Maynuo M9712b Always Displays 0.0000A, Won't Regulate Current

[The Doktor]

I just got a new Maynuo M9712b DC electronic load. It was listed as "new, not in box". My offer of $350 was accepted, I was happy Well, it arrived today, now I'm not happy

For testing, I hooked it up to my Power Designs 6050 (60V, 5A), which has current limiter. Tried CV mode on the load. Works great, holds voltage to set value (I tried 2V to 60V) with good accuracy. Current is shown as 0.0000A, power 0 watts.

Moved to CC mode (set to 1A), no good at all. Still shows 0.0000A, pulls supply down to about 1.25V as soon as it's switched on (pulling ~5A). Set current to 0.1A, no difference

Tried CR & CP modes as well, both fail as well. But 1 of those modes indicated 6.xxxxA, even if the supply was disconnected.

There is a BNC connector on the back which is supposed to give 0-10V, proportional to the percentage of the loads rated current it's drawing. Interestingly, that seems to work.

I'm going to contact Maynuo and see if I can get a schematic. I believe the problem is a dead op amp (or possibly the ADC?), but no idea how to identify which one. Anybody here have any suggestions or experience with these? I would prefer to repair it rather than return it.


Ed

[Armadillo]

Why don't you hook up external amp meter to confirm whether current is flowing in accordance to the setting. Doesn't your pro design power supply has a amp reading?
It has a multi-output transformer inside, prone to damage.

[The Doktor]

Current flowing is whatever the power supply is set to deliver (external meter agrees with PSU meter), load pulls whatever is available. It's basically just turning its transistors to full on. Would probably go up in flames if I tried current mode with a battery or server supply connected.

Which do you mean has a multi-output transformer, the load or the supply? The supply certainly does, but prone to damage? I've got 2 6050 and 2 6150 supplies, no transformer troubles.

[The Doktor]

Forgot to mention, on a couple of occasions it did show a current reading, and power as well. But only very briefly, less than a second. And not very often. The last 15 minutes or so that it was on it didn't even show current for a split second.

[Armadillo]

If you open up the Maynuo M9712b , you will see a multi-output transformer inside.

I don't understand what is "Current flowing is whatever the power supply is set to deliver"

In constant current mode, the DC load bank should be controlling the current, not your power supply.

Set maynuo CC mode to 1A

Set your power supply voltage to 12 V, current limit to 5A [let maynuo control the current].

I think maybe your power supply current limit contention with maynuo. Otherwise use a battery to test the maynuo.

[The Doktor]

Yes, I understand the load bank should be controlling the current, but it's not.

Supply set to 7V, 5A. Load set to 1A. Current = ~5A, voltage = ~1.25, current reading on load = 0.0000.
Supply set to 7V, 5A. Load set to 0.1A. Current = ~5A, voltage = ~1.25, current reading on load = 0.0000.
Supply set to 7V, 5A. Load set to 0.01A. Current = ~5A, voltage = ~1.25, current reading on load = 0.0000.
I could try 0.0001A, I'm sure it will do the same thing. The load is not aware it's drawing current, and therefor not able to regulate it. I don't believe there's any fuse, pretty sure using a battery will get me 8 burnt FETs.
You can't switch off the current limiting of the supply, but it does nothing until you try to exceed it's set point.

The supply can do 0-7V @ 5A, 0-15V @ 3A, 0-25V @ 2A, 0-60V @ 1A. I've tested a ~$10 10A 60V 60W chinese load with it before, understand how it works.

[2N3055]

Well, it has an internal short.
It might be reverse voltage protection is shorted, or one (or few) power transistors are shorted, or something in regulation circuit keeps mosfets constantly open, instead of regulating..

Since it measures no current, it might be something with current reading circuit (which would keep regulation trying to increase current to regulation point, and in effect keeping mosfet wide open) or there is a short close to output terminals, shunting all current through it...

It is broken, you have a choice of return or repair..



 

[Armadillo]

It would be hard to get a circuit schematic but certainly worth a try.

Gerry did a circuit on M9711, maybe it will be a good starter to troubleshoot.
http://gerrysweeney.com/maynuo-m9711-dc-electronic-load-unboxing-teardown-and-quick-test/

I think you are right, the op-07 and ADC around the current feedback network. It puzzles how the mosfets are all turned on?

But do check the multi-output transformer voltages.

Well, hope it helps.

[The Doktor]

 2N3055, definitely not a shorted mosfet or reverse current diode, it works perfect in constant voltage mode. Just doesn't display amps or watts.

 Armadillo, thanks for the link, although I already got there earlier. Spent a few hours searching this forum, then a few more searching the internet before posting. Perhaps I should try contacting Gerry. It would be very nice to know which op-07 sends the signal to the ADC (if he knows). If his diagram is correct, that op amp could easily be the entire problem, which is what I suspect. I think the only difference between a M9711 and M9712 is the 9712 has double the mosfets & heatsinks (and firmware). And the B version has higher voltage mosfets.

The mosfets are turned completely on because it's trying to adjust the measured current to the set current, but the measured current is 0. I'm 100% sure of that. I'm assuming there is a separate amp for the rear BNC output.

I suppose I could set up an external circuit to pulse the current on it's input on & off at a low frequency, then very carefully probe the inputs until I see the waveform, then check the output. I think maybe a variac, filament transformer, and diode would do the trick. Hook the scope up to the BNC, transformer and diode to the input terminals, and turn the variac up until the peak voltage on the BNC is close to 10 (to get the biggest output from the current measuring shunts). The try to find the distinctive half wave   rectified signal at op amp inputs. If I find the signal at the input and it's flatline on output, I've won Just talking to you guys gets me thinking a bit better.


Ed

[Armadillo]

Do check the multi-output transformer, mine was burnt.

[2N3055]

That's good.. So it is current regulation circuit.. It's not current shunt, so shunt voltage amplifier is the culprit.. Find current shunt and follow it to the op amp from there..
Of course, check power supply first if you suspect it...

[The Doktor]

I will, that's a very easy thing to check, much nicer than probing the op amps.


Ed

[The Doktor]

All fixed. Maynuo sent me a partial schematic showing the area I needed. Obvious problem was burnt U36. Nope. My power supply was giving about 5A. U36+ was ~.085V, - was 0V, out was ~11.6V. U36 was always about 11.6V when load was switched on, even with no input current. It was negative when switched off.

I marked how R144 & R147 are connected. Both ends of R144 & R147 were ~0V. I probed the right end of R144 with the meter on ohms, load displayed ~1.5A. I set CC to 1A, now if I probed with ohmmeter, load would quit drawing current.

Clipped a jumper from U36 out to right side of R144, current was displayed and regulated

Applied flux & tried to reflow U36 output pin (several times), no luck. Took picture to post, figured on adding a bodge with to replace the jumper.

When I looked at the picture, I saw the trace going to the via by the relay. Went back & probed the via. ~11.6V on the via, 0V on trace. Added flux & applied iron, all working  Added more flux & repeated, just to be sure. Readings now way off due to water based flux, cleaned with isopropyl & dried, all good now, even in spec

I'm a little paranoid about that connection failing at a bad time, may add a bodge wire from the via to the resistor. Otherwise all is good.

[Armadillo]

At the cost of a bit of solder. 

[The Doktor]

Sorry, was adding story as you typed, but you figured it out anyway. Posted first to see if pictures worked.

What do you think about adding the bodge wire, yay or nay?

[Armadillo]

Nay, marked the triumphant over circuit with no visible wire. 

[CJay]

so bad connection from via to track?

I think I'd be more confident in the reliability if I'd gone around some of the other vias on the board, maybe removing the solder from them and threading a short piece of thin wire through to be soldered to the surrounding pads.