Paul & Sabrina's cheap DIY 144v motor controller

[sawickm]

Originally Posted by MPaulHolmes

The only reason I was using the dspic was because I wanted it to have CAN, so it can talk with the bms and charger while driving.

Hey Paul,

If you want to add “CANbus” to the Atmega ReVolt Cougar PCB than check this out: CAN-SPI Adapter

You should just be able to interface it into the “Cougar” Rev.C SPI port.

If that would work ??? “CANbus” could be added on-board for next version of the “Cougar” Rev.E ???

-Mark

[MPaulHolmes]

Da** it Jim, I'm a substitute teacher, not a miracle worker. Sorry, my nerdy side got the best of me.

http://www.youtube.com/watch?v=bVhcB9ucmdg

Maybe there should be a SM version that just uses the old socketed atmega.

Hey! If I'm understanding this sheet right, the dspic30F Mean Time Between Failures is better than 17,000 years. hahaha. That can't be right can it?!
http://ww1.microchip.com/downloads/e...cs/00097AG.pdf

[mrbigh]

In the event that the dsPIC get deprogrammed or "crashes" and that it is realy not abnormal, what would it be necessary for the common Joe to re-program that device?

[MPaulHolmes]

It's basically the same as with the atmega. If the chip gets deprogrammed, then you just reprogram it through the ISP interface. That requires a $40 pickit 2 or 3. If the atmega gets deprogrammed, you need an AVRISP, which also costs around $40, or you can make a homemade one. I think they are pretty simple.

[mrbigh]

Originally Posted by MPaulHolmes

It's basically the same as with the atmega. If the chip gets deprogrammed, then you just reprogram it through the ISP interface. That requires a $40 pickit 2 or 3. If the atmega gets deprogrammed, you need an AVRISP, which also costs around $40, or you can make a homemade one. I think they are pretty simple.

PAUL, thank you.

This is the type of information that I was looking at for everybody information.

[DJBecker]

Originally Posted by sawickm

Hey Paul,

If you want to add “CANbus” to the Atmega ReVolt Cougar PCB than check this out: CAN-SPI Adapter

You should just be able to interface it into the “Cougar” Rev.C SPI port.

If that would work ??? “CANbus” could be added on-board for next version of the “Cougar” Rev.E ???

-Mark

There are several CAN controller boards advertised on the web. None are readily available in the U.S., and they are expensive for what they are. They all use the MCP2515 chip I talked about in post 4157
http://ecomodder.com/forum/showthrea...tml#post209940

As I mentioned in that post, the main chip is under $2 in single unit quantities. They *might* be able to justify their US$50 price if they have jumper options for an isolating transceiver and on-board voltage conversion. But they are just the MCP2515 and a basic transceiver.

(The MCP2515 is pretty much the only external CAN controller available. The few others are out of production, so you can pretty much expect that's the chip everyone is using. And it's available in a standard DIP package so it's easy to do a DIY build project.)

All of the boards require more signals than the Cougar brings out to its header. The Cougar doesn't have a SPI chip select there, and this chip works best with an interrupt line.

In my implementation I clock the chip from a timer output run at 8MHz, so that's a third additional connection, although with the Cougar you might prefer to use a 16MHz crystal. Once you are doing that wiring, you might as well do the rest of the assembly on a protoboard.

The challenge, as always, is the software to make it work. I have a pretty good first pass that should be close to working with the Cougar if someone wants to work on the testing with me. Note that you can only test CAN if you already have a working network, as it requires at least one other active node to set the acknowledge bit.

[Dwffy]

Hi folks,
I want to throw in a quick question here to see what you guys think of this idea.
When pulling away from being stationary, the acceleration in the vehicle is reasonable and it would be nice to get an extra boost on occasions. Now here's my idea. What if I had a contactor connected from Motor - connection to B- connection of the pack and used a push button to engage it, but, only after the vehicle was already in motion. This should give full pack voltage across the motor without effecting the controller. Could this be a real idea to get a great boost on take off, or would it only let out the magic smoke?
-Doug

[electricstorm]

Wow Paul! How'd you know I was a Star Trek fan.... Loved it!

If the MicroChip was to be placed and soldered to the board by the company making the boards for you (I think that's what the post stated), then if there is a socket for the chip you would not have to redesign the board. But, if it is not in a PLCC package, it would probably be a PITA!

Unless you do something careless to the control board, IC's are pretty reliable these days. So you may never have to replace it.

[mrbigh]

Originally Posted by electricstorm

But, if it is not in a PLCC package, it would probably be a PITA!

Unless you do something careless to the control board, IC's are pretty reliable these days. So you may never have to replace it.

Do you remember "Murphy's law".....................

[ga2500ev]

Originally Posted by Dwffy

Hi folks,
I want to throw in a quick question here to see what you guys think of this idea.
When pulling away from being stationary, the acceleration in the vehicle is reasonable and it would be nice to get an extra boost on occasions. Now here's my idea. What if I had a contactor connected from Motor - connection to B- connection of the pack and used a push button to engage it, but, only after the vehicle was already in motion. This should give full pack voltage across the motor without effecting the controller. Could this be a real idea to get a great boost on take off, or would it only let out the magic smoke?
-Doug

Doug,

A 5 second blurb on how motor controllers work. They always apply the full pack voltage to the motor. What varies is the amount of time that voltage is applied. This is known as pulse width modulation, or PWM.

A motor controller at 100% PWM is exactly the same as your proposed contactor.

So all that is really needed is a push button to the controller and a small bit of code that sets the PWM to 100% when the button is pushed.

But IIRC this is usually the same effect as pushing the gas pedal all the way down.

Now there is one problem with both approaches. Generally when the car is at a stop and full power is applied, the current from the batteries into the motor (and all parts inbetween) shoots through the roof. So the controller measures the amount of current and cuts off power when an overcurrent condition occurs.

The problem with your pushbutton to the contactor is that if the current exceeds the contactor rating, it can cause the contacts to meld together and not be able to release. Same for frying the electronics of the controller.

So be careful if you take this approach.

ga2500ev