Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

TeslaLog.com - Your hosted Tesla Data Logger - Announcement / Support threads

This site may earn commission on affiliate links.
Set this up today... Will play with it after my next drive... Would be good to add something if possible for changing your tesla log site email/pw at some point or maybe a web method to delete tokens.

any thoughts of effects on range, if this will be doing frequent pulls from the car isn't the onboard computer and modem going to be constantly awake which will increase vampire drain?
 
  • Helpful
Reactions: markn455
Set this up today... Will play with it after my next drive... Would be good to add something if possible for changing your tesla log site email/pw at some point or maybe a web method to delete tokens.

any thoughts of effects on range, if this will be doing frequent pulls from the car isn't the onboard computer and modem going to be constantly awake which will increase vampire drain?
For the moment, if you really need to change your email, just contact me.

As for deleting token, there is no way for me to tell tesla to disable a token (as far as I know), so it really best that you change your tesla password so all your token are reset. This way, you are sure that token become invalidated.

As for vampire drain, sadly yes of course it will happen ! You will lose about 60% of your battery charge over 30 days, so about 2% per day. Since my car is plugged in every night, that doesn't bother me much.

I currently query the car every minutes for it information.. I guess I could make a energy saving mode where the car is not polled as frequently. But if I do so, I won't be able to know when your leaving on a trip, so if I query every 30minutes instead, you might lose your whole drive.

You can manually disable remote access on the car which I think should stop the car from waking up if you intend to leave the car idle and need the battery energy.

- - - Updated - - -

One thing I'm disappointed at Tesla, is they didn't put any direct 12volt charger on the car. So even when the car is plugged in, the 12 volt battery will drain and force the main battery pack to then charge the 12volt battery. So once your car drop 3% of it predefined charging level, only then will the charger activate and recharge that 3%. I would have much more see the 12volt battery being recharged directly from the outlet ! (I guess I could remove the nose cone and connect a 12volt charger to the terminal, but that a little bit inconvenient hehe)

- - - Updated - - -

In the morning I REALLY want to have my 100% charge, what I do is I lower my charge rate so the charging last all night.

I REALLY wish I could change the charge rate with the API. I could then automaticly guess your morning behaviour of when your leaving, and adjust the charging rate so you get your charge complete just before that time.
 
I think the ability to start/stop recording via a simple web request (while also giving us control of the polling frequency) might be a good alternative to the frequent polling.

It would be trivial to automate this via Tasker/AutoMagic on Android, or you could even release a companion app which triggers based on phone sensor activity, or Bluetooth connectivity.
 
I'll add a option to disable the car recording.

I might be able to query tesla server without forcing the car to wake up so that I could still query every minute to figure out if the car is sleeping or not, then activate the full logging if it not sleeping... But that would require some experimentation. A companion app is also an option.

I could work full time on this project ! But I'm getting no income coming from this ! It actually cost me money to host. I'm estimated more than 150million data entry per month with the current number of car registred today, that I'll store and have to parse. Let just say, it the biggest database I've handled so far, and trying to be low cost on specs of the server is making my life difficult :)
 
Registered... Love it so far.

Feature requests etc...
* will this pick up my X when it joins my stable in a few days?
* Of course being me I want more stats... Run a regression through all of the factors effecting efficiency (elevation, wind, temperature, speed, HVAC, etc) to figure out what each costs and add them in so I can see what is causing what in each trip... The left over would be driving style and road conditions.
* more average and totals like average efficiency and total distance per calendar month (currently I track this manually)
* careful not to get Tesla complaining about using too much mobile bandwidth and hence money
* ways to reduce energy use when not plugged in. For me going to 15 minute intervals batten midnight and 6 when not plugged in wouldn't cause issues but I don't know how much battery it saves... Basically with this using 2kwh/day that is about the equivalent of one third of my house's entire base electricity load.
* fix typo in warnings... Your -> you're in one spot.
 
Thanks Muzzman1 !

I know it suck, any website could be exposed to be hacked. I've tried to use some good practice for user authentication, and most of the site is actually plain html files.

The frontend is 100% html only, so the actual page you request have no dynamic content into it, all query is done via API calls. For me, this was the best way to keep the backend code as simple and clean as possible. Also, you can actually see everything I've done, except the data gathering and database query for frontend API.

The actual backend API code have just little over 1000 line of code that generate dynamic contents which can easily be audited. That part is most clean section of the site :)

The frontend html and JS is a BIG MESS to anyone looking at it I guess... I've done web site back in 1998 and my html technique haven't updated much since then (I'm a table kind of guy :). So lot of thing have been experimentation which I'm surprised it worked so well across multiple browser and devices. And since it just frontend, it doesn't bother me much in term of security. (Though, there is a few things that I'll have to address before I give people ability to share their informations to others).

What worried me the most of is people with bad intent just to disturb the service. I don't control Tesla servers and there is some limitation in the action I can do to protect it without making it annoying for the rest of us. I know lot of way to actually do it. Tesla could decide to stop this service all together.

My actual biggest concern for the future is how do I scale the service as it seem new user registration are coming in steadily now. How will Tesla react if there is too much user using it (I've began contacting them). How costly will it be to host it, and at some point, it might be more than just a fun project and require much more time to handle the traffic...

I would love to keep this free, but I'm realizing now that it is not possible.

Since I store your data, I don't think you guys want that I let just anyone have access to the backend to help me out. So if I gather help, that mean I'll have to create a corporation, draft legal document and hire people.

I might own a Tesla (Well, the bank own most of it still), but the I need to be careful with my budget to shoes the car... So my funds are very limited !

So I got no idea where I should bring this next... As I write this post, 6 other people registred on the site. I want to share this tool to everyone who might find it useful. How would you guys react if I charge for it ? How much is ok ? And if I charge, how many user will want to keep using the service. How will Tesla react if I charge for something that connect to their API... Will they charge me an commercial API access fee...

I've been talking to one of the site user and first tester about integrating OBD data as well, which would give us more information about the car than the API provide, and as well allow to have more than just tesla being tracked.

Just sharing with you guys my current reflection on this whole thing...

It all fun and game until someone have to pay :)

- - - Updated - - -

Registered... Love it so far.

Feature requests etc...
* will this pick up my X when it joins my stable in a few days?
* Of course being me I want more stats... Run a regression through all of the factors effecting efficiency (elevation, wind, temperature, speed, HVAC, etc) to figure out what each costs and add them in so I can see what is causing what in each trip... The left over would be driving style and road conditions.
* more average and totals like average efficiency and total distance per calendar month (currently I track this manually)
* careful not to get Tesla complaining about using too much mobile bandwidth and hence money
* ways to reduce energy use when not plugged in. For me going to 15 minute intervals batten midnight and 6 when not plugged in wouldn't cause issues but I don't know how much battery it saves... Basically with this using 2kwh/day that is about the equivalent of one third of my house's entire base electricity load.
* fix typo in warnings... Your -> you're in one spot.

For your Model X (CONGRATS by the way ! ), the car will display probably a white model S on the site (Also, some of the new car color is not supported... I lost my documentation on how to extract the Image archive from the Tesla Mobile APP APK). But it should start logging correctly unless tesla revamped the whole API which I doubt. If not, contact me ASAP and I'll make sure it work of course, the earlier you get data recorded on your car, the most interested it is in the long term ! (Sadly, I don't have the first ~10 000km off mine :(

More stats are in the pipe line !!!

Reports of your average Wh/km(or miles) per month, week, and even depending of different temperature bracket is in the work. I also want to detect when the car is stopped at street light and look at the kW used by the car to estimate how much is used for heating and extrapolate info from it...

It is my goal also to create some competition for efficiency between us :) So have a leader board on different statistics.

Already trying to contact someone at Tesla HQ regarding the API and see how/if we can coexist happily !

I wish the car didn't use as much energy for it internal computer. Also, I wish I could get notify when the car start up... I got some idea I can experiment with, but that going to require lot of changed in the backend engine that capture the data, and it been very stable so I don't like to play much with that part of the code... But I'll get to it at some point.

There going to be lot of Typo to take care off... English not my native language, and I'm probably even worst in my native language (french).
 
Any stats thus far for how much bandwidth and storage are being used monthly server side?

With the current user base, based on pass statistic, I'm guessing around 12gig of per month of physical HD space per month and about 2000gig transfer.

By the end of next year, it might be in the ball park of 288gig per month (3456gig per year) of actual database space that need to be responsive if the user groth is similar to what I have now !

That about 1% of the number of tweet Twitter handle every year in term of the number of entry I'll have in the DB with under 0.0009% of it user base.
 
With the current user base, based on pass statistic, I'm guessing around 12gig of per month of physical HD space per month and about 2000gig transfer.

By the end of next year, it might be in the ball park of 288gig per month (3456gig per year) of actual database space that need to be responsive if the user groth is similar to what I have now !

That about 1% of the number of tweet Twitter handle every year in term of the number of entry I'll have in the DB with under 0.0009% of it user base.

I'm looking at starting something in the new year that might help with this, so that you can avoid some of the issues - it's a replay of some of the data federation work my company carries out for Telcos, Banls etc. and will help with data ownership/security issues ... like you I just need to find the time to progress it :D
 
I suggest that you prune the older data off of the database after a specific period of elapsed months, especially the data that consumes the most storage space.
For those users who would want you to retain the data for their own purposes then they could pay the storage fee.
I would imagine you need to understand how many users one server will support. If we call that U then U+1 will require a build out of another cloud server and its costs.

As far as what you could charge for this service... it really depends upon your overall costs and what the market will bear.
I think there should be an introduction period with a diminished price.

You goal would be to make your interactions as few as possible, i.e. you do not want to have to go in and manually delete data, users, etc....

BtW, have you considered selling/working out an arrangement with Tesla to run and host?
 
I find this service 'valuable' (actually more interesting than valuable) and would be willing to pay something. Given the uncertainty of the API (Tesla could revoke access/make changes at any time), I think a monthly fee is a better option than a one off payment (avoiding customers of yours looking for refunds if Tesla were to make changes that broke your service). As for how much I would pay.. a couple of dollars a month wouldn't be a problem. $5/month would make me stop and think, not sure if I get enough out of it myself at that price to justify it. Of course, you need to recover your costs (and make something). Once again, thanks for the amazing service. I signed up quite some time ago... didn't really use it for several months, but lately have found myself checking out trends, and am glad I have many months of data tucked away.
 
I would offer this as a "free" service with most of the bells and whistles, but data is only held for (Pick a number) one month.

If you want the premium service with unlimited storage (And maybe a few additional bells and whistles), that will cost $/month.

Just a thought.
 
  • Like
Reactions: robertvg
Some hooligans have made off with my wheels!!

Also, my odo and rated range are in KMs, not miles.

Really nice app. Thanks!

untitled4.png
 
  • Like
Reactions: robertvg
With the current user base, based on pass statistic, I'm guessing around 12gig of per month of physical HD space per month and about 2000gig transfer.

By the end of next year, it might be in the ball park of 288gig per month (3456gig per year) of actual database space that need to be responsive if the user groth is similar to what I have now !

That about 1% of the number of tweet Twitter handle every year in term of the number of entry I'll have in the DB with under 0.0009% of it user base.

Im curious how you got to this calculation and what your data storage architecture is.

Twitter handles more than 500 million tweets per day. Even just 1% of twitter volume is an absolutely massive architecture. Do you expect to be storing 5 million records per day in a year from now? How many users are you projecting to have?