QC Message Data Decoding

[garygid]

I had not heard that elsewhere, thanks.
Any time frame and how to get the specs?

A full public disclosure might give the Chademo
standard a real chance of surviving, if it is not
already too late.

The only practical way to deal with the shrinking
capacity problem is to have a good QC network,
I suspect.

 

[jclemens]

I believe the standard has been released in Japanese, you can buy a PDF copy of it for about 80$

http://www.webstore.jsa.or.jp/webst...TS+D+0007:2012&dantaiCd=JIS&status=1&pageNo=0

I have not bought a copy myself because the next step is to get a test cable, and since I am told they cost over 2k, i may just try and 3d print one myself.
IF i can get that to work (mechanically), then I may investigate further and start learning canbus code (i am not a programmer, more of a hardware guy)
there is an open source charger that would work well with chademo, but this guy is in the same boat as you guys and does not have the spec.

this would make a 12kw 240V portable chademo charger.
http://www.emotorwerks.com/tech/electronics
apparently this guy is making a 25kw version.

 

[garygid]

We are working on a 10 to 20 kW Chademo charger solution.

We have logged the QC (Chademo?) "conversations", and think that
we understand the BASICS (but not the many details) of the
QC control. A rough overview:

After getting some basic capability information from the QC
charger (MAX voltage available, and max current capability), the
car begins commanding the external charger, specifying the current
that should be delivered to the car's battery pack. These requests
start with about one amp, and at 10 commands per second, the
current request rises at 20 amps per second to the level that the car
wants (limited by what the charger said it could deliver).

The car usually checks that it is getting the current that it requested,
within some acceptable limits. Eventually, the car tapers off the requested
current, and eventually ends the session by requesting zero current.

So, the basic procedure is simple. Handling all the exception
conditions, in the expected manner, is still a mystery.

Even a rough "google" translation of the Japanese manual would be helpful,
to determine if the manual covers these details. It appears to be 70
pages long, so they have space for many details.

Can anybody find an English translation?

 

[garygid]

Downloading the Preview PDF fails on this tablet.
Has anybody gotten the preview, and translated it?

 

[garygid]

This EMW charger is about $2500, but still non-isolated,
and it appears that their design is not Open, since one
apparently needs to buy at least their kit before one
can see a schematic.

However, if they design an isolated-output version, and
are willing to share Openly, we are willing to work with them.

 

[GregH]

i may just try and 3d print one myself.

I would like to get involved in 3d printing a Chademo plug.. anyone know of a good source for the 9mm pins? Found something similar by ITT Cannon available on Digikey for $14 each
http://www.digikey.com/product-detail/en/030-8614-020/1003-2014-ND/2487681" onclick="window.open(this.href);return false;
... but the pins don't look nearly as deep as the Chademo port... The datasheet seems to imply that it can handle 120A at up to 100 degrees C! That should never happen and would surely melt the plastic.. at room temps (20-30C) it says it's good for over 200A. Works with AWG 2 wire.

 

[TonyWilliams]

I have access to a CHAdeMO cable/plug assembly that might be available for testing. I would really like an add-on CHAdeMO charge port for the Rav4.

 

[jclemens]

I started with the inlet, I thought it would be easier. This way, I can use this to plug into an actual chademo station near by and hopefully see how it operates mechanically.
Does not include pins of course.

edit: Jack Richard was able to get one of this including pins and cables for about 800$ He said it was from a yazaki distributor? he mentions it at 45min into his April 19th episode on http://www.evtv.me

http://www.thingiverse.com/thing:80475

 

[GregH]

I started with the inlet, I thought it would be easier. This way, I can use this to plug into an actual chademo station near by and hopefully see how it operates mechanically.
Does not include pins of course.

edit: Jack Richard was able to get one of this including pins and cables for about 800$ He said it was from a yazaki distributor? he mentions it at 45min into his April 19th episode on http://www.evtv.me

http://www.thingiverse.com/thing:80475

That's very cool!!
I'd certainly be interested in the plug.. I've got provisions on my new Dash Display to handle most of the control lines for Chademo as well as the CAN communication. The other side of it will be an isolated serial link to control an ABC-150 battery cycler to do the heavy work.. but I'd also be interested in smaller/cheaper say 10kW type solutions for roadtrips..

 

[jclemens]

This EMW charger is about $2500, but still non-isolated,
and it appears that their design is not Open, since one
apparently needs to buy at least their kit before one
can see a schematic.

However, if they design an isolated-output version, and
are willing to share Openly, we are willing to work with them.

How would the vehicle know if the charger is isolated or not? it isn't like the negative terminal would be connected to chassis ground! It couldn't be, because you would short out the line to the neutral and blow the whole thing up. (Neutral wouldn't be connected anywhere here, but at the utility breaker box it is grounded, so if you ground one of the hot lines, you short it out, even through a rectifier)
you would have L1 and L2 in at 120V each for 240V, it gets rectified, then filtered with a capacitor, then chopped and boosted through an inductor.
as long as ground from the rectifier is not connected to chassis, earth or any other touchable gnd, it should count as isolated. Naturally, it would be isolated from electronic communication ground which is easy to do, which is probably how the vehicle would detect if ground is isolated or not.

I'd love to have a dual 12kw charger, (total 24kw potentially) one master one slave, if you plugged in a single J1772 that was capable of more than 12kw (50A), a cable would carry over from the 1st to the 2nd charger and they run in parallel off the same j1772.
Conversely, if you have dual chargers, but surrounded by EVSEs that are limited to something less than 12kw (say, 30 amp is pretty common), you could plug in 2 chargers and run them in parallel. This would really make cross country driving much more feasible!

.. I've got provisions on my new Dash Display to handle most of the control lines for Chademo as well as the CAN communication. The other side of it will be an isolated serial link to control an ABC-150 battery cycler to do the heavy work.. but I'd also be interested in smaller/cheaper say 10kW type solutions for roadtrips..

GregH are you working on a chademo project??

 

[GregH]

This EMW charger is about $2500, but still non-isolated,
and it appears that their design is not Open, since one
apparently needs to buy at least their kit before one
can see a schematic.

However, if they design an isolated-output version, and
are willing to share Openly, we are willing to work with them.

How would the vehicle know if the charger is isolated or not? it isn't like the negative terminal would be connected to chassis ground! It couldn't be, because you would short out the line to the neutral and blow the whole thing up. (Neutral wouldn't be connected anywhere here, but at the utility breaker box it is grounded, so if you ground one of the hot lines, you short it out, even through a rectifier)
you would have L1 and L2 in at 120V each for 240V, it gets rectified, then filtered with a capacitor, then chopped and boosted through an inductor.
as long as ground from the rectifier is not connected to chassis, earth or any other touchable gnd, it should count as isolated. Naturally, it would be isolated from electronic communication ground which is easy to do, which is probably how the vehicle would detect if ground is isolated or not.

I'd love to have a dual 12kw charger, (total 24kw potentially) one master one slave, if you plugged in a single J1772 that was capable of more than 12kw (50A), a cable would carry over from the 1st to the 2nd charger and they run in parallel off the same j1772.
Conversely, if you have dual chargers, but surrounded by EVSEs that are limited to something less than 12kw (say, 30 amp is pretty common), you could plug in 2 chargers and run them in parallel. This would really make cross country driving much more feasible!

.. I've got provisions on my new Dash Display to handle most of the control lines for Chademo as well as the CAN communication. The other side of it will be an isolated serial link to control an ABC-150 battery cycler to do the heavy work.. but I'd also be interested in smaller/cheaper say 10kW type solutions for roadtrips..

GregH are you working on a chademo project??

Yes.. I had extra I/O on the Dash Display so I added an RS232 level transceiver chip (no isolation though, I'll do that externally.. This chip won't be populated for the production DD).
I also added an input to detect the car's "ok" signal as well as a FET to provide the negative ack signal after the CAN protocol exchange. Also a 12V sense input.. So I was thinking a physical switch to provide the local isolated 12V to the car (step 1) then after the CAN stuff and vehicle OK (forget the exact order of operations here) the FET can provide the negative side then the car will have both 12V and iso gnd to pull the contactor closed. At the end of charge when current is back to zero, the user would flip off the 12V switch. The DD detects this and withdraws the GND from the FET at the same time.

The car checks isolation between the chassis (ie the CAN bus) and HV+ and HV-... These are the only 3 signals it has access to. It has no idea what "ground" is thanks to the rubber tires. As long as the chassis ground, local 12V, and CAN are completely isolated from HV+ and HV- then we should be ok.

 

[jclemens]

The car checks isolation between the chassis (ie the CAN bus) and HV+ and HV-... These are the only 3 signals it has access to. It has no idea what "ground" is thanks to the rubber tires. As long as the chassis ground, local 12V, and CAN are completely isolated from HV+ and HV- then we should be ok.

I have been giving this more thought. Initially I thought I answered my one question because the AC mains has the neutral tied to ground at some point, so no, it is not isolated.
however. giving it more thought again, I wanted to know exactly just HOW the leaf detects a frame leak, i checked the service manual and it refers to a "on-board isolation resistance monitoring system" by using a IR SEN short pulse which then is communicated with the vehicle control module over canbus.
The monitor is completely internal to the LBC (Li-ion battery controller), which is inside the battery pack.
further research brought me to this link about how frame leaks are detected in EVs
http://www.tdipower.com/PDF/white_paper/Ground-Fault-White-Paper.pdf

it shows an intentional electrical connection from the battery pack to the vehicle frame with some high value resistors (very low current). the leaf has DTC for both "Signal voltage of the on-board isolation resistance monitoring system is too high" AND too low.
If either end of the battery pack has a slight connection to the frame, (ie; a person or excess moisture from A/C) it is like putting a resistor in parallel to the detection circuit resistors and thus changing the value, which changes the voltage, either up or down.
The leaf has both a minimum and a maximum value for leakage detection, the fact that it has a minimum means that there is always some level of leakage, without this, it would be an indication that something is wrong with the leakage detection circuit.

anyways, what is the point in all of this?

i am trying to answer this question:
if you have a transformer with a center-tap that is grounded, the high voltage ends of the transformer are rectified, boosted, blah, blah, then connected to the battery, what happens?
so far, i think nothing, especially since the vehicle frame is not connected to electrical ground.
ok, so what happens if you then ground the frame? (with a wire, not a person, if nothing happens with a wire, then nothing will happen with a person, even while standing, soaking wet, in rain)

I still think nothing will happen, because grounding the vehicle frame is NOT like putting resistors in parallel to the 2 resistors in the detection circuit. but only testing will tell us for sure.

 

[Ingineer]

I'm confident that you will trip the LEAF's leakage detection if any part of an external AC system is connected to the HV DC system at any point. The HV Leakage check is accomplished by sending a high frequency current limited square wave referenced from the HV system to the chassis. If there is even a nominal increase in capacitance in the HV system relative to ground you will trip it. On the early LBC firmware, the system was false-triggering just from stay leakage from the A/C Compressor. They reduced the sensitivity, but it's still very sensitive!

My first experiments with power injection into the LEAF's HV system routinely triggered the leakage detection, so I'm not just speculating here.

Also, connecting any part of the vehicle to a external power systems of any kind without a bonded ground connection is asking for trouble. Even something like a small amount of static could build up and destroy numerous systems in the Leaf.

Also, the CHAdeMO system performs a high-voltage leakage test of it's own by requesting a HV ramp-up to about 600 volts and watching for leakage from either side of the HV. It will not close the main contactors from the battery to the CHAdeMO socket unless this test passes.

I'm not sure if everyone truly gets the enormous danger the LEAF's HV system represents. Special measures must always be taken when dealing with these voltages and potential currents, or may you may very well find yourself or someone else killed dead, and/or suffer from extensive property damage!

I recently spoke to someone who owns a body shop and he told me that someone he knows was killed by electrocution from a LEAF that was being worked on. I haven't found out any details, so it could always just be some FUD, but if I find out anything more, I'll post the details.

-Phil

 

[jclemens]

Thanks for chiming in Phil, I hate having to make assumptions to cover the unknowns, in this case, I had to assume that the leaf's leakage detection was the same as the one i found in that white paper, which now, evidently, appears wrong.
It is inevitable that someone will try and reverse engineer the inner-workings of chademo, that is the purpose of this thread, so, as always, your guidance is always greatly appreciated, especially when safety is concerned. (I doubt anyone has died messing with this, as the "Fear, Uncertainty, and Doubt" people would eat it up and it would be all over the news)

I agree that connecting electronics without a bonded ground is a bad idea, I looked more closely at that, and I can now see that if either charge ports have a connector in them, the vehicle chassis is then grounded, or if the grounds are isolated, then there is a risk of ESD issues.

Ok, so it seems the battery pack has no direct electrical connection to ground (unlike white paper).
I am still trying to wrap my head around what will happen and why, if we do the following example:

battery pack 400V
capacitors connected to the chassis (just an equivalent circuit to represent inherent capacitance and no DC can pass through)
DC power source with +300 and -100V with respect to actual ground is then connected to the pack, what happens? the frame is not going to be lit up with that 100V difference, or is it? How would it, there is no connection to frame! but i know that there is capacitance between the frame and the HV lines, and the chademo charger will have the same...
So, how is any chademo manufacturer supposed to know the allowable capacitance that any particular vehicle manufacturer has setup for their frame leak detector? Especially if they keep changing it with software updates?

I'm confident that you will trip the LEAF's leakage detection if any part of an external AC system is connected to the HV DC system at any point. The HV Leakage check is accomplished by sending a high frequency current limited square wave referenced from the HV system to the chassis. If there is even a nominal increase in capacitance in the HV system relative to ground you will trip it. On the early LBC firmware, the system was false-triggering just from stay leakage from the A/C Compressor. They reduced the sensitivity, but it's still very sensitive!

My first experiments with power injection into the LEAF's HV system routinely triggered the leakage detection, so I'm not just speculating here.

Also, connecting any part of the vehicle to a external power systems of any kind without a bonded ground connection is asking for trouble. Even something like a small amount of static could build up and destroy numerous systems in the Leaf.

Also, the CHAdeMO system performs a high-voltage leakage test of it's own by requesting a HV ramp-up to about 600 volts and watching for leakage from either side of the HV. It will not close the main contactors from the battery to the CHAdeMO socket unless this test passes.

I'm not sure if everyone truly gets the enormous danger the LEAF's HV system represents. Special measures must always be taken when dealing with these voltages and potential currents, or may you may very well find yourself or someone else killed dead, and/or suffer from extensive property damage!

I recently spoke to someone who owns a body shop and he told me that someone he knows was killed by electrocution from a LEAF that was being worked on. I haven't found out any details, so it could always just be some FUD, but if I find out anything more, I'll post the details.

-Phil

 

[GregH]

I recently spoke to someone who owns a body shop and he told me that someone he knows was killed by electrocution from a LEAF that was being worked on. I haven't found out any details, so it could always just be some FUD, but if I find out anything more, I'll post the details.

-Phil

Wow! I know some garage guys aren't the sharpest tools in the shed, but given the precautions Nissan presumably goes through training the few select people that can touch the innards of a Leaf I find this amazing. Practically the first few pages of every section in the service manual (for the Prius as well) go through all the steps to avoid HV... Please do let us know if you find out any more on this.

P.S. Back at EnergyCS when we were working on the first lithium plug in Prius conversions I once accidentally crossed a 200V pack (from one hand to the other no less.. right through the heart!) and got a good jolt. I went into the office and told my partner Pete and he just smiled and looked at me and said "you look ok".. gee thanks. You can never be too careful!

 

[derdlim]

Hello leafers, the (english) chademo protokoll is now online.
You can download it for 160 dollar.
Now we can build a 20kw diy chademo charger.

http://www.webstore.jsa.or.jp/webstore/Com/FlowControl.jsp?lang=en&bunsyoId=TS+D+0007%3A2012&dantaiCd=JIS&status=1&pageNo=0" onclick="window.open(this.href);return false;

 

[TonyWilliams]

Hello leafers, the (english) chademo protokoll is now online.
You can download it for 160 dollar.
Now we can build a 20kw diy chademo charger.

http://www.webstore.jsa.or.jp/webstore/Com/FlowControl.jsp?lang=en&bunsyoId=TS+D+0007%3A2012&dantaiCd=JIS&status=1&pageNo=0" onclick="window.open(this.href);return false;

Great news!!

 

[JeremyW]

Great news indeed, looks to be CHAdeMO 1.0 which has bidirectional capability.

Edit: I'm not sure it includes that after reading further.

 

[garygid]

Has anybody successfully downloaded the English Preview PDF?

If one buys this document, what are the restrictions on
the use of the information that it contains?

Is there really anything useful in the document?

Thanks, Gary

 

[garygid]

The restrictions that you "agree to" when trying to purchase this
document are very restrictive. Your single copy of the document
cannot be on any networked computer, even if local network.
There are many other restrictions as well.

Since we are not going to agree to those terms, we have
chosen to continue our own investigations.

But, thanks for letting us know of its existence.