[brian_]

If you are going direct drive then you won't need a clutch pedal - with only two pedals you don't need as much space in the footwell - and the passenger never needed that much anyway

I would cut out the transmission tunnel and make a lightweight removable tunnel that the motor would live in

I understand the logic of enlarging the tunnel and forcing the occupants' feet outboard... but the tunnel is probably significant to the unibody structure, so I'm not sure some non-structural piece would be a suitable replacement.

Then you have the whole engine bay for your batteries

It certainly would be good to get the motor out of the engine compartment.

 

[brian_]

On my car I closed in the bottom completely - helps the strength and you don't need to let the drips escape - but that does mean that you have to get at the motor from inside the car

In an open-top car and with the access panel, that wouldn't be unreasonable at all.

 

[brian_]

The main body of the motor is around 280mm diameter by 330mm long. It is around the same diameter as the gearbox bell housing at its largest. Without any modifciation to the structure I should be able to tuck the motor in so that the front face is around 5-6 inches further back than the engine/gearbox interface.

If I understand this correctly, the output face of the motor, which is the motor's "front" face but is at the end of the motor toward the rear of the car, will be at about the current rear of the transmission bellhousing. That puts about 127 to 152 mm of the 330 mm length of the motor behind the original engine/gearbox interface, and the rest ahead of it. That sounds reasonable.

It would be really nice to put the motor entirely in the transmission tunnel, but without tunnel modification (or a custom car build) this rarely turns out to be viable.


That is one impressively short engine. I wish I could find something that short to swap into my Spitfire, but I don't think I would be happy with a Kei-class engine for that car. Thus, the consideration of other options, including EV conversion.

 

[brian_]

When I'm looking at battery packaging how much thought needs to go into cooling/heating? I believe most OEM EVs have the ability to cool their batteries. Is this generally done amongst DIYers as well? I'm just thinking about whether I need insulation and space for cooling around the batteries. I'm not planning to use the car in sub zero temperatures (celsius)...
My current plan is not to worry about heating or cooling but obviously just to make sure they are protected from water and dirt.

If you really don't want to worry about cooling the battery, use one which is designed for use without active cooling in the original vehicle - the common example would be the Nissan Leaf. Heat for a Leaf battery is electric, and optional.

I think batteries which are designed with integrated liquid thermal management - such as Chev Volt and Tesla - are usually hooked up to a radiator even in DIY installation, but the sophistication of the installations varies.

There's a forum thread on Working with Tesla Packs which started in November 2016 and is still active at 215 posts (and counting), with many side-tracks. The subjects of thermal management and housing are not trivial... happy reading!

 

[brian_]

Attached are some photos of the batteries. Each of these contains 48 cells.

So that a 4x12 array of the Swing cells, each of which is probably a pair of 18650 cells wrapped together and connected in parallel, so essentially an 8x12 array of 18650's... that means something greater than

65 mm tall
8 x 18 mm = 144 mm wide
12 x 18 mm = 216 mm wide
and 48 x 90 g = 4.3 kg
occupying about 2 litres​

... per module

Capacity is 4400 mAh per Swing unit, so 781 Wh per module (nominal).

I was planning to connect 16 cells togther in parallel and then 80 of these batteries together in series.

So 80s16p Swing units, or 80s32p 18650 cells, for 296 volts (nominal). That's 1280 units (or 2560 18650 cells) in 27 modules each configured as 3s16p, totalling 115 kg of bare cells plus module overhead, plus enclosures.

It's actually 26 2/3 modules, so perhaps 81s16p Swing units (or 81s32p 18650 cells) for 300 volts (nominal). That's 1296 units (or 2592 18650 cells) in 27 full modules.

Total volume would be about 60 litres, which doesn't seem unreasonable at all. The challenge is that it is a lot more difficult to pack a couple dozen rigid bricks than the same volume of liquid in a tank which can be shaped to conform to the available space.

I wouldn't expect all of that to fit in the front with the motor and controller, and the space used by the fuel tank is not going to conveniently fit the modules. I'm looking forward to seeing the pack design.

 

[brian_]

Before purchasing the Cappuccino I had considered a kit car or scratch built. I'm half considering this again using everything out the Cappuccino other than the bodyshell.

Since it's unlikely there's a kit in existence designed to use Cappuccino components, this would presumably be a scratch-built (although potentially following an existing general frame and body design, or a general pattern such as the "LoCost" style). Are the Cappuccino components really what you want in a scratch-built (maybe they are, as they would be light, have a sophisticated design, and use modern parts), and is that the best way to get value out of the car you have?

 

[brian_]

I've always fancied designing a car from scratch.
...
Another reason I'm tempted by the scratch built route is that I can design it around my components. I wouldn't have all the issues trying to cram the motor and batteries into spaces where they don't fit.

...
All the important bits are bolted to the subframes so it would be relatively easy to build.

While I hope the body repair works out, this sounds like a great scratch build.

I did a quick web search to see the Cappuccino chassis components, and they do look nice for integration with a custom tube frame, at least at the rear.

If building your own body and tubular steel frame, you might even be able to fit a small complete drive unit (motor with transaxle) into the rear.

 

[brian_]

They’re actually in a 6 x 8 array currently connected 6s8p so I just need to reconfigure to get 3s16p within each module. Although they are almost identical in size to two 18650 cells apparently they are actually one oval shaped cell.

I did some measuring up at the weekend and I think I can fit 12 battery modules in the engine bay in front of the motor, 10 modules in the boot, and 3 modules in the fuel tank area. This will give me 25 modules which should be enough.

So 25(3s16p) Swing cells, for 75s16p overall, for 278 V (nominal) at 70.4 Ah and 19.6 kWh (nominal). Sounds promising.

 

[brian_]

Putting the batteries in the passenger space would solve all of this. The only downside is the loneliness!

Quite a few racing cars have been built with a side-mounted engine, and it works well, for the same reasons (central mass position, large volume available) and others.

Only you can decide how important passengers are. Do you want to let people experience your car without letting them drive? If you do want to let them drive, do you want to ride along?

Here's an out-of-the-box idea: build it with the battery in the passenger seat area, get the EV system sorted out, then as a follow-up project do a scratch-built structure and body with more room for the battery and transfer all of the components (possibly including the Cappuccino chassis bits) so it can have two seats.

 

[brian_]

As someone who would want to make a Hayabusa V8 Cappuccino...

A Hayabusa V8 Cappuccino sounds interesting! Two hayabusas side by side?

Not likely two engines; it would be too wide and combining the outputs is awkward. There have been V8's built by welding together two blocks (and of course the original heads, etc), and some built with a custom V8 block (using two original heads, etc). Radical offers one of their cars with their production Hayabusa-based V8 from Powertec.

Motorcycle engines typically don't suit cars well anyway, due to the integrated and inappropriate transmission, so it makes a lot more sense to make an automotive block to use the motorcycle heads, pistons, and maybe connecting rods... and one transmission.

 

[brian_]

On this note, it baffles me that car manufacturers still haven’t produced a simple cheap EV (or an electric sports car for that matter). They’re all focusing on electric SUVs that cost the earth and have every gadget imaginable. I’m sure there would be a market for a basic small electric hatchback (in the UK at least) with a decent range at a sensible price (£15k not £30k+). Something like an electric kei car! I’m convinced it must be possible.

A low-performance low-speed car is possible, where it is legally allowed. The Euro "heavy quadricycles", such as the Renault Twizzy, are a good examples.

A basic car of conventional size and performance is much more expensive in battery-electric than engine-driven, so the £15k is just not possible, unless it is heavily subsidized. The vehicles which physically meet this description are much more expensive, and are generally purchased only where they are subsidized by government programs. Even a Kei-class vehicle (such as the Cappucino) is requires a substantial powertrain and so isn't cheap.

So to make any money - or even come close to covering costs - manufacturers target the high-feature market. I don't blame them. Ideally, the moderate volume of production of these vehicles (the fancy ones and the subsidized ones, and even the fancy subsidized ones) will bring costs down to the point that more reasonable EVs become economically viable.