tl;dr: the car has passed its MOT! We are now the proud, grin-wearing owners of a road-legal, home-made, fully-electric vehicle.
Getting to that point though, was something of a rollercoaster. Rewind to my last post. I was in the middle of a bunch of fixes to get the car MOT ready again. This included:
- Making a new coupler
- Bridging MG1 and MG2 to increase power
- Fitting cooling fans
- Upgrading the cooling pump
- Relocating the radiator to improve the bleeding of air from the system
- Redoing the motor cooling system
- Finally fixing the tracking
Let’s start with a quick update on each of those.
The lathe made this so much easier. I used it to turn down the two clutch centres to ensure that their inner and outer diameters were aligned, as well as to take any unnecessary material off them. I actually increased the diameter of the motor-end centre with some chunky welds so that I could then turn down a lip that would sit perfectly square against the – now square – tube. You can see that in this pic taken just as I started to tack things together.
Then I made an alignment tool by turning down a piece of bar to perfectly fit each of the centres. This meant the two centres were now aligned in three different ways:
- By their fit against the inside of the tube
- By their squareness against the end of the tube
- By their fit to the alignment tube
With all this in place I slowly built up welds at both ends before grinding them back and then finishing them in the lathe to get a (fairly) even finish.
The result is brilliant. It doesn’t look amazing, but refitted to the motor and gearbox (and packed with copper grease), it spins very smoothly and quietly. And thankfully, I didn’t need to remake the adaptor plate to suit the tweaked alignment. It just all slotted together with no issue.
Not much to say here. I finished making up copper wires and custom terminals to bridge the gaps between MG1 and MG2. I don’t seem to have taken any photos of the finished thing. Suffice to say I ended up using a hammer to flatten things down so that the lid would fit back on, and adding a lot of insulation to make sure the lid didn’t come into contact with anything. Not the prettiest job. And I’m yet to see what effect it has had.
I made up a new bracket to mount the top of the radiator almost level with the top of the inverter so that the filler cap just clears the bonnet. And the fans on their 3D-printed bracket are also now in:
Annoyingly I ran out of waterproof 2-way plugs/sockets before fitting these so they are hard-wired in (via a fuse). But I’ll order some more and put one in.
Note: the steam pipe is now cut and sealed at each end since this photo was taken for reasons I shall explain below.
My Bosch PCE pump from a Seat Ibiza arrived and once I worked out why it didn’t seem to be working (it has built-in dry run protection that seems to want pressure on the inlet and outlet), I got it installed.
Instead of having it feed into the charger first, as the old pump did, I moved the pump to feed straight into the inverter, oriented it on its side per the datasheet, and mounted it to the steel frame that carries the inverter with some cable ties and a 3D-printed bracket.
I also redid all the coolant pipes into a neater circuit. This caused no end of problems because lots of my cheap hose clamps gave up the ghost in the process. It’s all slightly bodged together with cable ties at the moment, but largely holding water. I have ordered and received some seriously beefy hose clamps (Mikalor) to replace the problem ones soon.
While I was trying to bleed the system I realised that the PCE pump is so strong it was creating a vacuum in the left hand radiator chamber because water wasn’t flowing through the core fast enough and it could then suck air through the steam hose from the other side. I cut and sealed the hose as above, topped up the coolant and since then it has been fine.
With a limited level of tune, all stall events have completely stopped. For now…
One of the major fails from the last MOT was a major hydraulic leak. I believed this to be coming from the motor, so decided to completely redo the oil cooling system.
I relocated it to the front valance, and while I had the motor out, made up a catch tray for testing and filled the motor with ATF to see where the leaks were.
I couldn’t find any, so I figured the leak must have been from the cooling system.
I went over the rebuilt system multiple times checking for leaks, adding PTFE to screw threads and replacing crappy hose clamps (I had good smaller ones for these 10mm hoses, just not for the 19mm water hoses). Eventually, I was happy there was no more leak.
I locked the steering wheel straight and adjusted the track rod ends to try and get the wheels also straight. But I soon realised that more adjustment was needed than the ends could offer. Clearly I had screwed up when I refitted the steering rack. So off it came again. I roughly aligned the wheels and steering wheel with the track rod ends centred, and connected it all back up. Then I did final tweaks to the rod ends. It still needs doing perfectly but it was good enough for the MOT.
I put everything back together and gave it a little test. It all worked – at least to drive…
So, last Monday (20th September 2021), I called the garage and booked in a new MOT. I still needed to try some tuning to be sure it wouldn’t stall again, but I thought the car was ready. I got a test date for the Friday.
That afternoon I went out to do some tuning. And found I couldn’t access the parameters for the inverter via the web interface. Somehow in assembly the newly moved WiFi adaptor had become detached from the main board. So I had to drain the coolant, disconnect all the HV cables and control wires, take the inverter out again, and dismantle it before doing it all again in reverse.
That done, throughout the week I did lots of little tests. I tried completely resetting the sync offset. But eventually I ended up going back to the parameters established by Jamie on his BugPlug. Then on the Thursday I thought I would go out and just top up the batteries….
First, I found I had a puncture and one of the front tyres was flat. I topped it up thankfully the pressure has held since.
Then I got distracted half way through charging and went into the house. When I came back out, I forgot that I had temporarily rigged up the DC-DC converter as the 12V battery was getting low and also needed a top-up. I opened up my laptop, which was still sending commands to the charger to pull in current, and turned on the car.
This meant that the DC-DC converter and the batteries started sucking in current through the pre-charge system, rather than through the dirty great contactors designed to handle that load. The pre-charge resistor went bang in a puff of smoke, totally melting the cable ties that held it in place.
Panic. But not for long. I pulled it all out, still smoking, and stripped it down. A few tests found that thankfully I had only destroyed the resistor and I still had a spare. I slotted this in, tied it down, reassembled the junction box (actually the BMW S-Box from one of my battery packs) that houses it all, and went back to charging.
It all worked until just as I got to the end of my planned charge. At which point my DC bus voltage dropped from over 300V to 70V and the charger cut out. WTF?
I pulled the junction box out again only to find that in my haste, I hadn’t tightened the nuts on one of the bus bars. With those nipped down, everything worked again. Except my nerves, which were shot.
Day of the test
So, the day of the test came and deep down I knew it was going to fail. I just had a total sense of foreboding. I was right.
The mechanic found a leak in the near-side rear brake union, which meant he couldn’t even test the brakes. More importantly, there was still a major hydraulic leak. He got me under the car on the ramp to show me ATF pouring down. And straight away, I knew where it was coming from.
It had never been the bloody motor. It was the high pressure line on the power steering pump all along.
The last fixes
This was an easy fix. The regulation metric f*ckton of PTFE was applied to the threads, albeit this meant disconnecting the hose at both ends and completely removing the power steering pump, taking yet another bath in ATF in the process.
I removed the brake union and re-flared the line, which soon cured the leak there.
While I had the car in the air, I also adjusted the front ride height as the mechanic had said it had been hard to get it on the ramp. It’s still very low but it clears my driveway entrance now.
Unfortunately they couldn’t retest it that day, so I had to book a retest for today, Wednesday 29th September. After a lot of waiting around and chewing of nails (and McDonald’s breakfasts) due to a miscommunication about the booking time, I finally got the pass.
And I’m still grinning.
This has been an incredibly long post. I’m aiming to do much shorter ones after this, with one post for each little additional job that I do on the car. So if you’ve got this far, thanks for sticking with me. And thanks for all your support.
What am I working on now? Well, a little temporary badge for the boot to replace the old ‘1.8’
A 3D-printed tablet mount so that I have easier access to gauges from the inverter and ideally readouts from the battery pack while I work on hooking up the fuel gauge.
And a heater. The weather has turned and happy as I was on the way back from the test centre, I was chilly. I have a new heater to play with. But that is a story for another day…