PM1000 console P.S. hum. Don't think it's the caps. Ideas?

[Jonk]

This is concerning a full console.

I'm getting some pretty serious 60 cycle hum out of my main outs on my PM1K.

The power supply electrolytics are pretty new, so I'm really doubting it's those. I think I've narrowed it down to the power supply though, or possibly a grounding issue.

Hookups are:
Only two things powered up/connected. The board, and my amp/speakers

I've swapped out amps. Not resolved there.

The hum isn't there when nothing is connected to the amp. It is only there when the main outs are connected to the amp, and their faders are pushed up. It disappears as the faders are pulled down. Also, if I pull the main buss strips out of the console, it disappears.

I've swapped out a total of 5 different main output strips, with varying degrees of newness concerning caps, with the same results. So I think we can rule out an issue in the main output strips.

Also, I have another PM1000 carcass that I decided to insert this power supply in, and same issue.

So, correct me if I wrong, but I think it's either this power supply, despite it's relatively new caps, or there is a grounding issue somewhere.

It's strange though, that both PM1000 console chassis' would have the same grounding problem.

Where should I start in trying to diagnose this?

Schematic is on page 9 here: http://www2.yamaha.co.jp/manual/pdf/.../PM1000E_2.pdf
_________________

[tINY]

Have you taken the supply from the carcass and tried it?

What about the output of the supply, how does it look on a scope?




-tINY

[Jonk]

Swapping in the carcass supply was the first thing I tried but it's useless. It's completely dead. I'm just trying to get 1 working p.s. for crying out loud.

Anyway, I'm curious. I have a scope, which I've used in the past only for measuring phase when aligning the tape deck.

What would the procedure be for measuring the output of the power supply on the scope? Where do I put my leads, and what settings etc?

[John Suitcase]

What are the specs on the power supply? You might try replacing it with one of the Power-One units. I have one that's 1 amp for my PM1000 channels strips, I think they make up to 4 amp models.

Here's one: Digi-Key - 179-2343-ND (Power-One - HE48-4-AG)

It sounds like something with the power supply, but I'm not techie enough to guess.

[ktf]

With the scope probe the power rails(+/-15) and any other voltages the thing puts out. Try to do it loaded - the psu I mean. Twist the scope down to the mv range and select AC . These rails should a be quiet fuzzy bar with no ripple.

[vince @ speck]

A couple of thoughts...

You say 60hz. Do you really mean 60 or is just a low hum. Could it be 120Hz?

Next thought: unplug everything (EVERYTHING!) from your mixer, including the connection to the speakers. Fashion a custom cable so you can connect headphones to the main mix outputs. Listening on the headphones... do you hear a hum at the mix out? How about the bus outs? You need to determine if the mixer is the source of the noise - or not.

One bonus thought:
Do you have any high power device with a strong magnetic field near/under the console? You mentioned that the noise was at a minimum when the stereo master was down. This makes me think that the source of your problem could also be input channels, bussing, or summing circuitry.

[Jonk]

Hey guys,

Thanks a bunch.

Vince, I'll fashion a way to listen with headphones to the buss outs, and get back concerning that.

Actually, the sound is more of 120 HZ. I guess I was assuming that there was a 60 cycle problem and that 120 Hz was resonating as a symptom of that.

Correct me if I'm wrong guys, but the main power rails I believe are +44V right?. I'm not sure where the +/- 15V is when I look at the p.s. schematic.

I got out the scope, but I'm still unsure what I'm looking at. Am I to hook up only 1 lead, measuring between the ground and the tip of that lead?

After I couldn't figure out what to do with the scope, however, I got out the DMM and started taking voltage measurements, and I think we may have a culprit here.

For these readings the console was powered on with the full load attatched. I'm getting +60.1 Volts DC at pins 7 and 2 of the DC connector. This is where the PCB says +44V on the bottom.

At pin 5, the meter lamps, I'm getting +14.2 (which is about right) Though the schematic says 13V, the pcb says 14V.

At pin 4, the headphone amps, I'm getting +13.4V, and the voltage should be +12V

At pin 1, the phantom supply, I'm getting +48.1, so it's fine.

The most obvious problem is that there are +60V for the main power rail when there should be +44.

Could over-sized electrolytic caps have contributed to this problem?

So, is this another can of worms, or does it rule out anything?

[Jonk]

Also, just to add, concerning your last thought, Vince.

The hum doesn't change when I have all the input strips pulled from their multipin connectors, so I think that would rule out the input strips. It does go away when I pull the buss channel strips out of their multipin connectors. I've swapped in and out about 5 different buss strips, all with the same effect, so either they all have the same problem, or it's not them.

No magnetic field that I know of around or under the console. I wish I had a magnometer.

I have a pretty beefy Furman balanced power unit (IT1220) but I get the same problems whether I'm using that or not. Right now I have it completely unplugged, and the console is plugged right into the wall.

[John Suitcase]

I don't know much about the power supply, but I believe it should be 44v, though I used 48v for mine and it works fine. I've read that the units can work with up to about 50v. 60v would be too much, as many of the caps and transistors are rated at 50v. I'm thinking there's something wrong there (in the power regulator), maybe those caps, or a blown diode?

Sorry I can't be more help!

[Matt Syson]

Hi
Looking at the schematic:
60 Volts is too high and is probably the 'unregulated' supply (what you get before the regulator circuit).
This will have lots of 120Hz 'ripple' on it which will get everywhere although as there is not a lot of gain after the master faders, the inherent 'noise rejection' built into the last stages is probably sufficient to make it appear 'kind of OK'.
Solution, fix the supply!
For starters check Q3 and Q4 as either may be short circuit (use either ohmmeter or measure if the emitter voltage (out) is exactly the same as the collector voltage (in).
Other possibilities are Q5 gone 'open', the preset resistor for adjustment (R14) gone 'open' or possibly broken during the recap process.
The desk will probably survive 60 Volts for a while but you are being rather hard on it so fix it ASAP.
The circuit is almost a 'classic' and is obviously at least 'OK' for this desk so you should not need to consider major expense of a new supply just to replace a few bucks worth of transistors, even if you just fitted 3 new transistors and the handful of 'passive' parts.
Matt S

[Snatchman]

Originally Posted by ktf

With the scope probe the power rails(+/-15) and any other voltages the thing puts out. Try to do it loaded - the psu I mean. Twist the scope down to the mv range and select AC . These rails should a be quiet fuzzy bar with no ripple.

" Try to do it loaded...the power supply I mean"... ( LMAO)...Good one!.....Good luck...

[Jonk]

Hey Matt,

Thanks a bunch!

I'm not visually seeing any shorts around Q3 and Q4. Do I take them out to test them?

Also, if I have to replace any of these transistors, should I look to Motorolas? These seem to be the most popular if I recall, for the channel strip replacements.

I always wonder about transistors, as you hear stories of only about 1 in 10 specing out correctly, but I don't know.

Thanks

BTW, I did rig up some headphones to listen to the buss outs with only the console plugged in and connected, and it's definitely coming from the console. Hum city.

[Matt Syson]

Hi
You would not 'see' a short visually as it is likely to be the inside of the transistor 'melted' due to an overload.
These is no 'short circuit' or overload protection apart from a fuse which is WAY too slow to actually protect a transistor. A fuse simply stops the tunit catching fire if things have gone horribly wrong.
The transistor application is not critical so as long as it has capability to over 80 Volts and suitable current rating most replacements will be fine.
The issue with transistors and ICs is the possibility of 'fake' (i.e not actually the right thing inside the box) or factory 'seconds' being sold as full spec devices. Counterfietting of goods goes very deep, not just a 10 buck Rolex! Buy from a reputable dealer.
Since the parts are so cheap it is probably best to 'blitz' and replace all at once as you may not have cought the actual culprit if you are not up with fault finding.
The original reason for failure may be simply you shorted the output while plugging something into the supply or a cap may have stayed charged while you were working on it.
Good luck
Matt

[Jonk]

Hey Matt,

Okay, I started measuring the DC voltage at the transistor legs. One of the 2SC485's has different voltages on each leg. The middle pin was 51.7 V, while the left and right pins were 68V and 59.6V respectively.

The other 2SC485 transistor has equal voltages at all pins, 68/68/68

I don't have the board in front of me, so I can't remember now if it was Q4 or Q5 that had the varying voltages. But I do remember that Q3 had the same voltage as the power rails on all pins, which is now 68V's. It went up from 60V now that I've pulled all the channels.

Anyway, so I guess replacing that transistor is where to start.

One other thing was, I starting turning the adjustable resistor R14 out of curiosity to see what it would do. I left it back in the general spot it was before I started turning it. Am I gonna have to re-calibrate it?

Also, what is the possible damage that could occur if I had the channels powered on too long with the voltage over 60V?

Thanks so much for your help so far.

[Matt Syson]

Hi
The transistor with the same voltage on all 3 pins is most likely to be dead so replace it. Replacing the other while you are there would be a good idea.
The adjuster should be set so that you have about 44 Volts out of the unit but I would not expect much problem if set anywhere between about 40 and 50 Volts.
In this vein I would not expect to see much problem with the caps in the desk even if they have been run a bit high for a while. Once you have it running OK just check for unexpected 'clicks' when switching things (indicating cap failure).
Matt S

[Jonk]

Hey Matt,

Just wanted to say thanks. It looks like it's up and running again. I think it was one of the diodes, as I replaced the transistors first, and nothing. But then after replacing the diodes it's down to 44V again. Sweet.

Thanks a bunch. It's dead quiet now.

Jon

[John Suitcase]

Glad to hear it's working again!

And that was a lot cheaper than a new Power Supply!

[Jonk]

Yeah,
No kidding. I think the power supplies in these boards were pretty dang good anyway.

I still am a little curious as to what could have caused the diode to go bad though. I wonder what the failure rate for diodes like this is.

[Matt Syson]

Hi
I think you said that it had been recapped in which case it could have been overheated if you were not quick at getting old caps out, physically cracked if you were violent or zapped if any of the caps had some charge left in them when working on it.
Glad you are happy now!
Matt S