scottf200
Well-Known Member
Module 13 is probably correct and all the others are being read inaccurately
My X example:
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
My X example:
You're probably right. I'm going to ask tesla to replace every module except the 13th.
Only known issue / weridness is on 70, 75, and possible 60 batteries, the last voltage of the last module reads all 0xFFFFFF, which gives something like 5 volts. The car sends this, the app detects and hides it. So you won't see that last voltage at all. In your case you have a module that is slightly out of balance, the BMS should constantly try to balance this, see how it looks at different SOC and if it changes over time/charge cycles.
Thanks for the info. I'll keep an eye on it.
Regarding balancing, I've noticed that even after my car is charged and the ambient temperature is around 70 it still keeps running what I think is a battery coolant pump. Maybe it's trying to rebalance, fails and just doesn't give up for quite a while. This goes on for hours.
Side note; showing the battery voltages from this app to the service center will elicit some very dirty looks from the employees there.
Regarding balancing, I've noticed that even after my car is charged and the ambient temperature is around 70 it still keeps running what I think is a battery coolant pump. Maybe it's trying to rebalance, fails and just doesn't give up for quite a while. This goes on for hours.
Side note; showing the battery voltages from this app to the service center will elicit some very dirty looks from the employees there.
Yes, I had a user (in this thread a few pages back I believe) show Scan My Tesla to Tesla to try to show his 12v battery draining, and have them fix it. They did some tests, and concluded it was the power draw of the bluetooth adapter that emptied the 12v battery. Then charged him for the troubleshooting and tests.
Story ends with him getting the car back, paying for the service, stopped using the app and adapter, then the week after, 12v battery died. Of course a few milliamps of a sleeping adapter (or even an operating one) does not affect this. But if you try to use the app against them, you can be sure they get defensive, blaming you for using 3rd party / non-original equipment. They always do with 3rd party stuff. I have my own experiences with this too:
- I was denied suspension troubleshooting, because I had 3rd party wheels. Came back on stock wheels, lo and behold they replaced all arms in my front suspension, because all bushings were worn out, and the car drove like *sugar* (Model S, just before warranty end at 100k km)
- had to pay for a new frunk actuator, it stopped working, and they blamed my 3rd party stiffer hood struts that opens all the way up. It was the 2nd latch, it doesn't know if you have stronger struts becuase it's the pop-up spring that contacts this latch in any case. Lost this case, couldn't get my car back without paying. (Model 3) Afterwards I read that this is a common problem on the Model 3's, and Tesla and many owners know this very well.
So whenever I go to their service center, more reluctantly each time, I make sure I remove all traces of obd2 adapters, wiring, apps or anything else that could look like I modified anything, just so I don't give them any excuse to get away from their warranty liabilities.
Unfortunately, their cars are so great that I have tried many times but always failed in finding anything else I want. And of course I need to keep at least 1 Tesla around to maintain the apps
.. hope the Cybertruck gets here soon, and that it comes to Norway at all. (And that I won't have to go to the service center at all
)But to be serious, the customer service is frightening, and only getting worse. I think it's very strange they haven't been burned by the customers yet, enough for sales to drop and they losing money. I know quite a few people who had Teslas, got sick of the service and will never have a Tesla again.
Sorry for the rant and complete derailment, did not mean to kill your hopes about fixing your battery. But the fact is probably that a 60 mv voltage imbalance means nothing, if that module is failing you will just have to wait until it fails. Then of course you'll get a new (or refurbished) battery, as long as you are inside the 8 year drivetrain warranty period.
If that module has less capacity than the rest, it should go lower on voltage as you discharge, and go to 4.2 long before the others when charging. You can also have a look at the CAC numbers, these are not in the production version of Android yet unfortunately, but can be read from the Android beta version + IOS version.
Scan My Tesla should not be used 'against' Tesla for this reason, and for another reason: If Tesla feels the app is causing them trouble, they could probably do lots of things to shut it down, both technically but also legally. I hope that does not happen, so I like to encourage people not to use the app in that way. There are also disclaimers in the user agreements saying you can't use data from the app for this.
What you should use the app for, is educate yourself on how the car works, and if it is misbehaving, again use the app for education, then point out the fault using other sources, such as: If you get a replacement battery, and you suspect that battery has less capacity than your old battery, don't use the app, but rather record your range numbers from the car screen before and after. If you have a module that has less capacity than the others, this will also eventually show in the range numbers, but I suppose you can't demand a new battery until it has lost a significant amount of range.
Still, it shouldn't be like this, customers should have the right to read some of this data, to have some artillery in warranty disputes. Tesla sees the same data, but they won't show it to you. So Scan My Tesla is a great weapon, but like all weapons, it should be used with care and respect, together with a healthy morality compass. It was never intended to be used as a weapon, just as a curiosity satisfier.
I imagine Tesla's reputation for lousy customer service is becoming known by now but it's like you said, what the hell else are people supposed to buy? A Bolt instead of a 3? A Taycan instead of an S? And for all I know this electric car thing is so new to them that their customer service would have similar problems.
Didn't really expect Tesla to fork out a new battery for me for an 80mv difference. I was hoping for a more technical assessment of what the potential problems of this anomaly could be. In retrospect this was not a realistic expectation. They're mostly guys that can turn a wrench and follow a service manual, not experts in chemistry and engineering.
I don't think Tesla's customer service is intentionally malicious, just a very low priority for them.
Didn't really expect Tesla to fork out a new battery for me for an 80mv difference. I was hoping for a more technical assessment of what the potential problems of this anomaly could be. In retrospect this was not a realistic expectation. They're mostly guys that can turn a wrench and follow a service manual, not experts in chemistry and engineering.
I don't think Tesla's customer service is intentionally malicious, just a very low priority for them.
aerodyne
2 of 3 EV's - Defect Free!
The App will show you what is running. In my case, the battery pumps run at 23-28% in the conditions you mention, when I am level 2 charging, then turn off after a while, and stay off unless the car is woken up.
However, since an update early this year, if I charge to over 78%, the same pumps go to 100% and stay on.
On your side note, fully agree. As Amund said, big mistake to leave the reader in the car when taken in for service.
My problem child string is number 2. I have seen as much as 40 mv low there. A few charges above 70%, and a few harder than normal accelerations seem to bring it down to under 20mv.
aerodyne
2 of 3 EV's - Defect Free!
It could be a top down approach. Since Elon as I recall is still in charge of Service, perhaps he knows exactly what he can get away with. As long as order books are full, and competition is failing (polestar recalls, for instance) every dollar spent on service is one less spent on the GF's or 4680 cell production.
Or it could just be greed. I really don't know.
I agree that one should be carefull in showing SMT results to the SeC, but having SMT is still valuable in another way. If it shows no problems, it can perhaps help when it becomes time to sell, esp with a out of warranty or high mileage car.
aerodyne
2 of 3 EV's - Defect Free!
Yes, I had a user (in this thread a few pages back I believe) show Scan My Tesla to Tesla to try to show his 12v battery draining, and have them fix it. They did some tests, and concluded it was the power draw of the bluetooth adapter that emptied the 12v battery. Then charged him for the troubleshooting and tests.
Story ends with him getting the car back, paying for the service, stopped using the app and adapter, then the week after, 12v battery died. Of course a few milliamps of a sleeping adapter (or even an operating one) does not affect this. But if you try to use the app against them, you can be sure they get defensive, blaming you for using 3rd party / non-original equipment. They always do with 3rd party stuff. I have my own experiences with this too:
- I was denied suspension troubleshooting, because I had 3rd party wheels. Came back on stock wheels, lo and behold they replaced all arms in my front suspension, because all bushings were worn out, and the car drove like *sugar* (Model S, just before warranty end at 100k km)
- had to pay for a new frunk actuator, it stopped working, and they blamed my 3rd party stiffer hood struts that opens all the way up. It was the 2nd latch, it doesn't know if you have stronger struts becuase it's the pop-up spring that contacts this latch in any case. Lost this case, couldn't get my car back without paying. (Model 3) Afterwards I read that this is a common problem on the Model 3's, and Tesla and many owners know this very well.
So whenever I go to their service center, more reluctantly each time, I make sure I remove all traces of obd2 adapters, wiring, apps or anything else that could look like I modified anything, just so I don't give them any excuse to get away from their warranty liabilities.
Unfortunately, their cars are so great that I have tried many times but always failed in finding anything else I want. And of course I need to keep at least 1 Tesla around to maintain the apps
.. hope the Cybertruck gets here soon, and that it comes to Norway at all. (And that I won't have to go to the service center at all
But to be serious, the customer service is frightening, and only getting worse. I think it's very strange they haven't been burned by the customers yet, enough for sales to drop and they losing money. I know quite a few people who had Teslas, got sick of the service and will never have a Tesla again.
Sorry for the rant and complete derailment, did not mean to kill your hopes about fixing your battery. But the fact is probably that a 60 mv voltage imbalance means nothing, if that module is failing you will just have to wait until it fails. Then of course you'll get a new (or refurbished) battery, as long as you are inside the 8 year drivetrain warranty period.
If that module has less capacity than the rest, it should go lower on voltage as you discharge, and go to 4.2 long before the others when charging. You can also have a look at the CAC numbers, these are not in the production version of Android yet unfortunately, but can be read from the Android beta version + IOS version.
Scan My Tesla should not be used 'against' Tesla for this reason, and for another reason: If Tesla feels the app is causing them trouble, they could probably do lots of things to shut it down, both technically but also legally. I hope that does not happen, so I like to encourage people not to use the app in that way. There are also disclaimers in the user agreements saying you can't use data from the app for this.
What you should use the app for, is educate yourself on how the car works, and if it is misbehaving, again use the app for education, then point out the fault using other sources, such as: If you get a replacement battery, and you suspect that battery has less capacity than your old battery, don't use the app, but rather record your range numbers from the car screen before and after. If you have a module that has less capacity than the others, this will also eventually show in the range numbers, but I suppose you can't demand a new battery until it has lost a significant amount of range.
Still, it shouldn't be like this, customers should have the right to read some of this data, to have some artillery in warranty disputes. Tesla sees the same data, but they won't show it to you. So Scan My Tesla is a great weapon, but like all weapons, it should be used with care and respect, together with a healthy morality compass. It was never intended to be used as a weapon, just as a curiosity satisfier.
good post, fully agree. With your experiences, I wonder if you have not given thought to what a weaponized version of SMT would look like.
Ahh. Well there's the problem. Elon has some impressive ambitions in life, reshaping transportation, going to the moon, stopping global warming. But a well run service department that's efficient enough to keep these cars running at a reasonable cost to consumers and a fair profit to Tesla probably isn't nearly as sexy in his mind as some of his other goals.
If he is in charge of service then I think the shareholders should demand that he delegates that responsibility to some boring accountant familiar with cars who gets hard just looking at positive yelp ratings for service centers.
I bought the app, cable, and tried to use it on my 2017 Model X with a bluetooth ELM327 I already had from 2013. I'm getting an error connecting and only one red light shows on the adapter. Do I just need to get a different adapter or is there another problem?
I installed the app on a Nexus 7 (2013) running Android 5.x. I tried to download the app on my Galaxy Note 10, but it seems like it will charge me again even though it's the same Google Play account?
I installed the app on a Nexus 7 (2013) running Android 5.x. I tried to download the app on my Galaxy Note 10, but it seems like it will charge me again even though it's the same Google Play account?
Attachments
No, 1 payment = all devices for that account, I don't think there is a limit for devices, double check which account is logged in.
For the lamp, see if you can see what it means, and if there's a lamp for bluetooth connection, there should probably be a blue one too if the app manages to connect to it. Check the bluetooth pairing in the phone's bluetooth settings, that you have selected the correct adapter in the 'choose adapter' menu of the app. If it connects to the adapter OK it should get to the next step of the status text, which is 'connecting to car', but it already fails while connecting to the adapter, for some reason. Also be sure to turn bluetooth off on the other phone while you test, often/usually the adapter stays connected to the last phone that was connected, and only connect to one phone at a time. Hope some of this helps.
PS. I have had an exact same looking adapter, it has worked great for years - but I also have reports from users with the same looking device that does not support canbus, so it's hard to know what's inside of 'no-name' adapters like this one.
Thanks for the reply! I tried again to download the app on my Galaxy Note 10 and this time it did not ask for payment. Seems like it took a full day to sync my purchase.
On the Note 10, the app connects to the OBD II adapter without error and it works fine! Several lights are flashing instead of just the solid red light.
@amund7 It looks like SMT misses some important temperature data available. No invertor temperature (EDIT: I said BS, they are here, but do we miss some others?).
Also, rear/front oil temps are not reflecting reality at all - I put actual sensors toeasure oil temperatures both sides. At a minimum rear looks more like front almost always.
Do you see how you can fix that? Thank you.
I need as many temps as possible to continue cooling development.
I wish I didn't have to do that,.but you can see oil temps are BS (bottom gauge is rear).
Also, rear/front oil temps are not reflecting reality at all - I put actual sensors toeasure oil temperatures both sides. At a minimum rear looks more like front almost always.
Do you see how you can fix that? Thank you.
I need as many temps as possible to continue cooling development.
I wish I didn't have to do that,.but you can see oil temps are BS (bottom gauge is rear).
Last edited:
aerodyne
2 of 3 EV's - Defect Free!
I think it is entirely possible for temps to differ depending on where the sensors are placed, esp in an oil system. No oil cooler, AFAIK.
Could it be the values SMT is "missing" are in fact not reported by the CAN bus?
Could it be the values SMT is "missing" are in fact not reported by the CAN bus?
If they are swapped front and rear in the firmware that has been suspected for a while, I have been waiting for good evidence of this. If I understand your post correctly, your pictures confirm they are swapped front & rear, that can be fixed in the app.
Where is the BS, and what is it you wish you didn't have to do?
They are not just swapped.
Imagine we swapped them.
Top gauge: 20C + 30C while R oil is 28C in both cases.
Bottom gauge: 32C + 46C while F oil is 43C and 41C.
I don't see any consistency whether it's reversed or not.
I wish that we had correct CAN data and I wouldn't need to use stupid gauges.
Last edited:
OK. From what I gather from the signal names (they are from JWardell's DBC) they are reported by the oil pump. Now I have no idea where the pumps are on the different drive units and where your gauge sending units are placed. Could it be explained by input/output/before/after heat exchanger? Or you said those were disabled in your test?
Another theory is, if you have an el-cheapo bluetooth adapter, they choke on the can traffic, and because reasons (canbus message priority and bitmask filters) if you have lots of fast packets in the same tab as slow ones, or just lots of packets in general in the same tab, the slow packets might not get to the app at all for long periods of time. If you have a slow adapter, run a new test with only the oil temps in a tab by themselves, and see if that changes anything. If you record a CSV log of this tab, you could see if this is really the case, because it records numbers only when they are coming in.
If this IS your problem, and you don't want to remove any signals, get an OBDLINK LX, it can capture ~1000 packets per second, and has individual can ID filters, so little or no data is ever lost.
Another theory is, if you have an el-cheapo bluetooth adapter, they choke on the can traffic, and because reasons (canbus message priority and bitmask filters) if you have lots of fast packets in the same tab as slow ones, or just lots of packets in general in the same tab, the slow packets might not get to the app at all for long periods of time. If you have a slow adapter, run a new test with only the oil temps in a tab by themselves, and see if that changes anything. If you record a CSV log of this tab, you could see if this is really the case, because it records numbers only when they are coming in.
If this IS your problem, and you don't want to remove any signals, get an OBDLINK LX, it can capture ~1000 packets per second, and has individual can ID filters, so little or no data is ever lost.
Both readings were done with radiators bypass.
Sensors are in the filter and the filter is just almost touching pumps.
In my case filter is relocated, but those are braided 10AN lines and they don't really cool oil significantly.
As you can see temperatures are not just all lower or higher. At different times they shifted differently.
I also don't see any effect of turning on the radiator at reported temperatures, while the gauge is clearly going 20-30C down within seconds and stator temperatures respond as well.
I use MX+ for logs, so no choking and the accuracy was 0.06 seconds.
Similar threads
