Maker Pro
Maker Pro

Problem on high voltage leg on triple output power supply.

He guys. Me again. I am in need of assistance once again with two AC/DC power supplies. The power supplies I would like to repair are 80VAC-120VAC Input ~ 390VDC/36VDC/24VDC output and I'm getting very high DC voltages (at start up, 160VDC then climbing steadily to over 430VDC) on one power supply and low voltage (>340VDC instead of 390VDC) on the other power supply. The bridge rectifiers seem to be fine and I'm also guessing that the Power MOSFET IRFP460A is functional as the power supplies have output and if the voltage was pulsing, which it is not, that would be a cap problem I would think. Any ideas on how to rectify this problem?

POWER SUPPLY 003 REV FW.jpg
 
As mentioned on another thread, these type of power supplies frequently require a minimum load in order to function correctly. With multiple output types, the regulation would be controlled by just one of the outputs which will most likely be the lowest one.
Try putting a load of 5% of rated output current on the 24V supply and see what happens.
 
I can connect an LED driver and a diode plate to the 24VDC output and adjust the output to the diodes from minimum to maximum output accordingly. Do you think that this will be appropriate to check the 24VDC output? I'm trying to get a grasp of the post and I am understanding that the regulation control of all the voltages is managed by whatever MOSFET or transistor controls the lowest output voltage, in this case the 24VDC output. Or is it that each voltage output is controlled by it's unique component?
 
If all the output voltages are generated from one transformer then all will be regulated from just one of the outputs. If there is no load on any of the outputs then no regulation will take place. Therefore the output that is responsible for regulation needs a load of some sort.
 
OK. Based on the comments, there is one common output, which is then divided into three separate outputs (24VDC, 36VDC and 390VDC). So I connected the LED driver and diode plate and tested voltages. With the LED plate's output at zero, the power supply's lower voltages had correct readings (24VDC AND 36VDC) but the high voltage side climbed to +420VDC. When I increased the 24VDC output to the diode plate, the high voltage dropped and so did the 24VDC output, down to 17VDC, causing flashing of the diodes so, based on what WHONOES stated, there's some issue with voltage regulation. Via one of the regulators I suppose? Would this mean that there is one specific regulator that could be pinpointed as the cause of the problem?
 
I ran the same load test on the second power supply and there were no voltage issues on the 24VDC or the 36VDC legs. The attached LED diode plate functioned well and there was no dipping or fluctuations on the 36VDC leg either. However, the problem persisted on the high voltage side (390VDC) as there was a lot of fluctuation between 348VDC and 360VDC but the output never peaked or stabilized at the required 390VDC so I'm figuring that this unit has a completely different problem. I also did some basic tracing of the low voltage side and it seems that the MBRF200CTG and SFF1606 and the 50V and 35V caps are the main components controlling the 24 and 36VDC outputs but I would like some confirmation before assuming I'm correct. I would also like to understand the high voltage side a little better. I know that the caps and rectifiers handle the low voltage side but how is the high voltage side created? There are no caps that I can see to smooth out the high voltage so I'm thinking that the high voltage is generated and controlled solely by resistors and rectifiers directly from the voltage from the bridge rectifier, specifically the FHF10N60 MOSFET and the main voltage controller for all outputs is the IRFP460A. Is my guesswork any where near vaguely correct? Anyone wants to follow up on these two power supplies?
 
It's somewhere between difficult and impossible to diagnose your power supply problems without reference to a schematic of some sort otherwise the nature of the design philosophy would be pure speculation on our parts.
Having said all that, it might be that regulation might be controlled from one of the other outputs. But, before trying that, put a resistive load on the most errant output and see what happens.
 
Top