Behringer xenyx 2442fx power supply?

[yan6]

Hi there I picked up one of the above mixers for what I thought was a great price. It was said to be working but when I got it home all I got was effects display to come on, the phantom power LED and nothing from the power LED; naturally it didnt pass any sound through it. I did some digging around on the internet and found the schematics. I took it apart and put a meter to the power, I get low voltages from and a hum from the supply when connected to the main board. When disconnected the hum goes away and I get +24v, -28v, gnd, +48v and maybe a 5v I cant remember. does this imply that there is a short on the power rails pulling down the supply when connected to the main board. If so what is the next step in trying to solve this issue. I would love to go out and replace it but cant afford it.

Edit: could there still be something wrong with the power supply causing it stop putting out when connected to the main board

cheers
Ryan

[The Dude]

You could have a regulator "giving up the ghost" or a short on one of the power supply lines. The thing to do would be to ohm out the line(s) that are going low to ground and see what your load is. That should tell you if you have a short/near short somewhere, or if your power supply is failing.

[Enzo]

That may be it, but I fear your short may be rail to rail, rather than to ground. +15 shorted to -15. Where? on one of the many ICs strewn throughout the unit. Leave it on a bit and you MIGHT find one getting warm. Of course it still could be a short to ground, but would more likely then be one of the million little bypass caps each IC has nearby.

AT YOUR OWN RISK, one method to find such a short is to apply something like +/-24v to those 15v rails, and let the bad little cap burst, revealing its location.

Yes, it is a serious pain in the ass.

The power supply is probably shutting itself down. It detects excess load, and stops feeding it. Yes of course the supply could have a problem under load, but I tend to think it is the mixer part. But you can test it. Disconnect the supply from the mixer section, so you get all the voltages. Try loading each 15v output to see if it holds up. Not a lot, maybe half an amp tops. Ohm's Law will tell you what size resistor will draw half an amp from 15v.

Further issues. ICs are circuits themselves, and all it takes is for a couple internal transistors to short, and the IC takes down the rail, but those shorts are usually on the other side of semiconductor junctions, so your ohm meter cannot detect them from outside the chip. And for that matter, "shorted" caps may really just be super leaky, but not until they see a few volts, so your mater may not be able to detect those either. But hopefully yours will turn out to be something with measurable clues.

The 15v regulators are pairs of discrete transistors. I wouldn;t normally expect both circuits to fail at the same time.

[yan6]

Thanks for the replies guys if it is a rail short, it sounds like near impossible to find and fix. I'm wondering why the smps is putting out +24v & -28v shouldn't it be +-15v.

Additionally I don't have access to a 8w resistor is possible to simulate the wattage at all with something like an automotive light bulb in parallel to a 1/4w 30 ohm resistor.

[Enzo]

24v probably because it is running unloaded. My suggestion of a half amp load is not critical. Load it with something remotely close. And clip your meter to it. You will know in seconds if the supply passes the test or not, long before it can heat the resistor up much. How about a 47 ohm 5 or 10 watt? Auto bulbs are not low current devices. A 12v auto bulb typically uses several amperes. All we need to see is if the supply can work at all. My money is on that it will. SO half an amp, quarter amp, 100ma? You must have something in your drawer. And as I said, even a 2w resistor won't erupt in flames in 10 seconds.

It is not impossible to find, just tedious. Once found, fixing is just soldering in a new part.


You have bench supplies? You could send a pair of 15v supplies from a bench unit into the mixer section, and see what it draws.

[yan6]

Enzo, thanks for the help thus far. I think I will try and work on this tonight; I will should be able to find a resistor to load the supply with But I'm thinking your right that a part has shorted the rails. I noticed when the lights were out that the power LED is very dimly lit.

I do have bench supplies to power the board with. If supplied with +-15v and I don't current limit the supplies shouldn't the shorted part give me a smoke signal, or do I risk burning other parts with it.

Cheers
Ryan

[Enzo]

If you have 100 things in parallel, and one is shorted, what danger does it present to the others? It essentially removes power from them.

Besides, go ahead and current limit it. The point is not to see if the thing works that way, the point is - at this step anyway - just to see if the thing is loading the supply or if the supply is faulty.

[yan6]

So after some testing, correct me if I'm wrong but its looking like the power supply.

I put a 46.55 ohm resistor across each of the 15V and ground and found the following:


+24.6v unloaded
+8.1v loaded

-29.7v Unloaded
-1.1v loaded

Additionally I set up a bench supply with +-15v and applied it to the main board and the power LED came on and it had 0.6A draw steady.

What would be the next step in repairing the power supply

Cheers
Ryan

[Enzo]

If the mixer board seems stable at 600ma, did you check it for passing signal while you were at it? Assuming it is fine, we would then expect the SMPS to be happy producing 600ma a side too. Your 47 ohms draws half that at 15v.

Did you try loading only one 15v rail at a time? SOunds like you did them both together. We may find the -15 is bad while the +15 is being dragged down, or vice versa.


PLEASE remember how dangerous it is to work on an SMPS. Everything on the primary side is HOT. NOTHING there is referenced to ground. The circuit common on the primary is just the negative side of the rectified and filtered mains. DO NOT clip a scope ground to that side. Preferable to use at least an isolation transformer.

The secondary side IS referenced to ground. But note also that the A ground and the D ground are not connected together. SO to make good readings of the 15v rails you need to use the A ground, not the D ground. Speaking of D ground, is your +12 close to voltage? Or is it substantially high like the 15v rails? If we find just one 15v rail is really bad, we can fix it. If we find both, then we want to look for common things, and the first one I see is the regulator based on the 431 and IC4. But if that is out of whack, then so should ALL your secondary voltages be.

[yan6]

I didn't try to pass any signal as I didn't have anything with to do so. Would it work with just the +-15v and analog ground, i presume it would also need the digital ground, which I'm not sure if I can provide with just my bench supplies?

I loaded each rail of the power supply individually to the analog ground

Where would I look for the 12v your are speaking of, I don't think I saw that off the main power connector and looking at the schematics I dont see any 12v from the supply

[Enzo]

All the mixer should need to work is +/-15. That gets you audio, and I think most controls. Without +5 you get no FX. +48 is only for phantom. And +12? Fan? But the basic signal path ought to work. ALL you need to test it is some signal source, presumably you have that. Then the output, just plug a cord from MAIN L or MAIN R, or for that matter any output, over to some sort of amplifier and speaker. I'd have thought you had amps and speakers to use with this unit.

Upper right corner of power supply schematic is +12. Looks like it goes to a two pin connector X2, the 6-pin cable is at X3. The two pin is why I think It might be a fan supply.

The difference in the grounds is just to keep noise out of the system, in testing you could hook them together.

[yan6]

Enzo, I do have amps, speakers and gear to generate signals. The problem is I am working between home and work, where all my test equipment is. I took a drum machine to work and powered on the mixer and put the signal through it; I only had my headphones but had output on the main output L+R and through the headphone amp on every channel. The channel one clip light was on all the time but I'll look at that later. The 12v supply is for an external lamp, it gives 18v when the power supply is unloaded and 12v when I load the positive side (+15v to ground). It stays at +18v when I load the negative side (-15v to ground), which seems sensible. Does that give any more clues as to where to look in the PS.

Is there a schematic of the power supply the schematics I have for the mixer only show the power section from a high level.

Thanks
Ryan

[Jazz P Bass]

Here is the power supply schematic.

[The Dude]

The channel one clip light was on all the time but I'll look at that later.

That is likely a clue you should not overlook.

Edit: To elaborate: It's at least plausible that the external supply you connected is capable of maintaining a voltage at a much higher current draw than the supply in the mixer. That clip light that's stuck on might be telling you that something is shorted on channel one and loading the mixer's power supply.

[yan6]

Jazz P, thank you for the power supply schematic.

I took another look at it this morning and the clip light was not on when I powered it on and signal came through just fine. I didn't realize but the solo switch activates the clip light, the solo switch on this channel is very sketchy, wiggling it causes the solo to come on and off. Its possible this is what had happened last night?

I noticed in the power supply C16 & C19, the caps that sit across the +-15v rails have swollen tops and there is a small amount of brown stuff under them on the PCB. These obviously need to be replaced, could they be the sole cause of this or should I replace the TIP31 & TIP32 while I am ordering parts.