Can anyone help me modify this power supply? EC-135 ( OSP-135 ?)

I purchased this HV power supply which is good for up to 550v and 500 ma, it is an electropheresis power supply, it has a safety feature built in that will not allow the supply to have power at the leads without a load, in case the leads are disconnected and left hanging. I would like to eliminate this feature, as the supply drops out with less than ~5ma and will not turn on with any voltage on the output. The supply was damaged in shipping, and IC 8 was squashed out of the socket. I was able to straighten the pins and reinsert the IC.
The supply seems to be a switching supply, without any computer control or major logic.The 4 rectifier diodes are mounted directly to the transformer,from there it goes through the large choke to the smoothing board and back to the main board outputs.

IMG_20171206_195619375_HDR.jpg IMG_20171206_195323399.jpg IMG_20171206_200249853.jpg IMG_20171206_195541146.jpg IMG_20171206_195638724_HDR.jpg IMG_20171206_195425425_HDR.jpg IMG_20171206_195446277.jpg
 
Last edited:
The power supply normally works this way, without any connection, you can hold the DC start button while setting the output voltage.and once released,(without a load) it drops back to 0v. In the constant current a resistance needs to be connected across the output then the DC start button is pressed, and the current level is adjusted, without a load the voltage will go to max while the button is held
 
Without schematic, it's hard to get into the internal circuit to disable the open circuit drop out feature, I think the best way is to load the output to draw current. Using resistor is ok, but when you adjust to lower voltage, you need to change resistor. This is a CCS circuit that you can use:

1) this is simple but need to experiment more and might drift more:

5mA CCS1.jpg

2) This is more precise, but more complicated:
5mA CCS.jpg

These are the two parts, the first is a 950V N-MOSFET, the second one is the opamp.

https://www.digikey.com/product-detail/en/stmicroelectronics/STF2N95K5/497-14271-5-ND/4515905

https://www.digikey.com/product-detail/en/texas-instruments/OPA705PA/OPA705PA-ND/416724


This is just an idea, if you are interested, you need to find a low voltage supply inside the power supply you can tap off. I see you have LM339 opamp or comparator and TC7107 LED display IC. Check the power pin what voltage.

Anyway, I don't want to spend more time unless you are interested. If so, I can work out the detail with you.

You need a heatsink as the transistor is going to get hot, no way out of this. You might be able to bolt onto the wall of the chassis. I choose this transistor because it has insulated tap, so you don't have to worry about the HV on the tap.
 
After looking at the components, I think I will look at the states of the quad latch and see if there is a pattern between enabled and disabled output, when the chip was pushed out of the socket, the DC power led was lit, but no power was present.
 
Last edited:
I was able to get it to work "properly" for a bench top power supply, the latch chip was extremely easy to bypass on the first try.
The first latch (0) is connected directly to the start / stop push buttons. I measured - 4.8v on the start(pin3) line during button push, and -4.8v on the on the reset line(pin4) unless there was a load present, I pulled the chip,and reinserted it without the reset pin inserted, and after the the start button is pressed, it will stay on. The output will stay on until I turn off the power supply.I was going to cut a trace, but it would make it harder to reverse.

So my $50 (with shipping) 500v 500MA power supply buy worked out pretty good, I may have to add more filtering, we'll see...

latch.png latch2.jpg
 
Last edited:
Without schematic, it's hard to get into the internal circuit to disable the open circuit drop out feature, I think the best way is to load the output to draw current. Using resistor is ok, but when you adjust to lower voltage, you need to change resistor. This is a CCS circuit that you can use:

1) this is simple but need to experiment more and might drift more:

View attachment 1060761

2) This is more precise, but more complicated:
View attachment 1060759

These are the two parts, the first is a 950V N-MOSFET, the second one is the opamp.

https://www.digikey.com/product-detail/en/stmicroelectronics/STF2N95K5/497-14271-5-ND/4515905

https://www.digikey.com/product-detail/en/texas-instruments/OPA705PA/OPA705PA-ND/416724


This is just an idea, if you are interested, you need to find a low voltage supply inside the power supply you can tap off. I see you have LM339 opamp or comparator and TC7107 LED display IC. Check the power pin what voltage.

Anyway, I don't want to spend more time unless you are interested. If so, I can work out the detail with you.

You need a heatsink as the transistor is going to get hot, no way out of this. You might be able to bolt onto the wall of the chassis. I choose this transistor because it has insulated tap, so you don't have to worry about the HV on the tap.
I wish I had your skills to "whip up" something like that on the fly.
 
I was able to get it to work "properly" for a bench top power supply, the latch chip was extremely easy to bypass on the first try.
The first latch (0) is connected directly to the start / stop push buttons. I measured - 4.8v on the start(pin3) line during button push, and -4.8v on the on the reset line(pin4) unless there was a load present, I pulled the chip,and reinserted it without the reset pin inserted, and after the the start button is pressed, it will stay on. The output will stay on until I turn off the power supply.I was going to cut a trace, but it would make it harder to reverse.

So my $50 500v 500MA power supply buy worked out pretty good, I may have to add more filtering, we'll see...

View attachment 1063233 View attachment 1063234


I'm awfully tempted by this supply Scott,as they are readily available,even in my neck of the woods,at reasonable prices.The footprint also looks to be very real-estate friendly.I like your no-load work-around,pretty slinky:)

I'm curious though: what are they using as the pass element? The only power devices (other than the TO-220 on the circuit board,which I surmise to be a low-voltage regulator for the chips) is the pair of TO-220's bolted to the chassis.Considering the current rating and available range of voltage,that seems a little ''light'' to me.Am I missing something here?
 
I think what they have done, is switched the primary on the power transformer, and used a feedback loop to control the switching (maybe PWM)
so there is very little current through the TO 220's on the chassis kind of like a class D amp.and no pass elements between the power transformer and the load.
It acts like it's controlling the power transformer with a servo controlled variac
 
I'll call it done,pulling heavy current, the transformer makes a slight buzz.
The power supply has some noise in the output, I will need to increase the output caps.
milliamps.jpg
volts.jpg
a little blurry, but it shows max power.
max power.jpg
 
Last edited:
Finished, I replaced the noisy original fan(sounded like an arc welder I used years ago) with a 12v processor fan, and connected it to the 5v supply, it is whisper quiet now, I also added another 2 100uf caps across the output for better filtering, it works fine now.
 
Back
Top Bottom