Yet Another Blown Velleman
Feb 13, 2006 at 6:46 AM Thread Starter Post #1 of 7

dsavitsk

MOT: ECP Audio
Joined
Aug 3, 2003
Posts
2,883
Likes
44
I have gone through my fair share of Velleman power supplies. The first (http://www6.head-fi.org/forums/showthread.php?t=109928) was my fault as the transformer was spec'd too high for the smoothing cap. The second (http://www6.head-fi.org/forums/showthread.php?t=114261) was a little less my fault and involved a LM317 dying. Some resistors also got toasted, and which was the cause is not clear.

Anyway, I had replaced pretty much all of the parts, and have been using the PS with a Pimeta for about a year. Well, today, I was listening to the amp, left the room for a minute (this always happens when I leave the room) and when I came back there was a stink, some smoke, the led's in both the PS and the Pimeta were off, and there were awful sounds coming from my headphones.

I shut it all down (oddly, the LED's came back on before I sould shut it down.) The phones (RS-1's) appear to be unharmed, but the TLE appears to have blown and the PS is giving about 50V. Nothing else in the PS is burned up, though.

I cannot figure out what could have done this. I had it connected to my new preamp as I was testing it out. It is a high voltage tube pre, but there are blocking caps on the output, and they are intact, so it seems unlikely to have been offset fom the source. The LM317 is a National, it does not run excessively hot (and should have shut down if heat was the issue) and has been runing for a year without incident.

The only thing different about today is that the pimeta's volume was most of the way up as I was testing the preamp and checking it for noise. But the volume was not high on the preamp. The only thing suspect about the Velleman is that it does not have the protection diodes on the regulator.

Well, any thoughts? Does a TLE usually blow due to overvoltage, or something else? Has anyone who has had a blown TLE have any thoughts on anything else that might be toasted? The caps are 35V and probably got more than that, so they may need to go -- though I don't see any leaks. I tested the Pimeta opamps (the L and R, not the G yet) in an opamp tester similar to Tangents. They test fine. Clearly I am going to throw away the Velleman board and replace it with something else, but this is a bit of a mystery.

?

-d
 
Feb 13, 2006 at 8:55 AM Post #2 of 7
Take the Velleman board out of the case enough to get at all electrical points and take some measurements. Examine the parts as well, including solder bridges or the possibility of it shorting out somewhere.

Basically, if you have no shorts visible and measure 50V before and after the regulator, it would then appear it has failed to a closed short.

As always, it would be nice to have big glossy poster sized pictures.
 
Feb 13, 2006 at 7:50 PM Post #4 of 7
For a 22 vdc output, 50 vdc is a high amount to input to the LM317. It has 28 volts across it and power changes from the wall can make it go higher. I would only put about 30 vdc into it so that it only has 8 volts across the LM317.
 
Feb 13, 2006 at 8:16 PM Post #6 of 7
LM317 are pretty robust but in theory it's possible to damage one. A typical scenario might be connecting the already-powered Velleman to an amp with a lot of capacitance. That creates a very large surge current, perhaps enough to damage the LM317 before the protection on it kicks in.

Just measure the parts, the regulator and diodes. I suspect one or the other(s) have failed closed, shorted out.
 
Feb 13, 2006 at 8:37 PM Post #7 of 7
Quote:

Originally Posted by dip16amp
For a 22 vdc output, 50 vdc is a high amount to input to the LM317. It has 28 volts across it and power changes from the wall can make it go higher. I would only put about 30 vdc into it so that it only has 8 volts across the LM317.


I bet that's it. I bet a +/-15V transformer was just simply too much here. It's a little unnerving that it made it quite so long though.
 

Users who are viewing this thread

Back
Top