Vendor Scan My Tesla, a CANBUS reader for Android

[Darren S]

I've been running this for a few days at home while the car has been sitting idle. I've been grabbing log files of 6- and 12-hr duration since not much is changing other than SOC. I was trying to get good readings of the "Usable full pack" at different SOC such as 100%, 90%, 75%, 50%, etc.

Sometimes when I check on the tablet that is running the app in the car it is fine every 30 minutes for several hours in a row. Other times, I will see an error message like one of these and I don't know how long it hasn't been logging. It seems that as soon as I open the door and wait a few seconds the adapter starts sending data again and the error messages go away. It might be fine for several hours or might stop logging soon after (but I'm not sitting and watching it so I don't know).

I haven't yet looked in the log files to see if it is only missing a few minutes of data or several hours but wasn't sure if this was a known 'feature.'

 

[amund7]

I haven't yet looked in the log files to see if it is only missing a few minutes of data or several hours but wasn't sure if this was a known 'feature.'

The car goes to deep sleep, and the canbus is silent.

You can (probably) manage this with the power saving settings of your car, if you want it to stay awake and log.

 

[supratachophobia]

I've been running this for a few days at home while the car has been sitting idle. I've been grabbing log files of 6- and 12-hr duration since not much is changing other than SOC. I was trying to get good readings of the "Usable full pack" at different SOC such as 100%, 90%, 75%, 50%, etc.

Sometimes when I check on the tablet that is running the app in the car it is fine every 30 minutes for several hours in a row. Other times, I will see an error message like one of these and I don't know how long it hasn't been logging. It seems that as soon as I open the door and wait a few seconds the adapter starts sending data again and the error messages go away. It might be fine for several hours or might stop logging soon after (but I'm not sitting and watching it so I don't know).

I haven't yet looked in the log files to see if it is only missing a few minutes of data or several hours but wasn't sure if this was a known 'feature.'

View attachment 326437​

Have you noticed yet how SOC use different wh/m at different states? In other words, not all miles are created equal.

 

[Darren S]

Can you confirm that a full charge rated range is calculated by using nominal capacity as opposed to usable full pack? And when the battery goes below 20%, it changes it's calculation to use usable capacity instead?

Get ready to have your world rocked. The 90kwh batteries (haven't checked any others) are hiding their range degradation by giving you miles remaining based on nominal remaining from 100-80%. From 80-20%, the car slowly adjusts to use a number between nominal and usable. Then below 20%, it starts to use usable for the miles remaining calculation.

Have you noticed yet how SOC use different wh/m at different states? In other words, not all miles are created equal.

Ha, I haven't forgotten about your question, @supratachophobia, and have been gathering data so I'd have a decent answer with info to back it up. I had been chatting with @amund7 about some of the readings and what to look for and how to get it and then messing with the RAW datalog and analyzing it.

I have some cursory info but haven't filled it in with the data from the past few weeks but have better data from the past month and just haven't taken the time to gather it. I wanted to take it to a 100% charge, then down to <2% charge, then back up to 100% charge and see if that made any difference on "Usable full pack" or such. I don't know if it is using a different Wh/mi for the calculation but I included that in the stats as I was curious what it was.

Here was what I had started to put together a month ago with the brief amount of info I had gathered (and it shows how behind I am on posting some of this as it is from July 8th -- oops).

These are readings with SOC at 100%, 50%, 25%, 10%, 5%, and 2% as I was wanting to see if the "Full rated range" was different at those SOC levels based on your comment about the calculation possibly being different. My calculation matched what the app / car was reporting except for SOC of 10%.

I'd like to take readings between 25% and 5% a little more accurately and see where this different calculation occurs and if it is <20% SOC or closer to 10% SOC. Perhaps when SOC is at a low value the calculation is a little off which is why it suddenly drops from ~260 miles to 253 miles (my calculation). I took additional readings at 4%, 3%, 1%, 0%, -1%, and -2% SOC so I have that data and just need to gather it together at some point.

SOC vs "Full rated range"

• At SOC of 100%:
  • Full rated range = 260 miles
  • My calculation = 260 miles

• At SOC of 50%:
  • Full rated range = 260 miles
  • My calculation = 260 miles

• At SOC of 25%:
  • Full rated range = 261 miles
  • My calculation = 260 miles

• At SOC of 10%:
  • Full rated range = 254 miles
  • My calculation = 270 miles

• At SOC of 5%:
  • Full rated range = 263 miles
  • My calculation = 263 miles

• At SOC of 2%:
  • Full rated range = 253 miles
  • My calculation = 254 miles

SOC vs "Wh/mi"

• At SOC of 100%:
  • Nominal full pack = 288.8 Wh/mi
  • Usable remaining = 273.5 Wh/mi

• At SOC of 50%:
  • Nominal full pack = 288.5 Wh/mi
  • Usable remaining = 273.1 Wh/mi

• At SOC of 25%:
  • Nominal full pack = 288.5 Wh/mi
  • Usable remaining = 273.8 Wh/mi

• At SOC of 10%:
  • Nominal full pack = 273.0 Wh/mi
  • Usable remaining = 259.2 Wh/mi

• At SOC of 5%:
  • Nominal full pack = 284.6 Wh/mi
  • Usable remaining = 269.4 Wh/mi

• At SOC of 2%:
  • Nominal full pack = 295.1 Wh/mi
  • Usable remaining = 279.3 Wh/mi

​

 

[DB 2]

@Darren S

Could you check the raw log when at 100% SoC and look at message 0x338? The second half of the message should be either 22016464 (in hex) or 21016464, but I can't tell which? I think we all expect 2201 (290), but it sure looks like 2101 (289) from your data.

 

[amund7]

Have you ever looked into the bluetooth 4.0 stuff? I would really love to get rid of the flawed MX adapter that is causing all the problems diconnecting.

Yeah me too, but would bluetooth 4.0/LE solve anything? Are there adapters there as good as MX, but without hanging?

 

[David99]

Yeah me too, but would bluetooth 4.0/LE solve anything? Are there adapters there as good as MX, but without hanging?

Good question. When I used TM-Spy I used a $30 adapter which worked very reliably. It was connected for many days. But I know your app transfers a lot more data.

 

[amund7]

Good question. When I used TM-Spy I used a $30 adapter which worked very reliably. It was connected for many days. But I know your app transfers a lot more data.

I believe any ELM327 cheapo adapter would also not hang the way OBDLink MX does. LX has it's own set of problems/sleep issues. I have tried, several times, to contact OBDLink, both for this, and for collaboration (They list many apps as compatible, but not mine). But I recieved no replies. Is there any way we can make them listen?

They should potentially have ~300 Scan My Tesla customers by now, and the curve could rise steeper if they/we all had a less buggy adapter. It is awesome and nothing else compares, except when it hangs.

We should really start searching for, or even building a better adapter. What about a Raspberry Pi, it has bluetooth, and you can get a CAN shield for it. Should be quite easy to program something that would emulate ELM327 or ST1110 for Scan My Tesla to work. But it would be bulky, slow booting, probably buggy.

It would be super cool if someone could manufacture something like that that plugs directly into the diagnostics plug!

Comma Panda? I have one, wifi/elm327 support is unusable for SMT, but with a USB connection it should be fast, stable, and could potentially read more buses simultaneously. But at $99 it's quite steep, I know many people already thing the MX is too much ($79).

Any other ideas?

 

[SongjinDK]

It failed on version 9 (39.7, MX 100D AP2.5). Any one got it working on v9?

 

[Jays200]

I'll let ya know, if we ever get V9 in Australia.

 

[Krash]

It failed on version 9 (39.7, MX 100D AP2.5). Any one got it working on v9?

Did you have it working on v8?

 

[SongjinDK]

V8 always worked, it worked right before V9 update.

 

[Krash]

Does anyone have working CANBus on v9? Bummer if in addition to Tesla disabling diagnostic port access on the new hardware they disabled access on the old hardware via new software.

 

[SongjinDK]

Good news, I got it working after disconnected the OBDLink LX and reconnected and repaired the Bluetooth. It's good that Tesla didn't disable it which was what I have worried about.

Inside and outside temp always show 188F, this is not new for V9, it has always been like that, but it would be great if this can be fixed.

 

[amund7]

Inside and outside temp always show 188F, this is not new for V9, it has always been like that, but it would be great if this can be fixed.

There is not much I can do with I believe only 1 report and no logs or data on it. The quick fix is for you to delete those readings, then start logging and digging for another packet ID with CanbusAnalyzer, then let me know what you find.

It would be great if more people can confirm/deny this problem with other model X's. Also, a raw log and/or canbus dump of this occuring.

 

[scottf200]

There is not much I can do with I believe only 1 report and no logs or data on it. The quick fix is for you to delete those readings, then start logging and digging for another packet ID with CanbusAnalyzer, then let me know what you find. It would be great if more people can confirm/deny this problem with other model X's. Also, a raw log and/or canbus dump of this occuring.

FYI, I reported it up thread on June 30th, 2018. -- Vendor - Scan My Tesla, a CANBUS reader for Android

 

[amund7]

FYI, I reported it up thread on June 30th, 2018. -- Vendor - Scan My Tesla, a CANBUS reader for Android

I stand corrected: It has been reported 2 times, no data.

I can remove them when they read 188F, or you can select them and delete them from that tab. I prefer the latter, until we can understand how many cars this is affecting, or the logic behind it (after a certain model year? You and the other guy has a dead HVAC module?). More data on this is very welcome.

 

[scottf200]

I stand corrected: It has been reported 2 times, no data. I can remove them when they read 188F, or you can select them and delete them from that tab. I prefer the latter, until we can understand how many cars this is affecting, or the logic behind it (after a certain model year? You and the other guy has a dead HVAC module?). More data on this is very welcome.

Sorry. When I reported it before I never got a response (request for logs) and I thought it was a common problem that you didn't want to deal with. I didn't know it was uncommon. Your app is awesome and I appreciate all you do and have done. I have no issues with my HVAC modules/system that I know of as my heating and cooling have worked perfectly since I got it in December here in Chicagland (ie. heat then) and throughout this summer (ie. A/C). Obviously, having log data if this is not a recreatable thing by you is important. (I used to support users in the mainframe development world). I'll add this to my todo list.

 

[amund7]

Sorry. When I reported it before I never got a response (request for logs) and I thought it was a common problem that you didn't want to deal with. I didn't know it was uncommon. Your app is awesome and I appreciate all you do and have done. I have no issues with my HVAC modules/system that I know of as my heating and cooling have worked perfectly since I got it in December here in Chicagland (ie. heat then) and throughout this summer (ie. A/C). Obviously, having log data if this is not a recreatable thing by you is important. (I used to support users in the mainframe development world). I'll add this to my todo list.

We now have about 350 unique users running this app. I find it very strange that only 2 have this problem. I suppose you 2 have a different HVAC hardware than the rest, or all the other model X'es of the 348 others just didn't bother to report it ? In any case, I am pretty sure your cars will have the same data on a different canbus packet ID, probably different message format as well. I have spent hundreds of hours digging out the signals I have discovered, these temps are one of those. What I am trying is to get you (plural: all you users) to help dig for more. I have made and released all the tools you need. If you can find these from your own CAN dumps I will not hesitate to add them to the app.

 

[scottf200]

We now have about 350 unique users running this app. I find it very strange that only 2 have this problem. I suppose you 2 have a different HVAC hardware than the rest, or all the other model X'es of the 348 others just didn't bother to report it ? In any case, I am pretty sure your cars will have the same data on a different canbus packet ID, probably different message format as well. I have spent hundreds of hours digging out the signals I have discovered, these temps are one of those. What I am trying is to get you (plural: all you users) to help dig for more. I have made and released all the tools you need. If you can find these from your own CAN dumps I will not hesitate to add them to the app.

Thanks. I recall mentioning Comma.AIs canbus tool and you indicated it did not have the EV oriented data you needed. I found this post with your tool. I need to add this to my todo: Tool post: Vendor - Scan My Tesla, a CANBUS reader for Android