Alelph 5 troubleshooting question

[SteveA]

My Aleph 5 has passed the smoke test (after having failed due to a 9610 in backwards, oops). I replaced the 9610, a couple of fried resistors and the MPSA's. Using a variac to bring the voltage up slowing (monitoring the rail voltage and the output voltage for both channels) I see the rail voltages expected (the good news)

The bad news is that for both channels I see the positive rail voltage appearing at the output. Which component(s) would be most likely candidates for replacement?

Thanks and Happy New Year
SteveA

[kilowattski]

Steve,

Which IRF9610 was in backwards? Was it Q1, Q2 or Q3? Which resistors burned? You need to give us some clues if we are to help you. There could be a host of other components that were damaged and need to be replaced but it is difficult to speculate because you gave us very little information. If it was one of the differential pair IRF9610's don't forget you need to change both because they should be a matched pair. Your output transistors should be matched also if you have not done so already.

[SteveA]

Hi Kilowattski-
It was Q2 which had been reversed, R15 (75R) which fried. I had several 9610's and output FET's on hand and had measured/matched them prior to placement. I am using boards that were designed by Mark Finnis for the Aleph 4 but decided to make an Aleph 5 (had a transformer on hand).

Mark suggested (I e-mailed him) measuring the Vgs on the output FET's. He said I should measure 3.5-4.5V (is this the same voltage from the matching process?). Based on this I think the FET's may have been fried also (none of the FET Vgs's were in that range). BTW, he suggested using 1/3-1/2 rail voltage for this. I could only go to 1/4 (at 1/2 R15 starts to smoke, will need to replace).

At this point I will replace R15 (which now are measuring ~60R), Q1/Q2 to ensure matched pairs, and all output FET's (ouch). Do you think I should check/replace any other components?

Thanks,
SteveA
PS, nice P1.7. I'm working on one myself. I am interested in comparing with the BOSOZ I built.

[Blues]

With the power off check the CCS output FETs ...most probably the D-S of one or even 3 of the FETS are shorted. You can use an ohmmeter to check a short (0 ohm or close to it)

[SteveA]

Hi Blues-
I'll check that out. It would be nice not to have to replace all of them.
Thanks
SteveA

[BDP]

Steve,

Sounds like you have a shorted output MOSFET(s) causing current to be drawn through R15.

You can test the front end by its self by disconnecting the output devices from the main board and power. Then you can supply a ground to the the gate of Q2 through R7 and R10. This will make the differential pair think all is well at the output. You can then check all your voltages around Q1,2,3 and R14 drain resistor. If this all checks out then I would hook up only one pair of output devices (One for the positive and one for the negative side). Of course you would use two good MOSFETs and sense lead from the source resistor. This way, if you have a board problem you only ruin a couple of devices. Like everyone else has said double check your layout.

I would be glad to correspond with you by e-mail if I can help.

BDP

[SteveA]

Blues/BDP-
Thanks for the much needed help. It would appear that about 2/3 of the FET's are bad. I will still replace R15/16 on both boards. Is it a safe assumption that (all) the FET's went south as a result of the backwards Q2?
SteveA