Large Drop in Charge When Parked in the Cold

[PRJIM]

If I might, I would like to add a small point to this interesting discussion. Some of the more savvy electronics experts can also address this if they think it is worthwhile:
Much of this thread speaks as if the battery in the MS is a single entity, however, we all know that it is constituted of many small batteries. It seems to me that the engineers at Tesla have used this to their advantage with algorithms that assure that there is appropriate rotation of load and SOC for each individual element of the system in order to assure that they are achieving maximum usability and life. Does it really make sense to "outthink" this process with crude choices such as not plugging the car in every day?

Several Tesla Roadster owners have been doing this for years- without repercussion. This is not our first time at the rodeo.

 

[JRP3]

Not to mention that all the cells wired in parallel are treated as a single large cell and don't really get individual treatment.

 

[stopcrazypp]

Chevrolet does not share this information with Volt owners. Not sure why Tesla needs to if GM does cannot do it.

There is this for the Volt:
http://gm-volt.com/forum/showthread...ent-system-temperature-band&p=48601#post48601

Granted it's not an official chart, but it was compiled from information made public by GM and GM engineers. We need a similar chart here.

 

[wycolo]

Same situation again today: 4*F '56' miles on speedo. Been plugged in @117v for days straight. Can't get it to resume charging, even by switching to RANGE from STANDARD which worked yesterday. Unplugged & took a spin around the lot in the snow- great fun. Parked & plugged. *Now* it begins a charge 117v 12A, estimate 12+ hours. Go out less than 2 hours later it shows 216 miles on speedo!! As I suggested in yesterday's post in other 'cold' thread, the '56 miles' is BOGUS value derived from deep sleep or ?? Should we call this the 'Memorex Range'? It sure ain't real. Say I want to take a 120 mile round trip to the City. If I guess that the '56' really means '212' then I'm golden. But if it ain't, I'm screwed.

Tell me where to hook up the 440v voltmeter (bluetooth?) then I can reference the nominal SOC. As experience increases relating Voltage to SOC we will be able to do our own correlations, thank you. Any remote sensing voltmeters to recommend? I've got little power-socket 12v DVMs on order to monitor the LV batt; this would nicely flesh out the Control Panel. Can't have too many voltmeters, esp in an electric vehicle.

Workaround would be to go outside in early morning a couple hours before take off & repeat the unplug, turn ON car, move the shifter thru the gears, re-plug, initiate charge routine. I could hit ENVIRONMENT so the car (& batt?) would be warmer, maybe with some REGEN by the time I left. This is doable.
--

 

[tezco]

I'm in and out of my garage frequently and usually have the fob in my pocket. My car, running 4.0, doesn't start a charge if it senses my fob and presents the door handles. It doesn't start a charge if I get in the car with the charge cable still plugged in. It will start a charge if I unplug and replug the cable from the charge port. It will top off every 24 hours if left plugged in (I have a NEMA 14-50 in my garage that routinely delivers 240 volts at 10kW during a charge).

Until this week, that is: on Tuesday I drove about 50 miles and parked the car in my garage mid-afternoon, plugging in for a charge. Looking at my whole-house energy logs I can see that it initiated a charge (but, strangely, only at 5kW instead of the normal 10kW: what's up with that?). On Wednesday I did not drive the car and it did not do a top-off charge at all. I should point out that my car has been doing daily top-off charges since about November 1, on whatever software version came to us before 4.0. Today I went into the garage and saw a Rated range of 230 miles on the center console, which is consistent with about a 5 miles-per-day vampire load. Because I have a long drive later today I did the unplug-replug thing to force a top-off charge. It charged for about 15 minutes at 10kW.

Yesterday's 5kW charge (which lasted almost four hours and included a 10kW spike for a few minutes just before it stopped charging) is the only time I've ever caught the car charging at less than 10kW on my home charging outlet. Weird.

Some owners have reported reduced charging rates if the charging cords aren't completely seated in the charging port.

 

[stevezzzz]

Some owners have reported reduced charging rates if the charging cords aren't completely seated in the charging port.

Ah, thanks for that, tezco.

- - - Updated - - -

Same situation again today: 4*F '56' miles on speedo. Been plugged in @117v for days straight. Can't get it to resume charging, even by switching to RANGE from STANDARD which worked yesterday. Unplugged & took a spin around the lot in the snow- great fun. Parked & plugged. *Now* it begins a charge 117v 12A, estimate 12+ hours. Go out less than 2 hours later it shows 216 miles on speedo!! As I suggested in yesterday's post in other 'cold' thread, the '56 miles' is BOGUS value derived from deep sleep or ??

Just a guess: a temperature-induced voltage drop on the pack is being interpreted as a lower SOC than is 'real'?

 

[VolkerP]

Just a guess: a temperature-induced voltage drop on the pack is being interpreted as a lower SOC than is 'real'?

Yes, sort of. At lower pack temp, much less charge can be extracted from the cells. The car heats up the pack, making consumption values skyrocket for some time, but then a much more realistic range value is displayed. Check the winter driving experiences from Doug_G.

 

[neroden]

Ah, thanks for that, tezco.
Just a guess: a temperature-induced voltage drop on the pack is being interpreted as a lower SOC than is 'real'?

It seems absolutely clear that this is happening. Unfortunately, it's hard to separate this out from the actual loss in charge in the cold, since that seems to be happening too. I wonder if Tesla can come up with a method to report the "frozen pack" situation properly on the screen.

- - - Updated - - -

If there really is an optimum storage SOC which is worth using, I'd think the way to do it would be just tell the car when you'll be requiring it again and at what state of charge, same as what should be in the timed charge. Just include a date as well as a time.

THIS. Compute how long it will take to recharge the battery from "storage level", using some sort of worst-case scenario, and then let the car drop to storage SOC until that many hours before the "scheduled use time". Trickle charge it to maintain storage SOC. This seems relatively straightforward to program, if Tesla employs high-quality programmers. (If not, my consulting rate is $100 an hour. And I'm slow. ;-) )

 

[JRP3]

It seems absolutely clear that this is happening. Unfortunately, it's hard to separate this out from the actual loss in charge in the cold, since that seems to be happening too. I wonder if Tesla can come up with a method to report the "frozen pack" situation properly on the screen.

Should not be that hard, they should have a temperature compensation algorithm to adjust SOC readings in relation to voltage at X degrees.

 

[wycolo]

>>> Just a guess: a temperature-induced voltage drop on the pack is being interpreted as a lower SOC than is 'real'? [stevezzzz]

>> Yes, sort of. At lower pack temp, much less charge can be extracted from the cells. The car heats up the pack, making consumption values skyrocket for some time, but then a much more realistic range value is displayed. Check the winter driving experiences from Doug_G. [VolkerP]

> It seems absolutely clear that this is happening. Unfortunately, it's hard to separate this out from the actual loss in charge in the cold, since that seems to be happening too. I wonder if Tesla can come up with a method to report the "frozen pack" situation properly on the screen. [neroden]

Strange that the Roadster sits thru this cold period exhibiting the usual mileage estimates, no weirdness. We had a 3-day warm spell into the 40s daytime & about +15*F at night, then 2 nights of -10*F (no more than 10*F in the day), then 0*F last night (18*F daytime peak). Expecting 10*F tonight & warming the next few days.

Good to know my S hibernates completely during serious cold. Makes no attempt to maintain battery temp until it is unplugged & then plugged back in. I'm curious to retrieve kwh usage records during next cold snap. With NO tech pkg this car might not be able to.
--

 

[vfx]

EXCLUSIVE Interview With Elon Musk on How Model S Manages Cold Weather