16s has died from a moments negligence.

[OldFartRides]

So I set it down while wheel was spinning. “ WARNING, Overload” says the wheel. So I turned it off.

 When I turned it back on, LEDs 1 and 6 are lit. Ewheels website says that’s a corrupted serial number, but when I connected with the app, I got a 205 - “Drive circuit not functioning. Restart, or replace drive circuit.

Arrrgh.

 

edit: went back on with the app. Diagnostics says (under Motor Phase Line)  Motherboard phase output short circuit.

 

Edited April 23, 2021 by OldFartRides

[RockyTop]

So we have a motor power short. That could be the board (motor output) or the motor wires or the motor. 

I would check the wires first. Then you would disconnect the motor wires and ohm out the motor. Looking for grounds or dead short between the wires. I am not sure what the ohms should be. 
 

The motors that I work with are good at .8 - 1.2. ohms bad at <.6  ( 600 volt) example only. 

I would think the ohms on these motors would be much much higher. 

[OldFartRides]

Took her apart. Found a  fried place on the board just to the right of the yellow cable. There are supposed to be twelve fine conductors (?) just like on the other side of the black block. I don’t know how I would fix that. 
 Seems a shame to have to get a whole new board. 

[OldFartRides]

Meh, now I’m not so sure. I picked the crust off. Need better magnification.

[.....]

I see a burned chip. def gna need a board. Sorry to see it man. Be sure to check what you can on down the line, as its not common a chip goes up like that, unless a short got to it. +1 on the advice to check the motor. A spike in power could be the culprit. Was it going really fast and you set it down pretty hard? Merely guesses, as I have zero idea how these board handle such power changes and demands.  The 'crust' you removed, was bits of solder and the legs of that chip. Proximity to the main power wire (yellow) may be a culprit during higher temp/amp demands.

Edited April 24, 2021 by ShanesPlanet

[Tawpie]

yeah, it's at least a new board for sure. See in the gap where the pins of the rectangular part are missing? There's supposed to be copper there, and you've got bare fiberglass/epoxy. Whatever happened vaporized the part's leads and the printed wiring part of the printed wiring board... not worth repairing especially since something else on the board may have died when that went zot (cascading failure).

[VolcanoSurfer]

If it were me I’d At least do a search for that chip that burned up in the picture, see if anyone has reported the same chip going out that might have done their own research, and at least kept the board around somewhere for a time I have a hot air solder tool and the time to change the part out.  Might not work, but often times those chips alone can be found for less than a few dollars or dimes.  
But that’s just what I would do.  
Good luck 

[OldFartRides]

Yeah, the heat has made the numbers illegible. Thanks guys !

Live and learn....or not.

 

edit update: Ordered the board. Could only find one seller that had it. 
maybe I’ll get that chip number off the new board and see about repairing this one.

Edited April 24, 2021 by OldFartRides

[RockyTop]

That is a shame. The resistor would be easy to fix. The chip... Not so much. 

[esaj]

I have my board on the table right now, as I was resoldering the hall-effect sensor wires... I had some intermittent "power loss" -moments last fall right before the season ended, likely due to losing the hall-sensor signals for a split second (at least other KS16S owners have reported issues with the board connector / wire crimpings getting loose with vibrations over time). Could be the connectors on the wires or the board connector, so I just removed the entire board-side connector and soldered the wires directly to it. I can check the chip model but mine's the "old" KS16S (2017 v1). My best guess is that the chip's a current sensor.

EDIT: The burned chip appears to be an ACS709T, "Hall effect-based linear current sensor with overcurrent fault output, which offers a low-resistance (~1.1 mΩ) current path and electrical isolation up to 2.1 kV RMS. The sensor has optimized accuracy for currents from -37.5 A to 37.5 A, and the analog voltage output is linear for current magnitudes up to 75 A. The ratiometric output voltage is centered at VCC/2 and has a typical error of ±2%. It operates from 3 V to 5.5 V, so it can interface directly to both 3.3 V and 5 V systems."

If you're in luck, just the power-side (circuit connected to left-hand side pins) is gone and it hasn't fried anything connected to the digital logic etc... the current measurement is also based on Hall-effect, so it's isolated from the rest of the circuitry. Worth the try if you have the means to exchange it, the chip itself is likely a couple of bucks. The chip resistor above it (R001) might be damaged too, looks like a 1 milliohm, better replace that also.

 

Edited April 24, 2021 by esaj

[OldFartRides]

Wow, thanks for that ! Having trouble finding one in stock in N.A., but definitely worth a shot. A repaired functioning board should be worth at least half the price of a new one. Or just “keep iton the shelf “ for backup.

Best,

edit: it is the same chip. With your pic to go on, I can just make it out on mine. Thanks again.

interestingly, there’s another 709T over by the green wire. 

Edited April 24, 2021 by OldFartRides

[Chriull]

1 hour ago, esaj said:

The chip resistor above it (R001) might be damaged too, looks like a 1 milliohm, better replace that also.

Seems like they put this resiator in parallel to double the current range?

So more than 75A blowing this ic up mean above 150A through 2mosfets (or 4 if the ks16s has mosfets paralleled?). Could have affected them a bit too as this is already above package limit if not paralleled or very close otherwise...

[esaj]

1 hour ago, Chriull said:

Seems like they put this resiator in parallel to double the current range?

So more than 75A blowing this ic up mean above 150A through 2mosfets (or 4 if the ks16s has mosfets paralleled?). Could have affected them a bit too as this is already above package limit if not paralleled or very close otherwise...

Possibly one of the mosfets or both are gone too. The current spike has probably been in reverse, if the motor has been spinning fast and then stopped very abruptly, so it might have also passed through the mosfet body diodes which can't take as much as the main channel. No paralleled mosfets in KS16S, they're all TO-247's instead of TO-220's, so they can take quite a beating, but maybe a good idea to take a peek at the fets themselves, if there's clear signs of damage, they'd have to be replaced too... if the damage extends further, then it's probably not worth it trying to save the board for other reasons than plain curiosity (the big fets are more expensive, possibly something like $10+ a piece). I think the exact model the KS16S uses been posted in the forums before.

Edited April 24, 2021 by esaj

[OldFartRides]

I don’t see any other obvious signs of damage. Since that 709T is cheap, I’m going to order some from Taiwan. Ten of them are like twenty bucks shipped. Seems to me someone in this thread was offering (lookin at you, V. Surfer !) to take a stab at it....

[OldFartRides]

Finally got the chip(s). One seller failed to deliver. Second seller did, but I had to buy ten of em.

I really doubt I could manage to surface solder something that tiny on successfully, even with practice.

thing looks like a tiny bug with twenty legs.

The soldering wizards on YouTube make it look easy. I guess I’ll try to find someone who can do that stuff.