Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[BrianTRice@gmail.com]

I appreciate the post, but I would like some claims that are easier for me to cite. So, I'm going to clean up and start over.

Your conclusion sounds valid but insufficiently explained. I need more than just "this looks like that, and it has ICs and coils, therefore".

I'll try to explain what I see, in order to establish a common frame of reference.


I'm assuming the board is laid out physically to maximize distance between heat-producing components, and that the routing roughly matches the component interaction/flow.


On the "topside" photo, I'll identify areas of components:
- On the AC input side (red/blue, rear left of the charger), there is a coil and then some Carli components that are likely power capacitors I can't find a photo match for.
- On the DC output side (red/black rear right of the charger), there is a capacitor bank.
- Under the heatsinks are presumably the core switching power supplies. I've uncovered them once and will try to do this more thoroughly to identify them.
- The 3 large capacitors in front are AiSHi LM series rated 450V (DC~) 330µF (datasheet at https://aishi.us/files/LM.pdf)

So what more do you want to know?

I want to work out what each area of components performs, and what risks the board design is susceptible to, given the location of this fire within the subsystems.


If there's anything an owner can do to mitigate design weaknesses in the board, that would be helpful, and also if there's basically nothing the owner can do other than use contact cleaner on the inlet and grease around connections.

But it is difficult to tell what failed first by those photos, because many things fried at once, which is the norm for switching power supply failures.  When one part shorts out, it usually  takes at least a half dozen other parts with it. The most common failures are shorted diodes and the next most common are electrolytic capacitors shorting out. But it's usually  difficult to even tell what blew first  between those two, because a shorted capacitor can blow a diode from the overload and a shorted diode can short out an electrolytic  capacitor from the excessive ripple.

That's the reason why the teardown doesn't label anything, because I can't triage that with as much damage as there is in the center.


For what it's worth, a teardown on this charger type is extremely rare. This might be our one chance to understand this problem and try to form a case to get Zero to obsolete this design in favor of something more robust (especially since this basic design was developed in 2013 for 2014+ models).

The third most common failure would be the high power transistors (underside photo, the two items with three leads each under the small heat sink, but I see nothing burnt in that area on yours)

Which area is that? I need specific labels somehow. Try taking one of the images and marking it up with colored rectangles and/or text. I can retrofit labels onto close-ups that I'm preparing for upload now.

[DonTom]

f there's anything an owner can do to mitigate design weaknesses in the board, that would be helpful, and also if there's basically nothing the owner can do other than use contact cleaner on the inlet and grease around connections.

I would say no, other than to use the OBC less. Use Delta Q-chargers more often, even if the OBC is also being used at the same time. That reduces the time the OBC is  on. Or better yet, only use external chargers when possible. Save the OBC use for when there is no other way to charge.

There could be a few design changes in the OBC that will increase reliability. For an example, if a capacitor is rated at 100  "working volts DC" and the same capacitor has 80 VDC on it, but still commonly fails, they could put a 200 VDC "working voltage" capacitor in there instead and they just increased the reliability. But a 200 VDC capacitor will be larger than a 100 VDC capacitor at the same MFD value.  If they do this with many parts, you now have a larger charger, so now a different design  problem. As well as add a little more cost (but most of the parts in a switching power supply are very cheap anyway).

The Delta Q charger has more room to be more reliable. But it is still a switching power supply as are all EV chargers (AFAIK).

Capacitors don't get hot if they are good, so no issues with the spacing from such. But the higher wattage resistors and power transistors can get very hot, and the spacing of such  can be important, especially if not cooled by some other means.

Burnt PC boards in switching power supplies is very common, as their design is such if  one part shorts out from a failure,  it causes a chain reaction that cause many other parts to fail and perhaps also short out and cause even more parts to fail.  Unlike linear supplies, which are heavy and usually have very few parts, the switching power supplies have many components but they are very small and light, The more components the more likely for something to fail, especially when in a smaller package.

-Don-  Reno, NV
 

[BrianTRice@gmail.com]

I think I just need to find an EE I can trust (no ego on the line) to pay for a proper lookover. I'll see what I can do with your input, but you're not filling in any blanks.

[DonTom]

Your conclusion sounds valid but insufficiently explained. I need more than just "this looks like that, and it has ICs and coils, therefore".

To me, it is as obvious as can be, and would also be with anybody else who knows about switching power supplies. It would take a month to explain it all, and I am not electronics engineer, I only know the basics, but that is more than enough to ID it as a switching power supply.

BTW, let me correct a small error on what I said.  IC's are NOT required for  a switching power supply, the older ones use no ICs just a  lot of transistors. But an IC is made up of many very small transistors.  It would have been more accurate if I said the many semi-conductors and small light transformers prove it to be a switched-mode power supply.

"Excessive sparking at the inlet" before your failure is a clue. It was drawing excessive current. I will take a guess at what happened to yours that caused the burn. But it gets to the "chicken or the egg" type of deal.

One thing that could have happened is a diode shorted. In most cases, that will make the supply not usable  right there, but that depends on  where in the circuit. The shorted diode can put AC on a capacitor until it decides to short out, and when it does, other components fail.

Or a  capacitor develops a high resistance short inside first  and starts getting warm (a good capacitor will NEVER get warm internally) and eventually shorts out, causes a diode to short, puts AC on many items and many more parts blow out.

Like I said, it's the norm for many parts to short  out and burn a PC board in a switching power supply when any one part shorts out. It is usually a chain reaction.

-Don-  Reno, NV

[BrianTRice@gmail.com]

I don't care what you (say that you) think or know. I care what you can communicate in a way that I can cite.

You make claims. Claims without an explicit train of reasoning must be treated as such. What I'd like is something I can confirm via an independent source that will hold up under adversarial scrutiny.

[DonTom]

I think I just need to find an EE I can trust (no ego on the line) to pay for a proper lookover. I'll see what I can do with your input, but you're not filling in any blanks.

I am not sure what blanks you need filled, put perhaps a few will be answered in the post I just typed out a few seconds ago.

-Don-

[DonTom]

I don't care what you (say that you) think or know. I care what you can communicate in a way that I can cite.

You make claims. Claims without an explicit train of reasoning must be treated as such. What I'd like is something I can confirm via an independent source that will hold up under adversarial scrutiny.

Good luck with all of that!  All I ask if if you do find the info. you are looking for that you post it here, so I will at least know what it was.

-Don-

[Crissa]

My rule of thumb (using switching power supplies in other equipment) the less sparking you can have, the longer they live.  Arcing creates signal and low-voltage situations where components are insufficiently charged, and you get weird things like backwards current through diodes and stuff which heats them up and stresses their ionic structure.

Which is why I chose to put an arc-resist switch into my home charger cord; reducing that arc reduces that stress.

Beyond that I don't really know much, since I only ever built a couple in my hs vocational classes and never really understood the math behind them.  I went into programming instead of EE as I went into university. ;

Since we're seeing more failure at 120v vs 240v also gives a clue where the component failed is; it would have to be on the side of the circuit exposed to the higher current.

-Crissa

[DonTom]

Which is why I chose to put an arc-resist switch into my home charger cord; reducing that arc reduces that stress.

There is a delay for the contractor to close, so there really shouldn't be much of a high current spark to begin with.

Nevertheless, I never plug a hot cord into the bike. I make sure it has no juice when I plug into the bike and plug in at the opposite end.

-Don-  Reno, NV

[BrianTRice@gmail.com]

My rule of thumb (using switching power supplies in other equipment) the less sparking you can have, the longer they live.  Arcing creates signal and low-voltage situations where components are insufficiently charged, and you get weird things like backwards current through diodes and stuff which heats them up and stresses their ionic structure.

Which is why I chose to put an arc-resist switch into my home charger cord; reducing that arc reduces that stress.

Beyond that I don't really know much, since I only ever built a couple in my hs vocational classes and never really understood the math behind them.  I went into programming instead of EE as I went into university. ;

Since we're seeing more failure at 120v vs 240v also gives a clue where the component failed is; it would have to be on the side of the circuit exposed to the higher current.

-Crissa

Do you recommend a particular arc-resist switch or know suitable criteria for an average owner to select one?

[Crissa]

There is a delay for the contractor to close...

...But the charger has been exposed to current the entire time.  The contactor protects the battery, not the charger.

Do you recommend a particular arc-resist switch or know suitable criteria for an average owner to select one?

Not really.  They will say arc-reducing or arc-resisting on them.  I used this switch https://smile.amazon.com/dp/B000LETGVY/ref=cm_sw_r_tw_dp_x_YTleFbYPY2AVA

You'll want something that says it has big, coated contacts and quick action.  Faster means less time arcing.  Sometimes arc-resistance just means that it keeps it enclosed, but the less time it spends arcing, the less time there's a weird random resistance on the circuit.

-Crissa

[Richard230]

I was thinking today that maybe Zero should offer a program where they send you a new charger every couple of years for a discounted price like the offers that you see on Amazon when you buy an item and they offer you a subscription service. 

The estimated time for the new charger that I ordered from AF1 Racing is about a month. Once I remove the old charger I will photograph it to compare it with the new charger box and also check it over to see if there are any burn marks on the case. I bet the new charger has an updated model number.

[BrianTRice@gmail.com]

I was thinking today that maybe Zero should offer a program where they send you a new charger every couple of years for a discounted price like the offers that you see on Amazon when you buy an item and they offer you a subscription service. 

The estimated time for the new charger that I ordered from AF1 Racing is about a month.

At $800, pre-ordering it is not really cost-effective, except maybe for a dealer to have exactly one spare on-hand: https://www.af1racing.com/store/Scripts/prodView.asp?idproduct=89280&sku=ZM45-08253

But I'd rather a dealer just take a charger off of a demo bike to give to a customer and use a Quick Charger for the demo bike until Zero HQ delivers a new one. Or maybe use it as a "loaner".

Once I remove the old charger I will photograph it to compare it with the new charger box and also check it over to see if there are any burn marks on the case. I bet the new charger has an updated model number.

I doubt anything will show on the case (through inches of rubbery potting), but I definitely welcome anyone documenting these objects a bit at their convenience. I live in SF near you, so you could just visit and drop it off so I can spend some time carving into the Calex charger like a voodoo doll to represent certain people.

[Richard230]

I doubt anything will show on the case (through inches of rubbery potting), but I definitely welcome anyone documenting these objects a bit at their convenience. I live in SF near you, so you could just visit and drop it off so I can spend some time carving into the Calex charger like a voodoo doll to represent certain people.


Sounds like fun.   

[BrianTRice@gmail.com]

Well, the first time, I was very careful because I didn't know what I would find inside a sea of rubber. The second time around, I know what I'm looking for, just not where it will be since Zero has likely revised the layout on that board since my 2016 model.