Behringer 3282A - Help with input board troubleshooting

[visoundguy]

I had a failure of my power supply phantom power regulator and guess what - the failure mode with the transistor is for the voltage to go from 48V to 80+volts! This quickly fried my blocking transistors and caused many other problems. I replaced all of the 47uf caps - replaced a couple of blown resistor - and believe it or not, have all but one channel working! I know I have 2 problems on this channel - the channel strip, and the input board. I am getting a visual inducation of sound (the peak indicator) off of the input board on channel 12 - but no sound to the mains. Now if I move this input board to the channel strips 13 to 24 - I get sound - but don't have any control of the gain pot. Same if I move it back down to channel strips one to 12 - it doesn't seem to make a difference where I place the gain pot - it shows the same indications on the peak indicator. So I know I have a problem in the channel strip board on Channel 12 - and a problem with the input board. The biggest problem is I don't have a schematic- so I'm very perplexed.

I have the following indications on the bad channel of the input board - Voltage measured across the zener resistors and resistors off of the 16V supply is 10.5 - on all other channels, it's 7.19. Voltage from ground to the lower side of the zeners and on the input to the transistors, and to the control of the op amp - is 5.4 volts! All other good channel strips show 9.1 volts. And the biggest mystery is that I have 4.6 volts at the gain pot - on both sides of the pot! On all good channels, I have less than .1 volts.

Now - I've replaced the dual op amp - (NJM 4580L) - replaced all 4 pnp and npn transistors, pulled up and checked the value of all resistors - and replaced all but the ceramic capacitors.

Still no luck! Very tough without a schematic!

Anyone have any ideas?

[Enzo]

OK, by channel strip I guess you mean the whole 12 channel mixer channel board. I/O I can figure out on my own.

The trim pot is just a 5k pot wired as a variable resistor, and that is the entirety of the trim circuit on the main channel board. from the pot, two traces back to the I/O board. The pot is probably OK, check it with an ohm meter.

With the whole thing assembled, plug a signal into the insert jack for any channels involved. The ring is the return, tip is send. From there the signal path is striaght down the ribbon to the channel board. Divide and conquer. You can use the tip send to monitor I/O if you like. From here you can tell what the channel card is doing.

Looking at channel 12 on I/O, that should be XLR X25. (Behr labels connectors with X numbers) You replaced four transistors, two NPN A06s and two PNP 1084s. T1,2,25,26. Look for C25, a 1000uf 6 volt e-cap. If any cap was vulnerable, that one was. 4.7 ohm R73 in series with the cap is likely OK. I worry about your e-caps, not the ceramic ones. A bad C25 could leave the trim pot useless.

I don't know from zeners and zener resistors. There are two diode in this circuit, plain old switching diodes. They are BAV21, which is just like a 1N4148 except 200v (or 250v depending upon vendor) And their anodes to -16v and cathodes to the input pins of the op amp, none to 16v, so I don;t know to what you refer clearly.

Since the gain pot is just a resistor, it is normal to find the same voltage on both ends.

The two didoes, D1, D2 each have a parallel resistor R112,113 825 ohms. So is the path to -16 intact from the common end of those two resistors and diodes? In other words, is there 825 ohms more or less from pins 2,3 of IC1 to -16v?


Now, you have a channel card with a bad channel 12? No sound through it? Even using the insert? But you get signal blinking on the green light? Push the low cut. Any signal now? Does PFL work for that channel? You have 8 AUXs, do any of those have signal? They are pre-fader. Is there signal at the channel fader? And through the assigns, are all the subs and LR missing this signal? All of that will help define where that channel is breaking down.

[visoundguy]

Thanks so much for your help. Unfortunately - the problem still exists. Since the input board behaves the same no matter if I install it on channels 1-12, or 13 - 24, I'll focus on fixing that first. I actually replaced all of the capacitors - with the exception of the ceramic caps. I also pulled up one end of each resistor on that channel and checked the value - all were good. Thanks for letting me know that those were switching diodes! I thought they were zeners! Anyway, I also replaced them with 1N4148s that I had and - no change. So I put the originals back in to have the higher voltage rating. So at this point - I'm down to the ceramic caps that have not really been tested and/or changed. Would a failure of any of these cause low voltage and a voltage reading at the gain pot? To repeat my readings - I'm showing 4.6 volts to ground at the gain pot (good channels show less than .1 volts to ground). And I did make a mistake with the 16V - it is -16V as you said. To be clear - I am checking the voltage from ground to each of the resistor/diodes. One side of them is connected to the -16V bus - and I'm checking voltage to ground at the other end. On all good channels, I get 9.2V. On this bad channel - I get 5.4V to ground - on both diodes and both resistors!

Oh ya, and I also checked the gain pot and it checks good!

[Enzo]

I'd go down the remaining caps with an ohm meter. If they are shorted, they ought to show up.

It isn't sufficient to check all the parts, it is necessary to have a complete circuit. Two perfectly good parts will not function if the connecting trace between them is open. DOn't assume problems are always bad parts.

And that is why I mentioned checking against the -16v rail. This whole circuit stretches between the +16 and the -16. If the current path between those rails is open, then things wanting to balance out at near zero voltage - your 0.1v trim pot voltage for example - will get dragged towards one rail or the other.

SO taking readings to ground doesn;t tell the story. It is also important to check the voltage across parts sometimes. Your diodes/resistors, if there is -16 on one end, and 5.4 on the other, that sounds like 21.4v across the thing. Looks like the good channel has 25.2v across it. From your readings, those diodes are sitting +9v over the trim pot voltage normally, but in the bad channel we have them within a volt of each other.

Now that I look at it, that doesn't make a lot of sense. The trim pot is connected at a more positive point in the circuit than the diodes, and yet your good ones measure less positive. If you are taking voltage readings at the trim pots with the I/O board ribbon disconnected, then that explains the lack of voltage there.

If you replaced the 1000uf, then so be it. I think I would go back through the four transistors. With power off, check junction drops across each, and compare to a good channel. I would be thinking one may have been damaged or the solder connection not made or pad damaged

[visoundguy]

I'm almost finished replacing all the ceramic caps. When I finish, I'll take some voltage readings and get back to you. My voltages before were from memory - so I'll write them down. Also, as with my assumption that he voltage at the top bus was +16 - I'm going to have to re-check the voltage readings with my meter connected with the proper polarity. Again, if I only could get my hands on a schematic!!!! Without it, it's a hope and a poke! I've vowed that this is my last Behringer product - not because their stuff isn't good - but because of the lack of support. But that's a topic for another forum!

[Enzo]

Their support is to provide service centers. That may or may not be satisfying.

Some Behr drawings are on the web, see if you can find any of them, the mic inputs tend to be the same. MX3282, MX2442, MX9000, UB2442, Xenyx502, Xenyx2442, PMP5000, PMH1000, all those use the same circuit. And I bet if I look through enough Peavey or Mackie stuff, I will see it there too.

[visoundguy]

Actually - service centers are a very important leg of customer support. But it doesn't work for everyone!

Anyway -here are my measurements:

All voltages taken from the -16V rail. All good channels show 7.2v to most components and 16.9v at the gain pot. Nicely balanced.

The following comparisons are between the 2 channels on the shared part of the board that's damaged - one is operating correcly, and the other is not. Note that it is very unbalanced!

Good side of shared channels

to switching diodes: 7.22v
to parallel resistors to diodes: 7.22
to transistor inputs: 7.8v
to op amp inputs 7.2v
to op amp output 0v
to gain pot 16.9v

Bad side of shared channels

to switching diodes: 10.7v
to parallel resistors to diodes: 10.7v
to transistor inputs: 11.4v
to op amp inputs 10.69v
to op amp output 10.69v
to gain pot 12.1v

[visoundguy]

As you suggested, I also checked the transistors junction drops - they are identical to a good channel!

I had replaced these transistors 2wice by the way.

[Enzo]

Your good channel shows zero at the op amp output? From the -16 rail? That would mean the output was pegged against that rail. That doesn;t sound healthy, nor does it sound like it would pass signal.

[Enzo]

Here, these folks have a Behringer collection. As a non-member you only get two downloads per day. Some are pdf and some zipped. Pick one on my earlier list that is also on theirs and download it. The component numbers will be different - R23,C47, etc, but the circuits are the same.

Elektrotanya.com

[visoundguy]

sorry - I messed that up - it's very confusing with the board being upside down and all. Corrections are in red.

Good side of shared channels

to switching diodes: 7.22v
to parallel resistors to diodes: 7.22
to transistor inputs: 7.8v
to op amp inputs 7.2v
to op amp output 16.39
to gain pot 16.9v

Bad side of shared channels

to switching diodes: 10.7v
to parallel resistors to diodes: 10.7v
to transistor inputs: 11.4v
to op amp inputs 10.69v
to op amp output 16.9
to gain pot 12.1v

[visoundguy]

Wow! Thanks for the tip on the schematic!

[visoundguy]

Update:

You won't believe this!

So here I was - Mr. Confident with a schematic in hand and a good foundation of basic electrial knowledge.

I figured I could have this licked in an hour or less. So I started with the schematic. Since I had basically spent a week replacing and checking each and every component - I figured I would check the trace connectivity. Trace by trace, I marked the schematic to verify that it was connected to the place it was suppose to be.

After an hour or so, every thing checked! I thought I would find it right away! But no such luck.

Then I started checking the voltage values on the board again - staring at the schematic to try to reason what could be causing one channel to be high and the sister low, and pulling the board to check resistance values.

I had one channel - still high on resistance. I checked the series of resistors, alternate paths, possibility of the "sister" channel affecting the readings, and as time went on - slowly grew more frustrated. So I set the board down for 2 days - then returned to it with a fresh perspective. But as time went on - the frustration again grew. I started kicking the dog - yelling at the kids, and became a real mess.

I was almost ready to give up - and live with a dead channel. Then I thought about the fact that every time I looked at the schematic, it led me right back to the transistors. Check the transistors, I thought! I checked and they were PNP and NPN respectively - seemingly a good substitute for the obsolete A1084s and A06s. As I stared at the spec sheets for them, however, I had the A1084s laying side by side with the replacement 2N3096, and I just barely noticed that the Emitter, Base and Collector didn't correspond! One must be viewed from an opposite perspective I thought - a PNP is a PNP, Right?

WRONG!!!!!!! As I studied it more, I discovered that the packages were different! The Emitter and Collector were swapped! So I twisted the pins to match the package of the A1084s, and - VOILA!

Success. Everything is balanced on the input board, and all channels work as they should.

Thanks for your help Enzo, and particulary for pointing me to the Holy Grail of schematics!

Now on to the lower board to see where the signal is being blocked! More later.

[Enzo]

Glad you sorted it out.

For future reference, it is a good idea to mention when you replace something with some other part. Otherwise we assume 1084 for 1084 for example.


USA type TO92 transistors, like MPSA06 and 2Nxxxx, are E-B-C across the front. Asian TO92 types, like 2SC1815 or 2SA1084, are E-C-B across the front. Parts may be equivalent, but they are not the same. Just like a 3mm machine screw is pretty much equivalent to a 6-32 machine screw, but they are not the same. SO it is a good rule of thumb to replace USA types with USA types, and Asians with Asians. Otherwise you have to do some sort of dance with the leads.

If 2SA1084 is not readily available, there are certainly many fine low noise Asian TO92s. 2SA733 is common and Mouser sells then for 9 cents.

[visoundguy]

More lessons

Here's the final chapter of fixing the bad channel on the Behringer 3282 board.

I obtained the schematic - and since I was getting the pfl light, but no sound out of the channel - it pointed directly to a 100uf 10v resistor.

I removed the resistor, but it tested good. Replaced with a new one anyway.

So now I started heading down the road with a blindfold - just swaping out the next likely component - and testing. Still, no luck.

At about 1am on Sunday morning - I was about to again throw in the towel. I set down the board - and noticed a very tiny shiny spot on the back of the board. I looked as closely as my eyes would allow and thought that it just looked like a minor scratch. But still, I wanted to be sure, so I picked up a magnifying glass.

To my disbelief, I noticed that the scratch actually severed 2 of the very small traces!

So I then check with the meter, and sure enough - they were diisconnected. I then jumped the traces - and reinstalled the board in the mixer.

Sure enough - the deed was done.

I now have a fully functional Behringer 3282.

Regards.