Power Supply won't work in Constant Current mode

[LoveLaika]

I'm trying to repair a power supply that's having issues in constant current mode, a SM 7020-D power supply by Delta Elektronika. For some reason, it's not consistent in constant current mode. Following my setup shown in the image below, I set my output voltage to 5 volts, and turn current all the way to 0. As I turn the current knob and view the corresponding current rise, the current suddenly jumps from a small value like 300-400 mA to 1.2 A. I'm not sure why this is happening.

On a related test, I also tried this in constant voltage mode by first setting the constant current limit to 2 amps. With voltage starting at 0, I slowly turn the knob and watch as current increases, and it does in a linear fashion as expected. I'm not sure what could be wrong with it, but as a start, I unsoldered the controls to the potentiometer being used for the current knob. It was still functional, 10 turns sweeping from 0 to 5k ohms, so that seems to work well. What else should I look for that could cause this issue?

[goaty]

What kind of Meter do you have in series ? Maby it´s not constant burden voltage causing this ? Unlikely but worth a thought.

[LoveLaika]

What do you mean by constant burden voltage? My meter is a Fluke 8846A bench multimeter set for the 10A setting.

[LoveLaika]

What kind of Meter do you have in series ? Maby it´s not constant burden voltage causing this ? Unlikely but worth a thought.

If it helps, I managed to find a schematic of my power supply at this link:

https://www.electronica-pt.com/esquema/lcd/power-supply-delta-sm1540-d-sm7020-d-sm3004-d-25395/

Page 33. The current control knob is R701. It seems that when I press down the CC/CV button, I can set the current limit to a fixed level, no problem (I think). The problem starts when I set the voltage limit and then I increase the current. Right now, I'm trying to trace back what could be causing this issue if there's nothing wrong with the potentiometer.

[TimFox]

The burden voltage is the voltage across the ammeter terminals.  At full scale, it may be larger than the full-scale current times the inherent full-scale voltage due to extra stuff in series, such as a fuse.  In general, it should be small or the power dissipated in the ammeter will be too high.

[goaty]

If it´s fixed 10A range, it´s not an effect of the meter. Just wanted to add that as a thought, but it´s not  an issue in your case...
You have to dig deeper then ;-)

You checked it´s not in programming mode somehow, so the manual controls really are in control, right?

That´s a hell of a schematic. What did they draw that with ?
(Guess someone was somoking stuff while drawing. Netherlands after all)

[LoveLaika]

Thanks for the explanation. I don't think that this is necessarily caused by the burden voltage. If that were the case, then perhaps switching to a larger resistor might help and nothing's inherently wrong with the supply. However, if there is something wrong with the supply, how should I try tracing the error back?

I started at the potentiometer (on P388, R701) and traced it to the current value display. Looking at it, the value being shown to the display is coming from white dot #8, but I can't find the corresponding white dot elsewhere on the schematic. I figured that this is as good as any place to start searching. The inconsistency error only seems to occur when you turn the current knob (after setting the voltage knob to a set value).

[LoveLaika]

Oh the joys of working with discontinued old stuff. Yes, manual controls are indeed being used. External programming mode is not being used at all, and the switches are set correctly.

[goaty]

I mean, even what does this mean ? Is this dutch code or what ?

So maby check Vref on IC605 and IC606 first that it´s stable on the potentiometer R701 ? Just an idea.
From there on I pass for now. No idea how regulation works ...

[LoveLaika]

This might sound kind of dumb, but even though a potentiometer may seem like it's working, is it possible that it can still be a problem? I unsoldered my pot that was controlling current, and using a multimeter, it showed the right resistance as I turned it. It doesn't seem broken or anything, but could it still have issues when in operation even though it seems fine?

[eblc1388]

On the front panel there is a button labelled "Display CV/CC Settings". Try changing the potentiometer setting while having this button pressed down will show you whether the power supply detects the user input. If the display changes in sync, then there is no reason to doubt that the potentiometer is faulty.

Whether the output acts accordingly is another matter.

[LoveLaika]

The thing about it is, I did. Initially, on my first test, I was able to set CC. When I brought it back to my workshop, I unsoldered the potentiometer and tested it. Once I found it was working, I soldered it back on and ran some more tests. Right now, it's in a state where I sometimes see the issue and sometimes I don't. I'm running more tests to exactly see how to cause the issue, though I am thinking of trying with the programming block to see if the issue occurs that way as well.

[LoveLaika]

Actually, in one instance of my tests, yes, it did. I was holding down CV/CC and was going to set CC to 3 amps. Going from 2.5 to 2.7 amps, for an instant, it jumped to around 3.5 amps before going back to 2.7 amps.  I don't know if I can replicate this again, but it was something that I observed.

[eblc1388]

The full current range is 20A so 2~3A is 10% of the range. Unfortunately, this is also the most likely current limit that is commonly used in testing other circuits. I think it is reasonable to assume that in the life of the power supply, the potentiometer has been repeatedly adj/set around this area and might have caused more wear than other parts of the potentiometer.

You can try the following:

1. set the output to 5V or 10V with no load connected

2. Press the "Display CC limit" button and turn the current limit adjust from 0~max and observe any jumps on the display reading. If so, you might need to replace the potentiometer. Chances are there is no jump after moving pass the 3A setting.

3. connect suitable load and current meter(DVM 10A range) to output. Check for jumps when you adjust the current limit from 0 to 10A.
 

[LoveLaika]

Thanks for replying. I tried what you said, changing out the potentiometer to a new one, and I also changed out an LM358 op-amp on the output board. Following the tests, working with a a suitable load and increasing the current slowly, I didn't observe a jump, though now I'm not sure if it was the potentiometer or the op-amp that was the problem. Either way, as I increased the current up to 9 amps due to the resistor, I didn't see any jumps so I think everything works out well.

Thanks for the help. I really appreciate it. I'm just glad this nightmare is over. If you don't mind me asking some questions, I'd like to get some advice from you regarding what you said and on another issue as well.

First, you theorized that the pot may have been repeatedly set/adjusted around this 3A limit causing some wear and tear. What are some good practices to prevent this from happening, or is this just something that happens as you move parts/knobs back and forth?

Second, with my link to the schematic below, I was wondering what you thought of the various black and white shapes littered throughout schematic? Parts aside, the shapes seem to represent connections to the other boards, but they don't seem to always be at the edge. Do you think the shapes are 1-to-1 with black # matching with black #, or black # matching with corresponding white #? It just felt kind of strange seeing these shapes, and it's hard to keep track of throughout the board, but it would be nice to know just in case this power supply breaks again.

[eblc1388]

...What are some good practices to prevent this from happening, or is this just something that happens as you move parts/knobs back and forth?

It just happens because current limit of 3A or more don't get used for 95% of the circuits one will be testing. You shouldn't worry too much about this as the new potentiometer will give you many years of service life before it worn out again.

Sorry but I have absolute no clue about the various symbols in the schematic as they are not commonly seen elsewhere. Instead, you are in the best position to find out exactly what they meant because you've got the physical circuit boards in front of you and you can verify your guess easily.