Zapino scooter with blown controller — my first EV!

[slam]

Hey folks!

I have blown MOSFETs. Should I attempt to rebuilt the controller, replace it, or upgrade it?

I picked up a 2008 Zapino scooter with known problems. This is my first EV and a bit of a learning curve coming from TDI VWs! My first time on this forum as well.

The scooter reportedly was stuck in "limp mode" going no faster than 10 MPH. Previous owner replaced the throttle (no effect) and the controller (apparently it wasn't the right replacement) before garaging the unit for a year. When I got it I hooked the controller up and... drove it home! The batteries were low but I was able to cart my 200 lb self about 2.5 miles at 15-25 MPH (depending on hills).

That night I have it a full charge. The charger LED never turned green (possible dead cells — tbd). The dash voltmeter was in the green so I took it for a spin. I got about 1 block before it started to stutter badly. I limped it home, sometimes stuttering sometimes not, then ran it on the center stand to evaluate the issue. Something started to smell like badly burning electronics and I shut it down.

I disassembled the controller and what a mess. At least five blown MOSFETs and one blown capacitor. Apparently the previous owner had opened up the controller to examine it and put it back together without fresh heat conducting paste and I suspect overheating caused the MOSFETs to blow, but I need your advice.



Here's the whole controller. From what I can gather this is a six-phase configuration? I have six 12 Ga/2.5mm2 wires to the motor not three as the manual wiring diagrams says. There's ten 22 Ga/0.3mm2 wires via a white connector from the controller to the motor (Hall sensors, and?). There's one 4-wire, one 3-wire, and one 2-wire connector (throttle, key set/braking switch, and LED diagnostic, but I don't know which is which), a pair of white wires (for the "high speed switch" I think), and the battery wires.



Here's the blown MOSFETs. The first four are blown and the seventh one has a fried connector.



Here's the blown capacitor on the flipside of the first four blown MOSFETs.

Is there any point to sourcing replacement MOSFETs? I'm handy enough with an iron to take it on, but I can't tell for sure why they blew or if there's more damage elsewhere on the board, or even damage to the motor.

Another issue I have is the bad wiring diagram. It's next to useless since it doesn't resemble my actual wiring:


I do have a slightly better wiring diagram in the printed manual.

Long-term I'm probably want to upgrade the SLA batteries with Li-Ion batteries with similar or greater total voltage. Short term I'd just like to get this rolling again.

Advice?

 

[slam]

Thanks eTrike! If I can't get it rolling again I'd be happy to sell it to you for what I paid. =)

My fear with replacing the MOSFETs is that I don't know what else was damage. I see one blown cap, but I could be wasting my time with this controller.

Replacing the controller might be a safer bet... but there's so many unknown inputs here. Perhaps I can do without the LED diagnostic output and the break switch power interrupt, but will another controller contain a thermal cutoff? I know nothing of controllers.

 

[Voltron]

If you happened to notice, what voltage are the capacitors rated at?

 

[slam]

Do you mean the MOSFETs? The MOSFETs are labeled IRFB4310. Spec sheet says 100V, 7 mOhms, 140 A.

Mouser has them for about $3.34 each. Not sure if I need to replace only the damaged ones, all 12 on one of two controller sides, or all 24 on the board.

The blown capacitor is tiny and totally destroyed. I'll have to guess the capacitance by looking at similarly-placed capacitors elsewhere on the board, but I see some of them have different ratings.

 

[slam]

If I replace this controller with two three-phase controllers, what wattage do I need?

It's a six phase 3000W motor; do I need two 1500W three-phase controllers, or two 3000W three-phase controllers?

 

[whereswally606]

my guess looking at the controller is that there are two rows of 12fets so 24fets total in the old controller. if you were to replace with two controllers you need 12 fets in each of similar ratings to the fets that are on the old board or better. hope that logic helps

 

[slam]

Thanks Wally. But can I get away with replacing only the damaged fets if I use exact replacements?

 

[Voltron]

Yes... Was referring to the main capacitors, as I have the same set up and thinking about upping the voltage a little, but guess I should just stop being lazy and open mine. I'm guessing replacing the FETS individually making hard to perfectly match them, even with supposedly exact replacements.

 

[slam]

I'm thinking of upping the voltage a little too, if/when I switch to Li-Ion batteries.

I wonder if the MOSFETs take power right from the batteries of if the voltage is stepped up somehow. I suppose I could hook the controller up to the battery and measure voltage at the fets.

 

[slam]

Stock battery config is 60V, yes.

The previous owner didn't detach the MOSFETs from the heatsink bar, but the bar itself is bolted to the controller case. That's where the paste was thin because the controller had been disassembled and reassembled without new paste. I think there was insufficient heat transfer to the case on just this one side; the other side looks fine. All blown MOSFETs are on one side and all phases on that side have at least one blown MOSFET.

But if not heat... then what else would have caused these to blow all at once?

Here's a photo of the MOSFETs, enhanced as best I could. They were all very faded.

 

[slam]

Each phase has one of these capacitors. They are all marked A475T U41KV.

All four fets (one whole phase) associated with that blown capacitor are toasted.

There's one more blown fet on the very next phase, the capacitor for that phase appears ok.

Blown cap: View attachment 1

Good cap on opposite controller:

 

[Dauntless]

All I've heard about Zapino for a decade is bad: Blown controller, burnt up, fixed it and it went bad again,etc. Why would you bother fixing that controller?

The thing to discuss is what would make an improvement? You could stay 60v or go 72v. What are the plugs on this thing, will the most common off the shelf plug in? So who has controller suggestions?

https://www.aliexpress.com/item/Fast-Shipping-3000W-72V-DC-36-mofset-brushless-motor-controller-E-bike-electric-bicycle-speed-control/32679609811.html

https://www.aliexpress.com/item/60V-72V-96V-120V-6000W-high-voltage-resistance-electric-bike-controller-controller-with-6pin-CA-plug/32607733533.html

 

[slam]

I supposed the main reason to repair would be cost. It's about $60 in parts to repair, plus my free labor.

Since there's six phases I'd need two controllers for a total of $300 (2x 1500W) to $600 (2x 6000W), unless I can find one 6-phase controller somewhere.

Plugs don't worry me; I can replace the plugs if I can figure out which is which. The question would be whether the inputs are the same — will I need to adjust for HALLs, throttle, ignition, etc.

 

[whereswally606]

if you are going to attempt repair i've read and seen youtube videos which use uv (blacklight) to see damage you cant see with the human eye. I would seriously think about taking some UV photos of the pcb before you just assume the only damage there is is what you can see.

 

[slam]

Good tip. I'm assuming there's more damage than I can see. There's a few traces that look smoked but show no visible damage.

 

[Voltron]

Totally guessing, but wondering if the damage might have been caused by some heavy load, low rpm riding that tripped the circuit breaker... Would that send a big voltage spike back thru the controller?

 

[slam]

I'm 215 lbs and The day before I road 2 miles up and down hills with a low charge.

After a full charge I barely got a block away, mostly downhill, at about 25 mph. Doesn't seem like a heavy load. It started to sputter like it was cutting out; this may have been the overheat protection kicking it. When I kept pushing it it got worse. Breaker never tripped on either run.

When I tested it at home on the center stand with zero load I smelled the electronics cooking. Breaker started to trip at this point.

So it's still a bit of a mystery as to what causes it to toast.

For all I know it's the motor.

 

[Voltron]

Hey again... Hope you're still rolling with the project. I just got a second one of those scooters with the Hubmonster. This one had been modded, so thought I'd share it's totally different two controller set up.



View attachment 1

Seeing the two controllers stuffed in there made me think, you could prob low power test your motor with two really cheap generic controllers, like $20 on eBay cheap... Just to see if it spins up.... I'm all enthusiastic about seeing every one of these motors run... Lol

Have you read up on shorting the phases and turning the motor by hand to see if it starts cogging? Could be a good start before hooking power up to it again.



The insides, if you haven't seen one..

 

[Voltron]

Well... It was actually $150

But it was a screaming deal..I think they inherited it somehow, and wanted it gone before it caught on fire! I just finished ripping out all the dead batteries and jumpering across to the other one for test power and the motor spun right up.

 

[slam]

I did see the cog test and I hope to run that soon. Winter is coming and I'm less likely to wrench outdoors in the rain. =)

Can I test this motor with one controller? Any reason it can't work on just three of the six phase as a test?

 

[slam]

Well, dang. Technically speaking I actually have half of a six phase controller. All the blown MOSFETs are on one side. If I desolder the power to the blown side maybe that would work. =)