Switchmode PSU Repair Advice for my RME converter

Hi,
The switchmode PSU in my RME converter stopped working, looked like a blown fuse on the SMPS so i replaced it and that one popped instantly.
After taking a look at the PSU I saw that a power resistor on the input side of the PSU had a hole blown in it so im presuming this is the startup resistor that has stressed and failed.


Its placed on the neg side of a big 400v 47uf cap on the input side of the PSU.

Looks like a 0.75Ohm to me, is this a carbon resistor?
I dont have one the same but i do have some other big fat 235Ohm resistors that look similar, can i drop one of those in or is the value vitally important in this part of the circuit ?

Any help would be very much appreciated, thanks !

Perhaps a side effect of the actual failed part... check the switching devices, IC?

JR

Ok I will try and take a look thanks, (although the whole rest of the PSU looks physically fine) should I just be testing that no pins are short that shouldnt be?
Any way are your thoughts on my question about putting a higher value resistor in as a replacement?

Thanks for your help

You want to know what I think of substituting a 235 ohm resistor for a blown .75 ohm?

I am willing to venture without much inspection that the difference seems too large.

JR

Thanks for your help, it looks like the Mosfet has gone short, so I have ordered a new one, there were 2 other smaller resistors connected to it that were blown too, i hope that's all that is wrong  :-\

I just fixed an SPX 90 with a shorted filter cap at the 5volt distribution point to the circuit cards.  This took out a number of parts in the path up to the  final cap.  Just sayin',  I'm sure you've looked at the chain.

Thanks, yes I planned to check everything else nearby, the Caps are fine the only thing I still need to check is the IC, I can do that while waiting for the delivery of the new Mosfet.

I also found that a resistor on the board near the mosfet's heatsink had burned through its leg and flash/shorted to the heatsink, see this pic:

I think you should replace any parts that are physically visibly damaged.  I am thinking the burned out resistors for sure and using the same values (although it is fine to combine values to get the startup resistor value, you need a high wattage there) I do not think that startup resistor is carbon just big due to wattage.

I found the following link useful when I tried doing this, but I have to admit, I have been defeated before by a broken SMPS (one for my Dell 30" display that Dell would not repair or sell parts for), they can be complicated beasts and if there are multiple voltages and regulators, and many different magnetics it gets complicated.

<http://www.repairfaq.org/sam/smpsfaq.htm#smpsmcp>

There are companies that will fix it for you, but they generally charge $200 up.  I could refer you to one in the US if that would help.  Nice guys, talked to me gave me my options, and let me know some ideas (check the diodes, check the fets check the electrolytics (they can fail, but they need to be checked for too high ESR not just value).

For a converter I imagine that you are talking probably 1 bipolar output (to drive opamps) and one 5 volt output.  If it was me, after I tried what you are trying now, I would try to figure out whether I could move the PSU external and bring in DC.  You would probably get a Noise level improvement in the process.


Sorry I can't be more help.

Thanks for the link,
Yes I have replaced the burned out small caps and checked the electolytics for ESR.
I have made up the correct resistance for the big resistor with some other higher wattage resistors
I will check the other components too.

I can just buy a new PSU from the RME dealers here in the UK for around £50 delivered (pretty pricey for what it is) but this is also an educational thing for me to try and keep up my electronics practice, so i dont really want someone to fix it for me, id rather learn to do this kinda thing myself whenever required.

This PSU is 3 rail 5, 12 and 24 V I think.

But if I were to give up I would certainly just make a linear supply delivering DC with a JLM Audio Powerstation, never liked the idea of this being a SMPS in my Analogue/Digital converter.

Well that's good RME sells the part, thank heavens for good manufacturers.

Dell wouldn't sell me a supply for love or money on a product only a year or two out of warrantee.  Eventually I had a friend of mine buy a used one in Mainland China. 

Linear supplies are bigger of course, my feelings is if you can get it outside the box, that is a big advantage (distance is natures Mu-Metal) bigger advantage perhaps than the linear/switched difference.  Also switched supplies have inherently high ripple, but the ripple is at such a high frequency that the removal of it is pretty easy.  Taking out 200K hz ripple is an easy, physically small and steep filter LC filter, whereas linear supplies you are trying to get rid of 50/60 HZ which is a larger physically and more challenging problem.