[egelmett]

My intention for the pair of MU524s was as a subtle buss compressor for mixes.

I noticed that there's a light, but audible, hiss/noise coming from these devices.

I understand this may be inherent to analog, especially tube/valve, devices, but is there anything to look out for that may cause this noise?

Thank You,

Ed

[JPK]

Can you measure the noise voltage at the output with the input pot at min and output pot at max?
Should be around 1mVAC.
If higher, try moving the unit to a different slot, further away for the PSU?

[egelmett]

JPK,

Thanks for the suggestions. As far as I can read with my meter, it was around 1mVAC.

I tried different slots, and a different 500 chassis with no changes. I was also sure to check, and then check again, the DC converters were different considering stereo usage.

However, in the process of testing one of the units today, while taking off the panel, I noticed a decease in the hiss I am concerned about. Through some more testing, I had to isolate two of the four screws that secures the cover plate from making contact with the plate. I isolated the points on the plate from the posts and screws. Those were the screws near U2 and the one near TP5. Both devices exhibited this noise, one way worse than the other.

It was odd though. Only the screw near TP5 did this at first, and once I isolated that screw, the screw near U2 started doing the same thing, so I isolated it as well. Luckily, there were no further surprises.

I did look under the main PCB to see if perhaps some unwanted contact was being made when tightening the screw. None found. I even tested this by taping off the whole underside of the cover plate thinking perhaps some intermittent contact was taking place along the plate. I couldn't find an issue there either.

In the end, I applied isolating electrical tape on the cover holes (both sides) near TP5 and U2 in order to isolate the screw from the plate, on both devices.

Any ideas why this might be?

Connecting this back into my system, I found upping the output of the MP556 and lowering the output of the MU524, reduced audibly, any remaining noise -  aka better gain staging.

[JPK]

Any ideas why this might be?
Does the noise also decrease if you remove the cover plate completely?

This kind of behavious sometimes come from a HF oscillation somewhere. Check the solder joints on all film and ceramic capacitors.

[egelmett]

Yes, the hiss would clear after removing the cover, and only reappear when securing the screws to the posts I mentioned.

[egelmett]

JP,

It's been a while since my last post on this. Even when used for Line In/Line Out (i.e. no preamp input), there is audible hiss and I'm not using them.

Along the lines of moving slots, I tried one in an Empircal Labs EL500...and the noise was even worse in either slot. I cannot use them with the EL500 in any way. The MPU566 I built are not noisy in the EL500.

I will go over the capacitors again, as recommended...but this is happening on two MU524s...surely, I can't be that inept at incorrectly soldering two units?

Any ideas?

[egelmett]

After taking your advice (reflow all film and ceramic capacitors), the odd noise related to the cover is gone. I did this to both units.

I did remove the transformer to examine C12 for any perforation. There were none. As a safety measure, I sanded down the closest transformer pins for clearance, and added electrical tape around C12...just in case.

Still, the EL500 case is too noisy to use the MU524 in it. The transformer picks up hum from the power supply, even when the MU524 is farthest away from the power supply.

And, analog being analog...there is still some inherent noise. On loud sources, not a big deal. For me, this is my opinion, too noisy for delicate sources like voice and acoustic guitar.

I see you have a solid state opto compressor kit! 

[JPK]

Thank you for the update.
Did you try with a different lunchbox?
We are currently using API, Lindell, IGS lunchboxes without a noise problem.
We also work at strong line levels which helps.

[egelmett]

I only have the Midas L6 and two Empirical Labs EL500s - one is worse than the other...it may be them? Hmmmm...

But, strong line levels may be the issue: recording soft vocals and finger-picked acoustic guitar leaves a lot of space for noise to be apparent.

They're a lot better than just after the build. With your advice, like I wrote previously, it helped to go over solder joints and make sure C12 wasn't impacted.

Now, I'm dealing with P1 having failed on one of the MU524 - asking you about the part number or if I can buy a couple from you, in another post.