Pull fuse F300. This will disconnect the main converter. The PFC section (with a UC3854 controller chip) has current sensing of the PFC and could be shutting down due to some excessive loading from the main converter or the AUX supply. See if the 312 VDC PF supply comes up then. Measure carefully at the P2 connector back. Check the F1 250V 0.5A fuse near the P2 connector again after powering down.
If pulling the F300 fuse didn't help, try pulling the two wire (Red and Blk) cable between P301 and P2 off (at P2). (leave the F300 fuse out for now.) That will dis-connect the AUX power supply also. See if the 312 VDC PF supply comes up now (measure at pulled P2 cable end carefully). Trying to isolate where the problem is. Hopefully this doesn't blow a fuse somewhere, not likely with F300 out already.
Check the F1 250V 0.5A fuse near the P2 connector again after connecting everything back up and powering up. (assuming we haven't yet found the problem)
The UC3524 main converter controller chip is the earlier version of the supply. The later UC3875 chip version added adjustable dead time control. (dead time avoids shoot through shorting in the main converter between + and - phases of the converter bridge. The UC3524 has fixed dead time control, I think, but it may not be quite enough to cover the worst case of the other IR2110 and IRFP450 power Mosfets tolerance variation in all manufacturing spreads. And the UC3524 likely has some dead time tolerance variation itself. There might be faster IR2110 and IRFP450 versions available by now, or a 'slower' UC3524 maybe. The IR2110 and UC3524 chips plug into sockets. )
Hopefully it's not a dead time issue. I have some UC3524 type supplies that work just fine. But, again, it's a question of manufacturing spread in the components possibly being too close. They must have encountered some problem cases to have changed the design.
If pulling the F300 fuse didn't help, try pulling the two wire (Red and Blk) cable between P301 and P2 off (at P2). (leave the F300 fuse out for now.) That will dis-connect the AUX power supply also. See if the 312 VDC PF supply comes up now (measure at pulled P2 cable end carefully). Trying to isolate where the problem is. Hopefully this doesn't blow a fuse somewhere, not likely with F300 out already.
Check the F1 250V 0.5A fuse near the P2 connector again after connecting everything back up and powering up. (assuming we haven't yet found the problem)
The UC3524 main converter controller chip is the earlier version of the supply. The later UC3875 chip version added adjustable dead time control. (dead time avoids shoot through shorting in the main converter between + and - phases of the converter bridge. The UC3524 has fixed dead time control, I think, but it may not be quite enough to cover the worst case of the other IR2110 and IRFP450 power Mosfets tolerance variation in all manufacturing spreads. And the UC3524 likely has some dead time tolerance variation itself. There might be faster IR2110 and IRFP450 versions available by now, or a 'slower' UC3524 maybe. The IR2110 and UC3524 chips plug into sockets. )
Hopefully it's not a dead time issue. I have some UC3524 type supplies that work just fine. But, again, it's a question of manufacturing spread in the components possibly being too close. They must have encountered some problem cases to have changed the design.
Last edited:
Late edit:
Always power the unit down and allow time for the caps to discharge before measuring the fuses.
Note:
Do NOT put the F300 fuse back in with the F1 0.5 A 250V fuse (AUX power) still out, since the main converter needs AUX power to operate and may enter a dis-functional state without it.
Always power the unit down and allow time for the caps to discharge before measuring the fuses.
Note:
Do NOT put the F300 fuse back in with the F1 0.5 A 250V fuse (AUX power) still out, since the main converter needs AUX power to operate and may enter a dis-functional state without it.
Last edited:
For possible future reference and any other broken Xantrex XHR's that may come along, I'll mention the next debugging steps for either the AUX Supply, or the Main Converter causing the problem.
If the AUX Supply appears to be not working, (leaving fuse F300 to the Main converter out for now), and with the unit safely powered down (discharged caps),
remove the one (or two) ribbon cables going to the XRA2 / XR02 control/supervisor board on the top side. Then remove the screws holding it on, and pull the board carefully upward to remove. There is a header connector beneath it, P3, going to the J3 Aux Voltages connector on the bottom side. (Aux voltages measured there earlier) (may need a screwdriver under the board to pry the connector off)
This will unload the AUX Supply. Re-connect the P2 to P301 cable, and with the F1 0.5 A 250V fuse near conn. P2 in place. (but keeping the F300 fuse out for now). Power up again and check to see if the Aux Voltages have come up steadily. If not, then the problem is in the Aux Converter section. If the Aux Voltages ARE back up, then the problem is over-loading from the XRA2 / XR02 board.
Proceeding to the XRA2 / XR02 board now. Solder suck the input V pin of the 4 TO-220 Voltage regulator chips. (3 have heatsinks) Bend the input pin upward enough to clear the board pad. Re-install the board. Skip the ribbon cable for now. Powering back up again, check to see if the Aux Voltages have remained up. Then push one TO-220 chip V regulator (at a time) input pin down enough to temporarily contact the board pad. Likely that one of these will cause the AUX Voltages to fail. Mark which one. Power down again.
Solder suck the OUTPUT pin of the suspect TO-220 V reg. chip and similarly pull that pin up to clear the board pad. Solder tack the suspect TO-220 V reg. INPUT pin down to the board pad temporarily. Power up again. If the AUX Voltages fail to come up, then that TO-220 V Reg. has failed. But it may still have a load short down stream besides. If the AUX Voltages do come up with the output pin disconnected, then there is definitely a short down stream of the suspect V Reg. chip. Power down.
Use a multi-meter Ohms or preferably Diode-Check mode across the suspect V output board pad and the TO-220 ground pin. Check both ways (polarities). Should charge up some 1V or so either way. If not, then there is a short on the board. Could be a tantalum bypass capacitor, or possibly a chip. The board I have in front of me does not have IC sockets on the XRA2 / XR02 board! You will have to go through the tantalum caps one by one and de-solder a wire lead to check for a shorted one. Lets hope you find a bad cap..... I would replace the TO-220 V reg. chip too, just for future safety. Repair, re-assemble, re-test.
------
If the problem earlier appeared to be from the Main Converter (pulling fuse F300 cleared the problem of the AUX voltages) then check the big electrolytic output capacitors for shorts. Then check the output diodes/rectifiers for shorts.
Using a scope ( fuse F300 in place and cable P301 to P2 in place), one can check for which Mosfets are switching or not. Sources moving on Q115 and Q117, and Drains moving on Q114 and Q116.
But this HF switching may only appear sporadically if the AUX Voltages are being prevented from coming up by overloading the PFC.
Otherwise, using process of elimination:
Check the four IRFP450 Mosfets on the heatsink (all adjacent) for shorts between any two pins on each (with fuse F300 still out). Any shorts found, then pull the IR2110 chips to see if that clears the short. If a short remains, then replace the Mosfet pair and the associated IR2110 chip. (U97 chip and Q116 and Q117) or (U95 chip and Q114 and Q115). Could replace the UC3524B controller chip too, cheap enough.
A Mosfet may however have failed to an OPEN condition. This will still lead to an overload since the converter XFMR could still be driven into saturation with just one functional polarity repeatedly pulsing it. In this case, pull one of the IR2110 chips and jumper pins 1 and 2 and separately jumper pins 6 and 7 on the empty IC socket. (small bare wires plugged into the IC socket, these jumpers keep the associated Mosfets turned OFF)
Re-install fuse F300 and power up.
Functional AUX voltages NOW would indicate the OPPOSITE Mosfet pair has an OPEN fail. (pairs: U97 chip and Q116, Q117 or U95 and Q114, Q115) Replace the failed Mosfet pair and associated IR2110 chip, and optionally the UC3524B controller chip.
If pulling that 1st IR2110 chip did not correct the AUX Voltages, then try the same procedure with the other IR2110 chip. (replace the 1st IR2110 chip in socket) If this 2nd chip pull (and socket jumpers) fixes the AUX Voltages on power up, then the OPPOSITE Mosfet pair has an OPEN fail. Replace failed Mosfets and associated IR2110 chip, optionally replace the UC3524B controller chip too. Hopefully one of the pair fixes will have restored operation.
Otherwise some more head-scratching....
Note:
The newer XHR version with the UC3875 controller chip may have different IR2110 chip and Mosfet pair associations than given here. Those would need to be checked by tracing/Ohm-meter first. I'm guessing they will be the same.
If the AUX Supply appears to be not working, (leaving fuse F300 to the Main converter out for now), and with the unit safely powered down (discharged caps),
remove the one (or two) ribbon cables going to the XRA2 / XR02 control/supervisor board on the top side. Then remove the screws holding it on, and pull the board carefully upward to remove. There is a header connector beneath it, P3, going to the J3 Aux Voltages connector on the bottom side. (Aux voltages measured there earlier) (may need a screwdriver under the board to pry the connector off)
This will unload the AUX Supply. Re-connect the P2 to P301 cable, and with the F1 0.5 A 250V fuse near conn. P2 in place. (but keeping the F300 fuse out for now). Power up again and check to see if the Aux Voltages have come up steadily. If not, then the problem is in the Aux Converter section. If the Aux Voltages ARE back up, then the problem is over-loading from the XRA2 / XR02 board.
Proceeding to the XRA2 / XR02 board now. Solder suck the input V pin of the 4 TO-220 Voltage regulator chips. (3 have heatsinks) Bend the input pin upward enough to clear the board pad. Re-install the board. Skip the ribbon cable for now. Powering back up again, check to see if the Aux Voltages have remained up. Then push one TO-220 chip V regulator (at a time) input pin down enough to temporarily contact the board pad. Likely that one of these will cause the AUX Voltages to fail. Mark which one. Power down again.
Solder suck the OUTPUT pin of the suspect TO-220 V reg. chip and similarly pull that pin up to clear the board pad. Solder tack the suspect TO-220 V reg. INPUT pin down to the board pad temporarily. Power up again. If the AUX Voltages fail to come up, then that TO-220 V Reg. has failed. But it may still have a load short down stream besides. If the AUX Voltages do come up with the output pin disconnected, then there is definitely a short down stream of the suspect V Reg. chip. Power down.
Use a multi-meter Ohms or preferably Diode-Check mode across the suspect V output board pad and the TO-220 ground pin. Check both ways (polarities). Should charge up some 1V or so either way. If not, then there is a short on the board. Could be a tantalum bypass capacitor, or possibly a chip. The board I have in front of me does not have IC sockets on the XRA2 / XR02 board! You will have to go through the tantalum caps one by one and de-solder a wire lead to check for a shorted one. Lets hope you find a bad cap..... I would replace the TO-220 V reg. chip too, just for future safety. Repair, re-assemble, re-test.
------
If the problem earlier appeared to be from the Main Converter (pulling fuse F300 cleared the problem of the AUX voltages) then check the big electrolytic output capacitors for shorts. Then check the output diodes/rectifiers for shorts.
Using a scope ( fuse F300 in place and cable P301 to P2 in place), one can check for which Mosfets are switching or not. Sources moving on Q115 and Q117, and Drains moving on Q114 and Q116.
But this HF switching may only appear sporadically if the AUX Voltages are being prevented from coming up by overloading the PFC.
Otherwise, using process of elimination:
Check the four IRFP450 Mosfets on the heatsink (all adjacent) for shorts between any two pins on each (with fuse F300 still out). Any shorts found, then pull the IR2110 chips to see if that clears the short. If a short remains, then replace the Mosfet pair and the associated IR2110 chip. (U97 chip and Q116 and Q117) or (U95 chip and Q114 and Q115). Could replace the UC3524B controller chip too, cheap enough.
A Mosfet may however have failed to an OPEN condition. This will still lead to an overload since the converter XFMR could still be driven into saturation with just one functional polarity repeatedly pulsing it. In this case, pull one of the IR2110 chips and jumper pins 1 and 2 and separately jumper pins 6 and 7 on the empty IC socket. (small bare wires plugged into the IC socket, these jumpers keep the associated Mosfets turned OFF)
Re-install fuse F300 and power up.
Functional AUX voltages NOW would indicate the OPPOSITE Mosfet pair has an OPEN fail. (pairs: U97 chip and Q116, Q117 or U95 and Q114, Q115) Replace the failed Mosfet pair and associated IR2110 chip, and optionally the UC3524B controller chip.
If pulling that 1st IR2110 chip did not correct the AUX Voltages, then try the same procedure with the other IR2110 chip. (replace the 1st IR2110 chip in socket) If this 2nd chip pull (and socket jumpers) fixes the AUX Voltages on power up, then the OPPOSITE Mosfet pair has an OPEN fail. Replace failed Mosfets and associated IR2110 chip, optionally replace the UC3524B controller chip too. Hopefully one of the pair fixes will have restored operation.
Otherwise some more head-scratching....
Note:
The newer XHR version with the UC3875 controller chip may have different IR2110 chip and Mosfet pair associations than given here. Those would need to be checked by tracing/Ohm-meter first. I'm guessing they will be the same.
Last edited:
Hi I have a Sorensen / Xantrex 600V power supply which had a short across STP4NB100 and it took shunt resistor (17) and 2 other resistors(11, 14). Photos attached.
The problem as always the schematics are unobtainium.
I would really appreciate if you could help me to identify the resistors
Regards Jack
The problem as always the schematics are unobtainium.
I would really appreciate if you could help me to identify the resistors
Regards Jack
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.