MS Estimated Charge Times

[MS-Learner]

I’ve had my 2015 MS (P85D), my first EV, for well over a year. Overall it has been brilliant and I wouldn’t want to go back to an ICE. Over this time I have experienced a constant flow of uncertainty both with respect to my MS and EVs generally and it goes on. I am hoping that the broad knowledge of the TMC can help me out.

Question:

Why is the ACTUAL charge time always way longer than the MS ESTIMATED charge time? Or is it just me?

An extreme example of this was at Birmingham St Andrews V3 Superchargers:

• I was the only car there
• Cold car and battery not sufficiently conditioned
• Initial SOC 14%
• Initial charge rate 2kW (this is not a typing error just 2kW!)
• Estimated charge time 70 minutes to 100%

Clearly charging conditions were far from ideal but surely the system has all the data/information it needs to provide a half credible time. It doesn’t take much to work out that 2kW for 1 hour (60 mins) = 2kWh will not result in 100% charge in 70 mins.

Are others experiencing not just long charge times (pain in itself) but gross errors in charge time estimates?

 

[Rocky_H]

Clearly charging conditions were far from ideal but surely the system has all the data/information it needs to provide a half credible time. It doesn’t take much to work out that 2kW for 1 hour (60 mins) = 2kWh will not result in 100% charge in 70 mins.

You're not doing "half credible" calculation. It is NOT going stay at only 2kW for the next hour, as you have calculated. The car knows that. It is diverting most of the Supercharger's power other than that 2kW into the battery heating system, and the car knows that within the next 5-10 minutes, it is going to be increasing the charging power, so somewhere around 70 minutes can very realistically be close.
Granted, I have seen people gripe sometimes of the estimates being off in both directions, so there's some ongoing variation that it doesn't always model the right way.

Quote: "Why is the ACTUAL charge time always way longer than the MS ESTIMATED charge time? Or is it just me?"

That is just you then. As I mentioned, I've seen people being frustrated with the car's estimate being too long, like 40 minutes, so they go off to get some lunch, and then the car gets done faster than the estimate, but they weren't checking or seeing the push notifications on their phone, so they were angry that they got some idle fees from the car being done sooner than it said.

(SIDE NOTE) Dang it! With the new formatting after the big forum update, I don't have a way to add another quote when editing a posted comment. I used to be able to quote it, and then I could copy and paste the HTML from that into my edited comment to make the quote look right.

 

[airborne spoon]

with the battery that cold you can expect your SC rate to be super slow for like 5-10 min. then it'll ramp up to proper speeds so at like 18% you should be around 73kW on an 85 pack.

 

[aerodyne]

@MS-Learner ....search Chargegate on this forum.

The 85D charging rate has been "Nerfed" by updates in 2019.

It will take 20% longer than it used to.

For me, 30-80% is 45 minutes.

Thankfully, there are tricks you can use to reduce the impact, namely drive down to 20% or less, leave at 70%, nav to SuC, etc.

Just enjoy your car. Don't let FUD get you.

 

[MS-Learner]

Thanks for all responses.

Apologies first to Rocky_H and others. My clumpsy calculation diverted from my point about estimated charge times. I have been logging recent charging events at 30 minute intervals and these results I think give a clearer picture which based on responses so far, suggest it might be just me:

20 Feb B'ham V3 (250kW) - Charge to 100% 1 car (Just me and 15 free stalls)
Stall 3B, 13 degs C
Charge time Estimate - 70mins to 100%
RESULTS Charge Estimated
Time, SOC% Chgrt Time +% +Chg Time left
16.35, 18% 61kW (0m) +0% +0kWh 70m
17.05, 50% 40kW (30m) +32% +24kWh 55m
17.35, 72% 29kW (60m) +54% +41kWh 40m CHARGING STOPPED.

So, after 60 minutes charging from 18% SOC to 72% SOC (Adding 54% and 41kWh) there remained an estimated 40 minutes to 100%. From my many charges adding 40 minutes to the already 60 minutes is generous giving a total actual time of 100 minutes compared to the 70 minute estimate. For me this is a normal charge at our relatively low ambient temperatures where the charge rate starts high and reduces as the SOC increases. (Battery Conditioned)

This next example is typical of when the battery is not fully conditioned with the charge rate starting low (22kW) and grandually increasing to a maximum (32kW) and then reducing as the SOC increases. Again, Estimated charge time 70 minutes (Never had an estimate longer than that) whilst the 105 minutes plus the estimated time left of 30 minutes equates to 135 minutes, almost double the estimate.

5 Mar B'ham V3 (250kW) - Charge to 100% 3 car (Me plus 2 and 13 free stalls)
Stall 3D, 7 degs C
(70mins to 100%)
RESULTS Charge Estimated
Time, SOC% Chgrt Time +% +Chg Time left
16.20, 23% 22kW (0m) +0% +0kWh 70m
16.50, 44% 32kW (30m) +21% +16kWh 55m
17.20, 63% 26kW (60m) +40% +30kWh 50m
17.50, 78% 21kW (90m) +55% +41kWh 35m
18.10, 85% 18kW (105m) +62% +46kWh 30m CHARGING STOPPED

My enquiry is not the actual time taken to charge but the estimated time error to charge.

And thanks for the recommendations for reducing actual charge time. Other than driving and charging in Spain (charging nor charge time was ever a problem) then driving the car seemed to be the most effective.

Also, my estimated charge time has never been less than the actual charge time. I wish.

As a closing note, I'm not really moaning I'm just leaning. I absolutly love this car.

Trust this helps.

 

[Greg29]

With my '16 S75 I have never had the car charge faster than predicted, always slower. I don't pay any attention to the predictions any more.

 

[aerodyne]

Good info is the enemy of FUD.

I have a CAN bus reader, SMT, and Teslafi.

Every day I check my SoC, Rated Range, cell imbalance, pack temps, and if driving, my max rate of DC charge which increases with high cell temps and lower SoC.

The cost for these items is less than $100 USD. Teslafi has a free two week trial...

 

[Rocky_H]

My enquiry is not the actual time taken to charge but the estimated time error to charge.

Oh, well, that's because you're doing this:

CHARGING STOPPED

That last little bit where it's at 99% or 100% and finishing up the balancing and such at the very top end can vary by plus or minus an hour. You are really going to get some wild mismatch if you're doing that. It can estimate reasonably well if you're filling up to 80 or 90% or something that fits in with the regular recharging process, but all that finishing up until "CHARGING STOPPED" is hard to predict ahead of time what it will need or how long it will take at the very end. As people have mentioned, there is no good reason to do that at a Supercharger. Even if you need about all the range you can get for a trip, (and I've had to do that), it's just getting ridiculous to keep waiting above about 97-98% or so at a Supercharger. It's just horribly slow.

 

[aerodyne]

As a fellow 85 pack owner, there is not much you can do about the following. You either have it or you don't:

Range gate
Battery gate.


Charge gate on trips using SuC can be mitigated by warming the pack, going to as low SoC as you dare, and leaving as soon as possible, 70-80%. "Nothing wrong" with going below 20% per Elon.

Regen gate can be mitigated by charging and or preconditioning the pack. Anything over 66 deg F and you should have full regen.

Pumpgate is avoided by not charging at home more than 78% SoC. Else the pumps run at 100% and will not stop, and the car can't sleep.

With this is mind there is very little else to worry about. Set energy display to percent, and keep the car between 30-70 and 50-62% for storage for best pack lifer per Elon.

If you are having pack problems near end of warranty, document with SMT or other tools, results of energy tests, and try to get the pack replaced.

You seem to be a quick "learner". Best of luck to you and your car experience.

Enclosed is SMT screenshot today

 

[MS-Learner]

With my '16 S75 I have never had the car charge faster than predicted, always slower. I don't pay any attention to the predictions any more.

Well Greg29 that's about where I am. My current quest is what alternative estimator/method to use though I suspect it is less of a problem on longer runs once the car has warmed up.

 

[aerodyne]

Oh, well, that's because you're doing this:


That last little bit where it's at 99% or 100% and finishing up the balancing and such at the very top end can vary by plus or minus an hour. You are really going to get some wild mismatch if you're doing that. It can estimate reasonably well if you're filling up to 80 or 90% or something that fits in with the regular recharging process, but all that finishing up until "CHARGING STOPPED" is hard to predict ahead of time what it will need or how long it will take at the very end. As people have mentioned, there is no good reason to do that at a Supercharger. Even if you need about all the range you can get for a trip, (and I've had to do that), it's just getting ridiculous to keep waiting above about 97-98% or so at a Supercharger. It's just horribly slow.

If you really need the last 5% of range, you won't get it because the cooling system will be running full bore and using up all the last few percent of energy as you drive off.

If short on range, charge lower, drive slower. The car will take care of you, and warn you if you cannot make the next charging station.

There is has never been, and certainly not now, a need to charge above 93% for balancing or calibration. You are just stressing the car for no good reason and wasting time.

 

[MS-Learner]

Oh, well, that's because you're doing this:


That last little bit where it's at 99% or 100% and finishing up the balancing and such at the very top end can vary by plus or minus an hour. You are really going to get some wild mismatch if you're doing that. It can estimate reasonably well if you're filling up to 80 or 90% or something that fits in with the regular recharging process, but all that finishing up until "CHARGING STOPPED" is hard to predict ahead of time what it will need or how long it will take at the very end. As people have mentioned, there is no good reason to do that at a Supercharger. Even if you need about all the range you can get for a trip, (and I've had to do that), it's just getting ridiculous to keep waiting above about 97-98% or so at a Supercharger. It's just horribly slow.

Thanks Rocky_H, I agree. 'In real life' I generally stop the charge at around 80% ish as you suggest. The 100% was specifically to assess the performance of the charge time estimator since, as you say, it can take forever.

 

[aerodyne]

Well Greg29 that's about where I am. My current quest is what alternative estimator/method to use though I suspect it is less of a problem on longer runs once the car has warmed up.

I suggest time the charge and get your own data. With pack temps above 25deg C, it is about 45 min from 30-80%.

Higher temps up to 40c, and lower SoC to say 20%, might shave off 10 min at most.

If you have a regular commute, you will be able to "dial in" for best results. If on occasional trips, just plan for more time.

A Better Route Planner is a free app that might help in this regard.

 

[MS-Learner]

If you really need the last 5% of range, you won't get it because the cooling system will be running full bore and using up all the last few percent of energy as you drive off.

If short on range, charge lower, drive slower. The car will take care of you, and warn you if you cannot make the next charging station.

There is has never been, and certainly not now, a need to charge above 93% for balancing or calibration. You are just stressing the car for no good reason and wasting time.

Thanks aerodyne, good down to earth nitty gritty guidance.

From your earlier post what is Pumpgate? And when you refer to 66 deg F, is this ambient temperature? In July here that is our summer average (42F in Jan). I have been searching for something integral to the MS that would give me an indication of the battery condition pre charging. A stab in the dark is monitoring my Wh/m on the way to the charger. 450Wh/m is bad, 320Wh/m is good.

 

[aerodyne]

Thanks aerodyne, good down to earth nitty gritty guidance.

From your earlier post what is Pumpgate? And when you refer to 66 deg F, is this ambient temperature? In July here that is our summer average (42F in Jan). I have been searching for something integral to the MS that would give me an indication of the battery condition pre charging. A stab in the dark is monitoring my Wh/m on the way to the charger. 450Wh/m is bad, 320Wh/m is good.

Pumpgate is the recent effect of last year's software update that turns on the the 4 pumps to 100% and keep them running even after the car is done charging, until the SoC is below 75%

This keeps the car from sleeping plus an additional 200-400w if vampire drain.

The temp I refer to is the average temp of the cells in the pack. Outside temps are similar if the car is parked for hours in a sheltered location.

SMT and other CAN bus readers will tell you this, see my screenshot above.

Say overnight, your cell temps or garage get to about 10deg C. You want about 18deg C to avoid regen loss.

If you charge up about 30%, but not more than 77% SoC, approx 4 hours on level 2, you will warm up the pack sufficiently with the pumps at 28-33% or less.

A warm pack is more efficient. Since you have to charge anyway, this a rare, "better, faster, cheaper" approach.

SMT will also show you if nav to the SuC is turning on the battery heater. Ideally, you want to avoid that and have the pack warm as a natural by product of driving.

 

[MS-Learner]

Pumpgate is the recent effect of last year's software update that turns on the the 4 pumps to 100% and keep them running even after the car is done charging, until the SoC is below 75%

This keeps the car from sleeping plus an additional 200-400w if vampire drain.

The temp I refer to is the average temp of the cells in the pack. Outside temps are similar if the car is parked for hours in a sheltered location.

SMT and other CAN bus readers will tell you this, see my screenshot above.

Say overnight, your cell temps or garage get to about 10deg C. You want about 18deg C to avoid regen loss.

If you charge up about 30%, but not more than 77% SoC, approx 4 hours on level 2, you will warm up the pack sufficiently with the pumps at 28-33% or less.

A warm pack is more efficient. Since you have to charge anyway, this a rare, "better, faster, cheaper" approach.

SMT will also show you if nav to the SuC is turning on the battery heater. Ideally, you want to avoid that and have the pack warm as a natural by product of driving.

This bus reader sounds like a useful bit of kit. I've found some info on the CAN bus reader. Should keep me out of trouble for a while. Thanks.

 

[aerodyne]

This bus reader sounds like a useful bit of kit. I've found some info on the CAN bus reader. Should keep me out of trouble for a while. Thanks.

It is. On trips, I put my phone holder under the visor, and it is like having a second IC with performance data I can customize.

My 0-100 kmh time is right at 4.2 seconds.

There is an extensive SMT thread on this board. If you get it, note instl is easy and

you do not need to cut holes, as shown above...

 

[gln103]

@MS-Learner, I have always had similar experiences with my 2013 MS 85 as you. I've owned the MS for 3 years, and when supercharging on long trips (i.e. 500+ miles), the supercharging stops were ALWAYS longer than the "charging time remaining" estimate on the screen. Typically, supercharging stops were about 2x longer. If the car estimated 30 minutes of charging, it would typically be 1 hour. Plus, keep in mind, these were in ideal temperatures (~60-80F).

Last week, when it was ~15F here in PA, I supercharged. Right after plugging in, the center screen said: 33 miles of range, charge rate = 5 kW, and "charging estimated time" to 70% = 30 minutes! It ended up taking about 1.5 hours. I get it. It charges slower when cold.

My problem is, with 10 years of data under their belt, Tesla be better at estimating actual charge time than being off by 2-3x. I'm not concerned about 5-10 minutes here or there, but my family routinely takes a ~10 hour trip. The extra charge time, above and beyond the original estimates of ~2 hours of charging, can be an additional 1.5-2.5 hours, which is pretty unacceptable.

 

[Yurand]

My problem is, with 10 years of data under their belt, Tesla be better at estimating actual charge time than being off by 2-3x. I'm not concerned about 5-10 minutes here or there, but my family routinely takes a ~10 hour trip. The extra charge time, above and beyond the original estimates of ~2 hours of charging, can be an additional 1.5-2.5 hours, which is pretty unacceptable.

Tesla knows exactly how long the charge session will take, but they choose to play dumb.

Supercharging Speeds: Who, what, where, why, and when

Supercharging speeds have improved dramatically over the past decade. The graph below provides some perspective. In order of descending charge rate: 2018 Model 3 LR (Blue), 2015 Model S 85 kWh (Black), 2012 Model S 85 kWh (Red)

teslamotorsclub.com

 

[aerodyne]

Tesla knows exactly how long the charge session will take, but they choose to play dumb.

Supercharging Speeds: Who, what, where, why, and when

Supercharging speeds have improved dramatically over the past decade. The graph below provides some perspective. In order of descending charge rate: 2018 Model 3 LR (Blue), 2015 Model S 85 kWh (Black), 2012 Model S 85 kWh (Red)

teslamotorsclub.com

Exactly! Why would they revise estimated charge times? That would be full evidence of charge gate.

Also annoying is how they round up the SoC. Says 70% as soon as the BMS goes from 69.5 to 69.6%

I use SMT to see the max charge rates before I plug in. I seem to have best results at pack temps of 95 to 105 Deg F, and at a 250kw stall. Sometimes I see high 80's to low 90's