Paul & Sabrina's cheap DIY 144v motor controller

[MPaulHolmes]

I don't think you got all the components that were listed on the wiki. It looked like quite a bit of freestyling componentwise from the pictures. And the soldering has to be very well done. Soldering to the ground plane requires a good iron. The soldering didn't look very good from the picture that I saw. I am not trying to be critical. This is from someone that made a couple control boards that didn't work when I first started soldering. The isolation rings are very thin, and solder bridges can be catastrophic. It took practice for me to make a board correctly, but the last 15 or so I have soldered (and I've only done about 17) have worked as expected. There are too many other unknowns to blame a bad NAND, which has a failure rate from the factory of something like 1 in a billion.

One of the failure modes of a controller is basically a dead short across the battery pack. A fuse for the pack is an absolute must. A disclaimer is a very good idea on the wiki.

[snakub]

There were no solder bridges on my controller and what I was trying to get at with the post was that on the wiki there should not only be a disclaimer about the fuse that's obvious but about the blinking light just because it stays solid doesn't mean everything's alright with the board. The board had no overcurrent for some reason the batteries are indicative of that and had I known the board was SO critical that even if one of the components were slightly off or not operational I probably wouldn't have purchased it. Only a couple filter caps were substituted the rest were either from mouser or from you.

[Tweety]

Well... I'm not a representive of Paul in any way, and I haven't even built a controller yet (planning, and getting parts for an EV, long term) But I work with manufacturing electronic components, and a lot of them one-off or prototypes... So I know how to check stuff out before first use... And I'd say warnings in the wiki or not, the builder ie you are always responsible for doing that in a safe manner, especially considering the fact that pack voltage can really do some damage...

The controller can be purchased as built and tested... Or as a kit... If one chooses the kit, part of the choice means saving money, part of it means you get to take credit for all blowups regardless of the reason for them... That's part of learning... And it's fun...

It's common practice in ALL cases, not just this, is to test any electric car you build up with 12V... ie stick just one battery on the engine and controller and check that it's working... Sticking a full pack voltage on to test things seems like a recipe for having something break, regardless of who's to blame...

Having a fuse on your pack isn't just common knowledge and good practice... I agree it should probably be in the wiki, but NOT having one is borderline stupid, whatever controller you are using, homemade or not, it's your life in the balance if something goes wrong...

[flores]

I agree with tweety.

[MPaulHolmes]

I still think that driving huge darlingtons at 16kHz with that mosfet driver probably isn't a good idea. It's made for mosfets, and works for some IGBTs as long as a few modifications are made. But darlingtons require completely different gate drive. You didn't use the control board's driver section, if I remember right. Or at least you added some sort of mosfet as part of the driver section. That sounds like an interesting idea, but it's departing from the original purpose of the board. An LED can't speak about the 200 things that could go wrong in the assembly process. As a side note, what parts did you get from me? I suppose it was at least the microcontroller and control board. Was there anything else from me?

For the wiki, we could say, LED blinks if throttle or current sensor is out of range. Use a fuse. It's designed to drive mosfets, but can be made to work for at least some IGBTs. What other warnings?

[Greg Fordyce]

Originally Posted by MPaulHolmes

For the wiki, we could say, LED blinks if throttle or current sensor is out of range. Use a fuse. It's designed to drive mosfets, but can be made to work for at least some IGBTs. What other warnings?

How about a section in the wiki for disasters. A page of problems that people have had, what caused it and how not to repeat it. The temptation some have with this project is to modify the design, but having followed this thread and built one of Paul's kits it may not be obvious why a particular component was used or the consequences of straying from the design. So a disasters page can help others from repeating costly mistakes and also give people an idea of what could possibly happen when straying from the provided design.

Of course the wiki is, by definition, open for anyone to contribute to, so maybe snakub could start the disasters page with his incident, changes he made to the design and what blew up. It's frustrating when you have a setback like this, I know from having unintentionally cremating one EV, but documenting the failure can help you and others understand what went wrong and how to prevent it in the future.

You could add a general disclaimer like the one on the bms project I have been involved with below. Adapt and use any of it you like.

'The BMS modules carry no warranty or guarantee of any kind! They are used at
'your own risk, and I make no claims as to their suitability for a particular
'function. Prospective users must evaluate the system before using it, and no
'liability will be entertained by myself in any shape or form whatsoever.
'The modules and software have been produced at low cost for the benefit of
'the EV & electronic community. The software is available free via the internet.
'Users may modify or adapt the system as they see fit. If you are not fully
'competent to work on potentially lethal battery systems and high voltages,
'then do not experiment with or use this system. Be aware that vehicle
'modifications can lead to invalidated insurance and warranty issues. You the
'end user remain fully liable for any modifications made to your vehicle.

Paul's design works out of the box, but only if you follow the design, the more changes that are made, the more risk of something going wrong and with EVs it can be rather catastrophic and costly when things do go wrong.

Greg

[MPaulHolmes]

I think a failure page is a great idea! I remember one was from using a contactor with the gas pedal when the pedal POT wasn't calibrated right, so pwm ran up to 100%. Another was swapping B- and B+. It would be nice to add pictures too.

[mrbigh]

Paul, where you able to send the new driver boards to be manufacture?

[MPaulHolmes]

Not yet. I live in my mother-in-laws basement. hahaha. The first board purchase is always the most expensive, because they add the cost of who knows what to it. From China, for 5 boards, it will cost $100.45 including shipping. Anyone want to pay $20 each for one? I'm broke. The good news is, for 75 boards, it's $1.52 each, so in the long run it could be cheap, for the board at least.

[wakinyantanka]

Originally Posted by MPaulHolmes

Not yet. I live in my mother-in-laws basement. hahaha. The first board purchase is always the most expensive, because they add the cost of who knows what to it. From China, for 5 boards, it will cost $100.45 including shipping. Anyone want to pay $20 each for one? I'm broke.

Hey Paul,
PM me I can help a little.