Allen and Heath MPS14 SMPS Repair

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Blissy

Well-known member
GDIY Supporter
Joined
Sep 5, 2020
Messages
67
Location
Australia
Hi All,

I am halfway through the repair of an Allen and Heath MPS14 and have reached point where I am stuck. I am not at all an expert in SMPS. Currently the issue seems to be on the primary side. R13, R76 and TR1 all get extremely hot under load after about 10 secs. When switching the power supply on Cold (supply caps drained) a screeching sound can be heard as the supply ramps up over 1 - 2 seconds then the sound stops. Hard to pin point which component is making the noise. I have an identical working supply which rules out the console as the issue and does not scream / ramp-up on power up.  The issue looks to be in the PFC, I could be wrong.
I have attached the schematic below, can provide any information required. 
The Supply does operate and supplies all the correct rails. The rapid heating of the mentioned components does not occur in the working supply.
I have replaced all the electrolytic Capacitors as a matter course.

Thanks any help appreciated. 


 

Attachments

  • MPS 14 .pdf
    240.6 KB · Views: 47
I hope you'll have fun enjoying all the spam you'll be getting from now on (due to your chosen username here)  :-X

Without some isolated oscilloscope probing, it's tough to pinpoint the exact cause of the issue you're encountering, but... Considering there are no esoteric components, i'd be willing to "shotgun" it, and replace all the silicon there (UC3854, the transistor, D6 and D28).

The noise will most likely be coming from the PFC inductor, but if the transistor is switching in an unruly manner, that would explain both that, and the apparent overcurrent situation there. If even the pair of current-shunt resistors gets hot, that would lead me to believe that something about the current-sensing of the PFC controller might not be entirely well.
 
You might consider replacing the guts of that PS entirely. The older SMPS aren't that good. They're noisy and inefficient. You could get 4 off-the-shelf enclosed SMPS and mount them inside. Make a PCB with CLC filters to take out switching noise. The only issue would be that there are no stock supplies that make 18V. But you probably want to regulate the +-18 outputs anyway so you could use some beefy mosfets and zeners to bring the voltage down like described here: https://sound-au.com/project102.htm
 
My first suspect would be TR1 is blown. R13/76 in parallel are 0.15 ohms so you need a LOT of current to warm them up. Obvious explanation is TR1 is either shorted or turned full on. Try removing the PF IC and using the shorting link for the gate of TR1 needed when the IC is removed. If the problem remains it is TR1. If it goes away it is most likely the PFC IC.

Cheers

Ian
 
squarewave said:
You might consider replacing the guts of that PS entirely. The older SMPS aren't that good. They're noisy and inefficient. You could get 4 off-the-shelf enclosed SMPS and mount them inside. Make a PCB with CLC filters to take out switching noise. The only issue would be that there are no stock supplies that make 18V. But you probably want to regulate the +-18 outputs anyway so you could use some beefy mosfets and zeners to bring the voltage down like described here: https://sound-au.com/project102.htm

Hi squarewave, Thanks for the reply I have considered this however struggled to find modules that wouldn't cost me a small fortune. Are there any in particular you recommend?

Thanks
 
If TR1 was indeed shorted, that would've blown it, and/or the fuse, right off the bat. And besides, the OP said "The Supply does operate and supplies all the correct rails. "

ruffrecords said:
My first suspect would be TR1 is blown. R13/76 in parallel are 0.15 ohms so you need a LOT of current to warm them up. Obvious explanation is TR1 is either shorted or turned full on. Try removing the PF IC and using the shorting link for the gate of TR1 needed when the IC is removed. If the problem remains it is TR1. If it goes away it is most likely the PFC IC.

Cheers

Ian

Hey, that's great news! You might've gotten off easy then :D

Yep! just found D28 and D6 shorted Thanks all, replacing all the silicon Just waiting on the UC3854.  Will let you know how it goes!
 
Yep! just found D28 and D6 shorted Thanks all, replacing all the silicon Just waiting on the UC3854.  Will let you know how it goes!
Failing as a short circuit is common failure mode with overheated diodes (silicon melts and becomes an always conductor).

Shorted diodes may be the cause, or may just be a symptom of overheating from some other overload fault.

Good luck.

JR 
 
Hi squarewave, Thanks for the reply I have considered this however struggled to find modules that wouldn't cost me a small fortune. Are there any in particular you recommend?
The MeanWell ones are known to work well and they're cheap. Just glancing at the website and datasheets it looks like RS, LRS, EPP and NPF series are possible matches. An RS-150 for example is $20 US. But again, you would need to know a bit about E to design a PCB for the CLC filters and two "regulator" transistors on the edge for mounting on a block of aluminum (which there might already be inside the enclosure). So it's not trivial if you're not experienced with E but it's something that could be explained here well enough. And if done correctly it could easily improve your noise floor.
 
Hi All,

UPDATE.
So parts arrived, I replaced IC1 (UC3854AN) and D28 and D6. Unfortunately R19 and then shortly after R20 Burned out. Trying to Narrow down what else could be Wrong. R22, C11 Tests good. 
Any help is appreciated. Getting short (2ohm) from TP1 to TP2 which goes away when IC 1 is removed. Luckily I ordered a few UC3845's!!

Cheers
Blissy
 
Blissy said:
Hi All,

UPDATE.
So parts arrived, I replaced IC1 (UC3854AN) and D28 and D6. Unfortunately R19 and then shortly after R20 Burned out. Trying to Narrow down what else could be Wrong. R22, C11 Tests good. 
Any help is appreciated. Getting short (2ohm) from TP1 to TP2 which goes away when IC 1 is removed. Luckily I ordered a few UC3845's!!

Cheers
Blissy
Did you check/replace the mosfet?

The TI data sheet suggests a 5 ohm resistor in series with the gate drive to prevent overshoot, although I presume it was working OK before.

JR
 
JohnRoberts said:
Did you check/replace the mosfet?

The TI data sheet suggests a 5 ohm resistor in series with the gate drive to prevent overshoot, although I presume it was working OK before.

JR


Hi John,

Yes can confirm that T1 checks good, and yes was working before

Thanks
 
UPDATE

Ok found that R33 and R76 where open... Obviously had melted open while testing under load. R19 and R20 are stable and do not Burn up! I have confirmed IC 1 is getting +18v in ref to TP 1 and the supply is stable under no load. I tested the supply shortly under load and there is a constant Hum(resonance) which sounds like its coming from IND1, R33 and R76 still Heat up.

Looking through the UC3854 Data Sheet pin 14 CT sets the oscillator freq, is it possible that the timing cap has failed and thus causing both the audible hum and heat?
or possibly C61 or other large film capacitors has failed resulting in higher noise on the rail causing the noise.

Wondering is any one is familiar with the UC3854 and might know where to start looking next

Once again, Thanks
Blissy
 
JohnRoberts said:
they are 1W resistors so probably run warm... Have you confirmed that the load is not drawing excessive current...

JR

Yes can confirm that the load is good. I have a working MPS14 that does not show any of these symptoms under the load. Further voltage at TP3 is 328 No Load and drops to 270 under load
 

Latest posts

Back
Top