Using clone ELM327 Bluetooth OBDII adapter with Leaf

[kubel]

I wonder how long one can leave the adapter plugged into the OBDII port before the battery drains. I noticed that even when the car is off, the OBDII port will keep the adapter powered on.

 

[JeremyW]

Here is a battery report for a new 2013 Leaf on the show room floor. The T3 reading is bad so ignore. I am sure that part of the battery was not that cold as it would be below absolute zero (-6244K).

Did anyone else notice the 101.67% capacity reading? :shock:

 

[sirfergy]

I wonder how long one can leave the adapter plugged into the OBDII port before the battery drains. I noticed that even when the car is off, the OBDII port will keep the adapter powered on.

I've not noticed any drop in battery, and I've had mine plugged in for over a week. I suspect it's such a small drain, the 12V battery doesn't mind. I would probably unplug when I go on vacation though.

 

[kubel]

Did anyone else notice the 101.67% capacity reading? :shock:

Wouldn't that imply > 281 gids? My guess is gids aren't quite as magically absolute as people have suggested. Either that or MY2013 is hiding a secret. Great, now I have Mr. Roboto stuck in my head.

'...secret secret, ive got a secret... with parts made in... tennessee?'

 

[sirfergy]

Did anyone else notice the 101.67% capacity reading? :shock:

Wouldn't that imply > 281 gids? My guess is gids aren't quite as magically absolute as people have suggested. Either that or MY2013 is hiding a secret. Great, now I have Mr. Roboto stuck in my head.

'...secret secret, ive got a secret... with parts made in... tennessee?'

You'll notice battery > 66, which seems odd.

 

[garygid]

The app could wait to get some first CAN passive message
before it tries to write to the car, requesting information.

So far, we are just guessing that 66 or so, maybe 67, is
the 100% capacity level of he Ah value, I believe.

 

[garygid]

This app's test (no-connection) cell-pair voltage data has CP 59 low
at 3.990v and CP 72 high at 4.920v, and the rest 4.000?

Trying the app on my Vizio Android tablet, and it displays just fine.
I will try with my ELM327 adapters tomorrow.

If it works, it is another item for the Show-n-Tell on Saturday in San Diego.
Edit: It Works!

Too bad we do not have (or have not found) GIDs on the CAR CAN bus.

Thanks, Gary

 

[sirfergy]

The test (no-connection) cell-pair Voltage data has CP 59 low
at 3.990v and CP 72 high at 4.920v, and the rest 4.000?

Trying it on my Vizio Android tablet, and I will try with the ELM327
adapter tomorrow.

If it works, it is another item for the Show-n-Tell on Saturday in San Diego.

Too bad we do not have GIDs on the CAR CAN bus.

Thanks, Gary

Aren't gids just soc * cap * 300?

 

[Turbo3]

The app could wait to get some first CAN passive message
before it tries to write to the car, requesting information.

Gary, This is not the case. The ELM327 can only see the CAR-CAN and the CAR-CAN does not need to be active before a pack request is sent. It is the EV-CAN that must be active to handle the pack request. You can send the pack request to a CAR-CAN that has no activity at all just as long at the EV-CAN is active. An inactive CAR-CAN can still pass the message over to the EV-CAN and the EV-CAN can route the response back to the idle CAR-CAN for the ELM327 to capture.

So sending the pack request and timing out is the only way to make this work. The way it is coded now you can wake up the EV-CAN by unplugging and replugging the charage cord to wake up the EV-CAN and then run the program to read the pack voltages without needing to get in the car and turn it on.

I will just add code so you can start the program up first before you do the unplug/replug and it will auto retry the request until the EV-CAN wakes up and responds.

Opening the door wakes up the CAR-CAN but does not mean the EV-CAN is active.

 

[DaveEV]

OK, I had to order an adapter off eBay, this is way cool! Very nice work, Turbo3!

What are the plans for the source for this? I'd love to start hacking on it (though I haven't done any Android dev yet) and am hoping this turns into an open source project... I'd like to get some of the values that are available on the LEAFcan/GIDmeter on here...

 

[Turbo3]

The test (no-connection) cell-pair Voltage data has CP 59 low
at 3.990v and CP 72 high at 4.920v, and the rest 4.000?

Yes, this is test data. The other two are set to test the auto scale upper and lower values and to give a 30mv difference. As people have noted, reading the same pack again always gives a slightly different value for all cells.

The data array is loaded with the test data at startup and it gets replaced with data from the car when it becomes available.

 

[Turbo3]

Aren't gids just soc * cap * 300?

The simple answer is no. But Gary can give more details. They are named after him ;-)

 

[Turbo3]

I wonder how long one can leave the adapter plugged into the OBDII port before the battery drains. I noticed that even when the car is off, the OBDII port will keep the adapter powered on.

In the near future, I will be measuring the current in the ones I have. But they do get warm to the touch.

 

[garygid]

Should GIDs be related to (soc * cap * 300)?

Probably, but we do not yet know "cap" well yet,
and Nissan was indicating that GIDs are related
to watt-hours not amp-hours. So, it seems quite
possible that GIDs are derived independently.

But, who knows.

With the CANary, we now have both the EV and the CAR
buses logged, with the 5 Groups Requested and included
in the log. Plotting the soc * cap * 300 and GIDs on the
same graph would tell us something.

Note that cap * 300 is essentially a constant at any moment,
and we have already plotted soc vs GIDs, and these to values
appear strongly related, but not really the same.

We should check the SOC from the EV bus with this "soc"
to see if the EV version is just a lower-precision copy of
this Requested "soc".

 

[DanBaldwin]

I wonder how long one can leave the adapter plugged into the OBDII port before the battery drains. I noticed that even when the car is off, the OBDII port will keep the adapter powered on.

In the near future, I will be measuring the current in the ones I have. But they do get warm to the touch.

I seem to remember from the ELM327 data sheet that it has a sleep mode. I wonder if putting it to sleep before exiting would have a significant effect on current draw.

 

[garygid]

Caution: many people have reported that some/many of
the inexpensive foreign clones of the ELM327 either
do not work well, fail to maintain a connection, or
just fail to work after a short time.

It would help to post some links to sources that
have proven to sell ELM327 devices that work well.
Thanks

 

[garygid]

I have two ELM327 mini (low profile, only sticks out of the
OBD connector about 3/4 of an inch) devices, one with
a Serial number stamped into the plastic bidy, and a "real"
mac address.

The other is a less expensive (probably a clone)
purchased on Ebay, where the seller pictured a device with
the stamped serial number, but actually delivered one
without any serial number. The mac address is also suspect
since it was something like AA:BB:CCD:EE:FF which is
probably not unique, but used in all the clones.

Edit: the MAC address in the clone is 11:22:33:AA:BB:CC

Others have mentioned this "same-mac-address" problem.

Now, the good news:
After pairing my Android tablet, the battery app managed to connect,
although it seemed to have more trouble connecting with the
"real" ELM327 device.

The clone did not appear to get data, so I tried the "real" one,
and eventuality got the counter at the bottom, and data displayed,
after I put the car into Ready mode.

Then, I tried the clone again, and managed to get data with it also.

Many Cool. :shock:

 

[GregH]

Did anyone else notice the 101.67% capacity reading? :shock:

Wouldn't that imply > 281 gids? My guess is gids aren't quite as magically absolute as people have suggested.

66.0Ah is not a constant.. nor is 281 Gids. People who divide these things by 66.0 or 2.81 do so at their own peril.. (IMHO)

As far as power consumption, my 128x64 OLED board is somewhat piggy at 40-50mA and I leave it plugged in and illuminated 24/7.. except when I leave the car parked for long (5+ day) trips. Isn't that little solar panel on the SL supposed to keep the 12V happy?

 

[sirfergy]

I wonder how long one can leave the adapter plugged into the OBDII port before the battery drains. I noticed that even when the car is off, the OBDII port will keep the adapter powered on.

In the near future, I will be measuring the current in the ones I have. But they do get warm to the touch.

I seem to remember from the ELM327 data sheet that it has a sleep mode. I wonder if putting it to sleep before exiting would have a significant effect on current draw.

You're probably thinking of wakeup messages, which essentially is a keep-alive. I'll try disabling that on mine, but so far I've not noticed any impact to battery life while the OBDII reader is plugged in.

 

[garygid]

My Vizio tablet (VTAB1008) has an 8" screen, and is running Android v3.2.1 HTK55.
I got the tablet at Costco for about $189 plus tax.

VERY NICE adaptation of the iMiEV app.
Thanks for the continuing Good Work, Turbo3.

Sincerely, Gary