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

[mochouinard]

I'm not sure using the lowest 90% value within a given charge actually makes sense, though.

If the API reports everything above 89.5% as 90%, but actually only cuts off charging when it really reaches 90%, the lower charge levels will actually be inaccurate. Wouldn't it be more accurate to use the last 90% value for any single charge?

Even when

Even when your charging stop at 90%, you could even see your % level go up to 91% after it stopped charging. It not an prefect science as I don't have the raw information, but as long as I use the same data over a long period of time, then it turn out pretty good. Like I said, I might get in the future a feature that show only complete charge of 90% and take the first % value of when it reached completed. But I don't think it would bring much more value.

Understand that I have data for a time span of over a year now which you can see in the video I made. If your just getting started, that data is going to be limited for you what ever chart you use. But as time goes by, more data will allow to see some correlation with range decrease.

- - - Updated - - -

I'm joining the queue of thankful users.

Some of my graphs are not displaying correctly, especially Charge levels & Temperature. Just getting the horizontal grid lines without values being plotted. Could be occasional null values (in Charge levels), although Temperature values don't show any nulls.

Otherwise perfect tool, something I've been dreaming to do myself. Thank you!

Sent a PM for more info

- - - Updated - - -

Bonjour Marc,

this is is a great tracking system and give a lot of visibility into the data being collected. The ability to export is great too.

thank you / Merci beaucoup!

Your welcome

 

[pappie]

Hi,

Great website, it is helping me understand more about my usage.

I have one question:
22:06 - 00:56 - thuis - 59.4/63.5kWh 90/90% Complete

What is the difference between the 59.4 and 63.5 kWh?

 

[mochouinard]

Hi,

Great website, it is helping me understand more about my usage.

I have one question:
22:06 - 00:56 - thuis - 59.4/63.5kWh 90/90% Complete

What is the difference between the 59.4 and 63.5 kWh?

63.4kWh is the number generated by me from calculating your Voltage and amperage over the course of your charging period.
59.4kWh is the number reported by the car that was put back in the battery. That number does not include any preheating the car is doing for example.

 

[Marcus Adolfsson]

Love the site, so much data!

Here is a Model X if you want to tweak up some the image assets. TeslaLog - Car Info

 

[Darthamerica]

Do you mind if I ask exactly how the "Avg. Dist Daily" and "Avg. kWh Daily" are calculated?

thanks

 

[mochouinard]

Do you mind if I ask exactly how the "Avg. Dist Daily" and "Avg. kWh Daily" are calculated?

thanks

Pretty simple, I take the total number of km and kWh(Energy put into the battery reported by the car) for the week (Monday to Sunday UTC) and divide that by 7.

 

[Darthamerica]

Pretty simple, I take the total number of km and kWh(Energy put into the battery reported by the car) for the week (Monday to Sunday UTC) and divide that by 7.

I see. In that case you won't get an accurate report until the end of the 7 day period. For example if I drive 7 miles or km total on Monday, the report will say my daily average is 1 mile/km if I check on Monday. (7 mi/km)/7 = 1 mi/km.

If I may suggest, instead of dividing by 7, divide by n + 1 where n = 0 for Monday, 1 for Tuesday, 2 for Wednesday, 3 for Thursday, 4 for Friday, 5 for Saturday and 6 for Sunday. This way you get the true average for any given day during the week.

This makes the report much more useful on any given day in my opinion.

Thanks in advance,
Brandon

 

[mochouinard]

I would have to do that for the first week ever recorded and on the last week being recorded. I've just took a different approached. Not perfect but based on my last +year it look pretty much the same except the first and last week recorded show more accurate value. But for example, if you disabled the logger for only 1 day during the week, it will only could that day (as before, it would average over the whole week making a very tiny distance per day. The goal of that graph is to see the overall of your stats. The problem is you have very little stats to see at this moment, but when you get to have a year worth of data, the first and last week info won't matter much per say

 

[Andyw2100]

Is the distance that you are showing in the trip log summary the distance recorded by the odometer or the GPS distance? I am guessing it is the latter.

Thanks!

 

[mochouinard]

Is the distance that you are showing in the trip log summary the distance recorded by the odometer or the GPS distance? I am guessing it is the latter.

Thanks!

Odometer

 

[Andyw2100]

Odometer

I guessed wrong then. But that's only because it doesn't match up with what the odometer is reporting in the trip log for the same trips. I mean it is close, but not a perfect match. Trips of a little less than 60 miles are sometimes off by a tenth or two between what the odometer reports, and what the trip meter shows. Definitely a bit odd.

 

[Darthamerica]

I would have to do that for the first week ever recorded and on the last week being recorded. I've just took a different approached. Not perfect but based on my last +year it look pretty much the same except the first and last week recorded show more accurate value. But for example, if you disabled the logger for only 1 day during the week, it will only could that day (as before, it would average over the whole week making a very tiny distance per day. The goal of that graph is to see the overall of your stats. The problem is you have very little stats to see at this moment, but when you get to have a year worth of data, the first and last week info won't matter much per say

I see. As long as you don't care about the first or current week being accurate then all of the weeks in between will be correct. You know, it's amazing how this car has us all "tinkering" with the API just like people used to do with old ICE cars!

 

[mochouinard]

I guessed wrong then. But that's only because it doesn't match up with what the odometer is reporting in the trip log for the same trips. I mean it is close, but not a perfect match. Trips of a little less than 60 miles are sometimes off by a tenth or two between what the odometer reports, and what the trip meter shows. Definitely a bit odd.

The distance in the trip page first graph are group by 1/4 fraxion. If you look at the Wh/km, then it I got 2 way to calculate distance. Either by GPS between the points, OR by using the car speed and doing the math over the whole trip to find the distance.

- - - Updated - - -

I see. As long as you don't care about the first or current week being accurate then all of the weeks in between will be correct. You know, it's amazing how this car has us all "tinkering" with the API just like people used to do with old ICE cars!

Correct, but I changed it.

So the first week and last week should show better result.

Imagine if we really had more data from the car ! I'm hacking stuff from what Tesla provide. Like finding the Wh/km(miles), I don't think much people expected that such data could be extracted from the current api informations.

 

[Andyw2100]

I'm hacking stuff from what Tesla provide. Like finding the Wh/km(miles), I don't think much people expected that such data could be extracted from the current api informations.

You know what might be kind of cool, now that you have a lot of cars logging data? A chart that shows, by model, the average wh/mi for each day. We'd see weather trends affecting the overall average. We'd be able to compare our numbers to the overall average for any day, and think what might account for the differences, etc.

Just so you get an idea what I'm talking about, the table might look something like the following:

Date 70D 85D P85D 90D P90D S85 S60 P85

1/25/16 332 340 375 360 380 320 340 355
1/26/16 337 345 377 369 391 322 339 358
1/27/16 328 335 365 362 375 315 332 347


The wh/mi would be the average of all the cars within that model that recorded trips that day.

If for some reason someone really did not want their stats included in the figures, I guess you could allow them to opt out, by having an opt-out selection, but I can't think of a reason anyone would object to their wh/mi being used to put together an overall average.

If you did put together this chart, you might also show the total miles driven each day, by model as well.

Just some ideas.

 

[villena]

Some of my graphs are not displaying correctly, especially Charge levels & Temperature. Just getting the horizontal grid lines without values being plotted. Could be occasional null values (in Charge levels), although Temperature values don't show any nulls.

Just for the record, this issue was fixed for us living in the positive timezones, thanks Marc!

Small suggestion, add brief descriptions of different variables in graphs, e.g. "kWh charged reported by car" or "kWh charged calculated from current and voltage". Having these on the bottom of the page should reduce many questions here. No need to describe all variables, mostly those that are not quite obvious.

 

[ArtInCT]

Moch:

I see you have been getting quite a few questions about specific data elements shown in your web site. Perhaps a data dictionary would come in handy for all as time allows.

 

[Darthamerica]

The distance in the trip page first graph are group by 1/4 fraxion. If you look at the Wh/km, then it I got 2 way to calculate distance. Either by GPS between the points, OR by using the car speed and doing the math over the whole trip to find the distance.

- - - Updated - - -




Correct, but I changed it.

So the first week and last week should show better result.

Imagine if we really had more data from the car ! I'm hacking stuff from what Tesla provide. Like finding the Wh/km(miles), I don't think much people expected that such data could be extracted from the current api informations.

It's much closer to what I expected to see now. Thanks!

 

[MarcG]

Hi Marc! Thanks again for adding the option to turn off logging.

Are you working on a way to schedule it, so that it automatically turns off at night then back on in the morning?
If not, would it be possible to please add it to your queue of enhancements? Ideally with a control to enter the time of day for turning logging off and on.

Merci!

 

[Darthamerica]

The distance in the trip page first graph are group by 1/4 fraxion. If you look at the Wh/km, then it I got 2 way to calculate distance. Either by GPS between the points, OR by using the car speed and doing the math over the whole trip to find the distance.

- - - Updated - - -




Correct, but I changed it.

So the first week and last week should show better result.

Imagine if we really had more data from the car ! I'm hacking stuff from what Tesla provide. Like finding the Wh/km(miles), I don't think much people expected that such data could be extracted from the current api informations.

I have a question, what values are being used to calculate the Ave. Wh/mi in the "Daily average weekly" report?

Thanks

 

[Darthamerica]

Other suggested features:

-Track vampire loss(between charges and when not driving)
-Track power consumption from activity other than driving if possible