[marc02228]

It's time to start my own thread, because I decided which donor to take and finally have some progress.

I'm gonna convert a Miata, even though I didn't buy one yet.


Your skill level with auto mechanics and fabrication.
I'm an electrical engineering student. Yesterday I finished writing my bachelor thesis about BMS and Capacity/SOC measurement methods.
So I'm quite sure, what to do on the electric side. Build many thinks before I started studying...
On the mechanic side I'm not sure how capable I am, when it comes to cars.
I'm sure that I can do it but I want to do build a good car, get a registration, etc.
So I might use some local help.

The range you are hoping to get (how many miles/charge).
Not sure about that. I purchased about 400 A123 20Ah cells, which I'm testing right now. The problem will be the weight of the car, because I have to stay as close as possible to the original curb weight due to the registration.
~100km would be nice..

What level of performance you are hoping to get.
Actually It's the same answer then the previous one. As much as possible. I want to use the Kostov 11 alpha (higher efficiency). But the problem is the weight, again. So I might switch to the Kostov 9 220V eventually.


What parts you've already considered, if any.
-Miata NA
-A123 pouch cells (up to 400)
-Kostov 11 alpha or Kostov 9 220V
-Soliton
-BMS (don't know which one, since i need a EMI certificate for it) actually I
wanted to use Orion or EMUS.
-Charger: didn't think about it, yet


Next Steps:
-continue cell testing
-get a donor
-throw out all the ICE parts and calculate what components can be used (based
on the weight difference)

I'm thankful for any criticism or suggestions.

Cheers
Marco

[njloof]

Marc0, good to see you. Our plans look similar so I will be paying very close attention

[Ziggythewiz]

Welcome Marc0! Curb weight shouldn't be much of an issue with a small amount of A123, what are they, 1 lb each?. Should make for a great low range vehicle.

[drgrieve]

To answer the question about the K9 220v - how much power?

Well I guess the answer to that is - For how long?

My thought is that the Solition Jr is a good match for this motor. Put your 400 A123 into 5p80s configuration and by using liquid cooling to the controller you can see peak of 600 amps. The A123's will keep voltage above 220v at 600amps so you should see 130KW power input to the motor.

More than enough for street use.

[somanywelps]

A123's have blistering 20C discharge though... it'd be a shame not to use a soliton 1 and beefier motor.

Can you explain the weight law?

Edit: Here is some math I did for the the motors for my build. "per 2" means dual motors in series. That miata is begging for dual 9"s with a soliton 1 due to that sweet A123 discharge rate.

Plug in Amps for 'x' and you'll get torque IN NEWTON METERS. I did the math down there in FOOT POUNDS.

To be clear:

Soliton 1 (1 motor alone or 2 motors in series)== 1000A

Soliton Jr (1 motor alone or 2 motors in series) == 600A burst (as long as cool, water cooled is long), 500A constant

Soliton 1 (2 motors in parallel) == 500A

True KW power is approx Amps*Volts*.87(eff.)=KW

Kostov 11 250V:

210A 77nm
500A 215nm

0.47586206896551725x + -22.93103448275862

334ft*lb per 1 @1000A
668ft*lb per 2 @1000A
193.67ft*lb per 1 @600A
387ft*lb per 2 @600A
158.57ft*lb per 1 @500A
317.14ft*lb per 2 @500A

Kostov 11 Alpha:
295A 500A
100nm 194nm
0.4585365853658537x + -35.26829268292684

312ft*lb per 1 @1000A
624ft*lb per 2 @1000A
177ft*lb per 1 @600A
354ft*lb per 2 @600A
143ft*lb per 1 @500A
286ft*lb per 2 @500A

K9 220V:

176A 500A
44nm 176nm

Torque=0.4074074074074074*Amps + -27.703703703703695

300ft*lb per 1 @1000A
600ft*lb per 2 @1000A
159ft*lb per 1 @600A
308ft*lb per 2 @600A
130ft*lb per 1 @500A
260ft*lb per 2 @500A

And I don't remember where I read it, but the K9 220V was listed for 600A overload for 10s, so 1000A is probably 5-6s. Do not do it for longer than that.

[marc02228]

Originally Posted by somanywelps

A123's have blistering 20C discharge though... it'd be a shame not to use a soliton 1 and beefier motor.

Can you explain the weight law?

Wow, thanks for posting all the calculations.

Yes, thats exactly my thought. I actually have to use a Soliton 1.

If you convert a car to electric in Germany, the electric curb weight should be at most the same as the original ICE weight. Due to stability of the chassis.
I heard from someone from the TUV, its not a problem, if you are a few kg heavier. But the heavier you get, the more likely you have to do some stability tests.

Thats why I am looking for a donor with a 1.8l engine because curb weight is 1040kg. Does anyone know if the 1.8 Miatas were built without Power steering, too?

Same with the power. The electric power shouldn't be above the ICE power. If I install a K11alpha, and cut back the power at the controller I hope it should be cool.
I don't know, what they think, if I install 2 motors (even if two smaller ones).

two K9 220V would be an awesome setup, but it's still 20kg more than the K11alpha.

In the attached picture I did some calculations for acceleration.

I put in a weight of 1150kg. So 1060kg for the car and 90 for me
Didn't find the Area... so I put in 0,645 for Cd*A, but it doesn't make much difference anyway.

0,85 is efficiency of K11alpha. Won't be this efficient at high powers but still better then 100%

One question I asked me, when I saw the plot:
-How does Tesla get an acceleration of 3.9s out of 180kW. The Roadster definitely heavier then 1150kg...

[Ziggythewiz]

Tesla uses a $30K+ battery pack, plus custom everything else. No sane DIYer can match that (though you can beat it for less).

If you're not allowed to exceed the original power, you're probably best off with a single 9". Do you know how they verify that? I would expect them to count your total voltage and controller capacity rather than trust you to limit it in the controller.

[drgrieve]

Weird requirements? Just tell them the batteries are rated for max 3C. Lol.

I would think that that could only take the nameplate power of the motor (ie it's continuous rating), not some arbitrary peak power which is really up to the owner to decide what overload that want to put the motor through.

In Australia we also have many requirements to pass - a big one is the 20G crash rating for the battery boxes. But at least the requirements are documented and freely available. Ask for the guidelines in writing. Otherwise if someone verbally gives you advice, make them send you an email so you have it in writing.

[marc02228]

Originally Posted by Ziggythewiz

Tesla uses a $30K+ battery pack, plus custom everything else. No sane DIYer can match that (though you can beat it for less).

If you're not allowed to exceed the original power, you're probably best off with a single 9". Do you know how they verify that? I would expect them to count your total voltage and controller capacity rather than trust you to limit it in the controller.

It's physically impossible to accelerate 1200kg from 0-100 km/h in less than 4 seconds (even with a million $ batterie) with just 180kW
So their motor has to have more power, at least for a short periode of time.

[Ziggythewiz]

My calcs say it requires 116kW to do that. Add in inefficiencies and you're easily around Tesla's numbers.

Also, the motor does not provide the power, the batteries do. The controller converts that power and the motor passes it to the drivetrain.