[blueblizzard816]

I'm looking for a way to make a cheap and easy to build (by theory) controller.

I've poked around on the forums and really liked the idea of the mechanical PWM controller (

http://www.diyelectriccar.com/forums/showthread.php/mechanical-pwm-46258.html

) but it looks like there are too many limitations for an EV application.

This sounds like a dumb question to me, but what the heck. Is there any reason why I can't just put a large potentiometer in series with the batteries and the motor and use that as my throttle to control the power output?

Have there been any other feasible attempts to build a simple (DC motor only) controller for EV applications?

My specialty is ME, not EE so try to explain it in a way that will make sense to me!

 

[peggus]

With the right motor you can control it with just a potentiometer and a contactor. Using a sepex motor you can control the field with a potentiometer and just connect the armature to the pack, via contactor.

The field power is just a few percent of the propulsion power so the losses in the pot are acceptable, you still need a sizable pot though, (rheostat).

With full field power (low pot resistance) the motor will spin very slowly, to speed it up- reduce the field power.

With a series motor, the power losses in the pot would be too great to deal with.

 

[major]

This sounds like a dumb question to me, but what the heck. Is there any reason why I can't just put a large potentiometer in series with the batteries and the motor and use that as my throttle to control the power output?

Not a dumb question. Sure, you could do it. The potentiometer would need to be about ten times the size of your motor. And it would waste half of the battery energy needed for every acceleration. And at any speed less than full speed, it would waste considerable amounts of power.

Good news, it could double as a cabin heater If there was any room left for a cabin on the vehicle after finding a place to mount such a large pot. See if you can even find a 25 kW pot.

This is a 7 kW pot. You'd need like 3 or 4 of these. It weighs something like 150 pounds.

Other than that, good idea

major

 

[DavidDymaxion]

This is how my car operates (for now, anyway). It is quite drivable this way for city speeds at 48V. Something I would add is keep the clutch and transmission -- you'll need to slip the clutch to get going to prevent a huge jerk. With sepex it is great fun to get regen on the downshifts.

I read some old golf cart controllers would have 24V + starting resistor, 24V, and 48 V as three steps. The starting resistor was meant to be used for just a moment.

With the right motor you can control it with just a potentiometer and a contactor. Using a sepex motor you can control the field with a potentiometer and just connect the armature to the pack, via contactor.

The field power is just a few percent of the propulsion power so the losses in the pot are acceptable, you still need a sizable pot though, (rheostat).

With full field power (low pot resistance) the motor will spin very slowly, to speed it up- reduce the field power.

With a series motor, the power losses in the pot would be too great to deal with.

 

[peggus]

As bad of an idea as the giant pot is, you can take comfort in the fact that someone came up with a worse idea and actually tried it.

Scroll to the bottom of the page for the salt water immersion controller.
http://www.poormansev.com/id24.html

 

[DIYguy]

I've got a $100 controller for you. It's a range switch out of an industrial DC weld power supply. I used this to move my truck around the garage..in and out etc. before I had controller stuff done.

It's like the home made stick shift...thingy...only a little more "refined" lol..

Want it?

 

[rillip3]

Also, there is an open-source controller for what you are seeking, made by DIY'ers. It's called the Open ReVolt controller. The easiest way to make one would probably be to start with a kit unless you've done something similar.

http://www.paulandsabrinasevstuff.com/store/page3.html

Note that I haven't and do not plan on doing this myself, but it's my understanding this is fairly cheap, i.e. < $300. Your mileage may vary.

 

[blueblizzard816]

I saw that on the budgetev link (http://www.poormansev.com/id24.html) the controller looks like it connects a varying series of batteries directly up to the motor through that "stick shift" to control the power output to the motor in the form of a finite number of power levels.

The first thing that popped into my head after look at it was if the batteries will drain un-evenly.

Should I even bother looking into this?

Thanks for all your guys'/girls' help! I am learning A LOT and your is very much appreciated!

 

[rillip3]

ignoring the sheer genius/idiocy of that setup for a moment, yes, unless you rotate your batteries, you will draw them unevenly.

 

[peggus]

A contactor controller may be an option for you, the better way to do them is to series/parallel the batteries so you get even discharge across the pack at all speeds. Some googling should get you started.

 

[exile]

Well, If you absolutely want to have mechanical controller why don't you do like the Prius?
Partially uses an electric motor(PM), generator and planetary gear and by controlling felid on generator (small current) to get the wished power.

 

[Lee Hart]

Not a dumb question. Sure, you could do it. The potentiometer would need to be about ten times the size of your motor. And it would waste half of the battery energy needed for every acceleration. And at any speed less than full speed, it would waste considerable amounts of power.

That's a bit drastic. Let's not scare the poor guy to death!

For many years, really simple crude low-speed EVs used exactly this type of motor controller. Examples include streetcars, golf carts, fork lifts, and many little personal vehicles. These were serious commercial products made by major companies; not hobby or experimental EVs.

The adjustable resistor in series with the motor is called a "rheostat" (not a potentiometer). At this power level, it is generally built with a number of fixed resistors, selected in by a big rotary switch.

This type of controller was only popular because it was CHEAP! It has no other attributes. It worked best when the vehicle was so under-powered that it was run at full power most of the time. At full power, the resistors are all bypassed and all the power goes to the motor. You ramp the rheostat up from off to full rather quickly (10-20 seconds) to accelerate from a stop to full speed; then cruise at full speed; then ramp the rheostat back down when it's time to stop. Since it's not in use most of the time, efficiency isn't too awfully bad.

Yes, there are still applications for such a thing. For example, I teach kids to build EVs out of old bicycles and other junk they can scrounge (see www.bestoutreach.com). A typical setup is a car fan motor, with a skateboard wheel pressed onto its shaft, arranged so it rubs against a bike tire to turn it. If you simply switch the motor straight to a 12v battery from a dead stop, the tire slips and quickly burns a hole in the bike tire. But if you put a rheostat in series with the motor, you can slowly ramp up the speed to get it going without slipping. The power of the motor is such that full speed is only 10-15 mph, so you just run it at full speed or off. You can't leave the rheostat at anything but full speed or off for more than 30 seconds or so, or the rheostat overheats.

At this power level, you can buy a proper motor controller for $10 or $20. But a rheostat costs $1 or $2. Guess which one the kids pick?

 

[major]

That's a bit drastic. Let's not scare the poor guy to death!

Just having a bit of fun and trying to make a point But the big pot is about what he'd need for an EV car.

The adjustable resistor in series with the motor is called a "rheostat" (not a potentiometer). At this power level, it is generally built with a number of fixed resistors, selected in by a big rotary switch.

Yep, but most buy a potentiometer and just use two terminals for rheostat applications. Even the PB-6 is this way. And the rotary switch with resistors approach would be on the same order of size as that giant pot with the same loss problem.

And sure, you can build EVs with resistor speed controls, or battery switching, but it is unlikely the outcome will be suitable for highway capable car performance.

Regards,

major

 

[Lee Hart]

Just having a bit of fun and trying to make a point But the big pot is about what he'd need for an EV car.

Point taken. Though, the pot you showed is bigger than the ones used in the streetcars, subway trains, and other huge vehicles that used this method.

most buy a potentiometer and just use two terminals for rheostat applications.

For small stuff, yes. But you really don't see high wattage potentiometers; they are all rheostats. Often they are specifically built for motor speed controls, and have a nonlinear taper so position is roughly proportional to speed.

And sure, you can build EVs with resistor speed controls, or battery switching, but it is unlikely the outcome will be suitable for highway capable car performance.

Of course it can provide highway-capable performance. You and I might not do it because it's not the "best high-tech" solution; but it is certainly still a viable approach. It's no worse than, say, driving a old car with a carburetor because it's what you can afford, and know how to fix.

 

[pyroboy1080]

I'm looking for a way to make a cheap and easy to build (by theory) controller.

I've poked around on the forums and really liked the idea of the mechanical PWM controller (

http://www.diyelectriccar.com/forums/showthread.php/mechanical-pwm-46258.html

) but it looks like there are too many limitations for an EV application.

This sounds like a dumb question to me, but what the heck. Is there any reason why I can't just put a large potentiometer in series with the batteries and the motor and use that as my throttle to control the power output?

Have there been any other feasible attempts to build a simple (DC motor only) controller for EV applications?

My specialty is ME, not EE so try to explain it in a way that will make sense to me!

I don't see why you couldn't drive your PWM from a circuit?
There are some nice schematics for PWM driver circuits here:

http://www.picotech.com/applications/pwm_drivers/

Or even perhaps an Arduino, but I'm not exactly sure if the Arduino would be able to work at the amount of current you would need to drive a nice big motor.
You might have to modify them a bit to better suit your needs, but I feel like you could make this happen, it would just take some planning. Also, I would consider a servo for your pedals if you didn't have any better plans for that. Servos have positional awareness, at a very precise level. Also, you'd be one step closer to car automation

 

[Lee Hart]

I don't see why you couldn't drive your PWM from a circuit?

Generating the PWM signal is comparatively easy. You can do it mechanically, with vacuum tubes, transistors, ICs, or even microcomputers.

The hard part is building some kind of switch that can switch the inductive load of a motor on and off millions of times and survive. This is one of those problems like riding a unicycle; an expert makes it looks easy -- then you try it, and "ow"!

It's possible to build your own PWM controller; but it's not easy. Don't take this route because you think you'll save time or money. Only try it if you're already experienced at designing and building high power electronics, or want to learn how it's done no matter how long it takes or what it costs.

For the purposes of this thread: The best way to make your own low cost DIY controller is a contactor controller. It's basically a few used or surplus forklift contactors, wired to switch the motor and/or batteries in various series/parallel arrangements to control the motor speed.

It also helps if you can find a motor with a shunt field (also called sepex, or separately excited). The shunt field makes them much easier to speed control -- all it takes is a moderately sized rheostat.

 

[pyroboy1080]

Noted. Also, cheers! That's great info!