TMC is an independent, primarily volunteer organization that relies on ad revenue to cover its operating costs. Please consider whitelisting TMC on your ad blocker or making a Paypal contribution here: paypal.me/SupportTMC

Cold Temperatures - how much does it effect charge times?

Discussion in 'Model S: Battery & Charging' started by wayner, Dec 9, 2015.

  1. wayner

    wayner Active Member

    Joined:
    Oct 29, 2014
    Messages:
    1,586
    Location:
    Toronto
    I am trying to write some code to determine when to start charging my car given a set end charging time - in my case 7am. One of the variables that can effect charging rate is temperature. In moderate temperatures my charge rate is roughly 0.33SoC/minute - so in 60 minutes I can charge my car from 70% SoC to 90% SoC - this is my typical day. Therefore I would want to start charging my car at 6am. But what if it is really cold - do I want to start earlier?

    Does anyone have a good feel for how much charging is slowed down by cold temperatures?

    Is it a (roughly) linear process?
    Or is it just a case of warming the batteries for X minutes at the start of charging?
    Or is it both?
    At what temperature does this start to kick in ? 10C? 0C?

    So far this winter I haven't noticed too much of a slow down but we haven't had too much cold weather yet - this morning my garage was 10C.
     
  2. arg

    arg Member

    Joined:
    Aug 22, 2012
    Messages:
    740
    Location:
    Cambridge, UK
    This is a mighty difficult problem! You really need to know the temperature of the battery, which has huge thermal mass and so can be quite different from ambient temp, and I believe there's no way to extract it from the car.

    My understanding of the behaviour is:

    - Above 10C battery temp, there is no effect at all: charging will proceed at full speed, the battery heater won't be used.
    - Below 10C and above another critical temperature (which might be 0C, not sure) maximum charging rate is reduced and the battery heater runs if the available power exceeds the permissible charge rate. At the higher end of this temperature range, the power limit is greater than is available through AC charging, so it only begins to bite at a slightly lower temperature.
    - Below the critical temperature, no charging is possible at all and the battery heater (only) runs until the temperature is high enough to start charging.
    - When running, the battery heater uses 6 or 7kW (I believe it is also capable of running at part power rather than just on/off). This power is taken from the output of the charger, so a single-charger equipped car is limited to a total of 10kW (EU:11kW) split between heating and charging, even if using an HPWC and more AC power is theoretically available.

    Given the lack of access to true battery temperature data, you probably need a more interactive algorithm than just starting the charge at a fixed time.

    I think you need something like:

    - Derive a formula to estimate the total time for a given temperature and starting SoC, which will have three phases: heat-but-no-charge, heating-with-slow-charging, full-speed charging.
    - From worst-case temperature assumptions (perhaps using weather forecast data for the 24 hours around the current time, though that may be pesimistic since you have a garage),
    determine the earliest start time that might be required.
    - Start a charge at that time, and monitor the current drawn from the wall vs. the charge going into the battery to see if the heater is running and/or any charging is happening. From that you can derive a more accurate estimate of the battery temperature, re-run the calculation, and if there's now time to spare, stop charging.
    - If you've paused charging, wake up again every now and then and update the estimated battery temp based on the previous estimate and the ambient temp reported by the car.
    - Start charging again when needed, based on the temperature estimate and the calculated charge duration at that temperature.
     
  3. mknox

    mknox Well-Known Member

    Joined:
    Aug 7, 2012
    Messages:
    8,556
    Location:
    Toronto, ON
    Well, I haven't studied it as closely as you seem to be doing, but I can give you some anecdotal data: I'm going in to my 4th winter with my car and really haven't noticed any difference in charge times summer over winter. When I was charging at 16 amps / 240 volts at home I would see it takes about 2 hours for every 10% gain in SOC and that was the same summer or winter. When I charge at 80 amps / 208 volts at work, I count on 1/2 hour for every 10% SOC gain and that too doesn't seem to change over the seasons.

    Now I'm just using rough approximations to come up with these "rule of thumb" numbers, and maybe it does vary by minutes as the weather changes, but that's too far down in the noise for me to worry about. I just look at how many "10 percents" I need and set my start timer accordingly so that it ends roughly when I want to depart.
     
  4. green1

    green1 Active Member

    Joined:
    Mar 25, 2014
    Messages:
    4,115
    Location:
    Calgary, Alberta, Canada
    There are a couple things to take in to account. If the battery is cold it will need to be heated to charge, if you have lots of current available, it will affect speed very little, but if you have less current it would affect it mroe (eg. charging from a 110v outlet, the pack heater can draw more than the outlet can provide, but a supercharger would never be able to notice the tiny draw of the heater)

    One other member on here somewhere wrote some interesting code that basically asked the car to decide. it started a charge, waited 5-10 mins for the completion estimate to stabalize, then stopped charging and waited until the appropriate time using that estimate to start again. That's probably your safest bet.
     
  5. mknox

    mknox Well-Known Member

    Joined:
    Aug 7, 2012
    Messages:
    8,556
    Location:
    Toronto, ON
    While no doubt true, I have not seen it impact my charging times materially. I do park in an attached garage, but it does get way below freezing out there. The only thing I've noticed that does quite obviously affect the charging time (at least the estimate the car provides) is if I'm sitting in the car with the heat on while it's charging. Even on my 80 amp unit at work it will bump the time to completion number up a fair bit.
     
  6. green1

    green1 Active Member

    Joined:
    Mar 25, 2014
    Messages:
    4,115
    Location:
    Calgary, Alberta, Canada
    I think the question is what the estimate takes in to account...
     
  7. arg

    arg Member

    Joined:
    Aug 22, 2012
    Messages:
    740
    Location:
    Cambridge, UK
    But do you think your battery gets way below freezing on those days? I seem to recall you have a long commute, so presumably you arrive home with the battery quite warm (do you ever have a regen limit still there when you arrive home?). Parked in an enclosed garage, it may be that your battery never gets cold enough before you start charging for the temperature to have a major effect.

    The case that's reputedly a problem is where you've cold-soaked the battery for some time (particularly outdoors where the wind can get under it) and then it takes some time to warm back up before charging can start.

    Admittedly, I have little relevant experience as it never gets that cold here...
     
  8. wayner

    wayner Active Member

    Joined:
    Oct 29, 2014
    Messages:
    1,586
    Location:
    Toronto
    #8 wayner, Dec 9, 2015
    Last edited: Dec 9, 2015
    The cold soaked battery with limited regen is the reason why I want to get a little more sophisticated with my charging time and end the charge as closely as possible to (1) The end of my off-peak electricity period, and (2) My departure from home. The idea was inspired by Doug_G's blog entry from about a year ago: http://www.teslamotorsclub.com/entry.php/194-Cold-Weather-Driving

    But maybe I am over thinking this and I should just not worry about temperatures for now, or only adjust when it is below 0C by increasing my estimated charge time by 10%.

    I could be more accurate by doing a shorter period test charge but that leads to more futzing around and more that could go wrong. For now I will just use my own estimate of charge rate which is roughly 0.33 SoC/minute.

    By the way - I use a 20kW HPWC - 240V and 80A. My daily commute is roughly 25km each way and my car spends the night in a garage attached to the house that is typically about 5 degrees above the outdoor temperature.
     
  9. AWDtsla

    AWDtsla Active Member

    Joined:
    Mar 3, 2013
    Messages:
    2,134
    Location:
    NE
    That's 20kW.
     
  10. wraithnot

    wraithnot Model S VIN #5785

    Joined:
    Dec 16, 2012
    Messages:
    475
    Location:
    SF Bay Area
    On a road trip earlier this year I learned that a cold battery has a pretty significant effect on supercharging speed. We pulled into the Rapid City supercharger with about 5% SOC, charged at over 100 kW until the car had enough range that I wasn't worried about making it back to the supercharger in the morning, and then we drove to our hotel about 3/4 of mile away and parked the car outside for the night. The temperature was below freezing that night (something like 28 F / -2 C) and when I drove back to the exact same supercharger plug, the car started out at a pathetic 30 kW. It peaked at 50 kW before the charge taper brought the charge rate down again. In retrospect, we would have been much better off staying at the supercharger until we had enough range to get to the next supercharger since supercharging on a battery nice and warm from driving is so much faster than supercharging on a cold soaked battery in the morning.
     
  11. wayner

    wayner Active Member

    Joined:
    Oct 29, 2014
    Messages:
    1,586
    Location:
    Toronto
    D'oh - you're right! I will edit to correct.
     
  12. green1

    green1 Active Member

    Joined:
    Mar 25, 2014
    Messages:
    4,115
    Location:
    Calgary, Alberta, Canada
    I have found that even a 2 hour charge at 10kW with me leaving immediately upon completion, is not enough to fully remove regen limits, and that's at about the freezing mark, as we drop in to actual winter I expect this to get even worse.
     
  13. mknox

    mknox Well-Known Member

    Joined:
    Aug 7, 2012
    Messages:
    8,556
    Location:
    Toronto, ON
    Yes, I do have an approx. 150 km (90 mile) round-trip commute. I don't recall ever having regen limits upon arriving home, but still, the car will sit for 12 to 14 hours before I use it again. Over the previous 3 winters when I was charging at 16 amps / 240 volts, I would try to time it to complete just before I left and would run the HVAC on shore power for 1/2 hour before setting out. Even then, I would have regen limits. Before I had charging at work, the car would sit outside in the extreme cold and I would have regen completely disabled when I left work (that is a scary driving experience when you feel it the first time!). It may take 5 or 10 miles until the regen even appears at all, then until most of the way home for the limits to clear completely. I do not use "Range Mode" in the winter (doesn't seem to help me at all).

    I installed 80 amp charging stations at work a bit over a year ago, and will charge at 80 amps (I have dual chargers) in the afternoon before departing. Even in last winter's cold, I don't recall it taking any longer to gain the same range as it did in the summer, and based on how it would completely cold-soak the winter before, I assume some pack heating must be going on. As I mentioned previously, there may very well be a difference, but it would be small enough so as not to affect my "rule of thumb" charging approximations (1/2 hour for each 10 % SOC at work).

    This year I have upgraded my home charging to 40 amps, but it has been so mild this year (it was 50 degrees F at 5:30 pm yesterday) that I don't really have any new data on cold weather charging stats.
     
  14. wayner

    wayner Active Member

    Joined:
    Oct 29, 2014
    Messages:
    1,586
    Location:
    Toronto
    Most Californians would probably consider this to be cold weather charging :smile:
     
  15. arg

    arg Member

    Joined:
    Aug 22, 2012
    Messages:
    740
    Location:
    Cambridge, UK
    I don't think the car tries to get the battery warm enough to fully remove regen limits - the first stage of regen limiting is a limit well above normal AC charging rates, so at that point it doesn't need to run the heater; after that it's only the waste heat from the charging itself - a few percent of that 10kW is only a few hundreds of watts.
     
  16. RiverBrick

    RiverBrick Active Member

    Joined:
    Mar 23, 2014
    Messages:
    1,369
    Location:
    Québec
    At 14 kW all regen limits are gone within 2 hours, even when charging outside after the car has been cold-soaked for 24 hours at -20C. It is disappointing that you can't clear regen limits at 10 kW. As another poster stated, it's because the car isn't trying and the heater has shut off. Since the L has optional supplemental heating for drag-strip performance, we should have it for cold weather.

    After a -20C cold-soak you have zero regen and power output is limited to 80 kW. The recharge will start at ~6kW and stay there for ~10 minutes with all energy from the charging station going to towards heating. After that it ramps up to 14 kW within 15-20 minutes of plugging in. At that point I don't remember if all the energy goes to the battery or if a significant portion is still being siphoned off for heating purposes.

    We're running at least 15C warmer this December than last, so I may not have a chance to retest soon and Tesla Motors is dodging a major bullet since Kingston to Toronto will trap many new 70D owners when it's -20C with a 20 km/hr headwind.
     
  17. mknox

    mknox Well-Known Member

    Joined:
    Aug 7, 2012
    Messages:
    8,556
    Location:
    Toronto, ON
    This doesn't seem to bother me as much as others. The car does have pretty good hydraulic brakes, after all, and it seems like a waste of energy if all we're doing is trying to eliminate re-gen limits. I'd rather save my money (on electricity costs). Before anyone jumps on me for the "trivial" amount of money, remember we are trying to convince people to change incandescent light bulbs for CFL/LED which is even more trivial by comparison.

    While it hasn't turned out to be a problem for me, I do kinda like reduced re-gen when the roads are slippery to reduce the risk of spinning out if the rear wheels "bite" too much slowing down. The cold temps sort of does this automatically for me.
     
  18. Rocky_H

    Rocky_H Member

    Joined:
    Feb 19, 2015
    Messages:
    882
    Location:
    Boise, ID
    I concur with @mknox on this. I would be a bit bothered with regen completely gone, but it's no big deal for it being partially reduced.
     
  19. mknox

    mknox Well-Known Member

    Joined:
    Aug 7, 2012
    Messages:
    8,556
    Location:
    Toronto, ON
    When it gets really cold, I have lost regen completely. That is a crazy, spooky experience the first time. You take your foot off the "gas" and it almost feels like the car is speeding up because it glides along so effortlessly. Fortunately a rare occurrence.
     
  20. RiverBrick

    RiverBrick Active Member

    Joined:
    Mar 23, 2014
    Messages:
    1,369
    Location:
    Québec
    It's more about range for me since last Winter I often had to drive in -20C temperatures to areas with neither Super nor destination charging. I'd rather warm the battery as much as possible while on shore power, to reduce draining the battery to do it later.


    At least there's a prominent warning when there's no regeneration at all. However, unless it has been fixed in 7.0, there's absolutely no warning when regeneration has been cut to miniscule amount above zero, so that's when it's most dangerous.
     

Share This Page