[El_Doug]

Finally getting the last delivery from Mouser, I was able to finish stuffing my sigma 22 board tonight.

I did a power-on, and was happy to see the LED light up. However, I only saw 300 or so millivolts dc on v-out. All of the sudden, Q15 sparked crackled and puffed. it smelled quite a bit, considering how little smoke came out...

of course I turned it off immediately. still, the LED remained lit for quite a while until the capacitors turned off.


Any suggestions would be greatly appreciated!

 

[sachu]

with it unplugged, check each of the MOSFETs..Use the multimeter in ohm mode and measure ressitance between each of the MOSFETs drain pin (middle) and the source pin (the rightmost pin if you hold the MOSFET facing you). If it reads a short or a 10s of ohms, then that MOSFET is pretty much dead.

Also check to see if your output resistors are okay..make sure they measure 0.47 ohms each.

 

[compuryan]

check out the .PDF at the bottom of the "initial check" page: The σ22 Regulated Power Supply for the specific voltage of your sigma 22. It can help isolate the area of the problem by checking values using a multimeter. AMB will probably have a much better answer, but poking around with a multimeter can't hurt until then. Good luck!

 

[El_Doug]

Q15 and Q16 both measure a steady 10.08kOhms from drain to source

Q13 and Q14 both begin measuring at 2.5kOhms, but do not remain steady, instead continually creeping upwards. i got bored watching the numbers slowly increase, as they passed 3.8kOhms. this seems troublesome to me... yet it was Q15 that produced the smoke

all 4 output resistors are measuring 0.47ohms, as they should


@compuryan - I already had that printed out, before I even powered on. however, I am hesitant to reapply power to test for those voltages, since Ive already had one big smelly puff of smoke

 

[El_Doug]

now i'm wondering if perhaps I have the transformer wired incorrectly into the s22...

with this transformer setup, which of the secondaries should be connected to each of the ~ G G ~ inputs on the s22?:

 

[sachu]

can oyu get some pics up?..how have you mounted the MOSFETs on the heatsinks? Have you used the insulating pads?

I suggest you remove the MOSFETs fromt he board , especially the ones on the side ( remove both on that side) where you saw smoke emanate from and then check resistace between drain and source pins again.

 

[sachu]

Quote:

Originally Posted by El_Doug /img/forum/go_quote.gif now i'm wondering if perhaps I have the transformer wired incorrectly into the s22... with this transformer setup, which of the secondaries should be connected to each of the ~ G G ~ inputs on the s22?:

blue and gray are both connected to G and red to one of the ~ and yellow to the other ~

 

[El_Doug]

I'll do the desoldering in a bit, I really cant tell you how much I appreciate your time.

I am indeed using Thermalsil insulating pads. Here's the image:

 

[El_Doug]

Quote:

Originally Posted by sachu /img/forum/go_quote.gif blue and gray are both connected to G and red to one of the ~ and yellow to the other ~

damn, that's how I had it wired

I was hoping against hope that it would be this simple of a fix

 

[sachu]

if you had wired your transformer wrongly you would have blown a lot more than just a MOSFET...your rectifying diodes and possibly stiffening caps would have blown as well.

 

[El_Doug]

then im glad that an incorrectly wired transformer was NOT my problem

desoldering these is proving harder than a simple resistor, possibly because of the much thicker joint. i guess ill have to pick up a desoldering pump monday - i'm getting nowhere with braid

does the mosfet mounting look correct?

 

[sachu]

I don't know how i made it this long without a desoldering pump...life saver i tell you..lot of heartache can be avoided with it. I still manage to lift a pad now and then.

 

[El_Doug]

ok, so despite my common sense telling me not to power up the board, I couldnt stand sitting there, looking at the s22 without any results. and this is actually quite interesting, and hopefully very encouraging!!!

I turned on the amp, the LED was stable, and I didnt smell anything. I tested the output, and low and behold:

V- is outputting -31.xx volts! V+ is outputting a grand total of... -3.1 volts. Note that this power supply is set up to output +/-27V.


Now I'm extremely perplexed. The negative rail includes Q15, which sparkled and puffed... yet it is stable and strong. However, my positive rail isnt working at all, and is in fact measuring a small negative voltage.

Any ideas?

 

[amb]

It doesn't make sense at all... the negative rail is supposed to track the positive rail.

Anyway, do the following measurements (power supply only, no load connected). Be very careful with your meter probes not to short circuit anything. Steady hands required here.

Check the DC voltage (relative to G) at each MOSFET's metal mounting tab. They should show the pre-regulated voltage (depending on the transformer, this could be somewhere in the upper thirties to forty-something volts range).

Then measure the voltages at the emitter pins of Q1 and Q2. This should be no less than a couple volts lower than the previous measurements (per rail, respectively).

Measure the voltage at R4, R5, R6 and R7 (doesn't matter which end). What is it?

Now measure V+ and V-. Screw down the screws in the terminal blocks before doing this, or you may not get a good reading. Still wrong?

 

[El_Doug]

I'm not certain which pins on Q1 and Q2 are the emitter pins

From ground, the mounting tab on Q13 was 45.2v, Q14 was 45.2v, Q15 was -45.2v, and Q16 was -45.2v

the resistor values were very very strange. they were never stable, and ranged from 0 to 20-something MILLIvolts


edit: the precise output numbers are: V-: -31.65v, V+: -3.026v