DIY Electric Car Forums banner

1 - 20 of 81 Posts

·
Registered
Joined
·
459 Posts
Discussion Starter #1
I'm researching using a Leaf as a donor for a classic conversion, and it seems like the state of things is moving so fast that even 2-year-old threads are outdated.

What is the current state? Like, instead of swapping in the whole car (which is apparently very difficult to do even after it is running on a bench), can I just use motor, reduction, and battery pack with some aftermarket controller?

What's the current easiest way to leverage cheap Leaf components in a car that never dreamed it could be electric?


Here is the megathread on Leaf donors:

https://www.diyelectriccar.com/forums/showthread.php/re-using-complete-leaf-drive-system-151458.html?highlight=leaf


Here's the megathread on a homebrew controller that may or may not exist as a purchasable product:

https://www.diyelectriccar.com/forums/showthread.php/another-homebrew-ac-controller-45909.html?highlight=johannes+huebner


Here is a thread stating all of the components needed for a complete Leaf swap to function:

https://www.diyelectriccar.com/forums/showthread.php?t=184785&highlight=leaf

The minimum we have found that works is:

Complete motor stack without AC (harness unmodified) including the park lock actuator.
Main ECU (VCM)
BCM
IPDM
J1772 inlet
Chademo inlet
Water pump
Coolant temperature sensor
Brake pedal switches (not the brake pot)
Accelerator HEPA
Start button and encoder
Two key sensors (both dash area, one above hvac and one in centre console)
Telematics module
Key security module (the bit welded to the dash crossmember)
Two original relay boxes from the engine bay harness
Repackaged HV battery with standard BMS
Heavily modified dash harness (with interior fusebox)
Body harness stripped down to just HV battery comms connector
The key
Prndl (gear selector)
Prndl light
Binnacle lower part
Sourcing a wrecked Leaf and using all of the above components doesn't seem that bad (though I would want to hide it all as much as possible), but nobody seems to have documented a full conversion, so clearly there are some difficulties I'm not seeing.
 

·
Registered
Joined
·
4,796 Posts
Here's a link to the Brain Board that I believe to be an end result of the megathread?

https://openinverter.org/shop/index.php?route=product/product&product_id=53

Worth noting is that the Leaf motor seems to be limited to 3,000 RPM, which greatly reduces its value/capability.
Where do you get that from?

In the Leaf it has got to run to over 8,000 rpm
If you read the linked page, you'll see that this controller isn't able to run the motor properly at high speed, so it - the Brain Board - is unable to run the Leaf motor over 3,000 rpm:
the synchronous motor control software does not support "field weakening" yet. This means it will only spin teh Leaf motor to 3000rpm. Field weakening is under development.
And the stock pre-2018 Leaf configuration maintains full power to 9800 rpm (where it is still 92% efficient at full load), then falls off rapidly to zero at 10,500 rpm.

I agree that this major functional omission reduces the value of the Brain Board for running the Leaf motor (or any synchronous motor)... to essentially zero, for now.
 

·
Registered
Joined
·
23 Posts
I'm keen to see what feedback you get on this thread as it is something I am also very interested in. I have been following this subject for a while both on this forum and elsewhere.

As you hinted at, there are quite a few parties out there working on different approaches to controlling OEM motors, with the Leaf being one of the key candidates.

At a high level they fit into four different approaches,

One is to retain all of the Leaf components intact and to basically have a leaf running inside a different shell.
While this may seem like the easy option on the face of it, the different systems within any modern car are very closely interconnected and while you may have all the necessary parts for the drive and battery systems, a missing sensor signal elsewhere in the network (brakes, airbags etc.) could result in the native VCM (Vehicle Control Module) logging an error and shutting down the drive system, or at least going into limp mode.

The second option would be to replicate the CAN messages to inverter to control it.
This does away with some of the concerns above, but not all of them. The inverter only looks out for a subset of the messages that the native VCM receives and interprets, so the fact that many of them are missing is irrelevant to it. The challenge comes in sending the right CAN messages to the inverter at the right time. This includes both those messages based on specific inputs (throttle, brake etc.) and those messages required to confirm to the inverter that everything else in the car is okay so it doesn't log any errors.

The thunderstruck VCM seems to be the main commercially available version of this and it looks like they have done a lot of the work to reverse engineer the messages the inverter needs to receive. There are also others that have put up bits and pieces on github detailing open source ways of doing this, the most complete seems to be the GEVCU repository from Collin Kidder. There is also a reasonable amount of CAN capture data for the Leaf out there, but I have yet to see a complete mapping of all the messages, so some further investigation would be required if taking on this effort yourself.

The third option, is the idea of replacing the native logic board of the inverter with a programmable version with software that can be configured or edited to control the inverter powerstage appropriately.
This option retains the major components of the inverter (IGBTs, Capacitors, Current Sensors and the driver and monitoring circuits) but replaces the board that interprets the messages coming in from the VCM. This removes the need to replicate the Leaf can messages and in theory opens up the option of pushing the power stage and motor beyond the limits that Nissan have put in place, should you want to try that.

The fourth option is to do away with the native inverter altogether and replace it with a custom made inverter. This would allow you to choose your own components and could provide further scope for driving more power through to the motor.

options 3 and 4 have gained ground over the last few years, with a lot of the work that happened in the option 4 space (software and pcb design for open source inverters) being carried across into the drop in logic board space. Different architectures exist from 3 or 4 groups including Paul Holmes in the US, Arlin Sansome in Canada, Johannes Huebner in Germany and Damien Maguire in Ireland all featuring. As these guys started to reverse engineer the native inverters some options have appeared allowing these logic boards designed initially for an diy inverters to be hooked into the native inverter powerstage.

This area has been moving pretty quickly recently with the various logic boards being developed coming out with new features as their designers figure things out. For example the inclusion of support for both encoders and resolvers for receiving in data from the motor has appeared on the different open source logic boards over the last six months or so.

As I understand it, different people in the field have taken slightly different approaches to the design of the logic boards and as a result have different combinations of the features that could be included. As more of these drop in options are built and these different architectures are tried out with various OEM inverters and motors, any features that may be missing are being identified and we are seeing software enhancements or even new board designs coming out to meet these new requirements.

In terms of Leaf specific drop in boards, the two available options if you want to purchase a complete board seem to be the Huebner and Holmes designs. As I understand it, the Holmes board already incorporates Field Weakening into the design, while the Huebner board is working on that functionality.

I am just starting a new conversion based around the Leaf drive unit and I will be doing some experimenting with some of the options above before I get anywhere near to dropping it into a car. I'm happy to feedback on any of my findings is that's of any help.
 

·
Registered
Joined
·
238 Posts
So, this was an important post for me. I went back and re-read the list of needed components and realized that in all my attempts to get my motor to turn, I had not hooked up HV connections to the AC compressor or the cabin heaters. After my last attempts, I worked on other stuff and let all of that sit for over a year.



Wednesday, I hooked those parts up and reconnected the 12V battery and fired up the "car". The first thing that happened was the water pump started turning and squealing. Got that sorted out, put a couple hoses on it and stuck the ends in a bucket of water. Just because.


Next, I dropped it into Drive and it actually dropped into drive. And the throttle works. And it goes into Reverse and Neutral and Park and I was pretty excited.



I still have 25% charge on batteries that haven't been charged since 2015 so I bought an EVSE that I never bothered with since nothing was working. I had borrowed one a couple years ago but couldn't get ti to charge. I'm hoping that has changed, too, but will have to wait until next Wednesday to find out.



Time to get serious again.
Bill
 

·
Registered
Joined
·
23 Posts
That's an awesome milestone.

I am hoping it won't be too long before I get to that point, though as I only have the following items from the original list, I know I am going to get creative.

Complete motor stack without AC (harness unmodified) including the park lock actuator.
Main ECU (VCM)
IPDM
J1772 inlet
Chademo inlet

Will let you all know how I get on.
 

·
Registered
Joined
·
117 Posts
Here's a link to the Brain Board that I believe to be an end result of the megathread?

https://openinverter.org/shop/index.php?route=product/product&product_id=53

Worth noting is that this seems to limit the Leaf motor to 3,000 RPM, which greatly reduces its value/capability.

Why is this only coming to light now, yes on his website it points out the motor will not run past 3000rpm (yet ;) )

I went ahead and built Pauls stand alone controller, when I did a RPM check the motor max is still only 6800rpm which is still too slow if you want to use the leaf gearbox, which is what I wanted to use,

Does Pauls drop in board take the motor upto 10000rpm, Can anyone confirm this

6000 rpm with leaf motor/gearbox is only around 60mph, which is a bit of a let down.

Arlo's said his board will go upto 15000rpm with the right components but he doesn't recommend it.
Sounds good but the price doesnt, looks like it could cost upto $3000 is you by all the parts from him and the is still alot of building to do.

looks like field weakening is still in its early stages.
 

·
Registered
Joined
·
727 Posts
Does someone know if this a RPM limit based on the maximum stator frequency(not the operating frequency) of a particular inverter/controller? I've been looking at some Curtis and Sevcon controllers for synchronous (permanent magnet BLDC) and asynchronous(induction) motors, and they list a maximum stator frequency limit of 300Hz and 500Hz respectively. Is this a limitation because of some component(s) in the inverters/controllers?

For higher pole count motors this must be a RPM limiting factor.
 

·
Registered
Joined
·
4,796 Posts
Does someone know if this a RPM limit based on the maximum stator frequency(not the operating frequency) of a particular inverter/controller?
In the case of the openinverter.org Brain Board, this is not due to a maximum frequency, it is due to the lack of functionality in the controller's logic. As explained by openinverter.org:
the synchronous motor control software does not support "field weakening" yet. This means it will only spin teh Leaf motor to 3000rpm. Field weakening is under development.

looks like field weakening is still in its early stages.
In these amateur and aftermarket controllers, perhaps, but phase shift (normally called "field weakening" or "flux weakening", which I think just makes it more confusing) is a normal synchronous motor controller feature.


I've been looking at some Curtis and Sevcon controllers for synchronous (permanent magnet BLDC) and asynchronous(induction) motors, and they list a maximum stator frequency limit of 300Hz and 500Hz respectively. Is this a limitation because of some component(s) in the inverters/controllers?

For higher pole count motors this must be a RPM limiting factor.
That makes sense, and may even apply to the Brain Board at some higher motor speed, but I have not seen any information about that.
 

·
Registered
Joined
·
260 Posts
Paul informed me that his boards are fine with 10k+ RPM which is more than I need at 120kmh. This is without field weakening which he says is still in development on his system.
 

·
Registered
Joined
·
4,796 Posts
Paul informed me that his boards are fine with 10k+ RPM which is more than I need at 120kmh. This is without field weakening which he says is still in development on his system.
So the inverter can put out a frequency which corresponds to 10,000 rpm, but can it drive a Leaf motor at full power and efficiently up to that speed, without field weakening logic? Great if it can, not very useful if not.
 

·
Registered
Joined
·
260 Posts
So the inverter can put out a frequency which corresponds to 10,000 rpm, but can it drive a Leaf motor at full power and efficiently up to that speed, without field weakening logic? Great if it can, not very useful if not.

My context was that of using the motor and stock gear reducer. My question to him was whether I would have an issue with his board losing power under those conditions. He did not believe it would be a problem. That said, I did say that I was not going to be driving faster than 130kmh. Others may wish to do a lot more.
 

·
Registered
Joined
·
4,796 Posts
My context was that of using the motor and stock gear reducer. My question to him was whether I would have an issue with his board losing power under those conditions. He did not believe it would be a problem. That said, I did say that I was not going to be driving faster than 130kmh. Others may wish to do a lot more.
Even with stock Leaf tires, 130 km/h would be 8900 rpm (presumably less with the LT28's tires)... so while not the motor's top speed, it's fast enough that the ability to maintain power and efficiency is important. I hope it works... but apparently it's not yet tested (at least under a realistic load).
 

·
Registered
Joined
·
260 Posts
Even with stock Leaf tires, 130 km/h would be 8900 rpm (presumably less with the LT28's tires)... so while not the motor's top speed, it's fast enough that the ability to maintain power and efficiency is important. I hope it works... but apparently it's not yet tested (at least under a realistic load).

I will also have 2 motors so that may help with the loading issue.
 
1 - 20 of 81 Posts
Top