# Trying to re-purpose high voltage power supply. Need help

Discussion in 'Power Electronics' started by ARMANDO, Jul 9, 2017.

1. ### ARMANDO

25
4
Aug 5, 2016
I reversed engineered the schematic but I don't know much about power supply's and how to build one.
I would like to know from looking at the schematic can someone tell me if it looks like something that's possible or even worth doing?
(never mind the red circles)

2. ### (*steve*)¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥdModerator

25,505
2,852
Jan 21, 2010
Sure.

Do you know what the output voltages are (and are you sure you can measure them safely?)

3. ### kellys_eye

4,717
1,312
Jun 25, 2010
It's a pretty simple bunch of unregulated DC supplies - you can measure the outputs if the circuit can be powered (CAREFULLY - there are potentially fatal voltages in there!)

If the circuit is just 'theoretical' you could still fabricate one. A dedicated transformer with multiple windings or separate transformers with appropriate secondary voltages is all that is required - apart from the rectifiers/smoothing etc. It's not difficult to replicate.

There are a mix of high voltage/low current and low voltage/high current outputs. A guideline to the outputs can be determined by the fuse ratings and smoothing capacitor voltage ratings.

BR1 seems to be a negative high voltage (around 100-120V) at a few 10's of milliamps.
BR2 could be around 30-36V at 10A
BR3 could be plus and minus 30-36V at 5-6A
BR4 could be 100-130V at a couple of hundred milliamps
BR5 could be around 60V at 10A+

all supply outputs are GUESSWORK at this stage based only on component ratings.

The separate diode at the top seems to add a high (??) DC voltage to the output from BR1

4. ### ARMANDO

25
4
Aug 5, 2016
I tested the transformer at its outputs the rest of the circuit has not been put together yet.
I have all the parts that were originally in it so I'm not trying to replicate it.
I just want to build a nice bench power supply that can handle high current loads, not really interested in high voltages. so my goal here is "can I re-purpose or even re-configure these parts to do that"

Output Voltages from top to bottom (on schematic)

Coil1 = red+red = 305vAC
Coil2 = green+green = 88vAC
Coil3 = orange+orange = 18vAC
..........................................................54v
.......................................................... / \
Coil4 = purple+white+yellow = 27v- -27v
.......................................................... ^
...................................................... C Tap
Coil5 = black+black = 62.6vAC
Coil6 = gray+gray = 46.6vAC

5. ### kellys_eye

4,717
1,312
Jun 25, 2010
Use the 27-0-27 taps to create a variable DC supply with + and - rails.

Don't bother with the rest of the tappings.

Unless you're after a particular volt span????

6. ### ARMANDO

25
4
Aug 5, 2016
I would like to have a broad range of voltages , but more specifically 12VDC at around 20-30A do you think i can draw 25A+ through this supply here's the transformer i got

7. ### ARMANDO

25
4
Aug 5, 2016
And what were these caps for? I assume they were connected to the AC side primary
(judging from the ratings of 120/240vAC and 60 cycles per second ) = (same as Mains)
but how?

Last edited by a moderator: Jul 10, 2017
8. ### ARMANDO

25
4
Aug 5, 2016
So After connecting up and installing the fuses for circuits BR2 And BR3 I got some Unexpected results for Voltages. on these two lines . Does anyone why?

9. ### (*steve*)¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥdModerator

25,505
2,852
Jan 21, 2010
What are you measuring the voltages with respect to?

I suspect that you are measuring voltages with respect to some point that has no connection to the voltage you're measuring and you're just getting capacitive pickup.

10. ### kellys_eye

4,717
1,312
Jun 25, 2010
If all you want is a high current 12V supply then salvage an old PC power unit. Using that transformer will require significant linear regulation that would make the result both cumbersome and hot - unless you used the DC to feed a buck converter rated to the current demand you have.

Those 'tin can' capacitors are actually 'line filters' intended to be placed in line with the incoming AC to filter noise going in and out.

I take it your meter was set to AC for the transformer winding measurements and DC for the voltages after the rectifiers?

11. ### ARMANDO

25
4
Aug 5, 2016
Thank you , (*steve*) and kellys_eye for the help

Well I did try several ATX power supplies that I had lying around, none had enough power (amps).
I tried one from a dell desktop that should put out 17A on the 12v rail. and that was insufficient.
I did however just recently buy a 1000w ATX from e-bay for around \$50 it should deliver 32A on the 12V rail.

Anyway wheres the fun in buying something when you can make and learn something along the way,
What did you make today?

When you say in Line with the AC do you mean like this?

#kellys eye Yes! that is how I did it .

# (*steve*)
No steve I dont think thats what i did. I had my meter tips on the two output lines.

12. ### davennModerator

13,991
2,018
Sep 5, 2009
NO definitely not

they will be between phase and neutral

what the heck are you trying to power ??

the power supply you have been showing us isn't going to be enough then !!

13. ### kellys_eye

4,717
1,312
Jun 25, 2010
Yes, that's how the filters are connected. Measure between the two terminals and report your results (resistance).

Measuring the AC at 14/15 on the transformer shows 18VAC (?) which, under normal circumstances, will rectify/smooth to approximately 26-28V DC across C5 (terminal 2 -ve, terminal 3 +ve). Check the 12A fuses.

Measuring 27VAC at 12/13 and 11/12 on the transformer should result in around 40V across terminals 25-/26+ and 24-/25+ (a total of around 80V across 24/26). Check fuses again.

There are many opportunities to obtain SMPS supplies on eBay - keep looking!

I'm fully onside with your policy of making stuff - check out this kit for making a variable PSU (0-30V 0-3A, current limited):

at \$6 (around £4) I got two!

It requires a 24-28V AC input.... how convenient.

14. ### davennModerator

13,991
2,018
Sep 5, 2009
show me a circuit where the filter caps are in series with the power rails !!
unless you think they are not caps

15. ### kellys_eye

4,717
1,312
Jun 25, 2010
I suspect it's an in-line filter module. The internal capacitors are likely connected to the casing.

Most such filters I've seen have two terminals at each end for the live and neutral in/out but IIRC the older, higher power, versions had one such for each line.

The resistance check should reveal the answer.

16. ### ARMANDO

25
4
Aug 5, 2016
Not sure who' right here davenn says NO! and
Is that not good?

I like K eye advice (
)
but I'm still not sure Who's correct. I need more info here Before I flip the switch.
I don't need any more fireworks here.

Thanks for the link K eye Ill check it out

17. ### kellys_eye

4,717
1,312
Jun 25, 2010
Like I suggested measure the resistance from one terminal to the other. If it shows a 'short circuit' then they go as I said.

If they measure open-circuit they go across the line/neutral as davenn says.

davenn likes this.
18. ### ARMANDO

25
4
Aug 5, 2016
I did the resistance test from the case to one terminal and then to the other terminal and I got a rising resistance on both? then terminal to terminal i got 1.1Ohms

19. ### davennModerator

13,991
2,018
Sep 5, 2009

seeing the name on them west cap .... makes me think they were caps .... maybe they are inductors disguised as caps
no cap value given tho

20. ### kellys_eye

4,717
1,312
Jun 25, 2010
I'd guessed they were a pi-type filter - capacitors across input and output with a series inductor between them.

The picture showing them in-circuit appears to show the neutral line connected to only one of them too.

davenn likes this.