TeamSwift

In another thread, Lihtan wrote:



Yup. Here goes:

(NOTE: since starting this thread, the image hosting service I was using (theimagehosting.com) for the pics changed its rules which broke all the links to the larger images. So now only thumbnails are shown below. If you want to see the "zoom" version of a picture, get the image URL (in Mozilla/Firefox: right click the image > view image ... in IE, right click the image > properties - then copy paste the image URL into your address bar) and remove the ".th" from the filename in your browser's location bar. The full image should open in your browser.)

This winter, a buddy and I decided we were going to make an electric car.

Why?

A. Mostly for an interesting project to pass the time. Also, there's a beer fridge in the garage.
B. Also because we both happen to like the concept of electric cars (and if you've ever driven one, you know about the "EV grin").
C. Plus, the price of gas is only headed in one direction, and we thought it would be cool to sever the ties to oil and still be able to drive around some.

The goals for the project are simple and modest:

1. Low budget (used parts where possible)
2. Low speed (we only want to be able to drive around in town, so 60 km/h / 40 mph would be fine)
3. Low range (the small city we live in is only about 10 km / 6 miles across. A 15-20 mile range would be OK, basically one trip per day.)
4. Low budget. (And did I mention low budget?)

The choice of car to use was obvious: some sort of Suzukiclone. Because they're cheap, plentiful, easy to work on, cheap, and most importantly, cheap. No, just kidding - most importantly: lightweight.

Light = more efficient. EV's don't usually carry much energy, so the same thing that makes a Suzukiclone attractive as a gas car makes it good as an EV for our purposes. Lightweight = better performance for less energy.

We would have preferred a MK 2 car, but they're just a distant, rusty memory around here. So the first car we bought was a '93 Swift 1.3:



For the electric parts, we bought a surplus 36/48V forklift from a local company, and had it trucked over to the house. We stripped all its parts. .. and then we came up with a name for the project: ForkenSwift...



The vehicle on the left weighs nearly 10 times as much as the one on the right.

Once the parts were out of the forklift, we sold the bones to a local scrap yard, and they came and trucked away the chassis.

Unfortunately, once we got picking & grinding at the Swift, we realized that the underbits were a little rustier than we thought. So we went out and found another cheap Suzukiclone with a little less rust: a 92 Metro. We've since mixed & matched the good parts from both cars into the Metro:



Cost for both cars: $250. (Extra parts sold to date: $170.) So far, so good!

More to come.

Kool!
But don't you mean Frankenswift?
Lookin forward to seeing this project get done.

no its forkenswift cause its a swift and a forklift

I know that I just am a big fan of young frankestein.

its a fork'n swift!

I like the approach with the forklift, did you get a charger with it too?

If you're going to need batteries, have a look at surplus UPS batteries are frequently discarded by maintenance policies at large institutions and data centers.

Right on both counts. (The second one is something like what I say when a wrench slips and I bash a knuckle.)

Really it's more Metro than Swift now, but the name stuck. And there's some Pontiac thrown in for good measure. (Why go electric when it's already a "hybrid" ) ...



Lihtan: no charger with the forklift, but since we're going to stick to 36 or 48V, we may just use 3 or 4 12v battery chargers.

FYI, for those who care, "dumb" chargers are not the proper way to recharge a battery on a regular basis. You really need to do some calculations and adjust both the volts and current as the battery SOC changes to avoid doing damage to the battery over repeated charge cycles.

As for battery source, the UPS idea is one, and there's also a truck service center in town that regularly chucks used 12V starting batteries. The trucks run 4 batteries each, and if one fails, all 4 are replaced. So we may be able to get some of those for cheap and/or next to nothing.

Flooded lead acid deep cycle batteries (like those in golf carts) are the way most EV's go for capacity and longevity. We may ultimately get a pack of those, if the car turns out well.

But our first pack will be used and cheap, so we're not too concerned with treating them gently with a proper charger.

Actually flooded lead acid batteries need a regular overcharge to equalize the cells. Dumb chargers are perfectly fine for these batteries - just keep them topped up. You don't need to compensate for state of charge - really. They're pretty forgiving. If you're going to float them, or you want some smarts - then you have to compensate for battery temperature. Golf cart batteries are great - thick plates and are designed for repeated deep discharges - and you can buy them at Sam's Club!

Gel cells are a completely different story and they're really not suitable for deep discharge, high current draw AND long life, not to mention they're pickier about charging.

We used new aircraft NiCd cells, 1.2V each, 40AH and built packs out of them. These were used for acceleration while our zinc-air batteries provided cruising power and to recharge the NiCds.

The Metro isn't an ideal conversion because although the vehicle is light - it doesn't have much of a payload carrying ability. Those flooded lead acid batteries are HEAVY!

We used DC traction motors from Advanced DC motors (they still around?) 1 small one, 2 small ones, or a BIG one depending on the application. We had a Curtis motor speed controller, though Motorola sponsored us one year and made a speed controller for a mini-van we built for Edison.

The boss had an electric dragster in his storage locker. Aircraft Nicads lined the sides, and there were 4 DC traction motors driving the tranny. Full size electric dragster - I never saw it run though (before my time there)

We did a CRX, Saturn SC-2, Dodge Caravan and at least one other that I cannot remember.

Here's a couple of our articles
http://www.findarticles.com/p/articles/ ... i_11010213
The next page talks about our Nicad pack and the race.


Here's a link to info on Solectria that converted geo metros to electric use.
http://en.wikipedia.org/wiki/Solectria_Force

Dave
(who used to work at DEMI, an R&D battery company where we converted a number of vehicles to electric use for testing and competed in a race against SOLECTRIA which retrofitted Metros to be electric back in the early 90's. I'm sure my story is in the archives somewhere.)

Hey Dave...



I've seen a number of people mention that before - and cheap too, around $50 US. Unfortunately, no Sam's club in Canada, so no cheap floodies...



Ah, but that depends on your definition of "ideal"

If your ideal is a "high performance" EV (long range & good acceleration) running on flooded lead acid, I'd agree with you because the pack would have to be large to get the needed voltage & capacity.

But if your ideal is a short-range, low speed, grocery-getter EV, the Metro will do just fine (6 or 8 batteries @ 60-70 lbs). (OK, maybe not in hilly San Francisco, but in the relatively flat town where I live, it'll work.)



Yup - they're still one of the most common brands used in conversions.

Thanks for the info.

Some vehicles that inspired Project Forkenswift ...

---

Greg's $800 US Yugo conversion ...



= 1640 lbs, 48V (4x12V), 35 mph top speed, 13 mile range

Classic DIY job. Parts made or scrounged from eBay and garage sales.

http://www.evconvert.com/article/yugo-greg

---

the CitiCar (a.k.a. ComutaCar) ...

<img src="http://www.aboatbuildingjournal.com/images/Sparky/formarkb1.jpg" width="300">

= approx: 1400 lbs, 48V (8x6V), 40 mph top speed, 30 mile range

These 2-seater electric "cheese wedges" were built in the US in the 70s and early 80s and have the distinction of being the most popular commercial EV to date (they sold a few thousand of them, most during the oil crisis). You can pick one up for between $1000 - $4000 US depending on condition.

http://www.didik.com/citicar.htm

---

Dr Larry's Metro Vert ...

<img src="http://www.evconvert.com/images/86.jpg" width="300">

= 1900 lbs, 72V, 70 mph top speed (right off the charger) - 50 mph is more normal, 40-50 mile range (not at 70 mph!)

This one cost about $4000 US to make - about 3 or 4 times more than I want to spend - but still a fraction of the cost of most custom EV conversions.

http://www.evconvert.com/article/larrys-ev

So, in March we bought a decomissioned mid-1980's Baker electric forklift for $500.

Batteries not included.

It was 18 km from the house, so we had to hire a trucking company to bring it:



Took the crash bar off it, hooked up 3 deep cycle batteries with jumper cables and backed it into the garage... well part way into the garage:



(That's what the neighbours looked at for about a week.)



(And that's what I looked at for about a week. Note the 20-ton ratchet action railway jack we borrowed to lift this puppy. 16,600 lbs without battery; 19,700 with.)

And after a week's tinkering a couple hours a day, here's the forklift loot:



clockwise from top:

- bunch of 3/0, 4/0 and other cables & control wiring harness;

- EV-1 controller manuals, and a box of miscellaneous goodies;

- Square D Co. direction lever;

- Square D potentiometer;

- two 8 inch pump motors and two pumps above;

- 12 inch drive motor;

- p. steering motor & pump;

- EV-1 control panel

For those who don't know about electric motor setup, the potentiometer is what you hook to the "throttle" pedal in an EV. It functions on the same principle as a lightbulb dimmer switch, or a volume knob, and tells the motor controller (the main card with the red stripe on the control panel) how much juice to give the motor. Most of the parts on the control panel we won't need - much of it is related to the hydraulics.

Once we were done, a recycling company came back for the stripped Baker chassis and hauled it away. Net cost for electric/electronic parts, after tax, shipping and scrap "refund": $343.95

Continuing the ForkenSwift story...

Part of May & June was spent fixing up the "host car". We swapped parts between the two cars to make one good one.

Mechanically, not too much was obviously wrong:

- driver's side window lift mechanism was bent out of shape, so the window wouldn't open (fixed)
- needed a control arm/ball joint (fortunately, the red car had relatively new ones on it)
- tires (put on a set of winter boots I had sitting in the garage)
- spliced a patch into a leaky fuel line near the tank
- a liberal application of muffler cement sealed up a couple of leaks in the exhaust system

The biggest problem was rust. It's amazing how when I looked underneath the car before buying it, I could only find one or 2 small holes. But by the time I finished grinding and peeling off undercoating, it was a different story...



Above: The passenger side was worse than the driver's side. And there were a couple of holes in the fender walls under the hood, ahead of the strut towers that needed fixing too.

So, I hacked off big chunks of the red Swift's rear quarter for sheet metal and learned how to torch braze:



Above: The red bit is paint i didn't burn/grind off the sheet metal from the red car. Most of what you can see is just the final layer tacked on to make the floor flat again. there's more work underneath that's fully brazed and more ... structural.)



Above: After... (fresh rubberized undercoating to catch fire next time the floor needs welding... I did set the rubberized undercoating on fire a few times doing this )

The good news is with the holes patched up and the mechanical fixes, my mechanic gave my work the thumbs up and we got the safety certificate.

Total cash outlay for our certified '92 Metro: about $79 CDN (when we factor in the purchase price of both cars, and the money recovered from selling some parts.)

Next stage: pulling the ICE, figuring out & fabricating the electric motor/tranny mount plate and shaft coupler issue...

Project is still going ... though slowly. Summer holidays, etc. You know how it is.

The adapter plate is cut (5/8 aluminum). This goes on the end of the tranny bell housing and the motor mounts to it.



Last week drilled the mounting/positioning holes, so it's ready to become the meat in the sandwich between this...



... and this ...



We're currently working on the electric motor-to-transmission shaft coupler. Probably the most important fabrication step in the project. It has to be aligned as close to perfect as possible or else we'll have vibration & bearing wear issues.

We were aiming to have the electric drivetrain put together and installed in the car in Sept. But that may be a little optimistic (based on the recent pace of the project ).

Man! We missed out on the deal of the century last month. The local cable company replaced its backup power supplies, and got rid of 100+ almost brand new 12v gel electrolyte batteries.

But we missed out! - Found out about it a couple of weeks after the fact.

Can you believe they sent them all to the recyclers? Arg.

But we do have a guy "on the inside" now keeping an eye out for similar opportunities.

Wow, this is the first time I've seen this thread. Hope you guys are able to get the car running. I'll be watching you!!!!!!
Josh

why not use a standard spider or sleeve coupling? the rubber spider or sleeve makes a dandy shock coupler to keep the initial shaft torque from hammering the gearbox. the rubber center also allows for less critical alignment letting the rubber flex a bit. instead of taking out a bearing, the rubber spider or sleeve becomes an inexpensive, replaceable wear item.

another thing that comes to mind is that when i was building an electric car in the mid 90s, i drew heavily from the engineering of a former MIT prof, walter kern (responsible for the early work in developing and testing true neutral steering and designing and building the saab quantum.) his early adaptation of front drive transaxle to electric power was sheer genius.

i was using a 1200 lb. monocoque chassis, front drive, a 98% mechanical efficiency gearbox (all ball bearings) and a fiberglass body with the lowest coefficient of drag available at the time. i also used 4.5j wheels with 155r 15 radial tires inflated to 55 psi.

missing out on the gel batteries probably saved you the pain and suffering of dealing with them. the best power storage option i've found is to find a maintenance and service company for forklifts who routinely replace the battery trays for those machines. they usually pull the trays when they still have 90% of their storage capacity. you can also find 220 vac single phase chargers on their way to the scrapyard from those guys, too. i got my 10 hp 90 vdc motor and control electronics from a scrapped forklift and went back every week to pick through their junk pile.

i had a 28 mile commute, 18 miles of it at 60 mph on a nice flat, rural interstate. that required a 3 hour and 20 minute charge cycle that i had automated by my home's energy management system to take advantage of lower off peak billing rates. i calculated that it cost me just over a dollar to drive to work each day.

i quit driving the electric car when it got really cold. although i had an electric element "defrogger" (it just wouldn't really handle frost ) when the temps got below -10 f the 20 minute one way drive was too uncomfortable.

i'm watching your geo electro progress.

t3: you have pictures of the car?



Actually, we're sort of doing that: adapting a Lovejoy jaw coupling with a rubber "spider" between the halves to fit the splined hub from the clutch.

But it's a bit complicated by the fact that on the other side we're using a pump motor with an internally splined shaft, so the corresponding half of the coupler will have to be adapted to use the shaft from the pump. We could dole out cash and order exact parts, and have an external shaft pressed into the motor, but that would violate the prime directive of spending as little money as possible on the project! It's a case of making do with what we have available.



We would have put up with the pain and suffering. Next to free, after all (see comment re: prime directive )



What did you do when it warmed up again?

We don't have any grand plans for heat either. Winging it - in case that's not obvious yet.

Project ForkenSwift update (so I guess the progress is going a bit slower than predicted. Whoops!)...

A machinist friend just completed the most important part of the puzzle this week: he modified our jaw coupler to mate the motor to the tranny's input shaft, and he also aligned everything perfectly on the adapter plate (see pic, above) and drilled holes to mount the motor to it.

I'm not totally sure the way he ended up putting a stub-shaft in coupler won't shear, but we'll test it hard once it's in, and try to break it. Just to be safe.

I know this may not be clear without pics to see what I'm talking about. I'll get & post some eventually.

So... theoretically, we now have everything we need to turn this puppy into an EV. A slower than a golf cart electric Metro! (Not sure, but we'll see.)

This weekend we start removing the ICE components.

Show 'n' tell time:

First couple of pics are of the coupler I described in the previous message...

Click to zoom...


Above: tranny with adapter plate & motor end bell in place



Customized Lovejoy motor/input shaft coupler



Eploded view... Ughh, bad choice of words.



The reason this was somewhat difficult: internal splines on motor (armature shown)

---

AND... the ICE is out!









---

This weekend: made up a cardboard version of our motor and reinstalled the tranny to check for positioning:




Above: reinstalled the transmission with the 3-cyl tranny mounts just to confirm earlier measurements. It all fits, but only 1 inch of clearance from the right frame rail (left side of the pic) = not enough room to re & re the motor without also taking out the axles & tranny. Not that we plan to be dropping the motor out every weekend, but you never know...



Much more comfy. Using the hardware from the 4-cyl car we get 4+ inches of clearance between the motor & frame. The only problem was the left side tranny mount (bottom right in the pic) wasn't located on the frame rail in the same spot in the 4-cyl car, so it had to be relocated...



The nuts the original mount must have been tack welded inside a the boxed frame rail during assembly, so we had to use a hole saw to gain access to the inside of the rail to put nuts behind the relocated holes. So it's a little less strong now. But lighter!

This week: need to make up a bracket to connect the right engine mount (lower left of 1st or 2nd pic) to the motor. Also, there's a bearing about a third of the way down the right axle that needs to be secured. (It was connected to the 4-cyl engine block.)

Oh, and install the metal motor. The cardboard one doesn 't conduct electricity very well.

Electric motor's in the car! Getting closer...



Above: Ivan's work - he adapted the original 2 "belly straps" that held the 2 motors in place under the forklift (shown at top, from under the beast) into an elegant 2-part motor mount bracket.



The e-motor shown installed on the car's original ICE mount bracket. The dimensions of the motor (length & width) are remarkably close to the ICE. Makes this stuff somewhat easier.



Above: shot from the back side of the motor, looking down and forward (firewall is bottom left, right shock tower @ bottom right) ... the hanger bearing (center) that stabilizes the longer right axle used to be bolted to the back of the ICE block. Ivan made up a couple of plates to connect it to the unused side of the tranny/ICE mount bracket.



Last thing we did was cut an inspection window in the top of the tranny bell housing so we can keep an eye on our coupler. This is the part of the works we think is most likely to wear out and/or break. Crappy pic due to glare off the adapter plate. But that's the Lovejoy in there...

---

Bought a used Curtis golf cart motor controller on eBay, and picked it up today. It'll be easier to set up for testing than the original GE EV-1 controller that came with the forklift. The EV-1 has better specs (like a higher current limit) & features (like a controller bypass circuit for turbo mode), but hooking it up is extra complicated.

So - we just need to tighten up a bunch of bolts and do a few other miscellaneous stuff. And there will be a dry run in the near future - the first test-spinning of the wheels under electrical power (on jack stands). If all goes well, next on the list after that will be the first electric test drive.

Well, after a couple of weeks delay in the proceedings to remove the motor, re-wire its field connections & realign the brushes so it'll spin the right direction for the Geo tranny (more info about that later)...

It's aliiiiiiiive...



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

awsome vid. nice music.

groovy man! shagadelic!

Sweet!

Thanks for the vid comments.

So, the reason for the long delay between the power-up vid and the previous post: the motor was turning the wrong way for the tranny input shaft, and it took longer than expected to sort it out.

Most EVs using DC power have 4 terminals on their motors: 2 for the armature and 2 for the field. Reversing polarity on either pair turns the motor the other way.

But the Forkenswift's motor only has 2 terminals, and was wired to spin in only one direction (as befitting its former life, turning one of the forklift's hydraulic pumps). I mistakenly assumed I could just reverse the polarity on the 2 terminals to make it run the other direction. Wrong!

Apparently (so I learned), series motors with just 2 terminals don't work that way. Reverse the terminals all you like, it's only going to spin in one direction. They're not like permanent magnet motors.

The fix was fairly straightforward: open the motor and reverse the field wire connections internally. Once done, the motor spun in the "right" direction. But now it was running noticably slower than before.

( FYI if you want to see pics of the motor and its guts:
http://metrompg.com/offsite/baker-pump- ... etail.html )

A check with the folks on the EV discussion list and I learned that the slow speed was probably from brush timing that was now retarded for the new rotation direction. Who knew that electric motors have "ignition" timing kind of like gas motors? Learn something new every day.

Long story short: I had to rotate the plate that holds the brushes, and advance them for the new direction the same amount they were advanced for the old direction.

I also learned I'm not the only person who's faced this brush timing issue. Here's a quote from a guy who ran his motor (in a Honda conversion) in the opposite direction without changing his brush timing. Burned up his brushes & commutator (the surface on the armature that the brushes press against):



So, that's probably too much boring info for this thread, but maybe someone cares.

---

Next on the to-do list:

- go to the junk yard and get half a dozen battery cables with terminal clamps so we can hook up a 48v string of batteries (without using jumper cables, like last time we did it ). Then we can try out the used golf cart controller I got on eBay ... and maybe try driving the thing!

Lastly: here's a clear shot of the motor in the car.