Power up version for complete Leaf

Nissan Leaf/e-NV200 drive stack topics
Post Reply
tomhanman
Posts: 13
Joined: Tue Apr 27, 2021 4:41 pm
Location: France
Been thanked: 1 time

Power up version for complete Leaf

Post by tomhanman »

Hello to all of you electronics experts. I am sadly not one of you. I will do my best to become like you, but I do not, in all honesty, hold out much hope.

I have been a kit car builder, errm thats a UK think, it means when we buy or build a chassis and bodywork and we use mechanics/electrics from a 'donor' car.

I've used ICE donors up until now (it seemed like a good idea at the time) but now I have purchased a 2017 Leaf as a donor (wall fell on it!)

Unlike your various projects where you convert an existing car to EV, I actually would like to use the Leaf system in as full as possible state because my chassis and bodywork has no dashboard or brakes or, well, anything. And that is the concept of a 'single donor kit car' which makes for the cheapest, and often most reliable (should that be least unreliable?).

So I will be trying, as others that have gone before me, to figure out which bits of the Leaf we need to keep to maintain functionality in dash, brakes with regen, inverter, charger, motor, battery (repackaged) and even 3G connectivity! But no airbags, multimedia or aircon etc. That will be a massive journey for me and will involve a baptism of canbus.

Anyway, what this basically means (at this stage) is that I don't want to use the open inverter to control the Leaf motor, I want to do it with the Leaf! BUT quite obviously I would very much like to have 130kW of power instead of the 80kW that pulls my daily driver Leaf about. Yeah, who wouldn't?!

So (finally he gets to a question) would it be conceivable to develop a version of this driverboard that interacts with the Leaf signals and does what a good old 'chipping' or 're-map' used to do. Hold on, I thought a re-chip was a new thing just a moment ago, doesn't time fly... So basically modify the torque requested by the Leaf so that the motor would give more? More, more and even more? Its always about more in the end!

Any ideas, advice, guidance dissuadance much appreciated. But please be gentle.

Much respect, and admiration to the lot of you,

Tom.
SRFirefox
Posts: 28
Joined: Thu Jun 25, 2020 2:28 am

Re: Power up version for complete Leaf

Post by SRFirefox »

The stock control board in the inverter will, to the best of my knowledge, never give you more than 80kW of drive. This is because, while the inverter has some nice headroom and the motor can handle it for a certain amount of time, the battery in the leaf cannot supply more power to the reliability standards of a production vehicle without severely reducing the lifetime of the cells. If you want to use the leaf inverter at the attained 130kW, you must replace the brain board with a Johannes brain, at which point it does not speak the same language as the original inverter control and would be effectively incompatible with the rest of the leaf systems. That said, if your kit car is significantly lighter than a Nissan Leaf, 80kW of go is still quite a bit for a lightweight electric car.
tomhanman
Posts: 13
Joined: Tue Apr 27, 2021 4:41 pm
Location: France
Been thanked: 1 time

Re: Power up version for complete Leaf

Post by tomhanman »

Thanks SRFirefox. That confirms my understanding of the situation. The question remains would it be possible to take Johannes brain board or elements of, to continue speaking Leaf?

I will be going ahead with the project, as a basis for a future series. And it will indeed be a lot lighter - I'm aiming for under 800kg, but the last one of these I built was 4wd had 180bhp and came in at 640kg. So this one will definitely be disappointing at 80kW.

I feel pretty sure that someone who very definitely isn't me, could in theory, take the required elements of the open inverter and piggyback them onto the original board. A bit like a man-in-the-middle canbus adaptor, but further down the line.

Candidates?
Cheers,
Tom.
User avatar
celeron55
Posts: 774
Joined: Thu Jul 04, 2019 3:04 pm
Location: Finland
Has thanked: 27 times
Been thanked: 110 times
Contact:

Re: Power up version for complete Leaf

Post by celeron55 »

It's not possible to modify the original board. It's all integrated into a processor there, the code of which we can't modify. If you knew the right person from Nissan you could get access (probably breaking company rules) but very unlikely anyone has such contacts.

However modifying a huebner based board to talk Nissan could be possible. It's never been done though. Even then there could still be a problem with the Leaf VCU if it's cross checking current or something from the BMS, in which case it needs to be replaced also, at which point talking Nissan makes no sense anymore.

All in all, it's probably not worth it but if you really want to get into it I guess you could hire someone to do what you want and hope for the best.

What I'd do is make sure you can charge without the inverter and if so, or after making it so (some reverse engineering), replace the inverter control board. If you want to use the dash, reverse engineer and adapt it separately using a MtM board or integrate control of it too to the new VCU. Lots of warning lights to "kill" in a modern dash. You need to sort out heater controls also, I suppose they go through the VCU originally. I guess alternatively you could keep everything except the inverter intact but cheat the VCU to think it's in control while it actually isn't.
User avatar
ZooKeeper
Posts: 507
Joined: Mon Mar 30, 2020 10:23 pm
Location: USA

Re: Power up version for complete Leaf

Post by ZooKeeper »

tomhanman wrote: Tue Apr 27, 2021 5:20 pm
So I will be trying, as others that have gone before me, to figure out which bits of the Leaf we need to keep to maintain functionality in dash, brakes with regen, inverter, charger, motor, battery (repackaged) and even 3G connectivity! But no airbags, multimedia or aircon etc. That will be a massive journey for me and will involve a baptism of canbus.
LOL, understatement of the month! :D
Anyway, what this basically means (at this stage) is that I don't want to use the open inverter to control the Leaf motor, I want to do it with the Leaf! BUT quite obviously I would very much like to have 130kW of power instead of the 80kW that pulls my daily driver Leaf about. Yeah, who wouldn't?!
That is "only" about a 50% increase, should be doable.
So (finally he gets to a question) would it be conceivable to develop a version of this driverboard that interacts with the Leaf signals and does what a good old 'chipping' or 're-map' used to do. Hold on, I thought a re-chip was a new thing just a moment ago, doesn't time fly... So basically modify the torque requested by the Leaf so that the motor would give more? More, more and even more? Its always about more in the end!
I believe it would be far simpler to use a OpenInverter VCU to run the Leaf controller as it does support CAN. Granted, one would need to know the messages required, but WolfLogics Dala and others have documented many of them.
Huebner VCU controlling a Gen2 Prius Inverter powering an MGR
"Talent is equally distributed but opportunity is not." - Leila Janah
SRFirefox
Posts: 28
Joined: Thu Jun 25, 2020 2:28 am

Re: Power up version for complete Leaf

Post by SRFirefox »

tomhanman wrote: Wed Apr 28, 2021 10:53 am Thanks SRFirefox. That confirms my understanding of the situation. The question remains would it be possible to take Johannes brain board or elements of, to continue speaking Leaf?

I will be going ahead with the project, as a basis for a future series. And it will indeed be a lot lighter - I'm aiming for under 800kg, but the last one of these I built was 4wd had 180bhp and came in at 640kg. So this one will definitely be disappointing at 80kW.

I feel pretty sure that someone who very definitely isn't me, could in theory, take the required elements of the open inverter and piggyback them onto the original board. A bit like a man-in-the-middle canbus adaptor, but further down the line.

Candidates?
Cheers,
Tom.
I think first step is to manage expectations as to EV power vs ICE power, the delivery of which is significantly different. For the arguments of race/kit cars, the biggest difference is going to show up at higher speeds. The top RPM of the Leaf motor is around 10krpm, if I remember right, which puts your top speed with a 24" tire around 90-100mph (~8:1 gear ratio). I'm aware of someone who's pushed this to 12-13k with a custom inverter in drag racing, but with the longer duty cycles of road racing that may not be a great idea. The efficiency and by extension continuous power would also be lower at those rpms.

At these speeds, assuming you're using most or all of the 80kW, a first gen Leaf motor will float around 140-150*C steady state after a few minutes of heat soak. You realistically can't run the motor continuously much harder than that; the enamels that keep the motor working will start to break down. This becomes very evident in the Tesla drive units, which will put out several hundred kW, but only for a few minutes before heatsoak renders them down to a more realistic 70-100kW continuous. There's a road race Cobra EV project report in the projects section if you want more details on that RnD.

If you want more power, realistically resigning yourself to adding a second motor is your most realistic option. Even something as simple as a cut down Prius transmission on the front axle would give you that extra 40kW you're looking for. Damien has a recent video on his youtube/vimeo showing progress on the Zombieverter project running a Prius inverter for MG1/2 using the stock inverter. Some clever code writing and you could probably get it to command both the Leaf and Prius inverters at the same time with a single ZombieVerter.

Finally a comment on the leaf's power delivery. Reference this Oak Ridge Natl. Labs report, page 12. If you look at the efficiency graph on that page it shows mostly constant ~275Nm of torque to 3500rpm and then constant 80kW to 10500rpm. If you run your motor at that 3500rpm point continuously, I would expect the continuous power capability to be less than 80kW. At that point the efficiency is about 90%, or creating 2.5 times the waste heat as operation at the benchmarked 7000rpm at 96% efficiency. So more than 80kW at anything other than that high efficiency zone won't be sustainable over long distances and times. More to the point though, once you hit 3500rpm, you have 105 shaft horsepower until you hit your top speed at 100mph. This is going to feel markedly different to an ICE engine power curve as you row through the gears as it peaks and drops and peaks and drops. My point being - find out if the stock inverter will do what you want before you go chasing higher power for road based applications. Especially as I doubt you'll see any extended term improvements on power as the motor reaches its thermal limit.

This was very much stream of consciousness of thoughts from the engineering applications side of my brain one of my mentors taught me. I know settling for less power is not a very car-person hot-rod thing to do, but it may work best in your application. Alternately, you may build your car, find out that I'm full of crap, and install the Johannes brain and squeeze out the extra 50kW. I hope the thoughts help regardless.
tomhanman
Posts: 13
Joined: Tue Apr 27, 2021 4:41 pm
Location: France
Been thanked: 1 time

Re: Power up version for complete Leaf

Post by tomhanman »

Amy thanks to all of you. It is unusual to find that nobody is even slightly optimistic regarding s project! I don't mean that in a bad way. I mean it in a way that suggests I may have to actually listen to some advice given, which has never been my strong point. Although in later life I am learning.

Thanks for your clarity on the original board Celeron. It's good to know absolute limits. Although, and I don't know enough about how things work to says this, but I'll say it anyway, that board analyses everything and then sends commands to control a few igbt (is that right?). So wouldn't the key to this being intercepting those control signals and scaling them? Of course, like you say, if there is teacher (VCU) checking currents fro the battery to compare at the same time then that could bring it all to a grinding halt. But is that concept of scaling a viable one or complete nonsense?

The other options you outline are well considered, and I will have to consider all options, but I would always like to keep it pure and modify the least number of things. One upside, there is no heater - two seater without roof, so I may use the seat heaters, but that'll be it for climate control!

Unfortunately hiring someone can only be done if one has funds. Unless you're a country spending your taxpayer's children's future sweat, then you can just blindly borrow.

And that is part of the basis of this, and every project I undertake, it has to be efficient. I understand the basic approach taken by the opInv community and suggested by ZooKeeper, But the Leaf has a 6kW charger, a DC fast charge solution, a dcdc converter, a BMS a digital dash (or two) and a half decent battery all included. It is soooo tempting to keep all that in place, and in the kind of projects I'm looking at it is almost vital. At 2500€ for the entire donor vehicle that makes an electric car actually viable. Not nearly as exciting as one based on Tesla components or a mix of Toyota and something else, but viable. Ding away with DC charging, and buying a charger and DC converter adds significant cost and just seems inefficient, etc etc.

SRFirefox. I like ramblings, and I appreciate you taking the time to write yours down, it can get very frustrating as (as Mr Musk points out) our interfered just isn't sufficient to keep up with the processing power. But, sadly I just don't think there's any escaping the power question. I currently have a Leaf, and one of my previous cars was a Citroen C4 HDi 110. It was faster. It was a turbo diesel and it felt and went faster. So same power and even with all the wonders we recount about electric torque delivery the Leaf (heavier yes) feels painful beyond about 60kph.

The previous ice incarnation of this vehicle that I made allusions to had a gearbox that limited top speed (6000rpm in 5th) to about 93mph. Which as almost exactly the same as the Leaf manages. So whilst it doesn't have to change gear to get there it doesn't have any hearing advantages. It will definitely be nippy at the low end, but it could only member be considered fast if it was to get that 50% power-up.

And then I'd really seriously question the idea of extended duty cycling on the road. This is not a racing car, it will not be tuned around the Nurburgring! It is for enjoying whilst keeping one's licence intact. I don't really see any way to require 175bhp for more than about 5 seconds. It just can't be done, the car would be at full speed, and probably only requiring about 80bhp to stay there. The motor has active cooling, so by the time the next requirement for silly acceleration comes around it should have cleared the heat to the rad. A Tesla can do over 150mph has a large battery capacity, is as wide as a bus and weighs, well, as much as a bus! So it will need to deliver very significant horsepower for uphill high speed autobahn cruising and then maybe heat does become an issue, if anyone would like to give me one I'll go find out and report back 😉. But in a car like this where it will be virtually impossible to run at over the 80kW for more than 4 seconds I would love to have the chance to worry about melting enamel and heat inertia once the thing exists! Does that seem silly? In the meantime I will definitely check out the cobra. I have failed to find stable upon it until now, so thanks.

It is always difficult to know just how one comes across when writing in forums, so I'd just like to clarify here, that I really do appreciate your input. You are all in another level with this stuff and I need to learn. I will go see what I can glean from Dala as well.

A very good evening to you,
Tom.
User avatar
celeron55
Posts: 774
Joined: Thu Jul 04, 2019 3:04 pm
Location: Finland
Has thanked: 27 times
Been thanked: 110 times
Contact:

Re: Power up version for complete Leaf

Post by celeron55 »

tomhanman wrote: Fri Apr 30, 2021 8:15 pm Thanks for your clarity on the original board Celeron. It's good to know absolute limits. Although, and I don't know enough about how things work to says this, but I'll say it anyway, that board analyses everything and then sends commands to control a few igbt (is that right?). So wouldn't the key to this being intercepting those control signals and scaling them? Of course, like you say, if there is teacher (VCU) checking currents fro the battery to compare at the same time then that could bring it all to a grinding halt. But is that concept of scaling a viable one or complete nonsense?
I don't know of anyone who has tried to do that. I guess in theory you could scale down the current sensor outputs using simple voltage dividers and it should result in more power. The inverter CPU has a very sophisticated and tuned algorithm that could somehow break if you do that, or maybe it won't. Nobody knows. The only way to know is to test. If you want to perform an experiment, that one would be interesting. If it works, you could even tune an otherwise fully intact Nissan Leaf using that method.
SRFirefox
Posts: 28
Joined: Thu Jun 25, 2020 2:28 am

Re: Power up version for complete Leaf

Post by SRFirefox »

tomhanman wrote: Fri Apr 30, 2021 8:15 pm It is always difficult to know just how one comes across when writing in forums, so I'd just like to clarify here, that I really do appreciate your input. You are all in another level with this stuff and I need to learn. I will go see what I can glean from Dala as well.

A very good evening to you,
Tom.
Ah! I misunderstood your intent and somehow got into my head that a car for racing was what you were building and ultimately running the motor hard constantly. If messing around on some nice streets and otherwise cruising and having a good time is the purpose of the car, then the 130kW limit of the stock inverter with alternate board would make perfect sense for you and make a sprightlier car and the duty cycle of such operation would make sense for your application! That said, Celeron55 has pointed out the only place I can think of to man-in-the-middle a signal that might increase the power output of the system.

The fundamental difference between MITM chips for ICE operation and EV operation is that most (if not all) ICE controllers are lookup table based, meaning they have giant pages of spreadsheets that correspond to RPM, fuel air, air intake volume, etc and give the engine its timing, fuel duty cycle, etc outputs to run on. In the 80s this was the only way to do it - even if the algorithms could be invented and tuned they'd take too much processing power on the microcontrollers of the day. An EV controller controls its motor algorithmically (look up Clarke and Park if you're interested in one way to do it), and messing with one of the inputs may or may not throw off the tuning of the internal algorithm depending on how it's set up. The only way to know for sure is to try it!
tomhanman
Posts: 13
Joined: Tue Apr 27, 2021 4:41 pm
Location: France
Been thanked: 1 time

Re: Power up version for complete Leaf

Post by tomhanman »

Thanks SRFirefox, I'm glad that some of my understanding is about right. And to the motor tuning section, that is what I was getting at. I felt that it wasn't going to be a look-up style affair because that would simply be too simple.

As the aftermarket demand for a tuning solution for the Leaf must be significant, and yet has is no aftermarket solution, it seemed likely that simple wasn't going to be on the table!

I'll continue my reading, thanks.
User avatar
johu
Site Admin
Posts: 5684
Joined: Thu Nov 08, 2018 10:52 pm
Location: Kassel/Germany
Has thanked: 153 times
Been thanked: 960 times
Contact:

Re: Power up version for complete Leaf

Post by johu »

I know that the BMS is sent motor torque and speed, so maybe it will make some comparison between expected and actual current. That said, I don't feed valid values to those variables and the BMS doesn't limit power. Maybe just some DTC and thats it.

If all else fails an openinverter board can be made to play with the Leaf if you fully reverse engineer the CAN protocol. Important bits have been done by celeron already. Maybe that's even sufficient.
Support R/D and forum on Patreon: https://patreon.com/openinverter - Subscribe on odysee: https://odysee.com/@openinverter:9
tomhanman
Posts: 13
Joined: Tue Apr 27, 2021 4:41 pm
Location: France
Been thanked: 1 time

Re: Power up version for complete Leaf

Post by tomhanman »

Johu, Celeron, many many thanks.

It is a fascinating, and yet frustrating area of evolution. Given that my first Arduino is still in my AliExpress shopping cart it seems highly unlikely that I could learn enough quickly enough (other obligations and brain power limitations). So can you think of anyone who might find this direction of development interesting enough to want to take experiments in that direction? And what could I do to enable that?

There seem two directions. 1: A mod board for the original Leaf environment which gives more power without changing any other functions. 2: a or several canbus mtm to enable the Leaf environment to run on the openinverter power board.

They both have advantages and disadvantages. The former would appear simplest, it could also have a large aftermarket potential it provides for charging. But in my planned installation I will need canbus mtm anyway for airbag deletion, smart key door sensor removal, climate control removal (no doors, no roof) and probably other bits I haven't thought of yet. But these will not be on the motive side. The latter requires some accumulation of all known Leaf canbus hacking as well as a whole lot more basically, and then it creates an ecosystem where potentially as many of the Leaf components can be used as required, but that would take genius levels of canbus,.or at least some serious dedication.

I would vote for trying solution A, and if it proves untenable then solution B. But for that it is not can hacking that is required it is full HV electronics and maths degrees. So who is game?

Maybe B is better... Or did I already mention both has pots and cons?!

Best regards, Tom.
User avatar
johu
Site Admin
Posts: 5684
Joined: Thu Nov 08, 2018 10:52 pm
Location: Kassel/Germany
Has thanked: 153 times
Been thanked: 960 times
Contact:

Re: Power up version for complete Leaf

Post by johu »

That said, a MTM should not be needed for B as the openinverter controller will just mimic the OEM controller on the CAN side. And HV electronics stay untouched.
Support R/D and forum on Patreon: https://patreon.com/openinverter - Subscribe on odysee: https://odysee.com/@openinverter:9
tomhanman
Posts: 13
Joined: Tue Apr 27, 2021 4:41 pm
Location: France
Been thanked: 1 time

Re: Power up version for complete Leaf

Post by tomhanman »

Ok, very interesting. I will now disappear for some time (as many do) whilst I try to read and learn. With a bit of luck I'll be back. Regards.
Post Reply