Repair: Blown SMD identification problem - Lipo charger/motor driver pcb

[tzaphkiel]

Good morning,

I find myself repairing a shaver with what was to be just a battery replacement. I accidentally inserted the LIPO in reverse polarity and what I suspect to be a protection diode or a cap near the motor blew.
I put here two annotated images: 1) of the whole board, 2) of the component that blew.

It is an SMD component that to me does not look like a diode, it looks more like a cap...
The circuit near the motor looks simple enough: a P-Channel MOSFET, one cap (orange to the right of the MOSFET), a blown component and some control electronics for the charging, switching speed of the motor, etc...

At the moment, I've tried different a diode in reverse bias with my lab supply to power the board with 3.7/4V 1A C.C. (to be safe). The red light lights up supposedly saying the battery is not charged.

I tested the motor separately and it still works. When I flip the switch, the motor twitches and then the red led comes on (batt low or something...).
I believe that even if the blown comp was a diode, the motor should still start turning...
If it is a cap on the other hand, maybe it is to give the initial impulse to the motor to make it turn but then with what cap value should I replace it with

Btw, the SMD component had no other markings than what I think is the brand: EDLEN .

Could someone please help me to:
- identify the blown component (SMD)
- possibly point me out to another problem as to why it does not work.
- as I'm a beginner, some pointers to further analyze what is wrong.

Thanks
Tzaphkiel

[amyk]

EDLEN? That appears to be a supercapacitor... consistent with it containing liquid (the electrolyte is corrosive, you should clean it off the board as much as possible) as you claim. Diodes do not contain any liquid.

[tzaphkiel]

EDLEN? That appears to be a supercapacitor... consistent with it containing liquid (the electrolyte is corrosive, you should clean it off the board as much as possible) as you claim. Diodes do not contain any liquid.

Thank you, that would make sense... I did clean the board straight away with a short brush and alcohol!

If it is a supercapacitor, why than have two cap in the circuit, one between MOSFET-src/-ve and supercap between MOSFET-drain/-ve
How could I identify the capacity? Would a simple cap do ?
So if it is a capacitor, is it to start the motor spinning ?

[metalphreak]

That looks more like a diode than a capacitor...

It's across the motor terminals to stop any back EMF from the motor feeding back. Reversing the battery polarity probably let a massive amount of current go through that diode and back through the reverse biased junction in the mosfet. You might find the mosfet is toast as well.

[tzaphkiel]

That looks more like a diode than a capacitor...

It's across the motor terminals to stop any back EMF from the motor feeding back. Reversing the battery polarity probably let a massive amount of current go through that diode and back through the reverse biased junction in the mosfet. You might find the mosfet is toast as well.

Yup, well that's what I thought until I saw liquid (electrolyte) coming out of it and after trying with a regular diode in reverse bias... So I'm not too sure about it being a diode.
On the other hand, whether it is a diode or a cap, having had a large current in reverse polarity indeed probably fried the mosfet...
Is there any way to test it ?

[metalphreak]

I just cracked open a similar looking diode (GS1G Vishay) and it has the same silicon die looking junction as your close up photo. I very much doubt its a capacitor as it is *after* the mosfet where it really isn't doing much help. DC (and other type) motors will almost always have a reverse biased diode across the terminals. If that's not the diode, there will be one somewhere else.

To test the MOSFET, you just need a voltage on the supply side there, and ground to the motor. Then apply 3-5V or so to the mosfet gate (if its N type) and see if the motor runs. If it conducts power to the motor it's ok. They usually have half decent diodes in them for peak current surges so I would not be surprised if the other diode failed first protecting it.


You'd be amazed at the goo and liquid that comes out of plastic when its exposed to high heat!

[tzaphkiel]

Thanks for the information.
I also already tried using a normal through-hole diode (as a test) and nothing worked.
Can I actually use any odd diode in reverse bias (N4148 or similar)?

The mosfet is a p-channel one, when I power the device, there is a +ve voltage at the gate; but than the circuitry determines that something is wrong or that the battery is not charged and the "red light" goes on... I suppose thus maintaining +ve voltage on the gate...

So I should probably need to ground the gate to see if it fires-up the motor... right ?
Can I just short it to -ve pole?

[amyk]

You'd be amazed at the goo and liquid that comes out of plastic when its exposed to high heat!

Having had experiences with catastrophic failure (and seen enough Photonicinduction videos ) any molten bits tend to become solid after cooling.

The package marking and presence of liquid suggests it's a supercap, the circuit and remains suggest it's a diode...

[tzaphkiel]

The package marking and presence of liquid suggests it's a supercap, the circuit and remains suggest it's a diode...

Let's say that the remains (molten or inner liquid) did not re-solidify afterwards... maybe they kind of gelled up but that's it.

... That kinda... doesn't help 

So to resume:
- mosfet: p-channel (base high when battery inserted & diode red: "batt not charged" - although I use now a lab supply; motor does not run...)
- burned comp:
  - from remains & brand & liquid: supercap from EDLEN
  - from common sense & circuit: diode...
    Strong feelings for the first of these two...
- from looking at the board, the comp sits between the terminals of the motor
- there is already a cap before mosfet between src&-ve

Hum...

[ConnorGames]

The +VE the gate could be caused by the MOSFET shorting G-S. They tend to fail with at least 2 terminals shorted, and failure is likely as the full current of the LiPo likely flowed through the body diode. So check for shorted pins on the MOSFET.

[tzaphkiel]

Yes especially with the Lipo, High current is probable...
I've just tested for shorts across the mosfet.
There does not seem to be any current flowing between the S & D... but between S & G there is some current flow (.9 ?scale?A) when on diode continuity testing on the meter.

I then shorted the mosfet's gate and the -VE supply, thus bringing the gate to zero Volts (p-channel mosfet) allowing flow from src to drain.
The motor started... So... either the mosfet is fried or some other control circuit upstream is not working anymore.

Thoughts ?

[Zapro]

One thing: Your won't find a supercap in a product like this - There is absolutely no need for a supercap.

That looks like a diode that blew up.

I'm repairing electronics that has blown up, on a daily basis - And plastic that has been subject to melting semiconductors quite often gets made into a liquid mess, and it doesn't re-solidify.

Also, the amount of liquid in a supercap (or any other modern electrolytic) isn't enough to even make the PCB wet. It's a very small amount of electrolyte that's in them)

// Per.

[tzaphkiel]

Thanks for the information...
The thing is that I've taken out the board and checked (visually first) both sides for burned-up components and this is the only one that is gone.

The problem is that there is probably :
1. a battery checking circuit
2. a battery charging circuit
3. a motor speed controller

The 3rd does not give the 'go' signal because I believe the 1st is somehow not saying it is ok... I've replaced the battery with a brand new one and I would thus think that is not the problem.

Concerning the diode, I've tried various diodes I've got near my bench and whether in forward or reverse bias, the behavior is always the same:
When I turn the switch on, the motor gets a very quick 'shock'/impulse than goes off almost immediately than the control light goes red...
On the other hand, when I place a cap there, a 1uF, the behavior does not change, when I put a 1000uF, there is obviously less of a shock on the motor but it still does not start.

It seems it wants to start but does not 'stay on' ...

I'm at a loss

[amyk]

Trace the schematic (and post it here). It'll give us and you a better understanding of how everything works, then we can suggest more signals to probe.

[tzaphkiel]

Will do, but it'll take me some time, it is a double sided board and fine traces that go under chips... Thx