A peek at the Leaf's Charger

[Ingineer]

To be honest, it wasn't even my idea to begin with. I think it might have been Phil's (Ingineer) idea a while back when he was tossing around the idea of installing a third party 10KW charger in the front under the hood.

Yes, this was my idea, and I've tested and confirmed it will work, however, the trick is finding an existing manufacturer of a 400v ~10kW charger that is isolated, affordable, and of reasonable quality. It will also have to fit, which is a problem with many air-cooled models. Water-cooling is doable, but more messy to install.

I've also verified that it is possible to charge the LEAF while it's in motion without extensive modifications, which opens up range extender possibilities. (both as in my Generator and auxiliary battery packs)

-Phil

 

[klapauzius]

Strange enough its really hard to find one...I assume the chargers built specifically for the large EV manufacturers will not be available to the public at reasonable prices?
A superficial survey of the EV-DIY world seems to indicate that most people there work with non-isolated topographies... Given that
the overall design of an isolated charger is not that complicated, one could come up with a custom built design, but that would add whole new can of worms in terms of safety etc.

Is the isolation really necessary? And could the Leaf be tricked into seeing the non-isolated unit as an isolated one?

All the difficulties aside, the idea of a portable J1772 to CHademo adapter would seem compelling, especially since some j1772 chargers can support large currents...it would be like a portable L3 quickcharger.

I wouldnt even want to have it permanently installed in the car...something you can drop in the trunk for longer trips would be fine. And 240 V 50 amp outlets these days are certainly easier to find than L3 chargers...

 

[jclemens]

I've also verified that it is possible to charge the LEAF while it's in motion without extensive modifications, which opens up range extender possibilities. (both as in my Generator and auxiliary battery packs)

-Phil

Any chance you can share what these modifications would be?

 

[Herm]

Is the isolation really necessary? And could the Leaf be tricked into seeing the non-isolated unit as an isolated one?

You could always get a large roll of electrical tape and wrap up the whole charger completely.. then it would be isolated

 

[planet4ever]

Is the isolation really necessary? And could the Leaf be tricked into seeing the non-isolated unit as an isolated one?

You could always get a large roll of electrical tape and wrap up the whole charger completely.. then it would be isolated

True, but only if you also cut the electrical wires going in and out of it and wrap their ends, too. :lol:

Ray

 

[TimeHorse]

All the difficulties aside, the idea of a portable J1772 to CHademo adapter would seem compelling, especially since some j1772 chargers can support large currents...it would be like a portable L3 quickcharger.

I agree but in addition to the Microcode you'd have to write to speak both J1772 and CHAdeMO the rectifying (charging, A/C to D/C) as well as installing a transformer for the voltage difference, you'd also have to licence the TEPCO plug which somewhere else in the forum I've read goes for $2300. Yes, $2300 just for the stinking CHAdeMO protocol TEPCO plug. What the #$^$%^ are they smoking anyway!?

Why must they make this so difficult. Man, not a day goes by it seems when I don't think my life would be a whole lot better with at least 6.6kW; to the point where I'm a broken record.

 

[klapauzius]

I agree but in addition to the Microcode you'd have to write to speak both J1772 and CHAdeMO the rectifying (charging, A/C to D/C) as well as installing a transformer for the voltage difference, you'd also have to licence the TEPCO plug which somewhere else in the forum I've read goes for $2300. Yes, $2300 just for the stinking CHAdeMO protocol TEPCO plug. What the #$^$%^ are they smoking anyway!?

Why must they make this so difficult. Man, not a day goes by it seems when I don't think my life would be a whole lot better with at least 6.6kW; to the point where I'm a broken record.

The microcode cannot be that complicated...after all, its only a charger...we are not trying to solve np-hard problems here...
Yes, these proprietary hardware/protocl issues are annoying. Especially since it did not take a bunch of rocket scientists to come up with
a glorified power plug.
Conceptually it seems straightforward, so again, in a few years you can just buy this stuff off the shelf.

 

[TimeHorse]

The microcode cannot be that complicated...after all, its only a charger...we are not trying to solve np-hard problems here...
Yes, these proprietary hardware/protocl issues are annoying. Especially since it did not take a bunch of rocket scientists to come up with
a glorified power plug.
Conceptually it seems straightforward, so again, in a few years you can just buy this stuff off the shelf.

My main concern is proper throttling of load being controlled by the BMS. If we can control the load, get the connectors and so on, Bob's your uncle, we're don! I agree ideally we just get a 100 or so TEPCO plugs and J1772 receptacles, a enough and variously sized rectifiers and transformers, encase and wire it water-tight and market them to a handful of hobbyists and disenfranchised LEAF owners. Can't imagine more than a 1,000 or so of these being sold though as the 201x LEAF (maybe 2, maybe 3, who knows) will have the 6.6kW but there would still be niche folks like myself with a desire for 19kW power. So in that case, we wait for the TEPCO plug to get mass-produced in China (or for an SAE J1772+2 Prong DC to replace it and for SAE Level 3 to TEPCO plugs to become ubiquitous and obviously then we could match the SAE Level 3 which may be licensed more cheaply). Then we test the 6.6kW rectifier, make it work, then look into 10kW and 20kW upgrades at cost + additional research and labor.

But I just can't see all this practically happening when the TEPCO plug itself is costing $2700. I'd expect the 20kW rectifier to be maybe that much but the plug? It just can't be economical if the plug is the most expensive component.

And that's my 2¢ and armchair engineering.

Edit: Thanks to EVDriver the TEPCO plug cost has been corrected.

 

[DaveinOlyWA]

The microcode cannot be that complicated...after all, its only a charger...we are not trying to solve np-hard problems here...
Yes, these proprietary hardware/protocl issues are annoying. Especially since it did not take a bunch of rocket scientists to come up with
a glorified power plug.
Conceptually it seems straightforward, so again, in a few years you can just buy this stuff off the shelf.

in the era of "patenting the best method for wiping one's ass" protocols are made proprietary and complicated for a reason and that reason is MONEY

 

[EVDRIVER]

The microcode cannot be that complicated...after all, its only a charger...we are not trying to solve np-hard problems here...
Yes, these proprietary hardware/protocl issues are annoying. Especially since it did not take a bunch of rocket scientists to come up with
a glorified power plug.
Conceptually it seems straightforward, so again, in a few years you can just buy this stuff off the shelf.

My main concern is proper throttling of load being controlled by the BMS. If we can control the load, get the connectors and so on, Bob's your uncle, we're don! I agree ideally we just get a 100 or so TEPCO plugs and J1772 receptacles, a enough and variously sized rectifiers and transformers, encase and wire it water-tight and market them to a handful of hobbyists and disenfranchised LEAF owners. Can't imagine more than a 1,000 or so of these being sold though as the 201x LEAF (maybe 2, maybe 3, who knows) will have the 6.6kW but there would still be niche folks like myself with a desire for 19kW power. So in that case, we wait for the TEPCO plug to get mass-produced in China (or for an SAE J1772+2 Prong DC to replace it and for SAE Level 3 to TEPCO plugs to become ubiquitous and obviously then we could match the SAE Level 3 which may be licensed more cheaply). Then we test the 6.6kW rectifier, make it work, then look into 10kW and 20kW upgrades at cost + additional research and labor.

But I just can't see all this practically happening when the TEPCO plug itself is costing $2300. I'd expect the 20kW rectifier to be maybe that much but the plug? It just can't be economical if the plug is the most expensive component.

And that's my 2¢ and armchair engineering.

$2700. Yes they are expensive. They are also huge.

 

[davewill]

... you'd also have to licence the TEPCO plug which somewhere else in the forum I've read goes for $2300. Yes, $2300 just for the stinking CHAdeMO protocol TEPCO plug. What the #$^$%^ are they smoking anyway!? ...

That's the beauty of putting it under the hood. You don't bother with either plug, you just tap directly into the wiring harnesses.

 

[klapauzius]

My main concern is proper throttling of load being controlled by the BMS. If we can control the load, get the connectors and so on, Bob's your uncle, we're don! I agree ideally we just get a 100 or so TEPCO plugs and J1772 receptacles, a enough and variously sized rectifiers and transformers, encase and wire it water-tight and market them to a handful of hobbyists and disenfranchised LEAF owners. Can't imagine more than a 1,000 or so of these being sold though as the 201x LEAF (maybe 2, maybe 3, who knows) will have the 6.6kW but there would still be niche folks like myself with a desire for 19kW power. So in that case, we wait for the TEPCO plug to get mass-produced in China (or for an SAE J1772+2 Prong DC to replace it and for SAE Level 3 to TEPCO plugs to become ubiquitous and obviously then we could match the SAE Level 3 which may be licensed more cheaply). Then we test the 6.6kW rectifier, make it work, then look into 10kW and 20kW upgrades at cost + additional research and labor.

But I just can't see all this practically happening when the TEPCO plug itself is costing $2700. I'd expect the 20kW rectifier to be maybe that much but the plug? It just can't be economical if the plug is the most expensive component.

And that's my 2¢ and armchair engineering.

Edit: Thanks to EVDriver the TEPCO plug cost has been corrected.

Check this out:

http://www.emotorwerks.com/cgi-bin/VMcharger.pl" onclick="window.open(this.href);return false;

Its a 10 kW charger...it looks pretty homemade (I guess it is...), but I would think it can almost do the job. Its non -isolated, so you have to modify the circuit design for an isolation transformer.

Here is an example:
http://e2e.ti.com/support/interface/industrial_interface/m/videos__files/264970.aspx" onclick="window.open(this.href);return false;

Note that these are low-power circuits, but its probably just a matter of sizing the components properly. The emotorwerks charger can serve as an example....

Component cost for these could be kept probably below 1 k$, so the isolated charger is really not the problem. The Leafs BMS communicates over the CHaDEMO port, so "if" that protocol were known, then all you need is a microcontroller to regulate the duty cycle of the inverter and the current to provide the voltage/current the Leaf wants.

 

[garygid]

I think ... that the LEAF's charger-controller is on the EV-CAN bus (and communicates with the Battery Pack's BMS) and the charger-controller also controls the QC charging session and is the only component connected to QC-CAN bus. But, I might be mistaken.

 

[Ingineer]

I think ... that the LEAF's charger-controller is on the EV-CAN bus (and communicates with the Battery Pack's BMS) and the charger-controller also controls the QC charging session and is the only component connected to QC-CAN bus. But, I might be mistaken.

You are correct. One doesn't have to worry about regulating power into the QC port, as the on-board charger tells you over the QC CAN Bus every 100ms.

-Phil

 

[palmermd]

One doesn't have to worry about regulating power into the QC port, as the on-board charger tells you over the QC CAN Bus every 100ms.

-Phil

It depends on how you define "one does not have to worry".

If you mean that there is a reference available that you can use to follow, then yes, the Leaf will tell you what to do.

If one interprets this as "you don't have to worry, just pump in some current", then it is incorrect because if you don't follow the Leaf's suggestion as to the amount of current, it will disconnect you via the main relay, and you wont get any charge via the QC port.

edit: I'm just trying to clarify so others to misinterpret and begin to experiment on their car.

 

[klapauzius]

I think ... that the LEAF's charger-controller is on the EV-CAN bus (and communicates with the Battery Pack's BMS) and the charger-controller also controls the QC charging session and is the only component connected to QC-CAN bus. But, I might be mistaken.

You are correct. One doesn't have to worry about regulating power into the QC port, as the on-board charger tells you over the QC CAN Bus every 100ms.

-Phil

So a DC charger which goes into the QC port can be "dumb" as everything is regulated on the Leafs side?

OK, just saw the other post.
So you need a charger that can regulate its output current/voltage as per the Leafs instructions. I assume that all this
protocol goes through the CHaDeMO connector?

 

[Ingineer]

Sorry if my post wasn't clear. Michael is spot-on, you must adhere to what the Leaf tells you, but as long as you adjust your output when the Leaf instructs you to, there will be no problem. All of this is sent as CAN frames every 100ms over the dedicated QC CAN bus present at the CHAdeMO port.

You absolutely must have 100% galvanic isolation from your voltage output to the car's body and the power line, or the QC session will be aborted, and your car likely will not restart until the DTC is cleared.

Using a simple non-isolated charger is not possible, and it is not a "work around" to simply isolate the vehicle body from the power line ground. Even the rather high resistance of a human touching the car while grounded will trip the leakage current detection, and would be a dangerous situation.

In any event, this is not DIY or hobbyist territory, there are extremely lethal power levels and you risk damage to your Leaf!

-Phil

 

[klapauzius]

Sorry if my post wasn't clear. Michael is spot-on, you must adhere to what the Leaf tells you, but as long as you adjust your output when the Leaf instructs you to, there will be no problem. All of this is sent as CAN frames every 100ms over the dedicated QC CAN bus present at the CHAdeMO port.

You absolutely must have 100% galvanic isolation from your voltage output to the car's body and the power line, or the QC session will be aborted, and your car likely will not restart until the DTC is cleared.

Using a simple non-isolated charger is not possible, and it is not a "work around" to simply isolate the vehicle body from the power line ground. Even the rather high resistance of a human touching the car while grounded will trip the leakage current detection, and would be a dangerous situation.

In any event, this is not DIY or hobbyist territory, there are extremely lethal power levels and you risk damage to your Leaf!

-Phil

I agree.

But just to explore the feasibility of something that would be immensely practical, i.e. a portable CHaDeMO to J1772 "adapter",
"all" it would take is

1) the CHaDeMO connector + knowledge of the protocol
2) an isolated charger capable of delivering ~10 kW
3) the J1772 connector + protocol

If there are no principal obstacles, one would hope that such a device will be available for sale eventually.
From the previous discussions about the charger upgrade from 3.3 kW to 6.6 kW it seems that there is no way anyone would ever be able to offer this at reasonable prices, but utilizing the QC port might enable us to bypass the design limitations of the built-in charger?

Conceptually, all the components seem to be there, so we just need to wait for someone to put them together and offer them for sale..


From what I have seen, non-isolated chargers are not that big and heavy (~ 30 pounds), so an isolated design, which has a transformer, might be a bit heavier but maybe not that much?

 

[Ingineer]

Isolated power supplies that are reliable and safe typically come in @ about a dollar a watt. Don't expect anything cheaper for something like this with the attendant technical requirements and liabilities.

-Phil

 

[klapauzius]

Isolated power supplies that are reliable and safe typically come in @ about a dollar a watt. Don't expect anything cheaper for something like this with the attendant technical requirements and liabilities.

-Phil

I guess that makes any of the above mentioned ideas economically unfeasible .