Paul & Sabrina's cheap DIY 144v motor controller

[johnrobinson457]

Howdy Paul, the question I have, what motor can be used with your ecomod controller? (this for a 48-60v motorcycle) I know I read it somewhere in the thread, but I can't re-find that.... brushless, brushed? got a recommendation for make and model? Like I said in the first PM to you, I ( and several of the other readers and adventurers) will need a fair amount of handholding in the assembly of the controller AND the obvious (to you) other parts needed.... anyway thanks for all your work... I've got a bike ready to strip and the local autoparts store guy is giving me a discount for any parts (like 12v flooded batteries) needed... got the elninja book and am assembling bike parts needed, waiting for the "official ecomod list" for the electrical parts, but figured a motor is gotta be needed so am lurking various websites for one....

[MPaulHolmes]

Not fiberoptics. These optocouplers are only like $0.95 or so.

You turn the PWM rectangle waves into on/off's of an LED that's deep inside a little Integrated Circuit. The LED beams it's light across the void of the inside of the chip (electrically isolated now!) to a thing that turns on and off whenever it gets hit with light.

http://www.mech.uwa.edu.au/NWS/How_t...to_circuit.jpg

[Roger Heuckeroth]

I like it, Paul. It sounds like improvements are coming at light speed (no pun intended). Opto's are great for taking care of ground issues. This collaborative effort is a wonderful thing to behold.

I went to the New York International Car Show in NYC this weekend. I really expected to see more electric cars. I was disappointed. The only car company really pushing electrics was Chrysler. They had two on display the Circuit, a two seat, sports car, and a Town and Country minivan. Each model displated on a pedistal with an attractive girl talking about its features. I have a bunch of pictures that I plan on posting soon for those that are interested. GM had a Volt on display, but it was kind of pathetic verses the Chrysler products. The Volt looked far from a finished prototype. It has stickers for fog lights, and they kept you 10 feet away from it so you wouldn't notice the lack of fit and finish. If this is the best that GM can do... they are going down.

Anyway I digress, but the one thing I realize is that its still going to be years before you can go down to the local dealer and buy one off the lot. Any many more years before you can get a good deal on a used one. So for the common man, this open source project is the way to be able to afford to get into an EV today. Making EVs affordable is the big challenge. EVs are the only way we can use renewable resources to power our transportation needs. I feel passionately about this, and Paul, you are a hero in this cause.

[Roger Heuckeroth]

Originally Posted by edown

Anyway, I have a question.. Paul uses 24, 200v, 470mf caps. Is it OK to use 3 large, 4400mf 300v caps? Is it the total “mf” that is important?
I want to just screw mount 3 of them straight to the buss bars omitting the PC board.. I can get 10 of them for $30 on ebay right now..

Evan,

That's a really good question. Paul already mentioned the importance of low ESR. It actually also comes down to heat dissipation. A bunch of smaller cylinders have more surface area for heat exchange than three larger ones, and the life of a capacitor is directly affected by temperature. The hotter they run, the shorter they last.

There is also another reason for going with smaller caps, and it has to do with geometry of the controller. You want to minimize the inductance between the capacitors, the mosfets and the diodes to avoid voltage spikes. Ideally, the connections between the capacitors terminals, the diodes terminals and the Mosfets drain and source are as short as possible. This is because of the massive amounts of current flowing between them.

Otmar, the brains behind the Zilla controller, has been quoted calling this the holy trinity. If you look at his design, these components are physically very close to one another. BTW, his 1000 Amp controller uses 12 - 330 uF caps, which is almost 1/3 of the total capacitance Paul is using. So, I believe Paul's design is very conservative.

[edown]

Thanks for the quick response.. I think I will gather up my remaining (7) 50A mosfets and build a small version of your prototype power section. I have a single 12500mf cap to try, and see if it gets hot..
I also use optocouplers.. I use the SFH620A (a 4 pin dip), it has 2 LEDs so polarity is not a concern. Also if you are using a crude PWM circuit (like a single 555 chip) they help clean up the signal.
One more question: I have Bob Brant’s book “Build your own electric vehicle” and he stresses using a large inductor (choke) in series with the motor to “control armature ripple current”..
“20 turns of 1/16” x ¾” copper on a 2” x 2” core” Do you think this would help, or is it just not necessary?

[Wherewolf]

This may be slightly off thread, but a word of caution to anyone deciding to acquire an electric forklift and haul it yourself. (Yes – this is experience talking….) I know you are thinking about it….. Or tow ANY heavy load for that matter.
I did such an act over this last weekend. It would seem obvious that we’re talking about a great deal of weight here – and it was. Total weight of this model was over 8k, add in the trailer (heavy!) and you get way up there in a hurry…..battery (48v steel cased) weighed over 3k by itself!.
NOW THE WORD OF CAUTION: Check your hitch bolts where the hitch connects to the frame. Mine were apparently loose (I realized in hindsight, probably from towing other really heavy loads) -- come on – really – once a class V receiver is bolted on the truck, who really gets down and inspects the bolts periodically? When I got home after a grueling hour of towing this monstrosity I noticed while unloading that the receiver tube itself was moving around some. Out of 6 bolts: 1 loose on the right side – other two GONE. 1 GONE on the left – other two loose enough to scare the crap out of me.
Needless to say, I replaced all with Grade 8 Fine Thread Bolts, washers, lock washers, and welded the threads below the nuts, so they can't come undone. By the way, this was a ¾ ton Ford with a 1 ton rearend. But I was “probably” over GVWR….But the moral of the story is to check your hitch bolts….it could have gotten real ugly in a big hurry. When the hitch comes off, so do your safety chains and the trailer would have gone wherever it damn well pleased.….whew!
a sidenote: I think I determined what was wrong with the forklift – found a wire hanging underneath, I think it’s a sensor that interlocks with the parking brake. This is an older Hyster, but in pretty good shape –26 foot reach - If I can fix it, slap some paint on it and sell it – buy a different unit to salvage some motors from and have towing fun all over again!

Wherewolf

[Blue Bomber Man]

John, This controller works with Brushed DC motors. Someone mentioned that the Warp9 motor would be a good fit. I would recommend something higher then 60 volts if you plan on going faster then 30+ but I am sure you have already considered that.

Good luck!

[Roger Heuckeroth]

Originally Posted by edown

One more question: I have Bob Brant’s book “Build your own electric vehicle” and he stresses using a large inductor (choke) in series with the motor to “control armature ripple current”..
“20 turns of 1/16” x ¾” copper on a 2” x 2” core” Do you think this would help, or is it just not necessary?

I don't think its needed in the motor circuit to protect the motor. Where I have seen this suggested more often is on the B+ side of the controller. In that location it smoothes out the current being drawn from the batteries. More specifically it acts as a low pass filter to keep the ripple current away from your batteries. Batteries get hot and experience shortened life when exposed to a lot of ripple current. The capacitors in your controller remove most of the ripple, but there is still some ripple at the batteries. A properly sized inductor will completely eliminate that ripple current from the batteries.

[MPaulHolmes]

There is also now a zener diode clamping the input voltage to protect the 12v DC-DC converter, just in case it was Roger's diagnosis of the controller failure that happened. I can't wait to test that DC-DC! Either way, it COULD have been the point of failure. It's really nice to eliminate it as a possibility in the future.

I made a full scale sketch of the mosfet driver board strip, and figured how many mill cranks every etch will be, so now I just need to quit typing and go out there and do it! Well, I'm hanging out with my son right now, and mills and 1 year olds don't mix.

[Christ]

I'm waiting for a son to hang out with... patience is a virtue, don't want him being born too early, now.

Maybe when he's been around for a few years, we'll take on a project together.