Power Supply Repair Help

[smd75jr]

Hello Everyone!!!!

For my first thread on the EEVBlog Forums, I request your assistance!

Yesterday i managed to pick up a Fluke 8000A multimeter, an General Radio 1340 Pulse Generator, and a Power Designs Inc. 5015T Variable power supply, for a total of 20USD.

The Fluke 8000A and the GenRad 1340 both work beautifully, however the power supply does not. after rigging up a fuse (the fuse holder is broken) it will only output a maximum of about 5.8VDC when its stated voltage range is 0-50VDC at 0-1.5A.

So i found the manual online and have started testing voltages starting from the line voltage switch and lamp (old style 115VAC lamp) and working my way in.

The first issue i have encountered is a transistor marked Q9 on the schematic (http://www.ko4bb.com/Manuals/09)_Misc_Test_Equipment/Power_Designs/Power_Designs_5015T_Power_Supply_Service_Manual.pdf). The transistor is a 2N4888 PNP can type. The manual says that it should have 0.6V between the base and the emitter which i have measured as 0.66V, and 0.9V between the emitter and collector which I measured as 0.03V.

I plan to continue my tests on the unit and update this post when I find a problem. If there is anyone who knows an equivalent to the problem components I would be eternally grateful!!



And now, for those of you who love old tech, pictures!! (pardon the imgur, the files were too big)

http://imgur.com/a/r0rLm



Problem Parts:

Q9 - PNP - 2N4888

Q3 - PNP - TI1028A

[alm]

0.66V sounds fine, but a collector-emitter voltage of 0.03 V suggests that they may be shorted. I would desolder Q9 and measure resistance from collector to emitter. The other option would be that I2 is shorted, but that doesn't sound likely.

[smd75jr]

0.66V sounds fine, but a collector-emitter voltage of 0.03 V suggests that they may be shorted. I would desolder Q9 and measure resistance from collector to emitter. The other option would be that I2 is shorted, but that doesn't sound likely.

I2 comes up clean, no short there, and I desoldered the 2N4888 and it reads open circuit between the collector and emitter.

With the negative on the base and positive on the emitter it reads 0.68 ohm and with the positive on the collector it reads 0.66 ohm

Accordion to this helpful guide (http://www.learnabout-electronics.org/transistor_faults_06.php) the transistor is dead.

Is there an equivalent part that is actually still made?

[Mr Simpleton]

Not fully convinced the 2N4888 is gone... you say high impedance betwen collector and emitter, fine and then you se a low impedance with positive on collector and negative on base or positve on emitter negative on base. This sounds perfect for a PNP transistor. You should reverse and measure the impedance with negative on collector/emitter and positive on base to verify high impedance.

Usually one diode do break down and shows short in both directions.

[Andy Watson]

Accordion to this helpful guide (http://www.learnabout-electronics.org/transistor_faults_06.php) the transistor is dead.

The page you linked to shows a NPN being tested. I believe the 2N4888 are PNP - so reverse the polarity of the tests.

[smd75jr]

Not fully convinced the 2N4888 is gone... you say high impedance betwen collector and emitter, fine and then you se a low impedance with positive on collector and negative on base or positve on emitter negative on base. This sounds perfect for a PNP transistor. You should reverse and measure the impedance with negative on collector/emitter and positive on base to verify high impedance.

Usually one diode do break down and shows short in both directions.

The reverse shows infinite resistance (i.e. open circuit)

[Mr Simpleton]

So the 2N4888 seems OK to me then

[smd75jr]

So the 2N4888 seems OK to me then

Yay!




And now the plot thickens!!!!

T2 should have 94VAC between pins 5 and 3, i measure 107VAC.

I measured S2 (a thermal cutoff switch) at 86VAC

I measured T1 between pins 5 and 3 with all settings at maximum at 108VAC

I measured C1 at 74.2VDC (should be 62VDC)

I measured CR6 at 1.92VDC (should be 2.5VDC)

[alm]

I don't have time to look at the schematic now, but a change in mains voltage from say 110V it may be designed for to 120V can easily be responsible for small differences in unregulated voltages, like on the primary or secondary side of a transformer.

[Andy Watson]

T2 should have 94VAC between pins 5 and 3, i measure 107VAC.

I measured S2 (a thermal cutoff switch) at 86VAC

I measured T1 between pins 5 and 3 with all settings at maximum at 108VAC

I measured C1 at 74.2VDC (should be 62VDC)

Have you clocked the text on the schematic that says "Voltages measured with 115VAC, 60Hz input, with 50V DC output and 1.5A load" ? The voltages quoted on the schematic are the loaded voltages. Small transformers can easily have regulation of around 10%, i.e. the voltage rises by 10% when off load. Your mains input might not be 115V either.

[smd75jr]

T2 should have 94VAC between pins 5 and 3, i measure 107VAC.

I measured S2 (a thermal cutoff switch) at 86VAC

I measured T1 between pins 5 and 3 with all settings at maximum at 108VAC

I measured C1 at 74.2VDC (should be 62VDC)

Have you clocked the text on the schematic that says "Voltages measured with 115VAC, 60Hz input, with 50V DC output and 1.5A load" ? The voltages quoted on the schematic are the loaded voltages. Small transformers can easily have regulation of around 10%, i.e. the voltage rises by 10% when off load. Your mains input might not be 115V either.

I did see that and have put a resistor (10kohm 5% 225W) across the output (not sure if that will help but it's worth a shot).



Did more measurements!!

CR8 reads 0.428VDC (should be 0.5VDC)

Q8 between base and emitter reads 0.28VDC with positive on the base and negitive on the emitter (should read 0.6VDC)
Q8 between the emitter and collector reads 18.71VDC (should read 13VDC)

Also, R37 is getting quite hot.

[alm]

A 10 kOhm resistor is not going to draw 1.5 A, unless this is a high voltage power supply. If the output voltage is zero, then no resistor connected to the output is going to draw any current. Small differences like 0.43 V instead of 0.5 V is unlikely to be a defect, most likely the drop across the diode is less because of the lower load. The only reading that might be suspicious is the 19 V reading where you expect 13 V, but depending on the circuit this may easily be caused by the lack of loading.

[Andy Watson]

So far you've established that the transformers and rectifiers work at least enough to fire-up the circuit. Have you a means of checking the main capacitors? Either by measuring their capacitance, measuring the ripple, or by substitution. If they've dried-up you'll be feeding the circuit with pulses - could lead to all kinds of crazy voltage measurements if you're expecting D.C.

[smd75jr]

A 10 kOhm resistor is not going to draw 1.5 A, unless this is a high voltage power supply. If the output voltage is zero, then no resistor connected to the output is going to draw any current. Small differences like 0.43 V instead of 0.5 V is unlikely to be a defect, most likely the drop across the diode is less because of the lower load. The only reading that might be suspicious is the 19 V reading where you expect 13 V, but depending on the circuit this may easily be caused by the lack of loading.

Not surprised that the resistor (looks more like a ceramic pipe, similar to this: http://www.digikey.com/product-detail/en/EWT225JB10K0/EWT225JB10K0-ND/2077384) wouldn't work.

Most of the voltages I've measured have been slightly (a couple volts at the most) above what is labeled on the schematic, which i believe are well within tolerance, especially since line voltage here is 120VAC. But i agree that that 19V is quite suspicious.

So far you've established that the transformers and rectifiers work at least enough to fire-up the circuit. Have you a means of checking the main capacitors? Either by measuring their capacitance, measuring the ripple, or by substitution. If they've dried-up you'll be feeding the circuit with pulses - could lead to all kinds of crazy voltage measurements if you're expecting D.C.

Sadly I've yet to invest in a capacitance meter (nor do I have the money for one). I do have a scope though, is that sufficient? And if so, what exactly would I do? (I'm still  kinda new to in depth circuit testing/troubleshooting)

[smd75jr]

Also, any ideas on a makeshift load that would meet the required specs? (50VDC 1.5A)

[megajocke]

It will not be possible to get that loading unless the power supply is working properly.

I'd suggest you start by checking that the strappings between RV2 to RV1, DC+ to S+ and DC- to S- are installed correctly. Then check voltages across zener diodes CR11, CR12 and CR14 and emitter-collector of Q3. These voltages should be close to the schematic values regardless of load on supply output.

That part of the circuit provides power to the control circuit and the reference voltage for regulating the output. If this bias supply is working properly the components around Q8 would be my next thing to check. That transistor turns on to increase the output voltage, and your previous measurements indicate that it is not turning on much. An open R27 or leaky/shorted C6 would be my guess in that case, considering your previous measurements. Does adjusting the output voltage control change the output voltage? And if it does, in which way does it change?