Project Soylent Green

Inquisitor

100 µW
Joined
Aug 17, 2012
Messages
9
I would like to introduce my project and goals. Soylent Green is a minimalistic commuter EV to be built relatively cheaply. I know it would be far easier to take a donor vehicle and convert it, but an important aspect of the project is my enjoyment of the design and building process and just figuring out things. I’m fortunate in that my job is only seven miles away in an area that is relatively flat. Basically, I can ride a bike to work (and have) but the traffic is very bad all the time and I feel like I’m taking my life in my hands every time I try it. So without further explanation…

· A minimal weight and price vehicle
· Reverse Trike configuration
· Carries two people occasionally
· 45 mph top speed
· Enclosed for aerodynamics and keep out the rain
· one hour duration or approximately 20 mile range


I’ve started accumulating some of the hardware necessary and running numbers. I currently have…

· the motor 130 Volts, 10 Amp which claims to put out 2.5 horsepower continuous. I’m hoping for closer to 4hp for short periods.
Motor1.jpg


· I’m going to use a microprocessor (Arduino) for many things, including motor controller, tachometer, speedometer, BMS, temperature sensors, and charge controller and anything else I can think of along the way. Here it is bread boarded to control and LCD screen and as a tachometer. Its showing the motor at 37 Volts running at 1248 rpm.
Controller.jpg


· The rear wheel, tire, brake disk, brake caliper, rear sprocket, bearings, spacers and axle
RearWheel.jpg


· Front suspension off a Honda ATV which includes brake disks and calipers
FrontSuspension.jpg


I’m not really an artist, being more of an engineer so I’m showing the guts first and will do the “artist impressions” later once I’m on a roll. Here is the first cut at the chassis and suspension system. At the moment, it is not fully triangulated around the passenger compartment and is not intended to be fully representative. I plan to start doing some Finite Element Analysis in the next couple of days to fully analyze the structure. But I want to see how weak this minimal, first cut is.
SG1-6.jpg



Cost and Weight summary.

Item, Cost, Weight (lbs), Notes
Motor, $50, 15.8, 130 V, 10 Amp, 2.5 Hp
Full front suspension, $162, 28.5, Includes brakes
Front wheels and tires, TBD, 30.0, Not found yet
Rear wheel, $80, 20.5, Tire, brakes and misc.
Microprocessor, $20, 0.1, and misc. electronic parts
Rear suspension, ?, 4.3, Steel tubing already had
Minimal chassis, ?, 30.4, Steel tubing already have most




Totals $312, 129.6 lbs, More to come.
 
I hope it’s not a problem that I posted this on both the diyelectriccar.com and the endless-sphere.com. One forum seems to do things on a far larger scale and the other does smaller and lighter. I’m hoping that I can get constructive feedback from both camps. I think both have a lot to offer.

Thanks.
 
dnmun said:
if you have articulated front hubs, why didn't you just keep the entire front suspension(frame mounting points) and half shafts and drive the vehicle through that?

Just a guess but I'm thinking they're from an ATV and too close together, he needs a wider track. There comes a point where trying to hang on to a bit too much just causes problems. If he does have the frame it could still help to emulate that.

tumblr_m6mcttXDPk1ra2s2ao1_400.jpg
 
If you have not already done so, I would suggest you do some thinking & calculating about your battery options, before you get too fixed on frame design.
Size, space, weight, location, etc etc will be fundamental to the basic design... not to mention costs ! :shock:
 
dnmun said:
if you have articulated front hubs, why didn't you just keep the entire front suspension(frame mounting points) and half shafts and drive the vehicle through that?

I bought the suspension on eBay and it didn't come with any frame pieces. It was a rare find to find it with the brake calipers. They alone would have gone way over $100. Dauntless is correct about the frame anyway. On that model of ATV, the frame between the A-Arms is only about 6" wide. I'm closer to 30". Also, it was a 2WD ATV and I wouldn't have wanted to try driving through a differential with this little motor. Too much friction, torque steer... and problems all around. I prefer the simple rear drive better.
 
Hillhater said:
If you have not already done so, I would suggest you do some thinking & calculating about your battery options, before you get too fixed on frame design.
Size, space, weight, location, etc etc will be fundamental to the basic design... not to mention costs ! :shock:

1. I've long punted lead based batteries of any type.
2. I next started looking at NiMH D cells as the next cheapest available. Doing reality checks on the Internet, was showing that 50C discharge rates are achieved with special batteries, but 20C seemed to be a realistic “norm”. Heck, my Internet research indicated that 5C for AA size batteries was no big deal. All I needed was maybe 2C. I thought I was good to go. Doing eBay searches mislead me in to believing that it was “easily” doable with the Chinese 10/11Ahr D cells. So I ordered a set of 10 Chinese NiMH D Cells. They came reasonably fast in 11 days. I’ve done several charge/discharge cycles and balancing. My first test, I got dismal results at a mere 3 amps (0.33C!!!!). I’m doing some more tests and will be documenting it on this forum when I get it done. Anyway, NiMH looks to be out.
3. So, it looks like I’ll be using lithium based something. It will be light and I wasn’t too worried about placement and weight. It looks like 30-40 lbs would cover my needs. Unfortunately, they’ll cost more than everything else in the vehicle combined! :(
 
in the end the commoditization of the lifepo4 prismatic pouches will prove to be your cheapest long term option. the offers from chinese sources on ebay now have reached new lows. there is a new guy on ebay sun-thing28 and he has 48V packs delivered in the US for almost half of what ping is charging now so i think you should be able to source cheap lifepo4 soon and this may be the low if the chinese push the build out of more EV options for their locals.
 
I've just started looking at these LiFePO4 batteries. These ones that sun-thing28 sells... they use the little flat cells that look like freeze dried food. Aren't these the same cells they use in production cars? And if so, how come cars have all kinds of liquid cooling around them, yet he's selling them all tightly wrapped up in brick? Won't they go nuclear or something? :shock:
 
If 130V, 10A is its rating, than that's what it is certified to deliver continuously. 130V * 10A * 75% efficiency = 1.3hp

You had better keep an eye on the operating temperature if you want to drive it at 4hp.

Chalo
 
Chalo said:
If 130V, 10A is its rating, than that's what it is certified to deliver continuously. 130V * 10A * 75% efficiency = 1.3hp

You had better keep an eye on the operating temperature if you want to drive it at 4hp.

Chalo

So, this is more Chinese marketing?

MotorSpec-1.jpg
 
Inquisitor said:
I've just started looking at these LiFePO4 batteries. These ones that sun-thing28 sells... they use the little flat cells that look like freeze dried food. Aren't these the same cells they use in production cars? And if so, how come cars have all kinds of liquid cooling around them, yet he's selling them all tightly wrapped up in brick? Won't they go nuclear or something? :shock:

sounds like you need to spend some time reading up on the "Battery tech" forum before you make any expensive decisions.
 
Inquisitor said:
Chalo said:
If 130V, 10A is its rating, than that's what it is certified to deliver continuously. 130V * 10A * 75% efficiency = 1.3hp

You had better keep an eye on the operating temperature if you want to drive it at 4hp.

So, this is more Chinese marketing?

MotorSpec-1.jpg

Beats me. Maybe they've figured out a motor that is 150% efficient. But I'm guessing either the watts rating or the HP rating is erroneous. Maybe both.

Chalo
 
Perhaps 2.5 is the HorsePuckey factor? ;)

Even if you used the full 130VDC on it from a power supply at 10A, no PWM, that's still only 1300W. One horsepower is about 750W, so it isn't even 2HP, and that's assuming 100% efficiency in converting watts in to HP out (which as Chalo points out in an earlier calculation, it won't).


Maybe what they mean is HP is a peak rating, and the other ratings are continuous; but I doubt it. I don't know what they actually mean by their ratings, and I wouldnt' trust any of the ratings on there without testing them, including monitoring commutator, brush, winding and magnet temperatures. (all but the magnets and brushes will be difficult to monitor as the rest are spinning on thsi type of motor).


It looks like a typical treadmill motor, most of which are rated at somewhere around 130VDC at 10-20A. Some are rated continous whcih means they can run 24/7/365 at that power level, some "treadmilll duty", whcih means they'll run ok at that power level for as long as a typical person could run on the treadmill, which is likely less than a couple of hours.

Either way, it probably cannot do what you want it to at those high power levels without pretty extreme cooling; high volume airflow thru the casing at a minimum; you'll probably want to filter the air before sending it thru there, and it's going to be very loud (probably like a vacuum cleaner or worse, to move enough air to keep it cool enough). You'd need to instrument it and test it to find out what kind of cooling it's really going to need, so you don't melt it down or overheat the magnets and demagnetize them, or burn up the windings or commutators.

There's other ways to cool it that would be quieter, but air cooling is a lot cheaper and less complex.

If they have overrated the motor and it already runs hot at the original power level, it may not be usable at higher levels. if they underrated it and it stays very cool at the original levels, you have a better chance of using it at the levels you want.

Some motors run fine at much higher power levels, for long periods; some do not. Over at DIYEC forums you'll find examples of each in various conversions, typically using forklift motors overvolted by up to several times.
 
Inquisitor said:
I've just started looking at these LiFePO4 batteries. These ones that sun-thing28 sells... they use the little flat cells that look like freeze dried food. Aren't these the same cells they use in production cars? And if so, how come cars have all kinds of liquid cooling around them, yet he's selling them all tightly wrapped up in brick? Won't they go nuclear or something? :shock:

nope, the prismatic cells like this can only produce about 2C and he builds them just like ping does. the pouches are assembled into bricks and the tabs are soldered into the connectors and they put these hard plastic shields around the entire assembly and then wrap tape around all of it to hold it together. his prices are just dramatically lower than li ping. but i am not dissing ping at all, he does build good batteries, and there is no way to know how reliable this guy is but from looking at the pictures of the pack it seems he has essentially the same components that ping uses. he ships by surface so the shipping is much cheaper so the pack is cheaper delivered.
 
I wanted to give an update on Soylent Green. I’ve gone through a little research looking for a free/open source Finite Element Analysis program. I settled on Z88Aurora (http://www.z88.de/). Although, it is sponsored by what appears to be college professors and their graduate students at a Germany university, they have been very helpful and are able to communicate with me in English. Also, all the documentation is available in English. Overall, it’s been a great experience. Per one of their primary goals, it is very easy to come up to speed using their software. They even have on-line video tutorials.

It has been twenty years since I’ve used FEA. I have been able to read through their documentation, build test cases and then build a first model of Soylent Green. I attribute this directly to their software. If you have ever used FEA software it is often a monumental effort just to get to first base… and that is after you have spent large sums of money and read through thousands of pages of documentation and/or a willingness to delve into Linux.

This first Soylent Green model is not for evaluating strength. It is to simulate the suspension geometry and motion it will go through in different load situations. For instance this video is under normal gravity, at maximum braking and maximum cornering. I need to make sure of the geometry for things like comfort, without bottoming out, tire position for cornering, and other things. Since springs “deform” a great deal as compared to steel tubing, only non-linear FEA programs can handle this situation. Z88Aurora, makes that very easy to accomplish.

Anyway, here is the video of the first model in action: http://screencast.com/t/ZsykEVolGrar
 
Hillhater said:
If you have not already done so, I would suggest you do some thinking & calculating about your battery options, before you get too fixed on frame design.
Size, space, weight, location, etc etc will be fundamental to the basic design... not to mention costs ! :shock:
Yea, second that on making sure you've fully integrated the battery into design, including substantial protection for the pack for collision. You'll be investing as much in the battery as for the rest the vehicle. Still recommending the A123 AMP20 cells that are available right now (but will disappear soon). Or you could wait for A123 to release their new EXT start/stop battery.
 
I've pretty much resigned myself to those A123 20AHr packs and have geared the front suspension box to hold and protect them. I needed more weight up front anyway. There's plenty enough room to hold about four times what I need for the range. And, yes, they'll be about twice the cost of everything else in the vehicle combined.
 
Back
Top