Paul & Sabrina's cheap DIY 144v motor controller

[bennelson]

OOOh! I like the "black-box" concept.

It would make it easy to build a standardized kit for the drive-board half!

[tech2]

Originally Posted by Intrigued

Oh, yes! Welcome, Tech2! Your screen name reminds me of the two GM scan tools I have managed to keep around the old hacienda.

Think my screen name is unique, wait till you hear what my password is!
It's .. .. . . ohh ...........duh!!!.........nevermind

[tech2]

Originally Posted by Intrigued

Hey, that is a pretty neat idea: using a heavy flywheel. We'll have to see if anyone with experience can punch holes in the theory. It would be a breakthrough if it worked!

The reason so many have stayed away from automatics is that you lose up to 15% of your efficiency to the torque convertor, which just sits there and turns the energy into heat when not locked up. There is such a delicate balance between carrying enough batteries to get any range versus being able to make the car able to carry all of the weight versus getting any speed out of the thing...etc that most folks just try to find a standard tranny.

Some of the automatic guys just connect directly to their input shaft - no flywheel or torque convertor. Saves weight to help with the balance of weight/power/speed that way, too.

Well, I want the smooth acceleration associated with automatics, and want to keep it as similar to a carbon burner as possible. I also thought of something else. A lot of the GM trannys have the capability of locking the torque convertor with an internal solenoid. That could be controlled with the software easily enough. I found that out on a Grand Prix the hard way when it locked in and stayed locked in. It will kill the motor when you come to a stop at a red light!
For distance driving, I thought I will "hybrid" the car with a very small one cylinder gasoline generator, adapted to output the battery voltage(+ a few volts for better charging), and just see what happens. Should increase the distance quite a bit.
cheers!

[MPaulHolmes]

Just so I understand:

Drive board is the power section with the mosfets, big capacitors, and big diodes, and control board is the low power stuff, right?

I think that's a great idea. Most controllers I know of sort of intertwine the two because of a DC-DC converter that turns full pack voltage into 12v and 5v for the control board. That's why I went with a 12v battery to power the controller. Full pack voltage becomes completely independent of the control section.

You could literally have 2 boxes! A control board and a power section. There is one thing to consider, though. The 12v output of the PWM signal from the gate driver that goes to the gates of the mosfets must be as short as possible. You could have the 2 halves of the controller sort of snap together making the connection.

The control board gets 12v power from the car's auxiliary battery that powers lights and blinkers and stuff. There's a 12v-12v isolated DC-DC converter inside the control board. The isolated 12v supply needs to share a common ground with the full pack voltage. So, the control board black box would need an input from pack voltage minus. Other than that, I don't think there would need to be any other interface required.

So, to sum up:
Control board black box has battery pack B- input pin, auxiliary 12v plus input, auxiliary 12v minus input, 0-5k input pin1 and pin2, and PWM signal output pin. It could also have an input for ISP (in system programming).

Power board has battery pack B+ and B- input, M- output.

Is that it? Or did I miss something? That would be nice and clean. Great idea!

[Intrigued]

Originally Posted by Intrigued

Some of the automatic guys just connect directly to their input shaft - no flywheel or torque convertor. Saves weight to help with the balance of weight/power/speed that way, too.

OOPS! Gotta correct myself there. Since torque converters also pump fluid and help determine shift points (plus I spelled the word "converter" wrong) there's something not right about that statement. I do remember someone using an old PowerGlide with no converter, but I think he cut the converter apart and tied to the part that fit over the input shaft.

Dadgummit, that's what I get for trying to help... If I had Paul's fork, I'd stick it in my eye!

[order99]

I noticed this the other day:

A Classic EV Design With Advanced Features

Looking at the Urba Electric design, three Controller designs are mentioned-

1) a modern controller
2) a Voltage Stepper-not as efficient but probably cheaper
3) a CVT transmission to keep the motor at most efficient RPM (possibly with a simple Amp limiter-no details on the page)

I don't have a copy of the plans yet(I plead Extreme Poverty at the moment) but i'm really curious on what effect a current CVT (either mechanical or electronic controlled) would have on a light EV...

[MPaulHolmes]

Originally Posted by Intrigued

Dadgummit, that's what I get for trying to help... If I had Paul's fork, I'd stick it in my eye!

Don't do it! The anger soon passes...

[MPaulHolmes]

I got the new tool thing from my brother today! It was a sort of a plexiglass thing that turns a dremel into a 3 axis mill, but it had a bit too much flex in it. It would probably work better if I added the 3 motors and software to make it computer controlled. I think a regular mill is going to be needed for heavy duty power board stuff. A decent micromill is maybe $500-$700, but maybe a regular mill would be better, so I could also make adapter plates that are actually centered and sell them cheap too. That's one of the other things that make an EV sort of expensive. It's just that they weigh like 500 pounds or something ridiculous. How the crap do I put that in the trunk of my Kia?!

I've been experimenting with small 0.1uF to 4.7 uF capacitors to eliminate some of the voltage spikes. The spikes are sort of small, but why not!? I cooked a tantalum capacitor that too high of an equivalent series resistance that I was using for spike removal. The top of it turned black. The MLCC surface mount capacitors would work better, but they are like 1 mm long! How in the name of Johnny Appleseed do you solder those things? Plus, it is magnetized! Each time I get a soldering iron near it, it gets stuck to it. What idiot thought that was a good idea? Makes me mad.

I don't like working. Sylvan at night, School during the day, family in evening, controller at 5-6am. grr... I need to go to sleep.

I'm going to take the electric car offline, and connect the home-made bike controller. I need to see how the bigger motor behaves with the software.

[tech2]

Originally Posted by MPaulHolmes

I got the new tool thing from my brother today! It was a sort of a plexiglass thing that turns a dremel into a 3 axis mill, but it had a bit too much flex in it. It would probably work better if I added the 3 motors and software to make it computer controlled. I think a regular mill is going to be needed for heavy duty power board stuff. A decent micromill is maybe $500-$700, but maybe a regular mill would be better, so I could also make adapter plates that are actually centered and sell them cheap too. That's one of the other things that make an EV sort of expensive. It's just that they weigh like 500 pounds or something ridiculous. How the crap do I put that in the trunk of my Kia?!

I've been experimenting with small 0.1uF to 4.7 uF capacitors to eliminate some of the voltage spikes. The spikes are sort of small, but why not!? I cooked a tantalum capacitor that too high of an equivalent series resistance that I was using for spike removal. The top of it turned black. The MLCC surface mount capacitors would work better, but they are like 1 mm long! How in the name of Johnny Appleseed do you solder those things? Plus, it is magnetized! Each time I get a soldering iron near it, it gets stuck to it. What idiot thought that was a good idea? Makes me mad.

I don't like working. Sylvan at night, School during the day, family in evening, controller at 5-6am. grr... I need to go to sleep.

I'm going to take the electric car offline, and connect the home-made bike controller. I need to see how the bigger motor behaves with the software.

Paul, use super glue to pre-align the caps on the board. Just a tiny drop applied with a tooth pick point will do. Let it sit awhile, and then you can solder it into the circuitry. B T W, I have a mill. If you go into production and need some product milled and drilled, maybe we can work something out. I will just need prints to get it right. Copper clad isn't cheap!
I also built a small 3 axis mill years ago, from diagrams available on the web, and bought some kits and built the stepper controllers, used old large printer stepper motors, and made a gear reducer from an old air ratchet gearhead..
SO..... I have a little experience with this stuff

[tech2]

Originally Posted by order99

I noticed this the other day:

A Classic EV Design With Advanced Features

Looking at the Urba Electric design, three Controller designs are mentioned-

1) a modern controller
2) a Voltage Stepper-not as efficient but probably cheaper
3) a CVT transmission to keep the motor at most efficient RPM (possibly with a simple Amp limiter-no details on the page)

I don't have a copy of the plans yet(I plead Extreme Poverty at the moment) but i'm really curious on what effect a current CVT (either mechanical or electronic controlled) would have on a light EV...

From looking at whats posted across the web, I would not even consider purchasing the plans. Everyone said the chasis was too weak looking and it's basically nothing more than an electric golf cart with add ons!
Modding a small car will be a lot more interesting than trying to completely build one