hum disappears.
- With the amp properly grounded, any other instrument, plugged into either input, does not bring out the hum (single coil guitars, humbucking guitars, basses).

- When the keyboard is plugged into other amps in the house, some hum is present, but not nearly as much. High-end fizz is much less.
-- This includes the other two Acoustic B200s (which are heads, not combos), plugged into a separate 1x15 cabinet. They're actually the quietest of all.
-- Connect the combo's speaker out to a separate cab, and the hum is still there.
-- Plug keyboard into PA head line input - same hum.

- Plug the SV-1 into a beat-up Peavey Reno amp (which has a horn), and the hum is even louder than on the B200; the fizz is the same.
-- Insert a direct box between the SV-1 and the Reno (in = 1/4" TSM, out = XLRF to 1/4" TRSM), engage the ground lift, and the hum disappears.
-- Plug the SV-1 and DI into the B200 combo, and the hum has a ton more HF fizz, the DI is really sensitive to handling noise (hold in hand and turn over, and it sounds like aliens eating - weird microphonic stuff, might just be a shitty DI). The overall signal level also drops considerably, and does not come back up with ground lift disengaged. Pad not engaged, so just plugging this DI in line with the keyboard significantly degrades the signal on this amp.

What solved the problem without switching amps, changing amp settings, or removing ground:
- Insert an Ebtech Line Level Shifter:
-- SV-1's mono out (which I understand to be unbalanced, on a TS cable, at -10) is plugged into the +4/TRS input on the LLS.
-- -10/TS output on the LLS is plugged into the Active input on the B200 combo
-- Now B200 is as quiet as with the ground lifted at the plug.

I'm not sure why this worked. The LLS is a passive box, which I understand means it just has step-up/step-down transformers inside.

- Was the loop broken because it uses a field effect connection instead of a wire - galvanic isolation?
- What happens to the ground in this case? There isn't a ground wire in the instrument signal, so are keyboard and amp still grounded as long as their internal grounding and/or ground connections at the mains are intact?

- Would installing a product like the HumX work as well? My guess is it just has a 1:1 transformer in the plug unit.
- Is the way it works different than the way the LLS works?

- The input on the LLS is expecting +4 dBu, and we're sending it -10 dBV. What does this actually mean in terms of signal integrity? There wasn't any noticeable difference in the sound, level or quality. I would have expected a signal drop since it's a step down.

- How did we have different ground potentials to create a loop? All of the amps and the keyboard are running on the same outlet.
- Why did using the ground lift on the DI work well on one amp, and not on the other?

- The SV-1's owner searched a few forums and uncovered similar problems posted by some users. Without cracking it open, no one can say for sure what the problem is, but is it possible that it's related to the power supply being inside the unit, next to the output block?
- If it is an issue with power supply, is that what caused the ground loop, ie interference from it changed the ground potential of the keyboard?
- The B200 combo is nearly identical to the B200 head, except that it has a control to engage and disengage the horn, which I guess means it has a crossover on its internal output. The external cab that the heads were attached to has a horn, but the crossover is in the cab; the output is full-range all the time. Killing the horn on the combo made it usable, though not ideal; if you were cracking it open, would you consider a poor quality or damaged crossover in the amp as a potential noise source?

[jhharvest]

Pure guesswork without knowing at all what the SV-1 is, I'd have said from the symptoms that it's incorrectly wired / broken. Measure with a multimeter the potential (and resistance) between the jack and the shell. Measure (carefully) the potential between the shell and the actual ground. If the jack is grounded to neutral, rather than ground, it's potentially lethal and would certainly explain your problems.

Your transformer in the signal chain would then break the incorrect grounding from reaching the amp. But as I said, pure guesswork. Would have to measure to know.

[Scodiddly]

Sometimes the molded IEC power cables that come with gear are actually miswired - I've found about 1% or so seem to be that way. I test all such cables before putting them into use now.

[Marc Alan Goodman]

Sometimes the molded IEC power cables that come with gear are actually miswired - I've found about 1% or so seem to be that way. I test all such cables before putting them into use now.

hahaha. Wow that is annoying. Good to know though, thanks.

[Nick Sevilla]

You broke a ground loop when you used that 3 to 2 ground lift plug.

Don't read too much into it...

[Gregg Juke]

The EBTech Line Level Shifter is a Magic Box.

There is only _one_ instance of nasty hum that I could not get it to rectify (iPod with A/C power plugged in, running direct into the PA; just can't seem to kill that one).

Otherwise, it has been an unbelievable purchase for me. We also have the Hum X.

I like to know why stuff works, but sometimes it's just more important that it _does_ (work).

GJ

[goose42]

Hey, all,
Thanks for your help! As soon as I can get ahold of a multimeter, I'll check out the potentials on the jacks. Good to know, too, about the potential for a miswired IEC; forgot to mention that we did replace that.

Sure enough, the player picked up a HumX and it works just as well. Works out better for him because he can just leave it on the end of the plug. He definitely didn't want to run the rig without a ground, so problem solved.

Transformers: more than meets the eye.

[themagicmanmdt]

there's a problem with that sv-1. be careful. it doesn't sound like it's grounded properly in the circuitry. you may get a shock if you touch the amp and any metal on the sv-1.... it's not the amp's fault, methinks.