Dual/Stereo mnats 1176 hum [SOLVED]

I built two stereo linked 1176 revD compressors into a 2u case some while ago and after finally starting to use it in my studio I'm experiencing a massive hum from both channels. Its there with the output potensiometer turned ccw, but it gradually increases when turning output cw. Same on both channels.

It's a low freq hum, typical transformer hum. My voltages are OK, and the compressors work flawlessly otherwise, stereolinked as well as dual. So I'm suspecting ground loops. I'm using the mnats "little psu", and when I built it I thought my grounding was as spesified according to the manual.

Has anyone built a similar stereo 1176 without hum? I'll post a description of my 2-1176's grounding soon!

Picture shows the unit while trying to troubleshoot. Both stereolink boards groundings detached (by the output transformer mounting screws), and also one detached ground wire from the left hand 1176 board. right board grounded to AC ground point. I've been trying attaching only one board to ground, different grounding points. Also I tried lifting both output pins1 to ground. Nothing helps so far..

EDIT: Solved the hum problem (was comming from psu) by lowering the 30V rail voltage, in my case down to 28V. The regulator (LM317) doesn't function properly when adjusting less than 3V. Check thread for alternative regulator!

I also tried detaching the VU-lights. not helping..

EDIT: I should also mention that all case panels are properly grounded to eachother. Top panel removed for troubleshooting,..

have you tried turning the toroidal around?
on the picture i see some connectors near the ouput transformers...
Try a better grounding scheme!
Thanks.

Thanks 3nity, tried turning the power trans around now, that didn't help. What would a good grounding scheme be for a dual unit sharing the same psu?

This picture shows the psu CT grounding point on the right hand 1176 board. Stereo link card grounded to the output transformer leg. 1176 ground lifted from former grounding point on outp transformer casing. I may be way off with this grounding scheme. Should psu ct rather go straight to AC inlet star ground? Hope someone with better knowledge than me can help out here..

have you tried searching the 1176 Rev D help thread?
Thanks.

3nity said:

have you tried searching the 1176 Rev D help thread?
Thanks.

Click to expand...

Yes I've been searching for a while, I found some postings regarding the psu not being sufficient for a dual build, so I tried disconnecting one channel. Still the hum is present in the connected channel.

I've battled hum quite a bit so hopefully i can help...

Hum is either coming from inside the unit: like the power supply isn't regulating, or the transformer is radiating.
Or its coming from contaminated grounds upon interfacing...aka ground loops.


With stereo units in one case sharing the same power supply its kinda of tricky..

But first make sure all your pin 1 XLRs are only connected to chassis right at the connector (solder pin 1 to that little tab) on both the inputs and outs. Don't connect ANY of the pin 1s to the PCBs at this point. Then connect the 1st Input connectors pin 1 (that is also firmly attached to chassi) to the Audio ground. A good spot would be the ground on the output of the power supply pcb.

Because pin 1 is the cable shield, that pics up all those radio stations and hum, you don't want that on the audio PCBs. You want to connect it directly to the chassis to make a continual shield. But you DO have to connect the audio ground and shield together at some point to keep them at the same potential. It seems simple, I know, but i took me a long time to get that simple concept through me head.

Then check for hum. If you still have hum, check your power supply, preferably with an oscilloscope, if oscilloscope you have none, and AC multimeter will work. AC voltage should be nearly zero on the DC rails.

If all this checks out and you still have hum, double check wiring connections, clean boards of flux, especially around the FET.


Ps: build looks really good, clean and neat. But get some shrink wrap on those Mains connections my friend!

abechap024 said:

I've battled hum quite a bit so hopefully i can help...

Hum is either coming from inside the unit: like the power supply isn't regulating, or the transformer is radiating.
Or its coming from contaminated grounds upon interfacing...aka ground loops.


With stereo units in one case sharing the same power supply its kinda of tricky..

But first make sure all your pin 1 XLRs are only connected to chassis right at the connector (solder pin 1 to that little tab) on both the inputs and outs. Don't connect ANY of the pin 1s to the PCBs at this point. Then connect the 1st Input connectors pin 1 (that is also firmly attached to chassi) to the Audio ground. A good spot would be the ground on the output of the power supply pcb.

Because pin 1 is the cable shield, that pics up all those radio stations and hum, you don't want that on the audio PCBs. You want to connect it directly to the chassis to make a continual shield. But you DO have to connect the audio ground and shield together at some point to keep them at the same potential. It seems simple, I know, but i took me a long time to get that simple concept through me head.

Then check for hum. If you still have hum, check your power supply, preferably with an oscilloscope, if oscilloscope you have none, and AC multimeter will work. AC voltage should be nearly zero on the DC rails.

If all this checks out and you still have hum, double check wiring connections, clean boards of flux, especially around the FET.


Ps: build looks really good, clean and neat. But get some shrink wrap on those Mains connections my friend!

Click to expand...

Thanks abechap024 ! Well I haven't electricuted myself yet but thats probably a good idea shielding those during trouvbleshooting yea!

The hum is equally loud on both channels so I guess I can rule out power transformer radiation, because all left channel components are located far away from the toroid. PSU not regulating; probably I should look into that, I'll go find my scope if I get no where regarding grounding.

Xlr pin1's are all connected to their tabs, and I've removed paint from chassi where they are screwed tightly so they should be grounded nicely.

Input audio hot/cold from XLR's are shielded all the way to 1176 pcb audio in's, and only connected to xlr input pin1. Like the manual says.

QUESTION: When you say audio ground, would that be the same as psu center tap?

This didn't work:
-CT from psu is grounded via two wires. One to each 1176 pcb.
-the two 1176 pcbs are both grounded to one of the two input xlr pin1 tabs.
-both stereo link pcbs are grounded to the same pin1 tab.

I also tried this which as far as I can see would be the right way according to your description:
-both 1176 grounds to psu ct, which is grounded at pin1.
-stereo cards are also grounded to pin1.

Maybe it's time to connect the scope, what do you think?

So I scoped the rails. -10VDC looks nice and clean as far as I can see. But the 30VDC rail makes me wonder..!! With scope set to AC, 0.2 Volts/DIV, Sweep time 5ms, see the picture!! (I had to draw the lines because of fast shutter digicam didn't catch it) Is this evidence of AC leaking into the DC rail??

check for backwards filter cap,

you should have 1 or 2 mv of AC, not 300,

CJ said:

check for backwards filter cap,

you should have 1 or 2 mv of AC, not 300,

Click to expand...

As you can see I'm really unexperienced when it comes to reading a scope. Do you mean that I have 300VAC running in the DC rail? That sounds crazy to me.. All electrolytics in the psu are connected correctly according to polarity. Could the electrolytics values be to low for a dual build?

I found out one more thing. When scoping pin 2 or 3 on the output xlr (unconnected) I see the same scope trace as pictured above. When output xlr is connected to mixer the trace dissapears as the hum enters the loudspeakers.

Pictured: Schematic for the PSU

Do you measure PSU voltage outputs while it is loaded ? (connected to 1176 pcbs)

I would try :

- CT from PSU connected to chassis, maybe at the same location where AC mains earth is connected
- GND OUT from PSU connected to both 1176 pcbs
- XLRs pins 1 connected to their XLRs ground tab and not connected elsewhere

That should be all for the ground connections, no need to connect wire to the transformer output frame as it is screwed to chassis

keefaz said:

Do you measure PSU voltage outputs while it is loaded ? (connected to 1176 pcbs)

I would try :

- CT from PSU connected to chassis, maybe at the same location where AC mains earth is connected
- GND OUT from PSU connected to both 1176 pcbs
- XLRs pins 1 connected to their XLRs ground tab and not connected elsewhere

That should be all for the ground connections, no need to connect wire to the transformer output frame as it is screwed to chassis

Click to expand...

Thanks keefaz! nice to see you all helping me out here!

Yes I've been scoping with 1176 boards and stereo link boards connected. Should I try scoping with the boards disconnected?

I see now that when I spoke about CT I actually ment "GND OUT", just because I used the CT-tab from the 1176 unused psu's to connect the ground wire from psu..

Should I really connect CT to anywhere else than the psu-board?

No, leave the CT as this in this case, but connect PSU GND OUT to chassis (maybe where AC earth is connected) and to the audio PCBs. At this point gnd connections are set, don't connect the gnd from pcbs to the XRLs pins 1, it is balanced inputs and output so the audio goes only from and to XLRs pins 2 and 3 (maybe twist the audio wires though)

keefaz said:

No, leave the CT as this in this case, but connect PSU GND OUT to chassis (maybe where AC earth is connected) and to the audio PCBs. At this point gnd connections are set, don't connect the gnd from pcbs to the XRLs pins 1, it is balanced inputs and output so the audio goes only from and to XLRs pins 2 and 3 (maybe twist the audio wires though)

Click to expand...

Thanks again! I tried this scheme now and its still humming.

Back to the scope trace found on the +30VDC rail. This really doesn't look right now does it. Is there any way I can troubleshoot the psu? I did a little research regarding usage of oscilloscopes, and as far as I can see the ripple found is approx 0,8 VAC. The amplitude is 4 divisions when Volts/Div is set to 0.2 V, so VAC must be 0.2 X 4 = 0.8VAC. Am I on the right track here? And is 0.8VAC and acceptable ripple or not? Hoping someone can help me out

And can the ripple be a result of contaminated AC from my wall socket? My other equipment does not suffer from annoying psu ripple, so I'm guessing not. But it occurred to me maybe the mnats psu is extra sensitive to contaminated AC

Spectral analyzer printscreen of 1176 output signal. This is with nothing connected to the input xlr's.

-Picture 1 : Output pots set to MIN (fully ccw)

-Picture 2 : Output pots set to MAX (fully cw)

Looks like you're in a 50Hz area,no?
The peak at 50Hz and the lower level at 100Hz (octave) normally indicates that the hum comes from pre rectifier.Maybe we should have a deeper look to the ground wiring including audio,psu amd safety earth?

Udo.

P.S.:Maybe you can add your location to your avatar/member/name (pardon,I don't know the correct word),that often makes it easier for the other members.

Did anyone say the words  " Star ground "  yet ?
where grounds meet at only one point [ usually pwr supply , but can be at inputs as well ]
Dan Kennedy or NYD ?  recommended using a piece of copper from house wire as a bus bar
, fits perfectly in xlr  , pcb  , psu all grounds meet there , good luck , it takes just one little connection
I was fixing a gtr tube preamp and only solved it once I took the  " Isolated " input jack off and back on again
that it must have been making contact without looking like it , use a meter to check continuity

if it's not too much work may want to completely disconnect one comp and make sure the other is working quiet before adding
the second in case your chasing multiple problems