I get knocked down…

So, things ended on a bit of a low just before I went on holiday a few weeks ago. The second run at an MOT ended in disaster. But my optimism was quickly restored and before I left, I posted some parts off to people for help.

One of those parts has come back sorted, albeit in need of reassembly. That’s the Battery Management System, now speaking CAN on all four possible interfaces (though I only need 2). This should allow me to charge automatically, rather than sitting plugged into the car controlling things manually. Bonus.


The other part was the coupler, with which my friend couldn’t help. But in some ways I was quite pleased by this. I like the idea of doing it all myself. So, I did what any normal person would do: built a machine shop in the back yard.

I bought a workbench and boxed it in with waterproof OSB and a felted roof into a sort of mini shed. Well, I say ‘mini’. You may not be able to get inside it but it is 2.4m long and now contains a pillar drill, lathe, chop saw, bench grinder, vice and welding station.

The lathe is nothing special: the most expensive, cheap Chinese beastie I could justify. First experiments have been reasonably promising and with some practice (it’s 25 years since I used a lathe) and some additional kit, I reckon I can make up a straight coupler.

The process is much like last time, just with a little more precision:

  • Cut and square a section of tube. Clean it up inside and out on the lathe
  • Machine a spindle that fits tightly through the Ford clutch centre (that fits the motor splines) and BMW clutch centre (that fits on to the gearbox)
  • Machine the outer diameters of the clutch centres so that they fit inside the tube
  • Assemble the tube, splines and centres to ensure that everything aligns
  • Weld it all together

I’m just waiting on a new profiling tool and some bar stock to make the spindle/alignment tool, then I can get this done. In the meantime I’ve already done a practice run. Even without the alignment tool, this is already much straighter than my first one. And thanks to the practice at welding and fabricating, much tidier.

Using my new dial gauge (it has been an expensive few weeks), this one is out by roughly +/- 0.4mm at the end of the tube. This is about the same as the amount of slop created by the splined connector at this range. If I can halve this I’ll be reasonably happy.


In the meantime I’ve been working on some other upgrades, starting with the inverter.

To date I have only been using half of its outputs – so called ‘MG 2’. But my control board supports me driving MG1 as well and bridging the outputs. This requires some simple soldering of jumpers on the control board (done) and then bridging the phase outputs on the inverter itself (half done).

I’m using short lengths of 20mm2 wire for this, and fabricating copper connectors on each end. This is a bit of a pain of a job, but I’m getting there. Five connectors down and one to go. And some tidying. I want to clean up all the connectors and maybe add some more insulation, assuming it will fit inside the case.

Improving communication

Perhaps it won’t surprise anyone that the range of the Wi-Fi connection from the inverter control board is a little bit limited, what with it being encased in an aluminium box. There has also been shown to be an issue with the current draw from the Wi-Fi module causing instability in the current sensor readings – one of many possible reasons for my inverter tripping out.

To fix both of these issues I decided to follow some other forum members and relocate the Wi-Fi module outside of the inverter housing and at the same time add a dedicated power supply to take the load off the control board. I had a cheap buck converter lying around so just wired this into the 12V supply from the AMPSeal connector. And there was plenty of space in my AMPSeal housing to site both the Wi-Fi module and the power supply. Some cloth tape added to the IDC cable should prevent any damage from friction.

Keeping it cool

Per my last update, I’m also working on keeping the inverter cool. This has multiple components:

  • relocating the radiator a bit higher to reduce the chance of air in the system
  • Upgrading the pump to increase the flow rate
  • Adding fans to maximise air flow

I haven’t started the first part yet as the inverter is in pieces. But my new workbench should make that easier (did I mention I also have a metal bending tool out there?).


The second bit has been a bit of trial and error. The Pierburg pump I was recommended is upwards of £300 new and over £100 used in any decent condition. They are also, I learned from a lot of reading, prone to failure, which means I didn’t want an older one.

Little Bosch PAD pump is too puny. Larger unit on the way.

I gambled on another BMW pump (a Bosch PAD unit) on eBay but I think it’s probably too puny to be an upgrade. So now I’m waiting on a Bosch PCE unit out of a VW that should fit the bill. These look nice and meaty and others have reported good results.


Being a hoarder with a history of PC-building, I have a lot of fans. I selected a couple of powerful 12V 90mm server units from my collection to keep the radiator cool. These are PWM controlled, though for now I’ll just run them at default speed. Down the line they can be made responsive to thermostat temperature, when I have a proper vehicle control unit (VCU) doing big brain stuff like that.

To mount them to the radiator I designed and printed a simple flat panel that bolts to the existing radiator.

The ‘To Do’ list

I took our two other cars (soon to be one) into the garage for MOTs the other day and the owner Jimmy told me I left another puddle of ATF behind at the last attempt. I think the motor itself is leaking, so that needs addressing before the next attempt. As well as redoing the layout of the motor cooling system to make it more accessible.

If I can get away with not modifying the adaptor plate this time, I will. I had planned to trim it down a bit and clean it up. If I have to take the gearbox out then I might still do this. But actually I think I might be better just making a new one and swapping them out once the car is running. With everything I know now I could make it much lighter and neater.

But the main thing now is to get this car on the road. Some changes in the kids’ weekend activities means that now we unexpectedly need two cars again, just as I have found a buyer for my Alfa. This means it might be challenging to do the body swap as planned this winter, unless I can take a week off and get the car stripped and road legal again in that time.

There is also the challenge of the garage. I got so close to getting one I thought, but the owner just stopped responding to my calls and messages. So I need to re-open that hunt. Certainly before I launch into…

The next project

Yes, I know, this one isn’t finished yet (and will likely never be truly ‘finished’). But somehow over the last year and a bit, I have accumulated pretty much all the parts to build a second EV. And I’m really keen to put everything I have learned to use. So at some point, I want a second donor car.

This will obviously need somewhere to live (a garage), so I won’t be buying anything until I have that. But I do have pretty much everything else. I picked up a rear motor from an Outlander for a good price over summer. I already had the matching inverter, a second battery pack, and most the ancillaries, including everything for a high voltage junction box (which I built for the original car but ended up not using.

Alongside getting the actual car running, I’ve started prepping these for assembly as well. This includes things like replacing the knackered connectors on the inverter, which looks like it might have been in a fire (though it seems to work fine – I tested it last year).

To this end I’ve designed up an adaptor plate to mount some cheap AliExpress-sourced waterproof connectors I’ve come to quite like.

This is just the beginning though. The plan is to get the second set up fully running in my workshop – with charging, cooling and everything – so that it can be transplanted straight into either the existing car as an upgrade (the new motor has perhaps better torque characteristics), or into the second car if/when I have space for that. This will also give me something to work on (because I’m so short of projects to consume my time) in the wet winter months when it’s hard to get out and work on the car itself.

Unless I can finally get a garage…

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.