Tascam M-3500 mooing like a cow, failing caps?

[AnalogApples]

Strange problem. And I don’t live near a farm.

When I power up the mixer it’s most pronounced even with every knob/fader down. There’s a humming-mooing sound that fades after five seconds.

I notice it also sometimes in any channel when there’s a percussive sound. There’s a “woooo” hum like reverb tail. It almost sounds like a bassy reverb.

Is this a sign that my main buss capacitors are failing?

Also, is it normal for every meter to show signal up to -10db upon startup, then fade after about ten seconds? The meter seem to correspond with the Moo but the moo is in my powered monitors (and headphones) even with the volume all the way down.

The moo problem isn’t noticeable 95% of the time but it’s pronounced at every power up.

 

[sweetbeats]

Can you post a link to a sample of the sound?

I’m not exactly sure what you mean by “main buss capacitors”…but if I had to take a stab at what your problem is I’d guess your power supply needs rebuilt. I’m guess the “mooing” sound is your audio power rails oscillating because they can’t keep up with the demand from the console. The meter issue is just showing the oscillation. Try this…can you force the noise to happen by opening up all the channels (i.e. unmuting the channels), raising all the channel faders to unity, assigning them all to the main buss and raising those faders up? If not that then add in assigning all the channels to all the subgroups and raising those faders up. Does it make it happen? Like a low frequency feedback but no mics are plugged in? If you can make that happen then I think I’m probably right, the supply can’t keep up, and the power rails are oscillating. This can happen for multiple reasons but the low hanging fruit is to start by recapping the supply. The PS-3500 has ample filtration, but I suppose if the main filter caps were out of spec enough it could cause an issue where the filter capacity falls below the the frequency of the mains rectifier circuit. And it will oscillate. I delay with this with a Soundtracs console I used to have…the power supply was under spec from the get-go and when the filters ages they couldn’t keep up. It was pretty easy to get a righteous LF resonance just by opening up the channels and pushing the levels up, and dumping each channel to all the output busses (subgroups and main buss). After a recap that included increasing the filter capacitance, and other mods that were called for it worked like a champ. The PS-3500 is better designed, so if it is the same problem likely a recap of the power supply would address the problem. But try my little test. And if you can, do post a sample of the sound.

 

[AnalogApples]

Can you post a link to a sample of the sound?

I’m not exactly sure what you mean by “main buss capacitors”…but if I had to take a stab at what your problem is I’d guess your power supply needs rebuilt. I’m guess the “mooing” sound is your audio power rails oscillating because they can’t keep up with the demand from the console. The meter issue is just showing the oscillation. Try this…can you force the noise to happen by opening up all the channels (i.e. unmuting the channels), raising all the channel faders to unity, assigning them all to the main buss and raising those faders up? If not that then add in assigning all the channels to all the subgroups and raising those faders up. Does it make it happen? Like a low frequency feedback but no mics are plugged in? If you can make that happen then I think I’m probably right, the supply can’t keep up, and the power rails are oscillating. This can happen for multiple reasons but the low hanging fruit is to start by recapping the supply. The PS-3500 has ample filtration, but I suppose if the main filter caps were out of spec enough it could cause an issue where the filter capacity falls below the the frequency of the mains rectifier circuit. And it will oscillate. I delay with this with a Soundtracs console I used to have…the power supply was under spec from the get-go and when the filters ages they couldn’t keep up. It was pretty easy to get a righteous LF resonance just by opening up the channels and pushing the levels up, and dumping each channel to all the output busses (subgroups and main buss). After a recap that included increasing the filter capacitance, and other mods that were called for it worked like a champ. The PS-3500 is better designed, so if it is the same problem likely a recap of the power supply would address the problem. But try my little test. And if you can, do post a sample of the sound.

I’m trying this right now. It’s not the same sound, more like a combination of high frequency buzz and some low frequency buzz.



Low frequency feedback is a good description for this sound that’s happening, it’s actually 120 hz noise



I replaced the very large 20,000 uF caps in the power supply as well as the three(?) that were right next to it. I’ll go ahead and replace every other cap in there since there aren’t too many.

Should I have maxed out all the mic preamp trim pots too? They were all up, but not maximum.

 

[rob aylestone]

You really need to determine what your replacements are doing. Ideally you need a scope. Then you can see the voltage on the rails and the ripple. Replacing capacitors is often needed when they are old, but to find out if they are good or bad, you need a tool. Do you know anyone who has one. Seeing a straight line with just a few little dimples where the rectifiers swap over really shows what happens - if you see big notches, then that helps indicate where the issues is.

 

[sweetbeats]

I’m trying this right now. It’s not the same sound, more like a combination of high frequency buzz and some low frequency buzz.



Low frequency feedback is a good description for this sound that’s happening, it’s actually 120 hz noise



I replaced the very large 20,000 uF caps in the power supply as well as the three(?) that were right next to it. I’ll go ahead and replace every other cap in there since there aren’t too many.

Should I have maxed out all the mic preamp trim pots too? They were all up, but not maximum.

Okay, so it would have been good to know you already replaced components. The key question now is did this problem exist before or after your partial recap?

And I agree with @rob aylestone you really need a scope at this point to see what’s happening at the output of the regulated supplies for the audio power rails before going further. My 2 pence on top of his.

 

[rob aylestone]

All joking aside, I hate repairing kit, always have, but year again, that’s what I did. If you have 70s kit, you really need 70s workshop equipment, and the skills to interpret what they tell you. You need a scope to see what you hear and you need a signal generator to squirt in audio for the scope to find, then you follow the sound through the circuit, power supplies are the easiest part of the gear to diagnose. In this case, if you have a hum, it points to the power supply, but if the hum is actually a honk, then something is oscillating and that suggests elsewhere. power supply hums are either rectifiers or smoothing, but honks much trickier and usually involve three components, inductors, capacitors and one other device.

 

[AnalogApples]

The problem was happening before and after I replaced the largest 5-6 caps in the power supply, no change.

I have a good digital scope but I’m still learning how to use it. Where do I connect the negative lead? Then I just test each separate positive wire going to the mixer, using the schematic, and look at the waveform?

That task might be a little too much too fast for a beginner oscilloscope user like me.

I’ll start by replacing the remaining electrolytic caps in the power supply, but I have to order them.

To be clear this problem is not usually noticeable when I’m using the board. It happens every time I power it up, and I’ve only heard it once in a while during use.

I plan to upload a YouTube video of it happening, so others with the same problem can get help too. My internet connection is terrible at home.

 

[sweetbeats]

I would not replace any more components until you get a diagnosis on the problem. If you start shotgunning components you may introduce a new issue and then you’re chasing your tail. The remaining caps aren’t likely to contribute to my hypothetical diagnosis, and it sounds like you already replaced the components that would and there was no change. So I hypothesize my hypothesis is wrong. But that’s the problem. We’re both stabbing in the dark, and the very best thing is to find out what is actually not right, and solve that issue. And when the problem isn’t obvious, 99.9% of the time the thing to do is to start at the head waters and make sure the power supply is doing what it is supposed to do.

What make and model scope do you have?

There shouldn’t be a “negative lead” with a scope…you have a probe, and that typically has a ground clip dangling off of it. Is that what you’re talking about? If so, the ground clip can clip to any part of the PS-3500 chassis. That clip is just there for noise shunting so you can get a better view on the scope. It’s not always needed.

You’re going to take that probe and, with the cover off of the PS-3500, touch it to the outputs of the + and - 15V power rails. They should be clean flat DC on the scope…a horizontal line. If there is a wave form to it then there are AC components in your DC power rail. That’s not good. And if those components make it all the way through the supply regulation circuit that means either the filter isn’t working, or the rectifier circuit isn’t working or something else. You want to look for low frequency wave forms and high frequency wave forms…if you see high frequency that means something is oscillating.

If you do go poking around in the PSU chassis with it powered, take great care…milliamps can kill, you can harm yourself or your equipment if you don’t know what you are doing, and not to sound uncaring but I am not in any way responsible if either of these things happen…proceed at your own risk. But I’ll try to help where I can. You’ll want to find a way to become familiar with your scope and how to use/understand one.

 

[AnalogApples]

It’s the craziest thing, I cannot get it to do the moo anymore, but here’s a vid of the meter behavior:

I’m turning everything on at once via a Furman power conditioner.

My scope is a Hitachi VC-5430 color lcd digital oscilloscope.

 

[rob aylestone]

most probes have interchangeable ends, so some will have little hooks that grip, or little sharp points. They usually have little test points too, so they might be labelled 5v and if you clip to these, you can adjust the gain so it reads five squares high, so you can then measure ac or dc waveforms. You still need a bit of common sense in terms of what you prod, 240v is a fair way away from 5v. Common misunderstandings are timebase adjustment. Whip the cover of a headphone jack off, and a naked jack in the line out, and adjust gain and timebase on some music to get a feel for what it looks like. It’s similar but different from a waveform display on a computer. The difficulty is learning to get it stabilised. wall wart mains adaptors are very interesting. They are supposed to be DC but will probably be worse looking at than your Tascam. As said, DC should be a straight flat line. The dinky psu units often look very odd. Series of little dimples. With a scope you’ll soon learn to see square, sine and sawtooth components and make the link with what you hear, so if you have a scrap but working psu you can get apart, that is worth using to learn. Firstly identify the mains side of the circuit and avoid it. Then measure and inspect the low voltage ac waveform. It will be a sine wave. Then look at the output of the rectifier. This will be a new visual to many. You should see a series of upwards going half sine waves, where the negative going ones have been flipped upwards. If there are gaps, one of the rectifiers have failed. Bridge rectifiers have four rectifiers, either as four separate diodes or four inside a package. then the capacitors try to smooth out the joins. With one rectifier faulty the dc will have a large hum component, too much for the capacitors to fix. When they’re all upwards going, the waveform has double the peaks so double the frequency, sort of, because it’s just the positive going component. Once you start poking about you start to build up a visual to audio sort of link.

 

[sweetbeats]

most probes have interchangeable ends, so some will have little hooks that grip, or little sharp points. They usually have little test points too, so they might be labelled 5v and if you clip to these, you can adjust the gain so it reads five squares high, so you can then measure ac or dc waveforms. You still need a bit of common sense in terms of what you prod, 240v is a fair way away from 5v. Common misunderstandings are timebase adjustment. Whip the cover of a headphone jack off, and a naked jack in the line out, and adjust gain and timebase on some music to get a feel for what it looks like. It’s similar but different from a waveform display on a computer. The difficulty is learning to get it stabilised. wall wart mains adaptors are very interesting. They are supposed to be DC but will probably be worse looking at than your Tascam. As said, DC should be a straight flat line. The dinky psu units often look very odd. Series of little dimples. With a scope you’ll soon learn to see square, sine and sawtooth components and make the link with what you hear, so if you have a scrap but working psu you can get apart, that is worth using to learn. Firstly identify the mains side of the circuit and avoid it. Then measure and inspect the low voltage ac waveform. It will be a sine wave. Then look at the output of the rectifier. This will be a new visual to many. You should see a series of upwards going half sine waves, where the negative going ones have been flipped upwards. If there are gaps, one of the rectifiers have failed. Bridge rectifiers have four rectifiers, either as four separate diodes or four inside a package. then the capacitors try to smooth out the joins. With one rectifier faulty the dc will have a large hum component, too much for the capacitors to fix. When they’re all upwards going, the waveform has double the peaks so double the frequency, sort of, because it’s just the positive going component. Once you start poking about you start to build up a visual to audio sort of link.

Excellent post ^^^^