Electric Vehicle Progress

Thought I would put a progress report here.

As I mentioned about a year ago I want to convert my Izuzu pickup to electric drive. I also have my eye on 1946 Dodge pickup which would be kinda unique.

I still have not completely discarded the idea of using a forklift motor and VFD for this but I think it will take time for the factories to change their mind over giving me access to VFD software.

Still saving money to do the project and pay off the replacement truck.

So here is how it looks at this stage. I think I have a good grasp on the questions / issues I have to deal with. Now all I have to do is learn more on how to answer / deal with them.

Have DOT center of gravity values for vehicle (both height and distance from front axle). Will weigh vehicle (each tire) and distribute load (mostly battery) such that when done the scale weights will be the same or lower and keep the same ratio between load on front axle and rear axle. Will mount batteries between frame rails and as low as possible to NOT raise CG and hopefully lower it. Any thoughts on how to make this easy would be great - more engineering I gotta learn.

Gas engine transmission driveshaft all removed, motor mounted to rear axle such that shaft is parallel to differential pinion, coupled with timing belt. Differendial pinion speed at 60mph about 3600.

Battery lead acid, total voltage 312 more than likely standard wet cells. Reason for this choice
1. Former submarine electrician - worked with 126 cell battery for four years. I understand them.
2. Have been in production for eons.
3. Watering is a pain but I do not trust sealed batteries.

Motor and VFD made by Azure (formerly Solectria).
Motor is a little oversized but that is OK - I won't beat it to death by running at full capacity all the time.

Motor AC55,
Max RPM 8000, nominal speed 2,000 RPM. Weight 106 Kg.
peak torque 280 NM, continous torque 140 NM (at nominal speed), Kw continuous 25 at 1500 to 2500 RPM (about 33HP), peak shaft power 59 Kw.

VFD DMOC 445, probably liquid cooled.
All voltages are DC -- minimum voltage 100, maximum voltage 400(on charge 450), minimum recommended voltage 312.

Has regen of course and (I think) all the "standard VFD stuff". Got some reading to do to get a better idea of what it can do for me. I also have to get the software manuals to learn more and write the factory with quesions. One of the reasons to post here is to learn more to enable me to ask better educated (less ignorant) questions.

Overview of VFD and Motor is on
http://www.azuredynamics.com/motors.htm
AC55/DMOC445 drive system

I want FVD and motor to do is NOT spin tires whether on ice, snow, wet hill (Seattle = hills + rain) This will require some method of torque control. I need to learn more about this so if someone can recommend good reading material I would appreciate that.

One of the things I find confusing on torque control is that torque is imposed on the motor by the load -- if the motor can deliver the torque it can speed up conversely if it cannot it will slow possibly to the point of stall.

Torque on start on flat road is X. Torque starting on hill is X times a function of road slope. If I am using same speed ramp I will have more torque on the hill than on flat. Can I control BOTH torque and speed? The VFD manual calls out for torque settings in the software.

I have lots of reading to do in the manuals but here is what I understand so far. They feature various starting ramps switch selectable. So I can have an ice start ramp, wet pavement ramp, dry pavement ramp. However this is speed control (I think ??) NOT torque although I am in a way controlling torque by controlling speed.

Thanks for help and devil advocacy.

Dan Bentler

I would look at something like this.

leitmotif said:

Battery lead acid, total voltage 312 more than likely standard wet cells. Reason for this choice
1. Former submarine electrician - worked with 126 cell battery for four years. I understand them.
2. Have been in production for eons.
3. Watering is a pain but I do not trust sealed batteries.

Click to expand...

http://www.a123systems.com/newsite/index.php#/applications/phev/pchart4/

Peter,

Have considered other battery types including thse. Will wait for year or two and let others do the field testing and debugging.

Dan

Did a little studyin
Looked up torque control - found bunch of listings - most helpful was from Dick DV.
Here is what I learned
VFDs come in 3 basic types
V / F control
where all they do is control output voltage as function of freq

Sensorless
where VFD does V / F control plus senses motor parameters to determine torque. This is done by sensing motor no load current (magnetizing current) and motor load current (magnetize current and torque current)

Sensored
where motor has a speed sensor (which may also be rotor position), This is termed encoder. VFD uses this signal in addition to mag current and torque current to control torque.

So I can
control speed
can control acceleration by means of a ramp (both accel and decel). Speed control is "dominant" (??) until torque limit is exceeded at which point speed is decreased.

As I understand it
I can have speed control
OR I can have torque control
BUT I cannot have both except as stated above ??

Have I got it right??

Dan Bentler

Dan, thanks for the compliment. This is kind of fun working thru some of these issues.

Regarding torque control, I think it is just to difficult to implement in this kind of application. On the other hand, having an adjustable torque limit makes it a lot simpler. I am aware of some full-featured drives having the torque limit brought out to an analog input for external control. This might be a good arrangement for you.

As I think about your starting torque issues (dry, rain, snow), it seems to me that you would want a level of torque that will at least start the vehicle moving regardless of the surface condition or grade. That is a fixed value. If the wheels spin under actual conditions, you are essentially stuck. Not the drive or motor's fault. If you were to set a torque level lower, you wouldn't spin the wheels but you might not move either.

On the other hand, if you consider an ordinary car's throttle, it functions more like a torque input than a speed input. I'd have to do a lot more thinking to determine which control method (speed or torque regulation) would be best.

More to your specific questions, as far as I know, you can have speed control with torque limits or torque control with speed limits.

Having both torque and speed under continuous control doesn't seem possible or logical to me.

Actually, you can limit speed and torque at the same time. Once you limit one you cannot control the other. You can control speed OR torque, but you can't control both at the same time. You guys don't know how difficult is is to get this concept across to some people.

It looks to me like the maximum torque is constant to 2000 RPM. After that the motor is power limited. This mean you should be able to accelerate at a constant rate until 2000 RPM after that the acceleration rate will diminish. This does not mean the motor must use full torque at all times. You can keep the torque from being excessive by limiting the acceleration rate. This is espeically true around 0 RPM. You will probably have a voltage or speed reference for the drive. You must be careful not to change that reference too quickly. As you stated, the motor and drive are design for much heavier loads and you could probably make your pickup do wheelies if it weren't for the lead acid batteries. I think the tires will suffer though due to spinning.

I think a small microcontroller/DSP can be used to limit acceleration rates and be the smarts that keeps the operator from tearing the car apart.

So 2000 RPM is going to equate to how many MPH? By keeping the gear ratio at the lower end of the 3 to 5 to 1 gear ratio range, you will reduce the max torque to the wheels. How much will the lead acid batteries weigh? What will be the power or torque to weight ratio? How much frictional force can a tire apply before it starts to slip? If you know the coefficient of friction of the tires and the ground then one can compute this. Obviously this will change with different surfaces.

OK you among several others say I cannot have it both ways - I can have torque control or I can have speed control.

SO can I have torque control with a speed limit and later on have speed control with torque limit.

I have watched electric trolleys spin the tires and chains in snow on Seattle hills. The high torque at zero to near zero speed of electric motors is actually a disadvantage.

It seems to me that torque should be the controlling factor at start (0 to say 5 mph). After that I would want to switch to speed control and have that predominant with torque limits. Can I switch controlling factor once I get rolling or am I stuck with either speed control OR torque control ALL the time?

That is OK with me since I can set accel ramps for varying conditions and have torque controlled (sort of) by controlling acceleration.
What I have in mind is an algorith which would double speed every 250 msec like this
0 is 0 f/s and 0 mph
250 msec is .5 f/s and .34 mph
500 msec 1.0 f/s and .68 mph
750 msec is 2 f/s and 1.34 mph
1 sec 4 f/s and 2.73 mph

At this stage the throttle would take control. I would still have another speed (accel) ramp to avoid instabiltiy and protect the motor. There would be torque limits to protect the motor in addition.

Have I got this right??

Dan

OK you among several others say I cannot have it both ways - I can have torque control or I can have speed control.

SO can I have torque control with a speed limit and later on have speed control with torque limit.

I have watched electric trolleys spin the tires and chains in snow on Seattle hills. The high torque at zero to near zero speed of electric motors is actually a disadvantage.

It seems to me that torque should be the controlling factor at start (0 to say 5 mph). After that I would want to switch to speed control and have that predominant with torque limits. Can I switch controlling factor once I get rolling or am I stuck with either speed control OR torque control ALL the time?

That is OK with me since I can set accel ramps for varying conditions and have torque controlled (sort of) by controlling acceleration.
What I have in mind is an algorith which would double speed every 250 msec like this:
0 is 0 ft/ sec and 0 mph
250 msec is 0.5 ft /sec and .34 mph
500 msec is 1 ft / sec and .68 mph
750 msec 2 ft / sec and 1.36 mph
where it is 4 ft / sec and 2.73 mph



At this point the foot throttle would be allowed to take control. I would still have to have another ramp for speed control at least to protect the motor (in addition to the torque limits ??)

At present differential input pinion speed (motor speed) is 3600 RPM at 60 mph. Motor RPM is subject to change of course.

Dan

I ain't the sharpest knife in this drawer but do not think y'all are going about this right. I think I would use a transmission to apply the torque, directly coupled to the differential will pretty much mean full torque at all times. This is like using a gas engine vehicle with a manual 3 speed transmission and only using 3rd gear, the motor would be screaming most times.

Another reason a transmission may help is the same reason cars have them, it may lower the torque requirements at higher speeds thus allowing less fuel (power) to be consumed.

This could eliminate the torque issues and allow you to just work on how to do speed control. May want to look at linear potentiometers.

You definitely want to have weight equivalent to the motor on the front end, especially if you put batteries in the bed. I would think about putting some in the front with rest in bed.

Rsdoran is on hte right track.

If your max motor RPM is 3600, use a 2:1 input belt transmission to get the rpm under 2000 and halve the amount of torque required to start the load.

You aren't going to want anything but speed control with full torque available until you brake, and then you will want programmable decel rate...proportional to brake pedal position...with a reverse torque limit or something similar to prevent locking up the wheels.

I doubt that you can achieve true antilock style regen braking, but I also doubt that you will need it.
You should be able to recapture tons of joules anyway though, by having the deceleration rate adjusted to work on the load just before the stock mechanical (failsafe) brakes take over.

I did the controls for a 1500 pound maintenance cart that would go up to 25mph and it never came close to spinning the wheels on takeoff or locking them up when decelerating.

The ramping was as smooth as a conveyor belt with an ordinary DC drive and forklift motor. I could adjust the max decel rate on the little DC controller with a pot to make it stop nice and hard, but not too hard. Due to variability in its stopping rate with both regen and mechnical brakes applied, if you slammed the brakes at full speed it would stop hard enough to strip the splined hubs (at the wheels). I ended up disabling the regen circuit to fix that since I did not have any torque control options. You could still regen brake by flipping the reverse toggle switch (tied to reversing contactors)

The hand built cart (by Harvey Cox, a hot rodder expert chassis fabricator) had the motor couple directly to a high ratio differential he said was made like a Ford 9". He "reappropriated" it from an old junked EZ-Go golf cart.

Another EZ-Go 3 wheeled front bed maintenance cart got the same controls and it had a silent chain 2:1 ratio input to it's differential.

It was faster, but also lighter and less durable...wish I could get pics of those machines, but my former employer does not allow cameras on plant site...

Dan, you forgot one more option, power.

When you step on the accelerator, what are you really doing? You are controlling the rate of chemical energy is converted to kinetic and potential energy. When going up a hill the car will slow down unless you step on the accelerator to compensate for the fact that not all the energy expended is compensate for friction some is going into potential energy. The point is that you can have your electric control provide a velocity signal to the drive but after the 'throttle' reading has been divided by the applied torque. This assumes you can get the current torque reading. This would be the most natural way for the operator to control speed.

Ron, the motor and drive are designed to work without a transmission. One doesn't want full torque all the time. That would provide maximum acceleration all the time and that wouldn't be good.

leitmotif said:

Did a little studyin
Looked up torque control - found bunch of listings - most helpful was from Dick DV.
Here is what I learned
VFDs come in 3 basic types
V / F control
where all they do is control output voltage as function of freq

Sensorless
where VFD does V / F control plus senses motor parameters to determine torque. This is done by sensing motor no load current (magnetizing current) and motor load current (magnetize current and torque current)

Sensored
where motor has a speed sensor (which may also be rotor position), This is termed encoder. VFD uses this signal in addition to mag current and torque current to control torque.

So I can
control speed
can control acceleration by means of a ramp (both accel and decel). Speed control is "dominant" (??) until torque limit is exceeded at which point speed is decreased.

As I understand it
I can have speed control
OR I can have torque control
BUT I cannot have both except as stated above ??

Have I got it right??

Dan Bentler

Click to expand...

wouldn't it be much easier, and more intuitive, to let the throttle pedal be essentially a glorified potentiometer... let the signal generated from that control the vfd via a feedback loop? (eg. 0-5VDC, 4-20ma, etc....).. that way you will always be in complete control. You can adjust the forward/reverse ramp rates of most vfds, to vary your acceleration rate during different seasons.
If you want to 'automate' the initial acceleration for various types of weather, you could 'snapshot' an actual acceleration, and store it as a 'macro.'
Just my $0.02...ymmv.. but have fun and good luck!

Basically the throttle would be a pot to supply varying voltage in proportion to desired speed to the VFD.

Maybe I am making this too complicated. I do not have automatic torque control with the 5 speed manual transmission and clutch -- just an educated pair of feet. This seems to work OK about 362 days per year.

I was thinking if the VFD has the capability then do it. Would save wear and tear on vehicle.
2nd - vehicle would be used as a demo so the auto torque / speed control would be to protect vehicle from drivers not used to vehicle. I don't think I nor they would like idea of taking off on a city street at dragstrip acceleration rates.

I think the main argument may be this
Just because you can do it does not automatically make it a good idea.
More studyin should enable me to better answer the question.

Thanks for help

Dan

leitmotif said:

Basically the throttle would be a pot to supply varying voltage in proportion to desired speed to the VFD.

Click to expand...

So the same throttle position would result in a constant speed reqardless of whether going up hill or down hill? That isn't the way a gas powered car works, but you wouldn't need speed control because it would be built in.

I think the main argument may be this
Just because you can do it does not automatically make it a good idea.

Click to expand...

The speed reference idea will work but one must ramp the speed reference up to avoid the wheelies.