M-Audio BX8 2002, Strange Hum From Studio Monitor Tweeter!

[SEGACDX]

I've got a pair for M-Audio BX8's the original ones from 2002. https://img.audiofanzine.com/images/u/product/normal/m-audio-bx8-3304.jpg



So I've been trying to solve this problem forever and put in on hold many times over a few years, I've done everything I can think of and nothing seems to solve the problem.



I thought it may be bad caps or and it may be but I've tired different caps with no luck, It seems the problem I've narrowed down is for now with the toroidal transformer making the tweeter side of the amp pick up hum/buzz from the mounting plate.



I own two of these from the same year 2002 in the same line of monitors, so I've been using the good one to compare, I thought it might be a grounding problem and it might be, so I added grounding to places missing on the unit with the tweeter hum/buzzing problem and while I did manage to make it match in terms of where the ground should lead to, testing the points on the board where it should've had ground and didn't before are fixed. It could be some kind of ground loop or something but I am not sure.



Here's the important bit, if I move the transformer away from the casing it's normally attached too the humming/buzzing goes from audible buzzing in the listening position, back down to normal levels of noise, gentle static if you put your ear up to it just like the good unit.



If anyone knows what may be causing that hum to come back only when it comes closer to the plate it's normally attached to it would be a big help.



I'll put some photo's to help clarify but if all else fails I may just have to leave it disconnected form the case when installing it back into the speaker box since it works fine other than the very annoying loud humming coming out of the tweeter when connected to the case.



Here are the images and a similar schematic, I cannot find the exact model, I can take more or better images if needed but they where to large to upload to the site so here is the link.



https://imgur.com/a/2BjIsFY

[shakalnokturn]

I assume the back plate is aluminium?
Have you tried scoping the noise at the tweeter? Its frequency and shape can give clues on what the cause is.

I'm a little surprised that there is no thick rubber gasket between the transformer and back plate.
Is the transformer winding actually pressed against the plate? Is the plastic film in good condition?

The most efficient way to avoid ground loops is to use star grounding meaning that any large grounded conductor is grounded in only one point.
Although a large conductor should only have one point tied to the ground reference point (usually bulk filter capacitors / transformer center tap) it may serve as a conductor to ground another component making the ground hard to follow.
It's usually also a good idea to have any mechanical and electrically conductive part grounded rather than left floating.

With this in mind, other points I'd watch on this design:
Heatsink/PCB/backplate ground linking.
Potentiometer shaft is it in good contact with backplate if it should be?

Shielding on smaller PCB how is that tied to ground on PCB?
One point only? Being magnetic and close to transformer if grounding is made in several points induced current could cause small voltage variations in the copper ground plane it's attached to. (Same idea as star grounding: You don't want a ground that's used for signal reference to be also carrying currents from elsewhere because that causes noise.)

If you have a megohmmeter around it could be worth measuring the black glue's resistance, if found to be lightly conductive check if any is spread on PCB's.

[SEGACDX]

The back plate is Steel. I wish I had a scope to test it with, but right now I don't, I do plan on getting a little one for the bench.

There is a rubber gasket it's just not shown, it's about 1.5-2mm thick. The film seems to be in decent shape.

The pot for volume is making good contact with the back plate and is grounded.

The shield was not grounded before I started poking around in it is is now, as for one point I believe so, it should be connected to the same point as the mains earth ground.

I'll check around with that black glue and see if it's conductive or not.

I've made a lot of progress on this and more or less solved the problem for now by adding in some steel sheading around the toroid and figuring out that the rotation of the toroid add or subtracts the buzzing noise, so it need's to be rotated slightly until the buzzing cuts out. To clarify not flipping it around but turning it back and forth while mounted will change the buzzing noise to be more or less. So was it's not audible unless you put your ear right up to it and turn the volume up to maximum.

Which is the same as the good unit, I've got it tuned right now so it's only audible by putting your ear to it at 0% volume and buzzing is not present at higher volumes past the lowest setting. Just the normal static noise is coming though which is good.

I'd like to understand why rotating it makes the buzz/hum get louder or be completely gone in the right position.   

[SilverSolder]

It is likely either capacitively or magnetically coupled noise.  By rotating the transformer just so, you are perhaps canceling out two opposite lobes of the field, or something of that ilk.

All part of the magic of analog electronics! 

[SEGACDX]

Yeah that sounds about right, I thought it may be doing something of the sort, now if only I could somehow see magnetic fields I could better understand what's happening.

Still I am glad I for the most part fixed the problem. Another small bit of knowledge to add to my repair toolkit. Still I wish I'd had most of my newer equipment to help solve the problem a few years earlier. One thing is for sure I need to invest in a oscilloscope to help aid me in my repair jobs.

All I can really afford is a cheap 120Mhz pocket scope like the MUSTOOL MDS120M but that's better than nothing.   

[bob91343]

What is the character of the hum?  Is it a smooth, clean very low pitched sound, or is it more buzzy or whiny?  If the former, it may be caused by magnetic pickup.  If the latter, it may be caused by ground loops or poor filtering.