Burning amp (unintentionally!)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi Everybody - once again I could use a bit of help. I have finally managed to solder a beast together that did not work at first flick of the powerswitch - guess everybody runs out of luck sometime :)

This is a P1.7 preamp, boards by KK-pcb. Powersupply alone works fine, and measures close to the specified 60 volts. Connected it this evening to the main board, and measured 0 volt on the output of the psu - after a few seconds the regulator FET on the PSU blew up (litteraly - there's hole in the plastic where it used to say IRF610!) in a mushroom cloud of nasty smellin' smoke.....damn! :(

Of course, I have checked the main board for solder blobs in the wrong places (lead free solder is a pest to deal with I think), capacitor polarity, short circuits between FET legs and FET heatsink - everything looks fine.

Can I measure the resistance of a preamps "power-inlet" (my physics says "NO!" - but I gotta ask) - any other idea of finding out what's wrong before I run out of IRF610 FETs?

Kind regards
Hans
 
Administrator
Joined 2007
Paid Member
Run it up with a low wattage bulb in the primary side to stop excess current.
If the PSU output was zero and it blew then it sounds like a dead short that it's working into, particularly if nothing it was supplying blew/smoked etc
 
Administrator
Joined 2007
Paid Member
I would run it up with the bulb... check for overheating of the reg all the time.
Looking at the circuit there is too much resistance in the preamp circuit to blow the FET like that unless those 1000uf caps were in wrong way round or faulty.
 
I've got 2.2 ohm on the dc input (two channels in parallel) - each of the four big 1000uF caps measure 5.4 ohms across. I only have one PSU though, I know the original uses two, but one should do - the circuit draws about 25 watt total, incl. the loss in the regulator.

Kind regards
Hans
 
Will try and get a picture online later tonight or tomorrow - double, triple and quadruple checked power caps polarity, and circuit polarity. Working on getting a bulb socket up and running, then going to test the two sides of the circuit once at a time.

Kind regards
Hans
 
Administrator
Joined 2007
Paid Member
If you look at the circuit you will see the only "low" impedance path is via the 22 ohm resistors.
You need to measure carefully. Remember too that DVM's can give strange readings with any voltage present on a circuit (charged caps etc) However anything reading a few ohms is definitely suspect.
Measure carefully the input to board, then measure again from ground to the positive end of those two 1000uf caps.
Nothing else can give such a low value. Maybe remove the caps and remeasure.
 
Think you are right about that last measurement Mooly - 2 ohms are wrong - its practically a short!

Just goot back home from a dinner with the Misses - could not stop thinking about the error - hardly enjoyed the food :) anyway, I removed the powerwiring when I got home, and now I measure about 700 ohms on both channels - with the ground probe on the positive power input - the other way around I have infinite resistance. Think the culprit is a short in the power cable somewhere - the heat from the solder probably melted the insulation (stripped out section of cat5 network cable - newer using this again!) Will try with better wire tomorrow - and a bulb - just got to scavenge the addict for a bulb socket :)

Thanks for the input and help - hope the problem is solved!

Kind regards
Hans
 
Since the DC input measures several Kohms after replacing the power wiring, I triedto switch on one channel - without light bulb - could not find a socket :(

The PSU slowly goes up to around 15 volts - then blows the 0,5A primary side fast blow fuse - should this be greater?


Kind regards
Hans
 
Administrator
Joined 2007
Paid Member
No :) 0.5amps plenty.
I usually solder wire to the bulb direct and (very naughty) have an old plug I can solder the wires direct to where the fuse would go.
Looking at link I posted it looks like the PSU connection is separate to both sides of the board.
So concentrate on one side at once. You either have to have the confidence now to know for sure that it's all assembled correctly or you have to go down the bulb route and take careful measurements.
I would (if it's easy) remove both ZTX 500's and those 22 ? (I take it it's 22 ohms, although to me 221 can mean 220 ohms 22 and one zero) it's unclear as 100 ohms are marked 100 so 221 means 220 so what's 22.1 and 33.1 ????)
Then power up and the full voltage should be present on the input.
Take it from there and remember to discharge via a resistor the PSU before reconecting things.
 
Hi Mooly
Yes, those are 22,1 and 33.1 ohm resistors - and they are a bit larger than the others (1watt) - however, judging from the size on the PCB, the first resistor (3,3 ohm) is only 0.25 watt.....hmmm

I can easy replace the ztx550 and disconnect the resistors - if everything then looks ok, you think I should go ahead and reconnect the 22,1 and 33,1 resistors, discharge the psu and then fire it up again (hopefully not litteraly! :))

I will keep the fuse - AND try the bulb :)

Kind regards
Hans
 
Administrator
Joined 2007
Paid Member
OK... I'm still confused :)

The 22.1 Is that a 2.2 ohm ? and 33.1 a 3.3 ohm ?

I am not familiar with the amp but they seem low... are you sure they are the correct values. 22 and 33 ohms would sound more reasonable. The ZTX transistors form constant current sources by turning on when 0.7 volts is developed across those resistors. 2.2 ohm gives over 300ma.
Can you actually measure those four resistors and confirm the values accurately.
 
Just unsoldered the ztx550 transistors - they all tested negative on the hFe test on my measuring instrument - replaced all four in both channels - all gave no reading. Still the same - voltage raises to about 15 volt slowly, then the fuse blows, so something seems like the current draw is way to big. Think I am going to replace all other semiconductors....might as well cut to the chase - or would that be waste for time?

Kind regards
Hans
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.