charliebruce

Hi, everyone.

I recently placed a bid on, and won, a mis-firing RX8, 2004 model, 192PS. I'm in the UK, and my plan is to tinker with the engine, and get it running if possible, but most likely convert it to an electric vehicle and sell the engine/anciliaries. I've got a fair bit of experience working with cars (mostly older though), electronics, and recently also electric vehicles, and am aware of the kind of money this'll likely end up costing.

I have a few questions, and this seemed like the best place to ask:

1. VIN checks

Before handing over any money, I want to make sure the vehicle isn't stolen/cloned/written off. I plan on getting an HPI check but this only offers me limited defence against cloned cars.

What I plan on doing is hooking up an OBD-2 device to the port and getting the VIN from the computer, so I can check that it matches the VINs on the chassis, engine and documentation. I'm assuming a cheap Ebay ELM scanner and something like ScanMaster or Forscan will get the VIN?

I'm also assuming that the VIN in the car computer can't be changed by thieves - is this correct? If not, is there another way of checking (besides the ones in the engine bay/computer? I know there are "hidden" VINs, is it worth searching for one around the door/rear wheels, or bringing a VIN inspector along? I don't know where they're located and it's obviously not common knowledge, so this might be troublesome)

2. Towing

It's only a few tens of miles from the car's location to my place, is it worth getting a beavertail truck for the day, or would a Golf 1.6 be powerful enough to tow it safely? The Mazda guide says to tow with the rear wheels up but I can't see why towing it in neutral with all wheels on the ground, someone operating the brakes, wouldn't be fine.

3. Modifications

Some more general questions here:

Do you think the transmission will hold up OK under about 250hp peak, if the gearbox is never being shifted and there's no clutch? (electric motors have a wide power-band so high torque and no need to shift - essentially planning on jamming it in 2nd or 3rd, the motor should rev up to 14k RPM giving a decent range of speeds. There are (I am led to believe) 50k miles on the box/diff, previous owner claims it's been well serviced, will replace the oil as soon as I get it just in case, and check for metal in the oil).

Will the airbags/safety systems work fine if the ECU doesn't believe that the engine is running, or are they only active when the engine is on? My vague plan is to replace the throttle/brake with a custom system but leave the rest of the vehicle electronics alone to as great an extent as possible. I can potentially see there being issues with the immobiliser/alarm, or the car computer hating me forever?

If I go down the EV route I'll need some way of controlling brake cylinder pressure - does anyone have a diagram of the braking hydraulics? I also understand that there's vacuum assist, which does require engine. My best bet might be to run the ABS pump into an accumulator then use 2 solenoids to increase/decrease the pressure in each caliper - will the ABS pump cope with this usage, or should I look into the Prius system instead (this is what the Prius does to control brakes, ABS and DSC)?

I'm assuming that the air con runs off the engine, so I'll need to replace the pump with an electric one? Is there anything else that expressly depends on the engine that I might not be aware of?

Sorry for the long first post!

Charlie

(Would this be better-off in Tech Garage/DIY section?)

godesshunter

Although long winded and somewhat well though out, I get the impression you don't really know what to do. I get the feeling you will be in over your head in no time.

Judging by how weary you are about vin checks and theft recovery, I assume you get the car incredibly cheap. Personally, I would just repair what is needed and either keep it as is or off it for profit.

You would need a wealth of expendable money and the parts,tools,and resources to do a lot of custom fabrication. It would be a shame to see the car get torn down to a dozen pieces and be left that way forever.

I can just see you start to sell off unneeded parts to fund the build and have it never get done. Then you are left with just half a car.

charliebruce

I've got more of a plan on the EV side of things (Ranger EV motor perhaps, custom controller, 20Ah 420V pack with a CB750 + water-cooled truck alternator, rewound for higher voltage as range-extender) - I was trying to keep things more oriented around the stuff that people over here will know about. I'll go into more detail about the electronics over at diyelectriccar or similar, once I finalise the plan.

The car was a little under £1k on Ebay, which is on the cheaper end but not completely out of the range for a 2004 RX8 with dead engine, as far as I can tell - it's my first major part of the project and I'd rather be safe than sorry, hence the caution.

I'm fairly wary of not finishing it/running into some major issues with it and being at a loss, but I've done similar projects on a smaller scale (electric moped - obviously not in the same league in terms of cost and complexity, but same idea). I'm still at the planning stage, hence the questions, but I wanted to have the vehicle on hand so I can get an idea of the space available.

Resources - I've estimated the costs and have enough expendable money to cover the parts - I have decent tools available through work (cnc mill, manual lathe, fairly extensive car tool collection, table and band saws, basic electronics stuff). I'll definitely keep the original components until I've at least got the motor mounted and test-driven it on electric. I've also got some good opportunities for sponsorship (still a student).

Thanks for the feedback in any case, I'm still at the planning stage so it's not out of the question to just repair it.

HiFlite999

iTrader: (2)

^ What he said.

In 30 seconds with Goggle, I came up with the factory service manual. Here you go: RotaryHeads.com - Mazda RX-8 PDF Technical Manuals FE3S

No the VIN is not accessable with an ebay OBD scanner. Only the dealer has that stuff and access to the security systems. Probably best to rip out most of the electrical systems and rewire the whole thing. The transmission will not survive 14k rpm. Tow with a VW for tens of miles?? You realize the towed car will have no power steering or brakes?? Etc.

godesshunter

Did a quick google search and came up with this. Maybe it will help a bit?

RX8 Conversion - DIY Electric Car Forums

elysium19

iTrader: (1)

Why jam it in a fixed gear? As long as you're retaining the drivetrain, and not putting individual electric motors on each wheel (which, frankly, seem to be the usual method for these type of EVs), then why not use the function of the tranny too?

I realize you got it for suuuuper cheap, but I would have thought you'd ask these question BEFORe buying it!

Your best bet is to try to get the gas engine working. if it really is just a "misfire" which could mean many things, it still proably isnt too difficult or expensive to fix. And the car was really cheap!

charliebruce

I've already got and read through that section of the service manual, no hydraulic diagram (specifically, how the ABS pump and DSC/ABS HU/CM work). I found this for the mazda3 but nothing similar for the RX8 - this is what I'm looking for.

Thanks for the advice re electronics/OBD.

What would fail in the transmission at that speed? I can see the engine having a problem if revved that high, but I don't see why the gearbox wouldn't cope.

When towed, I didn't know whether it would have power steering, if the battery was connected and the key was in the "On" position. I knew you wouldn't get the full brake force without the vacuum assist but there'd still be some braking force. Either way, I'll put it on the back of a truck, seems easier.


Thanks, I've already had a flick through that thread, I'm planning a similar project in many ways to that.

Fixed gear: No risk of burning out the clutch/issues with clutch pressure on the motor shaft, no shifting also reduces the wear on the gears especially with the higher power.

I've not seen one-motor-per-wheel setups used before at all?

Is a little under £1k GBP really *super* cheap for one with a possibly-dead engine, no longer under warranty? I saw prices anywhere between £800 and £1700 for comparable cars, this one had a few nicks and previous owners so it seemed a fair price.

I'm asking now rather than before buying because I knew that people had done similar modifications already, and that I would be able to find a solution one way or another. I knew that I could work around the ABS/DSC with parts from a Prius, and since other modders hadn't mentioned the airbags I'd assumed that the bags would also be usable - I'm trying to work out a more precise budget at this point by working out whether I can re-use the Mazda parts or will need to use the more expensive work-arounds.

If the petrol engine works fine or can be fixed economically, of course I won't tear the car apart completely.

j9fd3s

my cheapie auto engenuity PC based scanner comes up with a vin, so other scanners might too. i think you are correct that unless you have the dealership software you can't change the VIN in the ecu. it seems likely that since the dealership can sell new ecus, that they should be able to program the vin, but i don't actually know that.

as far as the towing, you guys in Europe tow completely differently than we do. i'm sure the 8 would be fine with a flat tow as long as the speeds were reasonable and you didn't go too far.

paimon.soror

iTrader: (15)

This is not true, you can certainly get the vehicles VIN from the OBD2 port. It is part of the OBD2 standard found in mode $09 pid 0x02. Most OBD2 scanners will show you this information including the cheap crappy one I have from Harbor Freight.

charliebruce

Yes, I thought that was the case but wondered if Mazda might specifically not support that part of the protocol (it's not a required part of the spec). My OBD2 tool (cheap ELM) can indeed display the VINs from some cars, hopefully also a Mazda - was just wondering if people had first-hand experience. I know that the VIN isn't writable without the IDS system, I have no need to do that.

Unfortunately, I'm not going to be testing it on the car I first mentioned - I took out an HPI check, and it turns out that it's a Cat. D write-off. I understand that isn't such a bad thing in practical terms, but the fact that the seller didn't declare it upfront, won't budge on price, and doesn't have any documentation to prove that the repair was carried out professionally makes me nervous, so I'm walking away from this one.

I'm still fairly set on using an RX8 as the donor car, and am also still fairly confident about buying one with a damaged or dead engine and repairing/replacing with an electric motor. I might have to commission an up-rated gearset or limit the top speed if I'm going to have a single-gear drive system (which does significantly simplify things elsewhere). I've also thought about the choice of "range extend" engine and realised that the CB750 is probably not a good idea due to noise and emissions, so I'm considering my options of small (>1L) car petrol engines instead.

Thanks to anyone who's given advice so far. I'd still appreciate any comments and suggestions if people want to throw their two cents in.

Nubo

Here's a good site to get a feel of what goes into electric conversions. The videos can be a bit long-winded but on the other hand if you're going to be successful you really are going to have to immerse yourself in the knowledge.

EVTV Motor Verks | Electric Car Conversion Videos

godesshunter

Any progress on this?

skooler

Just stumbled across this and thought I'd reply.

I have done this and have a reasonably well performing car.

To the OP, I would advise reading up on the diy electric car forums.

Here's my build thread for anyone that's interested.

Skooler's RX8 Conversion (Formally "A Good Motor???") - DIY Electric Car Forums

Worth mentioning that the rx8 stuff starts a few pages in.

Cheers,

Mike

charliebruce

The car I'd placed a bid on was a write-off, but the listing hadn't mentioned that, so I didn't continue with the purchase - can't trust the seller. I'm watching a couple at the moment, but also looking into the options for the electric motor in more detail - I really want to do this with a brushless (induction) motor, but they aren't nearly as cheap.

I've been reading around on the forums for a fair amount of time, I've had the idea to do an electric car build for the last ~3 years, but only recently found myself in a situation in which it'd be practical. Earlier this year I did a moped rebuild with a 6kW outrunner (waiting for the DVLA now), but I want something more suited to motorway/longer-distance driving.

I'm actually in a very similar situation, and my choice of RX8 was based on your build - I'm 18, UK, with decent mechanical, electronic, maths and programming skill, and access to decent tools and plenty of experts, should I need them.

My thoughts at the moment:

• Sponsorship to get a decent 200kw induction motor £? OR Ranger EV motor OR Warp9 or Warp11HV
• DIY control and charge system £1500
• 450v 20Ah A123 pack £2500
• Smart car engine as RE £700 - enough space in engine bay?
• Alternator for above £?
• RX8 231ps dead eng £1500 inc adaptors/mounts/mods

I'm trying to sort the question marks out before I commit though. The main reason I placed a bid on the one I did was that it was so close-by - there hasn't been one as convenient since.

I'm still considering the clutch vs single-gear locked option - it depends on if the motor has bearings designed to cope with the clutch force. If I go single-gear, I'm mainly concerned about stripping the teeth, or heating it too much at the high RPMs. What were your experiences with the gearbox, any problems since or has it all been fine?

I see you had the same idea about faking engine output with an Arduino - did that end up being necessary or not?

Thanks
Charlie

skooler

Hi Charlie,

What are your performance and range goals?

Why do you want to use a brushless motor? whats wrong with a series wound?

Looking at your component list, consider a Kostov Motor (cheaper than Warp)

Also, buy a controller rather than build, you wont regret it! My Soliton is excellent, also gives you more scope to get the outputs and data logging working with the car. got mine from home - Rebbl | We bring back the joy in driving

The DIY charger is fine. read 10kW / 60A DIY charger open source design - DIY Electric Car Forums

Don't Touch A123 cells if your thinking of getting them from China. They are not safe! Consider the large Prismatic Cells such as Sinopoly or CALB, I have imported these many times before if you want a hand with that.

Also, replace the mechanical alternator with a DCDC converter, cheap and easy. below or similar work well.
EV DC to DC Converter Electric Car Voltage Reduce 144V volt to 13.8V 35 Amp 480W | eBay

Otherwise it looks good, things that I'd have done differently are below...

I have thought about mounting the motor in the transmission tunnel with a small 2 speed + reverse gearbox. I only use 2nd and 4th on the original box. that then gives the entire engine bay for batteries and components

I'd also use all of the voltage available to the controller, in my case 350V.

I'd also of purchased a brand new motor such as the kostov 11alpha.

If your after more ideas then start a thread on DIY Electric Car

My conversion has cost about £7,000 (including new tyres, brakes etc). If you want to go lithium the you can easily double that!

Cheers,

Mike

skooler

Oh, and arduino has not been needed, keep an eye on my build thread for why.

Gearbox has been solid, no worries with using it although i would mount the motor either directly to the diff or with a 2 speed box in the transmission tunnel if I were to do it again.

charliebruce

Hi,
Sorry for the late reply, I've been quite busy over the last week.

Why AC induction - I guess I just prefer the concept/efficiency of a brushless motor, provided I can actually get my hands on a decent one, of course. If I can't do so, I'll need to do more research re brushed DC motors - at that point I'll consider a Kostov and bought controller. If I can do it AC I want to use my own controller though - I wouldn't be able to afford an AC controller otherwise. Control systems and electronics in general are fascinating, and I'd like to think that I could pull it off - I might do a lower-power ~100v 120A controller first for my other EV, to see if I'm actually capable of building a complete controller.

Performance and range - I'd like circa 40 miles on battery, then a range extender petrol engine for 250mi range on a tank, capable of motorway driving and occasional fun bursts. 20Ah at around 400v, a smart car engine and an alternator should be enough to achieve this (I wasn't meaning to use the alternator to lower the voltage - was already planning a 1000w DC-DC conversion stage).


Stuff that I'd love to hear about:

China A123s - I haven't heard anything about the Chinese A123s being dangerous - you mean Victpower ones? Would it be sufficient to give each cell a few cycles in some test machine and reject badly-performing ones? Or do you mean unsafe as in sending money but never receiving anything? If so I have another couple of ideas (sponsorship to get genuine cells, or playing it very safe when buying).

The reason I was looking at A123s was to get the high peak power for performance driving, without breaking the bank or taking up too much space/weight, especially since the petrol engine is there as RE. To get the same peak power from Sinopoly cells, I'd need about 5 times as many cells.

Transmission - I guess this all depends on whether or not I can get an AC motor which revs high enough to manage direct drive at a wide range of RPMs. Fitting it in the transmission tunnel would be ideal but I suspect a 2-speed box like that might be expensive or hard to work with?



I'm still firmly in the planning stage at the moment - I'll open a thread on an EV forum fairly soon (once I've contacted a few people regarding motors and batteries). You could also email at charliebruce at (google's public email service).com if you like - your feedback has been really useful.

Charlie

kcochran

Just as a note, these days for electric conversion, LiFePO4 prismatic cells are the way to go. Yes the energy density is a bit lower, and you need to up the amp hours a bit to get your current, but they are far safer and much more resilient to temperature, charge cycles, unbalancing. Buy a few extra, or even better try to get a set that's well matched (internal resistance especially). I've done some work on a conversion of an SUV with 100 AH cells, 150 mile range and the thing weighs a ton. It runs at around 360V and can outrun an RX-8 hands down.

Don't DIY a controller, it's your life at stake. Consider how you are going to power the steering and brakes, they have boosters. GFR's and E-stops are strongly advised/required

Just remember, a car is a 4,000 pound weapon. Don't be testing it on public roads. We've had brake failures, steering cutouts, cars going in reverse and skid steering on electric car conversions.
LiFe Pristmatic Cell 3.2V 20AH | eBay


good luck!

skooler

Whatever you do please don't touch the A123s if your thinking of getting them from China. They are factory rejects and are not safe. I know of at least 3 failures using these cells. If you are using A123's buy them directly from the USA, if you can find a supplier.

Look at Sinopoly, CALB and maybe Hipower prismatic cells.

As has been said. the controller is a key safety feature and needs to work perfectly. add to this that all of the standard open source controllers will only do about 100HP peak. My Soliton Junior that cost £1,700 will do 200HP all day.

You will find DC much simpler and cheaper to build with more power.

If you want some inspiration feel free to come and have a look at and drive mine.

Cheers,

Mike

xexok

Just saw this on youtube, took me a second to figure out why they were making no noise lol. Video even says what each car is running for batteries and such.

volttan

Hi,I'm working on a similar project.
I am looking for help with can bus mazda

Brettus

iTrader: (2)

Wow ...you actually did it !

TeamRX8

iTrader: (25)

Love that gas gauge on empty

volttan

I have a little problem with him.
I am trying to start DCS, undergoing a calibration procedure, but I do not want to turn off this yellow light from slipping. I removed PCM and tries to recreate its operation on my processor.
I am looking for help from mazda can-bus


more about my projecthttps://www.diyelectriccar.com/forums/showthread.php/electric-rx8-voltttan-202095.html