TeslaFi.com

[Cyclone]

View attachment 221785 How does Teslafi compute the KWh used and added and what is the difference between the two? Is it a guestimate or is it actual usage?

I am trying to figure out why my electric bill is so high so I signed up for Teslafi and installed a Sense. There is a discrepancy between Teslafi and Sense as far as total KW used. For example, this morning the charging numbers for the total amount of KWh my house has used is at 44kwh. There is no way I used up so much besides the Tesla. It's also possible the Sense is off a little with total usage.

TeslaFi is taking the numbers from the car's reporting via the API. Used is how much AC power the chargeport is taking in and added is how much DC power is put into the battery. There are conversion losses as well as losses to the wiring and the overhead of the charging systems (like the battery heater/cooler and other components).

 

[mbhforum]

TeslaFi is taking the numbers from the car's reporting via the API. Used is how much AC power the chargeport is taking in and added is how much DC power is put into the battery. There are conversion losses as well as losses to the wiring and the overhead of the charging systems (like the battery heater/cooler and other components).

Thanks. In your opinion, is my experience normal with Teslafi KWh usage vs another source? Or do you feel I have an electrical problem?

 

[Cyclone]

Thanks. In your opinion, is my experience normal with Teslafi KWh usage vs another source? Or do you feel I have an electrical problem?

I am unfamiliar with Sense. Assuming it is measuring usage somewhere along the power supply to the plug (or at the plug), it should read higher than TeslaFi because of line losses before even getting to the chargeport. There is resistance in wiring (which is what makes it warm over time) and thus, the more distance between the Sense and the chargepoint as measured in length of wire, the higher the line loss you will see.

As for the 95% efficiency per TeslaFi, at 36 amp/237v, that sounds about right. Going above 36 amp, if you use the Tesla mobile connector, will reduce efficiency due to heat losses. That is why I charge my car at less than 40 amps even though I'm on a NEMA 14-50. In fact, I think I may need to try something around 36 amps instead of 30 since you are charging with more efficiency than I am!

 

[AWDtsla]

As for the 95% efficiency per TeslaFi, at 36 amp/237v, that sounds about right. Going above 36 amp, if you use the Tesla mobile connector, will reduce efficiency due to heat losses. That is why I charge my car at less than 40 amps even though I'm on a NEMA 14-50. In fact, I think I may need to try something around 36 amps instead of 30 since you are charging with more efficiency than I am!]

95% is not achievable with the current hardware. This is a measurement error.

It would be quite difficult for i^2*r losses to exceed the auxillary overhead of charging (~400W), especially if you're only measuring from the car itself. The more amps you draw, the more efficient your charge will be. Also running the chargers at less than full also hurts efficiency

Model S Gen2 Charger Efficiency Testing

 

[Cyclone]

95% is not achievable with the current hardware. This is a measurement error.

It would be quite difficult for i^2*r losses to exceed the auxillary overhead of charging (~400W), especially if you're only measuring from the car itself. The more amps you draw, the more efficient your charge will be. Also running the chargers at less than full also hurts efficiency

Model S Gen2 Charger Efficiency Testing

If I charge at 40 amps, my UMC becomes quite hot to the touch. Not burning, but in the hot range well above warm. At 30 amps, I do not perceive a temperature difference, though I have not taken a thermal camera to it. There must be some losses from the wall to the plug for that heat, so I just dropped to 30 amp and called it a day over a year ago.

 

[AWDtsla]

If I charge at 40 amps, my UMC becomes quite hot to the touch. Not burning, but in the hot range well above warm. At 30 amps, I do not perceive a temperature difference, though I have not taken a thermal camera to it. There must be some losses from the wall to the plug for that heat, so I just dropped to 30 amp and called it a day over a year ago.

It gets hot because it's made out of insulating material. The temperature doesn't indicate how much energy is being lost, only how much energy is being retained. 40A will be the most efficient.

 

[cvrcv]

@Jdeck Something is up with the efficiency/inefficiency calculation. It does not change even if you update your correction factor in settings, but even with 1.0 correction, it still seems wrong. i.e 331 Wh/mile = 3.1% inefficiency in a P90D? Shouldn't that read 331/310=6.8% inefficiency?

 

[Jdeck]

@Jdeck Something is up with the efficiency/inefficiency calculation. It does not change even if you update your correction factor in settings, but even with 1.0 correction, it still seems wrong. i.e 331 Wh/mile = 3.1% inefficiency in a P90D? Shouldn't that read 331/310=6.8% inefficiency?

The efficiency is simply the miles driven dived by the rate miles used. 50 miles driven / 100 rated miles used = 50% efficiency. Changing the kWh factor will not effect the efficiency %.

 

[Jdeck]

After the sleep fix (uh, it'snot actually a fix but) my car shows Florida USA location when GPS is not available. At my home, at the office, at underground superchargers, etc. Is it possible to fix?

There's a thread somewhere on here about this but I can't seem to find it. It appears to be a bug in 8.1 but for some reason it's not affecting everyone. I enabled the 'latitude' column in your raw data page and you'll see the latitude jumps to -91.0. I was hoping it would be fixed with 17.11.10 but apparently not. The API is actually reporting the same -91.0,-181.0 for all users that are affected. I can't seem to find any rhyme or reason to it, it just comes and goes randomly. I'd give a reboot a shot though, someone told me it helped temporarily in 17.11.03.

I just put a change in the data logger to replace any data points with these bad coordinates with the last good coordinates received. This should help a lot. Any new data points will reflect these changes. Please let me know if it helps.

 

[Jdeck]

A minor but somewhat confusing bug: I've been annotating recent Supercharges with stall information (why I don't know, I just felt like it). From the daily view, I click on the notepad icon for the Supercharging session in question, enter my comment, and press Submit. When the page refreshes, that one session that I'd just edited seems to have replaced all of the charging sessions for that day. If I navigate away from that daily page and back again, everything's back to normal.

Probably easiest if I just show what's wrong, so here's a screenshot:

View attachment 221762

In this case I've just added an annotation to a charging session at LA/Hawthorne. The next two charging sessions should show Oxnard and Atascadero, but they've both been replaced by the LA/Hawthorne entry. (There are other charging sessions before and after these that were similarly affected.) If I reload the page, all is well.

Bruce.

EDIT: PS. It also seems like if I've annotated any of the charging sessions on a day's page, all of the charging sessions for that day have the notepad icon underneath their data, regardless of whether I'd added an annotation for that particular charging session or not. This quirk persists across page reloads.

Thanks. I'll look into this.

 

[cvrcv]

The efficiency is simply the miles driven dived by the rate miles used. 50 miles driven / 100 rated miles used = 50% efficiency. Changing the kWh factor will not effect the efficiency %.

Hmm, how are you calculating kWh and Wh/mi then? If I work the math backwards, you are using 321Wh/RM for a SP90D? Should be 310.

 

[clostridium]

Can't seem to get the car to go to sleep - it will sometimes get to the last minute of the 15 minute time frame and then wakes up again when I look at the raw API data. I just end up with 24 hours of idle each of the days. The car has 17.11.10 installed. I've only been using teslafi for a week or so and only on that firmware version so don't know if it would have worked before the last update.

I've got always connected ON and power savings ON. The car is sitting without charging for several days outside. The vampire power loss of ~3% per day isn't a big deal for this trip but I want to get this sorted out for a trip next week that will be longer.

I've got Teslafi sleep mode on (though the footnotes make it seem that this may not work if always connected is ON) with the default time periods. I've also got scheduled night sleep mode on. The location is tagged as being OK for all sleep modes.

I don't have access to my car to change any settings that can't be done remotely. I'm wondering if you can sleep at all with always connected = ON - and does anyone know if you can change that setting remotely? Could Remote S and/or the Tesla phone apps be messing this up? I'm not opening them during these failed time periods. I just set them to no background refresh on my iphone as an experiment so will see how that works.

Here's a screenshot of the raw data. The car gets to the 30 minute waiting period and then starts the 15 minute countdown and always right at the end it fails. There's a couple of these each day however interestingly it doesn't consistently keep trying. It also doesn't ever do the night time sleep.

Thanks

 

[hiroshiy]

I just put a change in the data logger to replace any data points with these bad coordinates with the last good coordinates received. This should help a lot. Any new data points will reflect these changes. Please let me know if it helps.

Thanks for your fix! Not sure 100%, but the bad coordinate while charging still seems to have that Florida USA? Will check tomorrow again.

 

[Jdeck]

Hmm, how are you calculating kWh and Wh/mi then? If I work the math backwards, you are using 321Wh/RM for a SP90D? Should be 310.

I'm not using any rated wh/m. It's all based on battery size and how many rated miles the vehicle has at a 100% charge. Then the rated miles used in the drive determines the kWh and Wh/Mile used.

 

[skitch23]

Can't seem to get the car to go to sleep

That's weird. Mine is falling asleep on a razor's edge since the latest update.... often within 15 mins. If I notice that it's not asleep when it should be, I open the app, let it connect, then close the app. It falls right asleep shortly after.

 

[clostridium]

That's weird. Mine is falling asleep on a razor's edge since the latest update.... often within 15 mins. If I notice that it's not asleep when it should be, I open the app, let it connect, then close the app. It falls right asleep shortly after.

I can now confirm that shutting off the background refresh on the apps doesn't help my situation. The car just did the same 15 minute countdown and then failed right at the end.

When you say you close the app are you quitting it completely on your phone or just changing to another app on your phone?

 

[hiroshiy]

Hmm, how are you calculating kWh and Wh/mi then? If I work the math backwards, you are using 321Wh/RM for a SP90D? Should be 310.

You need to adjust the kWh & Wh/m(km) Factor in the Settings. Type in 1 into the setting, and calculate the following value:
[Typical or Rated km] x [Typical or Rated Wh/km for your car] / [km Driven] / [TeslaFi calculated Wh/km]
Now in my case the number is approximately1.07 so I changed the value to 1.07.

 

[cvrcv]

I'm not using any rated wh/m. It's all based on battery size and how many rated miles the vehicle has at a 100% charge. Then the rated miles used in the drive determines the kWh and Wh/Mile used.

Well that probably explains the weirdness and having to set different values for the teslafi correction factor. A rated mile should be worth exactly the value from this thread, and does not change over time: Calculate usable battery capacity based on rated miles values Tesla has had several versions of the 90 pack, some with more energy than others, so can't use that. So if I get 260 miles at full charge, the BMS is really saying that I only have 260*.310=80.6kWh available, on the first gen 90 "kWh" pack.

 

[cvrcv]

You need to adjust the kWh & Wh/m(km) Factor in the Settings. Type in 1 into the setting, and calculate the following value:
[Typical or Rated km] x [Typical or Rated Wh/km for your car] / [km Driven] / [TeslaFi calculated Wh/km]
Now in my case the number is approximately1.07 so I changed the value to 1.07.

I have to use .93, so clearly some numbers are off or calculation is missing, as in the post above.

 

[skitch23]

When you say you close the app are you quitting it completely on your phone or just changing to another app on your phone?

When I open the app, I am launching it for the first time. If I already had it open from earlier, I kill it and then reopen it fresh. When closing, I just minimize it and go back to the home screen on my phone. It's still running in the background, just not actively being used. I'm on iOS.