Luke's 2001 NB MX-5

[necrogt4]

So I started collecting parts a couple of years ago while saving for a house so things we're pretty slow going. I've bought the house and have a lot of time at home these days so things are moving along much much quicker.

The first few posts will be a bit of a catch up, here's the short and curly of the project so far...

• Vehicle: 2001 NB MX-5 (LSD, 6-speed manual, 1090KG wet).
• Motor: Netgain HyPer 9 120v.
• Gearbox: Retained standard 6-speed with a 4.1kg aluminium flywheel and Exedy cushion button ~400nm clutch.
• Batteries: 115V @200A 120x 3.7V 50A CATL NCM cells ~200Wh/Kg (maybe, I've also considered Tesla modules and LG Chem pouch cells. Currently testing a batch of the CATL cells).
• BMS: Batrium (not yet purchased).
• Charger: TC 6.6kw (not yet purchased).
• DC/DC Converter: Undecided, the TC charger has DC out but very very low current. Not enough to run much but I'm not sure of load yet either.
• Cooling: Custom designed (by me) chill plates for battery packs.

The aim of the project is to come in at spot on 1,000kg. Which with the help of lighter seats, brakes, flywheel, wheels, less cooling etc should be achievable. Weight is a consideration with each and every change. I've actually weighed every single item that's come out and that's going in to have a fairly accurate measure of final weight.

Along with the standard EV components I've also replaced all of the wear components: polyurethane bushings, tie rod ends, steering ball joints, bearings, sway bar ends links, sway bars (at some point soon), tyres once running and brakes (willwood calipers, discs, brake lines all round). While the power increase won't be substantial from the ICE powerplant, sharpening up the handling to go along with the smooth power delivery should make for a much more enjoyable drive. This conversion is about efficiency and fun rather than outright power. Better use of what power you have in all circumstances.

A couple of photos of just after I purchased the car:



And another just before I moved, none of the ICE hardware in the chassis.

[necrogt4]

As I was saving for a house and the HyPer 9 motor I decided to get busy with convenience electronics and teach myself how to use Arduino. I'm not a fan of manually holding power window buttons down, it's a thing. So now I don't have to...

A couple of four relay boards 12V -> 5V signals into the Arduino with 5V -> 12V signals out of the Arduino. Powered by a USB charger. I also have two hall effect current sensors (one for each power window motor).

The power window switches in the MX-5 are super simple and run the full current from the battery, through the switch and into the motor. So the Arduino is spliced into these wires and holds the circuit open via relays until a present time is reached (fail safe), the button is pressed again or the current sensor detects a current spike (i.e. the window has reached top/bottom).

My old workbench (now retired, I'll post a workbench build later):


The big mess of wires and circuits that make it all work:


And a video of it all working, it looks boring and that's the point. It just works like any other car does and nothing else, exactly how you'd expect an OEM feature.
https://drive.google.com/file/d/1ANTsEA ... sp=sharing

[necrogt4]

I also mentioned lighter seats, there's two reasons for this.

• I'm too tall for the standard seats
• These seats are half the weight and look twice as good (IMO).

[brendon_m]

I assume the custom battery boxes are going underneath where the fuel tank was? Any going in the boot or under the bonnet?

[necrogt4]

There will be a large battery box in the engine bay that sits behind the front axle that holds 80 cells and another smaller box in the fuel tank cavity that will hold a further 40 cells, the charger, contactor, fuses and DC/DC converter.

The fuel tank cavity (see below) is pretty small. One of the reasons I'm not looking at Tesla modules anymore is I would have had to cut an opening into this cavity which would have meant a chassis rigidity test (~$2k).

There will be no batteries going into the boot. Except maybe the 12v replacement battery (made from the same cells, or LTO possibly) in the original location behind the drivers side rear wheel in the space between the chassis rail and body. Which to be honest seems like a terrible place for a battery. I'll try to squeeze this into the fuel tank cavity if I can. Anything behind the rear wheels makes a noticeable negative impact to handling (even groceries!) so I've taken the spare wheel and jack out.

This is the front of the cavity can be seen here, it's normally covered with a few panels that adds a bit of extra strength and carpet.


Behind, that cutout on the left is where the fuel tank filler neck used to go. Perfect to run the charge cables from the original fuel tank inlet.


Underneath:

[CZal]

Nice looking car Luke!

[necrogt4]

Nice looking car Luke!

Thanks! It'll look even better once it's moving under it's own electricity...

[necrogt4]

I decided to get serious about tools and workspace a bit over a year ago now and upgraded from my Ikea kitchen bench to a set of cheap Ebay shelves and some Bunnings supplies. I wanted something standing height that would have everything I needed within arms length.

So from this:


To:


A blank canvas for tool hanging:


And more recently the addition of a workshop vice and 4x 27 litre drawers. Now my only wish is that it was all bigger


All up I think it's cost about $450 (no way I could buy an off the shelf bench for that much) which includes:

• $85 for the bench
• $27 for the particle board top
• $29 Pegboard
• $16 Pegboard backing support
• ~$30 Paint and fixings
• ~$30 Pegboard pegs and zip ties
• $40 for a surge protected 8 outlet powerboard with individual switches
• $64 for an adjustable monitor arm 20kg capacity
• $30 castor wheels with locks
• $20 3000 Lumen florescent light
• $15 Shelf railings (C channel aluminium bars) + screws
• $48 4x 27lt Sistema tubs for shelves

[jonescg]

Hi Luke - hope this project is getting the attention is deserves? Setting up the workshop is a priority though. I bought a house with a big shed and had a hoist installed so I could convert the Prelude.

[necrogt4]

Hey Chris, it's moving along well at the moment, I think I've finally locked in my battery choice (simple and robust but a wee bit expensive) after iterating on a heap of different CAD designs.

I've also after being screwed around with delays by two separate suppliers of gearbox adapters (~8 months of waiting) and having to have a coupler remade locally as my supplied one didn't fit my clutch I've finally fitted my Hyper 9 into my MX-5. So things are starting to pick up.

On the workspace side, now that I've moved into my new house I'm putting thought to building a bigger workbench that fills the entire rear of the single car garage. I want more tools to make more things LOL.

[brendon_m]

Great to see.

First thing I thought after building my workshop is "I need a bigger workshop"

[necrogt4]

Haha, it's always the way. There's really not much I am not confident in having a go at in fabricating, and the back and forth with engineers for parts that you can mostly make yourself seems like a good reason to make a bigger workshop!

[jonescg]

For me it was the hoist. That was $3600 well spent from day one. I can't put a price on my sanity crawling underneath cars in the dark...

[necrogt4]

If only I had the room for a hoist. Being able to "stand" while working under a car would be awesome. One day I'll have a shed bigger than the house to fit all these things in.

How many friends have asked to use your hoist since you've put it in?

[whimpurinter]

Hi,
When I had my crummy early EV, I had been warned never to tromp the accelerator in low gear but never told why.
So I was U-turning on a hill and a car came around the corner in the distance. As usual, not wanting to intentionally inconvenience another driver, I rushed to reverse the car and it went into I'm-not-going-to-drive-readily mode (one of a few modes available )
I think this was most likely a gearbox issue and I just want to ask if you know for sure that the gearbox you are going to use can definitely handle the torque in any gear and at maximum torque?

[necrogt4]

The standard 1.8lt from my series MX-5 is rated at 118kw peak and 180nm peak and the Hyper 9 going in is ~115kw peak and 235nm so it's only a 30% increase in torque. If it does break I do have the option of using a gearbox from an RX-7/8 with a little bit of work.

That said, if the motor is strong enough to lunch a gearbox I probably don't need one! I'm not even sure if I'll end up keeping the clutch, I do like the idea of a straight coupler (which will save ~10kg of rotating mass) but I have a feeling on the track I'd like to be able to select a low and high gear with relative speed on the shifts rather than relying on the syncros to do the work.

I guess I'll find out soon enough!

[OzSpider]

Nice project Luke. MX-5 is a good choice.

Was the hyper 9 an obvious choice for you? Or did you dwell on it for a while? What else did you consider? Only asking as I'm just starting my own journey and also have a rear wheel drive 2 door that ways about 1 tonne.

[necrogt4]

Nice project Luke. MX-5 is a good choice.

Was the hyper 9 an obvious choice for you? Or did you dwell on it for a while? What else did you consider? Only asking as I'm just starting my own journey and also have a rear wheel drive 2 door that ways about 1 tonne.

Well firstly it was a budgetary constraint, I wanted to keep the motor cost within reason but also have something with decent performance/weight. I also considered a Netgain Warp 9" due to their popularity, simplicity and availability. Being something fairly common it makes finding parts (e.g. adaptors) and info easy. But I ultimately chose to use the HyPer 9" due to regen, weight and the fact that it came with a fairly feature packed controller. It did push the budget cost up a bit from a DC motor but having regen plus being a fair bit lighter than it's DC counterpart is was a pretty easy choice.

Another plus, for me anyway, is that the motor's rev range is very similar to the original engine so it shouldn't be a huge departure from the revvy feel of the original. Keeping the power/chassis true to the lightweight agile nature was probably the most important thing as oppose to say outright power (e.g. placing a Tesla motor in the rear and removing the entire original drivetrain, which I may do for a daily at some point in the future).

[necrogt4]

As I mentioned a few posts ago, I've finally settled on a battery choice after much much design and testing. I was pretty hard set on using CATL NCM 50Ah prismatic cells, I even ordered sample cells from two different distributors in China (which I'm looking to sell now) and put them through some performance testing in which they lived up to the distributors claims in both capacity and output.

My first design which I ended up moving away from to try and manage front end weight distribution a bit better was a 21s4p arrangement with a further 10s4p in the rear battery cavity. With the modules clamped together to a central chill plate.Initially I had envisaged the busbars being laser welded to the cells but eventually moved toward having the busbars incorporated into the clamping system holding everything together. This can be seen in the second evolution of the battery box in the next post.




And with an updated busbar design:

[necrogt4]

So this design has a few less modules (18s4p) with more to go into the spare wheel well. I've used acetal as an insulator between modules, on busbar clamps and also for the enclosure. Aluminium is used for the frame, busbars and chill plate. There is also a 3mm thermal pad sitting on both sides of the chill plate. While this might seem excessive for a 3d design, it did allow me to calculate a fairly accurate weight estimate of ~80kg which is pretty good. But the complexity of the design pushes the cost and time up somewhat in line with say Tesla or LG modules which are far less complex.

[jonescg]

Not sure how happy they will be living upside down, but it's an innovative approach!
I'm also working on 18650 based battery packs right now and while it might be a bit late, they do have one advantage - completely customisable shapes and sizes. Have a Look at the Prelude thread and see what you think.
They can feasibly have a cooling plate sandwiched between them too, a bit like Rivian are doing with their R1T battery packs:
https://cleantechnica.com/2019/06/15/ri ... y-storage/

[coulomb]

I'm not a fan of battery cells living right on top of the motor. Electric motors get hot. On Weber's MX-5, most of the worst cells (before his recent upgrade) were on top of the motor. The closer to the middle of the block (where the motor was underneath), the worse the cells were.

So if it's possible, try to keep the cells away from the motor. I know it's really hard in an MX-5.

[necrogt4]

Not sure how happy they will be living upside down, but it's an innovative approach!
I'm also working on 18650 based battery packs right now and while it might be a bit late, they do have one advantage - completely customisable shapes and sizes. Have a Look at the Prelude thread and see what you think.
They can feasibly have a cooling plate sandwiched between them too, a bit like Rivian are doing with their R1T battery packs:
https://cleantechnica.com/2019/06/15/ri ... y-storage/

Well, I have moved away from that design now but the engineers working for the distributor I bought the cells from there shouldn't be any issues to running the cells upside down.

I actually considered using new Tesla 4680 5Ah cells but to get the desired I would have had to double layer (vertically) modules in the engine bay and it would foul on the bonnet. That combined with the added assembly complexity and points of failure (vs the LG cell pouch modules I'm looking at which are sealed units) put a nail in the coffin for that idea an 18650 pack. Though seeing all the nice work you've put into your pack does make me think.

So what happened to the pouch cell pack that you'd already made for the Prelude? I thought the pack was pretty much done?

[necrogt4]

I'm not a fan of battery cells living right on top of the motor. Electric motors get hot. On Weber's MX-5, most of the worst cells (before his recent upgrade) were on top of the motor. The closer to the middle of the block (where the motor was underneath), the worse the cells were.

So if it's possible, try to keep the cells away from the motor. I know it's really hard in an MX-5.

It's incredibly hard. I'm trying to keep the batteries as far back and low as possible so that means compromises in other areas (e.g. motor and batteries close together). Though dealing with heat is easier than dealing with the confines of a small car. I don't have a completed 3d design yet but my current mock up designs do have the front battery pack sitting in line with the steering rack that's just in front of the front axle but far behind the original engine and much much lower. The rear pack completely fits inside the fuel tank cavity without any body/chassis modification.

In terms of heat, I plan to place chill plates under both front and rear packs. That will work as an insulator between the motor and front pack, plus I can always go one step further and use a chill wrap/blanket around the motor. I'll have a better idea of how much heat is generated once I get things rolling...

[jonescg]

The original Prelude pack with liquid cooling plates was working well until coolant (mostly water) leaked into one of the modules and the resulting condensation spread throughout the pack, wetting everything completely. So while two modules were damaged, the rest were trouble with a time delay. So far @brendon_m has been using them for his home storage system and they've not burned his house down yet. Hopefully that trend continues.
But yeah basically it was a very expensive failure due to a very small weakness. The new pack won't be liquid cooled, but it could feasibly be done with a cooling plate sandwich.