200SX conversion

Post by **MikeG** » Wed, 24 Sep 2008, 16:41

Has anyone heard of a conversion that fits the motor direct to the rear diff, mounting it under the back seat facing backwards?

Any comments on this? Ovbiously this would have electrical safety issues to be sorted out, and perhaps cooling issues, but it makes sense the ditch 30kg of driveshaft if my motor is going to be small enough.

I'm considering building up an Electric s15 (not, not my current ride) so I'm doing the homework on electric powerplant, so yes the AC powerplant should be capable of up to 7000rpm if I connect it directly to the drive shaft - which will yield a top speed of around 210km/h (plenty).

This would make for interesting weight distribution too... With loads of battery space (liIon) low down and behind the front wheels in the engine compartment, and a new cargo space on top (in theory).

I dare say it would be possible to modify the back seat so that it can be retained, although it will probably not fit as-is even with the back seat's bucket shape with pronounced center ridge.

According to my father, retired lift engineer, there are a number of interesting AC motors out there at the moment, some of which are flat rather than cylindrical, so maybe this will be an option.

MikeG

Post by **Richo** » Wed, 24 Sep 2008, 22:26

Flat? you mean a pancake motor.
More of an issue than electrical would be motor mounts to handle the torque.

Post by **MikeG** » Wed, 24 Sep 2008, 22:48

I dare say some cage development in the underseat area would be required, for mounting purposes.

I've been doing the research on some of the packages available, such as the Azure AC55, and it looks like I'd need dual gears or a change of final drive ratio to get the car over freeway speeds (for racing). The std final drive tops out at 115km/h due to wind resistance, whereas a 1.6 final gear ratio gets to 145km/h at peak power output (2000rpm).

Given the diameter of most motors, I'd have to look at repositioning the rear diff to allow the drive shaft to elevate to the motor, so the motor doesn't hang down 6" below the drive shaft.

Post by **Richo** » Wed, 24 Sep 2008, 23:57

AC55

Sure you don't mean the AC90.

Post by **EV2Go** » Thu, 25 Sep 2008, 00:34

If your tailshaft weighs any where near 30kgs you need to substitute it for one NOT made out of lead

I think the S15 would make a good donor vehicle but unless you have one with a clapped out SR20DET in it, your not going to find them cheap to buy.

I thought about buying a S15 after my 180sx got written off last year but no one sells just the body at a cheap price.

You might need to also reconsider your expectations, by the time it reaches 200kph you will need to recharge your batteries it will suck the juice that fast.

Post by **Richo** » Thu, 25 Sep 2008, 02:49

Another reason not to directly connect the motor to the diff.(no tailshaft)
The back wheels/ diff assembly will be moving up and down with the road.
Imagine going over a speed bump with a 190kg motor strapped directly to the diff.
That would be nasty.

Keep the tailshaft.

Post by **zeva** » Thu, 25 Sep 2008, 04:01

FWIW the 200SX has IRS so direct coupling to the differential wouldn't be adding any sprung weight.. But yes, it'd certainly be a bad idea for any vehicles with solid rear.

MikeG wrote: .. if my motor is going to be small enough.

I think this is the biggest issue with coupling the motor direct to diff, that the diameter of (most) motors is greater than the diameter of your diff, so is potentially the lowest point on the vehicle and hence at risk of damage.

I was originally planning something like this for my MX5, using two 6.7" motors (small diameter motors, but quite long) coupled straight to the diff and extending down the transmission tunnel. But even such small motors didn't quite fit in the drive tunnel of an MX5.

Also you have to make very sure that the motor housing and diff housing are *very* securely coupled so the motor shafts don't experience any transverse loads.

Anyway, how to work around the diameter issue.. Raising the diff a little to compensate for the diameter of the motor would adversely effect CV joint angles, maybe not a showstopper but not ideal. Rotating the diff a little so the motor is angled up.. may work? But would definitely eat into your back seat space and need some panelwork on the drive tunnel.

I think Matt Lacey was scheming something like this for his MR2 conversion.. Matt?

Post by **Richo** » Thu, 25 Sep 2008, 04:47

zeva wrote: FWIW the 200SX has IRS

Ahh ok - that's nicer and more doable.

Post by **MikeG** » Thu, 25 Sep 2008, 16:43

Thanks for the helpful replies guys. I slept on what I said yesterday, and realised that an AC55 would be pushing schmitt up hill (I'm used to 220kW now (180kW to the wheels) so moving to an AC55 would be like undoing all of the modifications and the turbo - why bother?!

I'm trying to discover why the EV community goes with DC motors all the time, given their extra cost and lack of motor braking and regenerative capability. Sure the controllers are cheaper for DC, but that's due to the added complexity of the AC motors. The Tesler and most hybrids use AC, so my gut feel is that DC is a more simplistic way for owner converters to do the job.

Given my penchant for elegant solutions, I'll be going for the most bang per buck, and simply engineering around the shortfalls.

I'm now considering retaining the drive shaft (or taking out the solid steel one and replacing with a carbon jobbie) and going to a larger motor. I want the capability to cruise around at 110km/h, and do short spirts to 180km/h if I so choose.

Ian, we had a presentation from one of the guys in Canberra Branch about your car. I think you've done well but I have a few things I'd do differently.

1. Battery location - I'd want to keep the weight as physically low as practical, so I'm thinking that a custom battery tray in the boot like you've done, but I want to avoid placing them anywhere but between the axles. A tall order I know, but what with the fuel tank being located under the back seat, and room behind the back seat, it seems like the logical location.

2. Curtis controller? No thanks, I'm thinking something seriously over-engineered to last... If I'm pulling say 20A, then I want a controller that handles 50A. I know you're designing your own controller, so I eagerly await news of how it stacks up

The alternative is to put the motor deep in the boot and link to the diff via a flexible drive shaft (I heard that someone used this to avoid mating the motor to their gearbox directly) - this frees up the under side of the back seat for batteries...

Unlike the MX-5, the 200SX has a fair few places it can fit batteries apart from the boot and under the bonnet. Though, it's still fairly slim pickings...

MikeG

Post by **acmotor** » Thu, 25 Sep 2008, 17:42

IMHO if you start from the premise of lots of action for your money then you have little choice other than DC at this point in time.
Then you too can build another retro EV !

Sorry, unfair, use lithium batteries and get on with it. You will be as happy as your budget allows.
Controller --- think 2000A.

Post by **MikeG** » Thu, 25 Sep 2008, 17:47

P = V x A

P = 114 x 2000

P = 228 kW

I wish... But good point, I just picked 20A out of the air to make a point...

Post by **antiscab** » Thu, 25 Sep 2008, 18:00

on my mr2 conversion i have a really big structural beam in the way between my engine bay the the centre tunnel where the fuel tank used to be.
a real shame, as the centre tunnel is huge. 2x warp9s would fit, but alas, i will have to settle with putting 48v worth of 90s in there instead.

i wont be tilting my diff any, i replacing my diff with a BMW E30 unit (as it has a speedo hookup built into the diff).
the diff will be at the original angle, so no tilting for me.

Mike:
If you are thinking of performance, try and get one of the last zilla 2k. These will do 2000A at 320V, 700A continuous.
they cost aroudn $6k by the time they get here.
oddly i didnt have to pay any tax on mine (usually you would have to pay gst on anything you import over $1k.)
atm, they are the most powerful controller there is, and they are going out of production.

big power from batteries is either very expensive, or you have very short range.
200kw out of thunderskys requires a pack that weighs around 600kg and costs $30k. (but your range would be in tesla territory)
200kw out of optima yellow tops is 300kg and $5-6k, but range is only around 30km.

Matt

Post by **acmotor** » Thu, 25 Sep 2008, 18:03

Almost.
P=VA where V is likely to be 68V at that load.
(1.5V /cell or thereabouts at 10C on 200Ah TS lithium c.f 3.2V, over the top for these girls) BTW LA wins here.
FWIW I would use higher voltage anyway.

Now 68V x 2000A x motor eff at that load x controller eff x losses to the back wheels and you are likely to be looking at only 60kW.
Even then, with that reduced voltage you would have trouble pushing 2000A at any revs, so battery volts need to go up in the first place.

A123 or similar lithium or my preference, LiPo, would perform better but cost more again (less voltage droop under load). You'd have to upgrade the motor as well if the battery and controller could deliver ! and on it goes.

Post by **MikeG** » Thu, 25 Sep 2008, 18:29

From what I've read, Lithium Polymer are the way to go if you have a big budget, as they are alot smaller & lighter than Lithium Ion types. I've heard one guy say you should avoid these because they explode, but I think that is ill informed based on the laptop battery scenario where they were under specced for the current drain.

I'm thinking that retaining the gearbox solves alot of problems with motor bandwidth, and the fact that you need higher power at higher speeds to overcome wind resistance (the way ICE works, but the opposite of the way that DC motors work).

A CVT would be ideal because you could have it set for maximum torque up to say 70km/h and then switch to maximum power from then onwards...

I'm also thinking that such a setup would need to be adaptive depending on throttle position... Flat out, the above setup would work, but cruising at 110km/h you don't want maximum power output, you'd want maximum efficiency for the minimal power usage that meets the power output needs.

It's all bloody complicated...

Has anyone get powerxrpm graphs for the warp 9/11 motors?

Post by **acmotor** » Thu, 25 Sep 2008, 18:54

The first full size electric power fixed wing aircraft to fly (earlier this year) used LiPo. Most models do too. As far as I can see, Lipos are far more suitable to mass production than most other battery systems. To me, they need better housing / mounting for serious EV use.

A DC motor's torque curve actually matches a vehicle's requirements quite well (rev dependent torque). Gear boxes are really only required if the motor is too small for the job torque wise in the first place or not designed for the required rev range. 'Electric' gearchanges e.g. series parallel switching of dual motors or voltage switching of battery packs is sometimes used. Most DC controllers are a bit like a CVT in their current multiplying (constant power) function anyway.

This is only a general response. Others will probably throw in graphs to prove the point. Zeva have motor graphs on the web site zeva

Post by **antiscab** » Thu, 25 Sep 2008, 18:59

Hi Mike,

Lithium Polymer is a contruction type, and is still lithium ion.
Thunderskys are lithium polymer.
Its the chemistry and C rating of a pack we're interested in.
the explosion capable packs are of the lithium cobalt variety, and yes can be made 'safe'.
lithium iron phosphate can explode aswell....it just takes alot more to do it.

unfortunately, there isn't any large format high C rate LiFePO4 batteries, or any other lithium chemistry.
there is an outfit called Life Batt that build large format battery packs out of 26650 size cells. for example of pricing:
http://www.lifebatt.com/

you could build a pack out of K2 cells, but again very expensive.

a gearbox will solve torque issues, but not power issues.
its all about the ratios.

CVTs dont really deal well with high torque (even a little 6.7" at 500A has 200Nm and a warp 11 at 2000A is around 900Nm).
They also aren't all that efficient, which performance wise means less peak power.

on my setup with a kostov 11 (if i can get hold of one), i expect the max power plateu to start at 22kmh, and finnish at 100kmh, since my batteries aren't anything like powerful enough to push my zilla 2k and 11" motor to full power.

the reason white zombie and the like have such good 0-100kmh numbers is because they are still getting full power at 100kmh.
Its only above 100kmh, that a gearbox would improve power output.
the type of gear box you would want for that situation is one that decreases the total ratio, ie, slows the motor down.
every automotive gearbox ive ever seen increases the total ratio.

Matt

Post by **MikeG** » Thu, 25 Sep 2008, 19:27

> a gearbox will solve torque issues, but not power issues.
> its all about the ratios.

Isn't that what a gearbox does? Sort out your ratios, rather than having a single gear that compromises on your final ratio.

If I want a quick 0-100 time the ratio is going to be different to the one I want for cruising at 110km/h. To keep an electric motor in its efficiency zone, you'll want to vary the motor speed. If you want an underpowered motor to stay within it's power curve, you'll need to vary the motor speed (downwards, espcially with DC motors)

The work around is to massively overspec your motor and to pick a final ratio that works for high speed & accelleration. My calculations put the motor spec at 570.1467 Nm for a 0-100 time of 5s (about what it does now) and 6514 motor rpm to attain 210km/h with the std final ratio in the s15.

Going an AT diff and a lower top speed would enable me to reduce the specs for the motor needed...

> the reason white zombie and the like have such good 0-100kmh numbers
> is because they are still getting full power at 100kmh.

I thought it was because they effectively gear the motor for max torque (the twins are wired in parallel) for launch and then switch it for max power at higher speeds (series) - there's a vodeo floating around on youtube where they explain it... You can do similar things with AC motors, with star and delta setups.

> Its only above 100kmh, that a gearbox would improve power output.
> the type of gear box you would want for that situation is one that
> decreases the total ratio, ie, slows the motor down.
> every automotive gearbox ive ever seen increases the total ratio.

Funny, my gearbox on my 200SX runs 3.692 in 5th gear (1:1) and 2.831764 in 6th gear (0.767:1) - in fact I'd wager that's what most gearboxes do...

sorry...

To illustrate, in 6th gear, at 180km/h the motor spins at 4282rpm (rather than 5583rpm).

I havn't looked into batteries yet, as it's all academic without a drive system, and you don't know what sort of current drain you'll need to support until you've designed everything else.

I'm hoping LiSu cells come onto the market soon, reducing the price of the LiFe packs.

I hear the Telsa roadster makes use of literally a-ton of reconfigured laptop batteries

MikeG

Post by **acmotor** » Thu, 25 Sep 2008, 19:30

My TS are not LiPo ? was that a typo ?

Post by **antiscab** » Thu, 25 Sep 2008, 19:35

no,
The thundersky construction method (from what i have seen) is to encase parralleled smaller LiPo cells in a high temperature rated plastic enclosure suspended in a phsae change solution.

I have a link to a dis-assembled LCP cell, ill find it in a moment.

Matt

Post by **zeva** » Thu, 25 Sep 2008, 19:39

MikeG wrote: Ian, we had a presentation from one of the guys in Canberra Branch about your car. I think you've done well but I have a few things I'd do differently.

1. Battery location - I'd want to keep the weight as physically low as practical, so I'm thinking that a custom battery tray in the boot like you've done, but I want to avoid placing them anywhere but between the axles. A tall order I know, but what with the fuel tank being located under the back seat, and room behind the back seat, it seems like the logical location.

Agreed. Unfortunately the MX5 didn't give me much choice for battery location due to space constraints, but if I were doing another conversion with performance in mind, I'd start with a slightly more spacious donor!

MikeG wrote: 2. Curtis controller? No thanks, I'm thinking something seriously over-engineered to last... If I'm pulling say 20A, then I want a controller that handles 50A. I know you're designing your own controller, so I eagerly await news of how it stacks up

Agreed again, the Curtis 1231 are rubbish. I blew Rob Mason's Curtis within 2 hours, after borrowing it to get the MX5 going! Still working on my own controller, though it'll be more an alternative to the 1231 than a competitor to the Zilla!

Right now the Zillas are the way to go for high performance DC EVs. But maybe in a year or two I'll be able to build something remotely comparable!?

MikeG wrote: Unlike the MX-5, the 200SX has a fair few places it can fit batteries apart from the boot and under the bonnet. Though, it's still fairly slim pickings...

If it were me, I'd take out the back seats and put the whole battery pack there. Kind of a pity to lose the back seats, but it is pretty much the optimum location for batteries. So.. how high a priority is performance to you?!

Post by **antiscab** » Thu, 25 Sep 2008, 19:41

found it:
disassembled LCP cell

EDIT: started new thread here:
viewtopic.php?t=621&p=5187#p5187

Matt

Post by **MikeG** » Thu, 25 Sep 2008, 19:52

Good question Ian. The timeframe I'm looking at to build this is a couple of years, prob a year before I decide on a design...

Performance? I don't think I could go backwards, although realistically I could probably make do with a top speed of 130km/h

I want 100km range from the car...

--

So how does the MX-5 perform? Range? 0-100 time? Top speed?

MikeG

Post by **zeva** » Thu, 25 Sep 2008, 19:59

MikeG wrote: So how does the MX-5 perform? Range? 0-100 time? Top speed?

Range is about 120km. Its direct drive so the motor would hit rev limit at ~150km/h.

0-100, let me answer this in a few months, when the new battery pack and Z2K are in there..

Post by **acmotor** » Thu, 25 Sep 2008, 20:01

Matt,
Can you start a new thread under batteries with that.
Its gone off topic here. I will follow. I am interested in what you have found out.
BTW that link tells me zip about 'LiPo' . If it were simply LiPo then 20C would be norm and o/c volts would be 3.7 not 3.2

Post by **MikeG** » Thu, 25 Sep 2008, 20:01

What's your best guess?