| Electronics > Repair |
| HP E3610A Power Supply - Current not displaying / Voltage not changing |
| << < (4/5) > >> |
| F4:
--- Quote from: nvmR on December 15, 2024, 01:25:08 pm --- I once saw one that like to give some 3V AC along with the required DC. fun fun fun --- End quote --- Checked that, no AC at the output |
| F4:
--- Quote from: pqass on December 16, 2024, 08:53:53 pm --- --- Quote from: F4 on December 15, 2024, 01:05:11 pm ---In going thru the traces I decided to replace R29 and R30, which looked worse for wear. After removal I started to test continuity and saw that pin 7 and 8 of U5 were incorrectly repaired. The trace from R28 was attached to pin 8 instead of 7. --- End quote --- R29 and R30 are just used to limit the current to the CV and CC LEDs. U5 (LM393 comparator) drives their output (pin 1 or 7) to -12V if +Vin (pin 3 or 5) goes below +9.8V (center of R25+R41 divider); lighting up the LEDs at 24mA (24V/1K, hence the 1W resistors)!!! Whichever the Current Error Amp or Voltage Error Amp is controlling the output voltage (via CR4 or CR5 switching), the other will output close to +12V which will disable its corresponding LED. R28 is in the CREF parts; you probably mean R29 was incorrectly tied to pin 8 (+12V). That would always keep the CC LED off. Given potentially many issues with this board, if you haven't already done so, clean the board with generous amounts of IPA to remove all gunk/film. Also inspect all obvious (non-factory) changes or damage/cracks. --- End quote --- Yes, R29. I've inspected and inspected, ha ha. --- Quote from: pqass on December 16, 2024, 08:53:53 pm --- --- Quote from: F4 on December 15, 2024, 01:05:11 pm ---I put everything back together to see if this remedied everything, it did not. On the plus side, the output at the terminals is now -17v across both the + and - terminals. Since its a 15v supply, this is progress. Unfortunately it does not respond to the knobs for either voltage or current. --- End quote --- If the output at the terminals is -17V (assuming your DMM on DCV leads were reversed), then this is an improvement. The resistor divider formed by R15 (14.3K) and R37 (50K front-panel voltage pot) would produce 17V output IF the wiper on R37 was open/disconnected or stuck at max range. Q: With the unit powered-off, if you isolate R37 from the circuit (disconnect wire from PCB to one pot terminal), does it still vary from 0 to 50K (wiper still attached to one pot terminal)? Put the wire back onto the pot terminal once you've checked R37s resistance. This next test assumes that one CR4 lead is still lifted off the PCB, thus current limiting won't be affecting the output voltage and only the Voltage Error Amp is controlling the output voltage. Q: Although the voltage setting knob has no effect, can you apply a small resistance to the output terminals (try in the range of: 330R, 220R, 100R, 50R) and does the voltage stay the same? ie. can it regulate regardless of attached load? Don't use <20R @17V as that is 850mA and >10W. --- End quote --- I have not performed this test yet, see below. --- Quote from: pqass on December 16, 2024, 08:53:53 pm --- --- Quote ---Q2 was replaced a while ago, but put a new one in just in case, same behavior. U6 remains hot. I double checked the caps for polarity. U6 is drawing current, would the CC set be the area drawing that current it? U6 was also replaced previously, by the way. --- End quote --- U6 (-12V regulator) being hot is unusual. I can barely feel it in my E3611A. Q: Let the unit cool down then remove all three ICs U1, U4, U5. Turn the unit back on (output terminals will be at 26V, that's okay). Does U6 (-12V regulator) get hot? If no, then turn the unit off and put back U1 and turn it back on. Does U6 get hot? If no again, then the issue is the other U4 or U5 or in their parts of the circuit. Leave U4 and U5 out for now and concentrate on getting voltage regulation working. If U6 gets hot with only U1 installed it could be U1 or Q2 again. Or, maybe something else (non-factory) pulling on -12V. --- Quote ---I tested the TO3 transistors (Q1,3) in circuit with the diode setting, I got proper beeps from them with testing across CBE. I also went thru and took out and tested all the black DALE resistors, all looked good. Should I pull Q1/3 and test it further? --- End quote --- Troubleshooting should be methodical. After checking the obvious stuff like (a) all power supplies producing the correct voltages, (b) obvious damage areas or faulting components (bulging caps), then you need to isolate sections and work your way from the (measured) good parts until you find the unexpected. Yes it's good to replace problem components like electrolyic/tantalum caps and swapping ICs for known good ones, but just random checking or replacing rarely solves the problem(s). This is why I've asked you to lift one end of CR4, in order to isolate the current limiting parts from controlling the voltage at the output terminals. So we can concentrate on fixing the voltage regulation first. Sometimes schematics are hard to read, so I hope I've given you enough clear information on the Voltage Regulation loop for you to find the "unexpected" in it. For a detailed explanation of how the voltage regulation loop works see a previous post of mine here. --- End quote --- After removing the IC, the same behavior persisted, U6 is hot. I detached CR4 again, same. I grabbed my FLIR and looked at it and saw that CR6 was also hot, and Q2. I decided to change CR6 (7, 10 and 11 as well) just to be safe. No change in the behavior, these components are getting hot. CR6 is part of the voltage error amplifier, but I am still too novice to understand why it's drawing current. I am continuing to look at the board, but wanted to share that CR6 is also hot. |
| pqass:
--- Quote ---After removing the IC, the same behavior persisted, U6 is hot. I detached CR4 again, same. I grabbed my FLIR and looked at it and saw that CR6 was also hot, and Q2. I decided to change CR6 (7, 10 and 11 as well) just to be safe. No change in the behavior, these components are getting hot. CR6 is part of the voltage error amplifier, but I am still too novice to understand why it's drawing current. I am continuing to look at the board, but wanted to share that CR6 is also hot. --- End quote --- CR6 and CR7 are protection diodes; keeps U1 +Vin to -Vin difference to below 0.7V. (FYI: JP1 is open, JP2 is shorted) Neither one should get hot. But CR6 can get hot if the Voltage Control pot (R37) is fully CCW and the loop isn't regulating possibly due to a break between U1 output to Q1,Q3 base. CR6 and R13 (1K) can get hot because they're between Q1,Q3 (letting too much voltage through at [+S]) and R37 (at low ohms when CCW) to TP4. For now, just keep the Voltage Control pot (R37) at mid-point; don't dial all the way down until we get voltage regulation working. a. Check if CR5, CR6, CR7, CR8 are still diodes with your DMM on diode-check; lift one leg to be sure. Confirm working diodes before proceeding to the next step. b. Keep R37 pot at mid-point and one CR4 leg lifted. Then lift CR5 cathode off the PCB. Turn the unit on, you should see 26V on the output terminals. Why? Because Q2 is off pulled up by R5. And R3 turns Q1,Q3 fully-on. Now, with an alligator-clip jumper wire, momentarily (<20 seconds) connect -12V (TP8) to the lifted (unconnected) CR5 cathode. What should happen is that Q2 should be fully turned on, which pulls the Q1,Q3 bases down (shutting them off) and the output terminals should read 0V (or close too it). If this occurs like I describe then the problem is before CR5 toward U1. If not, the problem is after CR5 toward Q2. Report back. |
| F4:
--- Quote from: pqass on December 17, 2024, 09:02:42 pm --- --- Quote ---After removing the IC, the same behavior persisted, U6 is hot. I detached CR4 again, same. I grabbed my FLIR and looked at it and saw that CR6 was also hot, and Q2. I decided to change CR6 (7, 10 and 11 as well) just to be safe. No change in the behavior, these components are getting hot. CR6 is part of the voltage error amplifier, but I am still too novice to understand why it's drawing current. I am continuing to look at the board, but wanted to share that CR6 is also hot. --- End quote --- CR6 and CR7 are protection diodes; keeps U1 +Vin to -Vin difference to below 0.7V. (FYI: JP1 is open, JP2 is shorted) Neither one should get hot. But CR6 can get hot if the Voltage Control pot (R37) is fully CCW and the loop isn't regulating possibly due to a break between U1 output to Q1,Q3 base. CR6 and R13 (1K) can get hot because they're between Q1,Q3 (letting too much voltage through at [+S]) and R37 (at low ohms when CCW) to TP4. For now, just keep the Voltage Control pot (R37) at mid-point; don't dial all the way down until we get voltage regulation working. a. Check if CR5, CR6, CR7, CR8 are still diodes with your DMM on diode-check; lift one leg to be sure. Confirm working diodes before proceeding to the next step. --- End quote --- Yes, they are. CR6, 7 and 8 have been swapped already and all 4 check out on the DMM. --- Quote from: pqass on December 17, 2024, 09:02:42 pm ---b. Keep R37 pot at mid-point and one CR4 leg lifted. Then lift CR5 cathode off the PCB. Turn the unit on, you should see 26V on the output terminals. Why? Because Q2 is off pulled up by R5. And R3 turns Q1,Q3 fully-on. --- End quote --- I put back CR7 and 8, did as instructed here, but got 0 volts at the output. I thought it was the ICs missing, so I put those 3 back but the behavior was the same, 0v. --- Quote from: pqass on December 17, 2024, 09:02:42 pm ---Now, with an alligator-clip jumper wire, momentarily (<20 seconds) connect -12V (TP8) to the lifted (unconnected) CR5 cathode. What should happen is that Q2 should be fully turned on, which pulls the Q1,Q3 bases down (shutting them off) and the output terminals should read 0V (or close too it). If this occurs like I describe then the problem is before CR5 toward U1. If not, the problem is after CR5 toward Q2. Report back. --- End quote --- With the result of the previous test, I went ahead and did this one anyway. I did get 0v but am not sure this is relevant as to whether the problem still between CR5 and towards U1 considering I did not see 26v on the output terminals? When you say CR5 towards U1, does that mean pulling and checking all the parts in this red blob? |
| pqass:
--- Quote from: F4 on December 21, 2024, 12:02:27 pm --- --- Quote from: pqass on December 17, 2024, 09:02:42 pm ---b. Keep R37 pot at mid-point and one CR4 leg lifted. Then lift CR5 cathode off the PCB. Turn the unit on, you should see 26V on the output terminals. Why? Because Q2 is off pulled up by R5. And R3 turns Q1,Q3 fully-on. --- End quote --- I put back CR7 and 8, did as instructed here, but got 0 volts at the output. I thought it was the ICs missing, so I put those 3 back but the behavior was the same, 0v. --- End quote --- If CR4 and CR5 each have one leg disconnected from the PCB, and there's STILL 0V on the output terminals, then that means Q1,Q3 are turned off or permanently open (broken). Or, assuming R3 and R5 pull-ups are working, then the only possibility is Q2 is always conducting (broken). For now, we need to work from Q1,Q3 thru Q2 to U1 output. Upstream of U1 can wait until the driver transistors are confirmed working first. a. Start by confirming R3,Q1,Q3 are working by pulling-off Q2 from the PCB (watch out for the ferrite bead on the emitter leg and mark the orientation on the PCB). You should see 26V on the output terminals. If not, STOP here. b. If previous step sees 26V on the output, then we want to see if we can shut off Q1,Q3 by using an alligator jumper wire shorting TP3 with the PCB pad that Q2 emitter occupied. You should now see 0V on the output terminals. c. If the previous step sees 0V, then test Q2 as it may be permanently shorted. Replace Q2 (you may get away with a 2N3906 or other small signal PNP transistor temporarily) and re-do the b. test and following paragraph again (from my previous post). --- Quote --- --- Quote from: pqass on December 17, 2024, 09:02:42 pm ---Now, with an alligator-clip jumper wire, momentarily (<20 seconds) connect -12V (TP8) to the lifted (unconnected) CR5 cathode. What should happen is that Q2 should be fully turned on, which pulls the Q1,Q3 bases down (shutting them off) and the output terminals should read 0V (or close too it). If this occurs like I describe then the problem is before CR5 toward U1. If not, the problem is after CR5 toward Q2. Report back. --- End quote --- With the result of the previous test, I went ahead and did this one anyway. I did get 0v but am not sure this is relevant as to whether the problem still between CR5 and towards U1 considering I did not see 26v on the output terminals? --- End quote --- Correct' it's not relevant. This test is premature given that we didn't get 26V on the b. test. We now have to work toward Q1,Q3 instead. --- Quote ---When you say CR5 towards U1, does that mean pulling and checking all the parts in this red blob? --- End quote --- Yes but hopefully we won't need to pull and test everything in the loop. This is why I started with D5 (mid-point in the loop). I was hoping at least half of it (Q1,Q3 to D5) was working but since we never got 26V on the output terminals (with the loop broken at D5), then we have to work toward Q1,Q3 until we can be sure those components are working. Once everything from Q1,Q3 to D5 works as expected, we'll then work from D5 to U1 output (and U1 input components too). |
| Navigation |
| Message Index |
| Next page |
| Previous page |