openinverter forum

Here it is! Spinning it with just manualid for now. But I know I can go find the proper syncofs value for this motor from here now.

Nice!

Today was syncofs tuning day. I was able to dial it in suspiciously quickly. Unfortunately, I managed to leave my multimeter turned on for the past few days so the battery ran flat. This meant I couldn't figure out the wiring of the BMW e60 throttle pedal I'm planning to use. So, I'll need to come back to this to confirm that I have the right syncofs value.

(on reflection, I'm sure I have a pot lying around here I could use to test this out)
Next, I tried to do a quick test to determine how much of the Lexus CVT will fit into the transmission tunnel. Apparently the mustang tunnel is notoriously narrow. The original transmission mount point is a piece of U-shaped steel welded to the floor pan. It links into both of the frame rails. I don't want to touch this as it's helping to strengthen the body. The lower part is just bolted on, so this can be removed.
Next I did a little CAD work. I trimmed an A4 piece of paper until it matched the curve of the transmission mount. This is the most narrow part of the transmission tunnel. I then traced this on to a piece of card to cut out the negative version of this. The distance from the face of the firewall back to the transmission mount is around 400mm. Ideally all of the CVT would go into the tunnel to leave the entire engine bay free for batteries and other parts. I measured 400mm back from the front face of the CVT and placed my cut out over this area.
It's a little tight, but it looks like it could fit OK. One concern is the shifter arm. You can see it on the right of the picture above. I want to keep this and link it into the original shifter control from the car. It's a little tight, but I think it's doable.

I also measured the height of the CVT at this point to see how close to the ground it would reach. From my rough measurement it would be about in line with the underside of the frame rails.
But really, the limiting factor might be how much the tunnel flairs out. It looks like the CVT might be wider at the front than the tunnel. I need to be sure the CVT is far enough back that it won't interfere with the steering rack.

This is all just speculation anyway. I just need to crane the CVT into the engine bay and check it for fit directly!

I popped a fresh 9V into my multimeter and it’s back in action! This allowed me to go through the process of figuring out the wiring of my BMW e61 pedal. There are 6 connections in total. According to the wiring diagram, the pedal has two sensors in it. Each sensor takes a 5V, a ground, and has a single output signal. I tried a few combinations of applying the 5V + ground and then probed around with the pedal in different positions to find the correct signal. I found the right combination which gave me close to 0V with the pedal fully ‘out’ and close to 5V with the pedal fully ‘in’.

With the pedal properly wired up to the inverter, I tried putting the inverter into normal run mode and pressing the pedal … anddd …. nothing. Just buzzing and a little shudder.

As I suspected previously I probably had the wrong value for syncofs. So, I started over finding this value. This time I went especially slowly and stepped through small increments when changing syncofs. When I thought I had found that it was around 22k, I went up to around 26k and came back down to try and ensure I was zeroing in in the right spot.

With a value of 22k I tried again in normal run mode and …



For now that’s just MG2 spinnng. I have spun both MG2 and MG1 in manual mode. I just need to tune in the correct syncofs for MG1 and both motors will then be spinning the output shaft.

Did a first fit of the motor in the transmission tunnel this evening. It fits better than I expected to be honest. Not perfect, but not terrible.
The oil pan will stick out a bit below the frame rails, but not by much, maybe 30mm or so.
The steering rack bolts to the frame rails here. You can see three bolt holes on the driver side and two on the passenger side. The motor more or less needs to sit behind these bolt holes. There’s a little room to move the motor forward a bit, which I’ll need to do as it’s actually touching the transmission tunnel right now.
A bit of an issue here though. I would like to keep the original shifter, but there may not be enough room.

Some rear battery box design this evening.

I love a bit of cad

Well done with the motor, slightly jealous of you

rstevens81 wrote: ↑Tue Sep 07, 2021 8:45 am I love a bit of cad

Well done with the motor, slightly jealous of you

Haha, yeah, turns out a bit of old cardboard is a great tool

Doing a few sketches to see how I might mount the front part of my motor in the engine bay.

At first I thought I might just put a 50mm x 50mm piece of box across the front and mount it to the frame rail on either side. The problem with this is that I want to keep the original steering rack as-is and this would be in the way.
Whatever I do, I’ll want to put a flat plate over the face of the motor. I’ll take the splined shaft from the flywheel and fix it to the inside of the plate. This will lock the ICE input shaft of the CVT in place so I can use both of the motors in the CVT to produce traction. The plate will also be part of the mount which will hold the motor in the car.

So I’m looking at two different designs. The first reaches back a bit to hit a spot on the frame rails where there is space to attach. The cone is a simplifed representation of the CVT. The cylinder sitting between the two brackets is a rubber damper to soak up vibrations in the motor.
The other option is to reach forward and link up with the original motor mounts.
I still have to consider how I’ll attach the rear part of the CVT under the car. The mounting area of the CVT will be much further back than the original mustang mounting point.

Hi Christian,
thanks for all your published work on this conversion. I think this and related RWD units are going to be great for budget RWD classics conversions, I'd like to try the AWD L110 in a 4x4, should be fun! Excellent work getting this one to spin.
Your work is several steps ahead of mine viewtopic.php?f=11&t=1466&p=24521&hilit=locost#p24521 which has been halted for a while till now for family reasons. I'm using the same CVT in a Locost frame and it seems to fit well. I was stumped for a while just trying to program the wifi modules !
For what it's worth, I think the former front mount is good unless you combine the motor mount with a battery box ?
keep up the good work!

great build, I'm using the same battery pack for my GS450H

chrskly wrote: ↑Thu Sep 02, 2021 7:03 pm I popped a fresh 9V into my multimeter and it’s back in action! This allowed me to go through the process of figuring out the wiring of my BMW e61 pedal. There are 6 connections in total. According to the wiring diagram, the pedal has two sensors in it. Each sensor takes a 5V, a ground, and has a single output signal. I tried a few combinations of applying the 5V + ground and then probed around with the pedal in different positions to find the correct signal. I found the right combination which gave me close to 0V with the pedal fully ‘out’ and close to 5V with the pedal fully ‘in’.

Could you tell us the right wiring of the bmw pedal, please? Just 1-6 pins as they are signed

chuuux wrote: ↑Thu Nov 11, 2021 5:20 pm

chrskly wrote: ↑Thu Sep 02, 2021 7:03 pm I popped a fresh 9V into my multimeter and it’s back in action! This allowed me to go through the process of figuring out the wiring of my BMW e61 pedal. There are 6 connections in total. According to the wiring diagram, the pedal has two sensors in it. Each sensor takes a 5V, a ground, and has a single output signal. I tried a few combinations of applying the 5V + ground and then probed around with the pedal in different positions to find the correct signal. I found the right combination which gave me close to 0V with the pedal fully ‘out’ and close to 5V with the pedal fully ‘in’.

Could you tell us the right wiring of the bmw pedal, please? Just 1-6 pins as they are signed

Sure, I popped out to the garage there just to see what way I have it wired.

Pedal pin 1 <-> Gnd
Pedal pin 5 <-> +5v
Pedal pin 4 <-> MG2_ACCEL (inverter)

Here are some pics so you can double-check for yourself.
I only have one of the two pots wired up here of course. When it's in the car I'll use both pots most likely.

chrskly wrote: ↑Thu Nov 11, 2021 6:35 pm

chuuux wrote: ↑Thu Nov 11, 2021 5:20 pm

chrskly wrote: ↑Thu Sep 02, 2021 7:03 pm I popped a fresh 9V into my multimeter and it’s back in action! This allowed me to go through the process of figuring out the wiring of my BMW e61 pedal. There are 6 connections in total. According to the wiring diagram, the pedal has two sensors in it. Each sensor takes a 5V, a ground, and has a single output signal. I tried a few combinations of applying the 5V + ground and then probed around with the pedal in different positions to find the correct signal. I found the right combination which gave me close to 0V with the pedal fully ‘out’ and close to 5V with the pedal fully ‘in’.

Could you tell us the right wiring of the bmw pedal, please? Just 1-6 pins as they are signed

Sure, I popped out to the garage there just to see what way I have it wired.

Pedal pin 1 <-> Gnd
Pedal pin 5 <-> +5v
Pedal pin 4 <-> MG2_ACCEL (inverter)

Here are some pics so you can double-check for yourself.

IMG_0017.jpg
IMG_0018.jpg

I only have one of the two pots wired up here of course. When it's in the car I'll use both pots most likely.

Thanks! It works.

I've found the second pot (0-2V) wiring :

Pedal pin 2 <-> Gnd
Pedal pin 3 <-> +5v
Pedal pin 6 <-> MG2_ACCEL (inverter)

chuuux wrote: ↑Thu Nov 11, 2021 7:30 pm I've found the second pot (0-2V) wiring :

Pedal pin 2 <-> Gnd
Pedal pin 3 <-> +5v
Pedal pin 6 <-> MG2_ACCEL (inverter)

Nice, thanks

This might be a little late, but I gathered some information from the BMW Schematics and publicly available parts catalogues about the throttle pedals in the E46 and E60/E61: https://openinverter.org/wiki/BMW_Elect ... ttle_Pedal

It would be nice to have precise measurements of the output voltages at 0% and 100% throttle.

I just dug through the BMW Wiring Diagram System and the nominal voltage ranges are actually specified. I added it to the wiki too.
If anyone is looking for it: it's only in the DDE5, DDE6, ... sections. The pedals for gasoline and diesel engines are the same though, so the information applies for both.

With a socket finally wired up in my garage for my welder, I'm getting stuck into making up the rear battery box. I'm making a frame out of 3mm x 30mm right angle steel. I then plan to line it with 1.5mm sheet. Don't look too close at my terrible welding! This is the first thing I've ever welded.

The rear battery box will go in the space where the fuel tank used to sit. In this car, the fuel tank has a lip running around the outside edge. It sits down into a square opening in the chassis - the lip holding it up. The top of the fuel tank acts as the floor of some of the boot space.

At one point, I tried cutting the top off of the original fuel tank to try and re-use it. But it has rounded edges which make the usable space too small to fit all of my batteries.

So the plan is to make a square-edged tub which sits down into the opening in the chassis and has a lip like the original fuel tank.

I started at the top, making the lip part. I then welded on the four vertical pieces into the corners. The total height of the box will be 200mm, which is a little taller than the original fuel tank, but there is plenty of clearance back there. Before putting the bottom of the frame on, I tested the fit against all six modules on the bench. There is only around 30mm clearance in the width and length of the battery box, so I wanted to make sure I didn't screw up the dimensions somehow.

There will also be plenty of space for cables to exit the battery box through the side/rear wall. I'll need this in particular for the high voltage cables which will travel under the car to the high voltage junction box in the front.
Here is the frame in place - with two battery modules very carefully balanced. It fits very well.
I'm also working on the front motor mount. I've got a sheet of 6mm steel which I plan to use for this. I'm doing a little CAD (cardboard aided design) to figure out the dimensions for this plate.

You ever get this finished, @Christian?

aaronmcg wrote: ↑Fri Mar 10, 2023 5:27 pm You ever get this finished, @Christian?

Look into Christians signature for his blog.

Yeah, I'm still working away on it. The problem is I broke the first rule of EV conversion club : don't pick a rust bucket I've mostly been working away on rust repair, but there have been some developments on the conversion front. It's been a while since I've posted an update here, so, will post something later on.

I made up some of the front mount to hold the motor in the transmission tunnel. I cut it from a big 6mm steel plate with an angle grinder. Probably overkill.
These are the upper brackets that will bolt to the big plate.

Another major piece of work has been around the BMS. I have two BMW PHEV packs that I plan to put in the car. They'll be paralleled together. If I were using just one, I could just use SimpBMS and call it a day. But since I'm using two, things are a little more tricky. The two packs will not be permanently paralleled together, just when driving/charging. I could use two SimpBMS, but I would still want to have something monitoring both Simps to get the overall status. I'm thinking here about situations like where the packs get out of sync with each other for some reason.

So (because I like to do things the hard way apparently) I'm developing a custom Simp-like BMS.

It has three CAN ports. One for each battery pack and one for communication with the rest of the car. Because I'll be using two identical packs the CAN ids from the CSCs will overlap. Hence the need for one port for each pack.

I'm using the Raspberry Pi Pico as the microcontroller.

This is me testing v3 of the hardware (from the comfort of the sofa) with one BMS module.
I started with a blank slate, code-wise. But, I adapted a few chunks of code directly from Simp around polling the modules - Yay opensource! You can see the hardware design files and code over here. I'm trying to keep it as clean and understandable as possible. I'm going to have to fix issues in the future and I want to at least try to save my future self a lot of pain.

The latest (and hopefully final) version of the hardware is v4. I saw someone in another project using a very cool transparent case and I decided I had to have one too.
This version also has 6 digital inputs and 6 low side switches.

The low side switches will be used for the following:

1. Turn on a battery heater when I'm charging in cold weather. It won't come up often in this climate, but better to have the option there than try and retrofit it later.
2. Generate a DRIVE_INHIBIT signal.
3. Generate a CHARGE_INHIBIT signal.
4. I'll also have one BATTERY_INHIBIT signal for each battery.

The digitial inputs will be used for the following:

1. IGNITION_ENABLE
2. CHARGE_ENABLE

I've been trying to assimilate safety considerations from lots of different sources. This is unlikely to be everything, but this is what I'm thinking so far.

I want the BMS to be able to prevent the car from driving under certain circumstances.
1. When the battery is empty
2. When the battery is too hot
3. When the car is currently charging (drive away prevention)

Opening the contactors when driving is not a good idea. As the magnetic field in the motor collapses it will generate a surge of energy which would normally get absorbed by the battery. If the contactors are open, it has nowhere to go. So, when driving, if there is an issue which means the car needs to stop, then the BMS will turn on the DRIVE_INHIBIT signal instead of opening the contactors. How this will work is that the forward and reverse signals going from the gear selector to the inverter will pass through the normally closed side of a relay. The DRIVE_INHIBIT signal will turn this relay 'on' which will turn off both of these signals and effectively put the car into neutral.

But the BMS will have the ability to open the contactors in certain circumstances. Basically as long as the car is not in drive mode, the BMS can open the contactors. This will work in a similar way to the DRIVE_INHIBIT signal. The contactor control line will pass through the normally closed side of a relay and the BMS can turn that relay 'on' to disable one or other battery pack.

The design here is to work by default. If there is a problem with the BMS (or I think I know better than the BMS for some reason) I can disable it and the car will work.

I also want to disallow the contactors from closing when the voltage difference between the two packs is above a certain level. I don't want a big, uncontrolled current flow from one pack to the other.

This design can allow some other nice things. If your packs voltages are different, you can choose to close the contators for one of the packs and still drive the car. Also, if you close the contactor on the 'high' pack you can drive off the extra energy and when the two packs match again, close the contactors on the other pack. So, they can automatically resync.

The same thing applies for charging, you just need to deal with it the other way around. When going into charge mode, you close the contactor on the 'low' pack. Then when the low pack is charged up to the level of the high pack you can close the contactors on the high pack and continue charging.

The BMS will also send out max charging limits based on temperature that the charge controller will need to listen to. Which I'll need to 'discover'.

It will also need to deal with low temperatures as well. Don't want to damage the pack by charging below freezing. This is where the heater control and CHARGE_INHIBIT signal will come in. The BMS will get the CHARGE_ENABLE signal from the charge controller. The BMS will turn on the CHARGE_INHIBIT signal and the heater if the temperature is too low (say, below 3 degrees). The charge controller will pause charging until it sees the CHARGE_INHIBIT signal go low again.

There's probably other stuff I'm forgetting ...

Was interested to take a look at your code but can’t find it?