5
$\begingroup$

I was attempting to install new TMC2208 chips on my MKS Gen L v1.0. Everything was going swell, I just had one more thing to do.

I was checking and tuning the Vref for TMC2208 by measuring voltage across the GND pin and the through-hole shown in this pic (from Stepper drivers - TMC2208):

Checking current motor setting

However during very last measurement on the X-axis stepper I slipped and the probe on my meter shorted the GND and VM pins on the TMC2208!! Pic below:

Illustration of shorted pins

Immediately my board turned off, LCD went out, then I unplugged the printer as fast as I could, but the damage was done. My board now does not power on completely when connected only to the PSU.

The LED D7 will flicker on and off, as will the LCD screen. However, if I connect the board over USB I am able to connect and interact with the board just fine.

I have meticulously checked the board for blown components but can't see any damage. The two fuses near the 12 V power on the board are not blown.

Is anyone able to offer some insight on this? Does it sound like a salvageable board?

Comment responses:

@Oscar: Thanks and yes I've (attempted) to power the board without anything plugged-in/attached to see if it works, but no dice.

@Tom: There is a lot packed into your comment that required searching for info to even respond. The 5 V voltage regulator on my MKS Gen L v1.0 board looks to be an MP1584en chip. Seen in pic below from MP1584_r1.0.pdf:

MP1584en pin out

When I connect the board via USB and test the Vin pin, I see a steady ~4.2 V . When I remove USB and connect straight to the power supply I see a unsteady Vin between 6-7 V .

So I would like to trace from this MP1584EN back to my 12 V power source on the board. However, after searching it appears that the schematics for the MKS Gen L v1.0 are not available openly. Unless there is an avenue to find the schematic which I have overlooked. Anyone that can link me to them would be much appreciated.

@R: OP's level is a green-horn baby kitten at its very first rodeo. However your suggestion appeals to me because it sounds like a quick easy soldering fix. To better understand, do you mean to take my 12 V from the power source then connect it to different regulator (usb car charger) and then on the secondary side of the car charger, connect it back into my board?

If you're willing to give a bit more of a description this is something I'd definitely be willing to try.

Thanks for all the responses.

Improve this question
$\endgroup$
4
  • 5
    $\begingroup$ 12 V power is not getting to the 5 V regulator. Try tracing out the board schematic, tracking back from the regulator to the power input. Somewhere along there, something is blown. $\endgroup$
    – Tom van der Zanden
    Commented Jul 20, 2019 at 6:58
  • 1
    $\begingroup$ It might be possible to just bypass the regulator on the board (disconnect it) and add an extra DC-DC converter (e.g. USB car charger) to power the logic even without much electronics experience. Not sure what OP's level is. $\endgroup$
    – R.. GitHub STOP HELPING ICE
    Commented Jul 20, 2019 at 13:55
  • 1
    $\begingroup$ Did you ever get to the bottom of the issue? I know it has been a while, but if you have a solution, it could be useful for someone else... after all a simple slip of the probe is an easy (and surprisingly common) mistake to make. $\endgroup$
    – Greenonline ♦
    Commented Aug 8, 2021 at 8:00
  • $\begingroup$ The OP said he still had the same issue with nothing "plugged-in/attached" however I still question if that included removal of all TMC2208 modules. An answer might want to explicitly include a note of this. I think that this is important as it's easily likely a module was damaged in a event like this and is dragging down the 5 V rail. Also if true would be one of the easiest ways to fix a issue like this. Checking the resistance of 5 V(unpowered board) rail to ground might also assist in troubleshooting. $\endgroup$
    – Kezat
    Commented Aug 8, 2021 at 20:13

1 Answer 1

Reset to default
1
$\begingroup$

However, after searching it appears that the schematics for the MKS Gen L v1.0 are not available openly.

Here is the schematic.

Anyway, the problem is the MP1584 regular is toast. There is no reverse biased protection diode going from output to input (arrow pointing towards input) in the schematic.

Most buck or linear voltage regulators aren't meant to be operated with Vout being higher than Vin, even briefly. There are internal parasitic diode structures inside that will get turned on by voltage in the wrong direction (higher Vout than Vin). Once turned on, the output capacitor happily dumps its charge through these diode structures and out to Vin.

This kills the regulator.

What results is typically an internal short of some sort in the chip.

So when you shorted Vm, that was also shorting the input to the 5V regulator and the input capacitor there, so suddenly there was 0V at Vin and ~5V at Vout. Poor chip never stood a chance.

What results is typically an internal short of some sort in the chip.

This is not your fault, this is poor design (or a reckless cost saving decision), as accidentally shorting Vm is very easy to do on accident. A suitable diode would have cost ~1.3¢, not even two whole pennies and would have made accidental shorting of Vm be harmless. There is no valid reason not to include such a protection diode on the board.

Regardless, your only options are to either desolder the MP1584 and solder a new one onto the board, or, since you can talk to the board via USB (which provides its own 5V power to the board), desolder the MSP1584 (since it is shorting the input power to some degree) and simply keeping a USB charger of some sort plugged into the USB port (I wouldn't leave it connected to your computer since it doesn't have any USB isolation) to provide the 5V power rail instead.

The main problem is that bad chip. You gotta get it off the board before anything will be able to work unfortunately. Desoldering is relatively easy, especially if you don't care about the chip you're removing. Just be sure there are no shorts between the pads after you're done.

Improve this answer
$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .