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

Charging in High Temperature environments 13A vs 32A vs 70A: cost of air conditioning

Discussion in 'Charging Standards and Infrastructure' started by markwj, May 23, 2011.

  1. markwj

    markwj Moderator, Asia Pacific

    Joined:
    Apr 10, 2011
    Messages:
    3,656
    Location:
    Hong Kong
    Tetsous raised a question to me at a recent meet-up. He charges at 13Amp 240volt and is seeing dramatically more energy being used to charge the car than being consumed by the car on the road (according to the info the car provides). He asked me if I was seeing the same as I charge at 70Amp 240volt. The suspicion is that due to the high temperature (around 30celcius +/-5celcius), the air conditioning used during charging to cool the battery pack is consuming perhaps as much power as the car itself (doubling the cost of a charge).

    I got three data sets. The outside temperature was not the same on the three days, but was dropping over the week (perhaps 28celcius at 70A down to 24celcius at 13A) - so that should favor the 32A and 13A charges.

    All charges were in standard mode.

    First day: I drove about 44km on my commute. Car says I used 7.4kWh (rounded to .1 kWh). Recharge was 50minutes at 70Amp, using 8kWh (rounded to 1 kWh). P.S. This puts my commute at about HK$4 each way (about 1/4 the Prius I replaced, and 1/10 the Land Rover the Prius replaced).

    Second day: 10.0kWh usage (commute plus a couple of school runs) took almost 2 hours and 12kWh charge at 32Amp.

    Third day: 9.2kWh usage. Charging at 13amp (using HPC dialed-down by the car) needed 4hours 25minutes and 15kWh energy.

    I'm not sure if the net energy usage figures include regen (I assume regen has already been taken away from the figure). For example, for day 3 the car showed 9.2kWh usage and 1.9kWh regenerated. On average, the regen numbers are about 20% of the 'net energy used' numbers (or about 17% if net energy used already includes regen).

    IF (big if) the aircon used about 1kWh, the numbers would work out. The 70A charge took 50 minutes (and the diff is about 600Wh), the 32A charge took 2 hours (and the diff is about 2kWh), and the 13A charge took 4 1/2 hours (and the diff is about 5.8kWh).

    The power to run the aircon doesn't seem to affect charge time.

    Does this make sense? At 13Amp, we seem to be paying about 50% extra for the charge.
     
  2. markwj

    markwj Moderator, Asia Pacific

    Joined:
    Apr 10, 2011
    Messages:
    3,656
    Location:
    Hong Kong
    On a related note, what would the impact of the aircon be on charge time?

    13A @ 240v is 3.1kW
    32A @ 240v is 7.7kW
    70A @ 240v is 16.8kW

    If the aircon is 1kW, presumably that would use 1/3rd of the available 3.1kW @13A, and extend charging time appropriately (but have negligible impact at 70A)?

    I have the logs for these, if it helps...
     
  3. tetsuos

    tetsuos HK #488 (2.5)

    Joined:
    Apr 10, 2011
    Messages:
    32
    Location:
    Hong Kong
    As far as I can tell, Net Energy Used indicates the reduction in available power in the battery pack (i.e. gross energy used - regen gained = net). You can see this in action if you watch the trip computer while going down a long hill with regen (it helps to reset the trip counter so you can see Wh instead of kWh).

    I am still waiting on some feedback from Tesla Motors on their own experience.
     
  4. markwj

    markwj Moderator, Asia Pacific

    Joined:
    Apr 10, 2011
    Messages:
    3,656
    Location:
    Hong Kong
    Tesla Hong Kong started in the autumn last year, right? I think they also charge at 70A at those HPCs in Hopewell.

    As the summer heat up, I think we're all in for a learning experience ;-)

    At least the AirCon seems to be holding up. Even at 28celcius (a week or so ago) I've only got it on the first click to blue and it is still damn cold.

    P.S. Really glad I got that HPC. Noisy fans and AirCon on run for an hour or so at night. 32A doesn't seem too bad though, and easy to install.
     
  5. tetsuos

    tetsuos HK #488 (2.5)

    Joined:
    Apr 10, 2011
    Messages:
    32
    Location:
    Hong Kong
    Keep driving... I have already experienced the aircon override during "spirited driving." Performance Mode is now almost always RED for me, especially after any 0-100 kph "performance verifications." YMMV
     
  6. Sparrow

    Sparrow S105/ Roadster 189

    Joined:
    Dec 14, 2010
    Messages:
    566
    Location:
    Marietta, GA
    I look at the air conditioning running during charging as a very positive thing to prolong the life of my pack and therefore I don't care that much about the energy it uses during charging. My goal is to make sure that charging goes on at least long enough for the car to decide it no longer needs to cool the pack any further during the charge cycle. So under normal circumstances where I am charging overnight, I like to extend the charging time somewhat to insure the batteries cool down as far as need be. Also, once my car is safely in my garage, if I notice the air conditioning is cooling the batteries and I have the time, I'll leave my car on while parked in the garage to keep the cooling process going. Anything to insure the batteries are happy.
     
  7. kgb

    kgb Member

    Joined:
    Sep 22, 2009
    Messages:
    605
    Location:
    Houston, TX
    I find Performance most of the time when I am driving in performance mode. The VDS says outside temperature is >85F, but that is probably off by 10F. I assumed that the battery/PEM/engine was overheating, but when I switch the VDS to the temperature monitor, they are all in the BLUE range.

    I'm not sure why my car stays in performance limited mode so much.

    P.S. My service tech says that the "outdoor temp" is measured close to the pavement, so that is why it seems to read higher than air temperature.
     
  8. markwj

    markwj Moderator, Asia Pacific

    Joined:
    Apr 10, 2011
    Messages:
    3,656
    Location:
    Hong Kong
    If 240v @13A is 3.1kW and the AirCon really takes 1/3rd, the situation at 110v must be appalling. Charge duration 3 times the already long normal.

    I agree that AirCon while charging is a good thing. At 70A it has minimal impact on charge duration or energy used. But, the issue is at 13A (which is why tetsous raised it).
     
  9. richkae

    richkae VIN587

    Joined:
    Jan 15, 2008
    Messages:
    1,917
  10. markwj

    markwj Moderator, Asia Pacific

    Joined:
    Apr 10, 2011
    Messages:
    3,656
    Location:
    Hong Kong
    Richkae,

    Thanks for the link. I hadn't seen that before, and it makes fascinating reading.

    Tom says "all charging was done overnight in cool weather" and:

    Charging in a hot environment definitely changes energy consumption during charging because the fan and A/C will kick on to cool the battery pack. It's harder to control for ambient temperature across multiple charges, but it would be interesting to collect data and see how things change. I would not be surprised to see a significant penalty for charging at higher current if that pushes the temperatures high enough to require the A/C during the charge.

    I guess we need to see how much current the aircon draws and that should give us an idea of its impact.
     
  11. scott451

    scott451 KWH-PWR#1349Sprt,S Sig#96

    Joined:
    Apr 2, 2009
    Messages:
    251
    Location:
    Palo Alto
    #11 scott451, May 24, 2011
    Last edited: May 24, 2011
    Mark,
    Can you and tetsuos send me a log file and I'll take a look?

    My theory is that the higher ambient temp and lower charge current causes charging to take longer and the AC to run more. At higher current, the overall battery temperature is higher, but it finishes sooner so less AC is used. Send me the logs and I'll run the plots. It should be easy to tell. Please give me the specific day/time the charge was done. Take a look at the charge plot in this post, notice how the battery starts out very hot and is slowly cooled to an acceptable temp range.
     
  12. markwj

    markwj Moderator, Asia Pacific

    Joined:
    Apr 10, 2011
    Messages:
    3,656
    Location:
    Hong Kong
    Scott, I'll send you mine tonight. I've got the data points for 13A, 32A and 70A (albeit at slightly different ambient temperatures).
     
  13. tomsax

    tomsax Member

    Joined:
    Jun 19, 2008
    Messages:
    867
    Location:
    Sammamish, WA
    I think the right thing to measure is Wh per ideal mile (or SOC %) gained. That's probably the best way to compare charging efficiency in different scenarios.

    I would also be interested to see your log file to do some analysis.
     
  14. Doug_G

    Doug_G Lead Moderator

    Joined:
    Apr 2, 2010
    Messages:
    15,850
    Location:
    Ottawa, Canada
    To a first approximation, it shouldn't matter much how fast you charge it. About the same amount of waste heat will be generated by the process, for the same amount of energy added to the battery.

    Of course any power converter (i.e. the PEM) has a peak efficiency power level, so presumably charging really fast or really slow will be slightly less efficient. But that's probably a matter of a few percent. I don't expect the battery behavior would change much, because they are temperature controlled.

    As for the air conditioner, it will cycle on as needed to remove the heat. If you charge slowly, it will cycle less often. But will the total amount of time it runs, and therefore the total energy it consumes, be any different? The difference might be negligible.

    Unfortunately determining the optimum high temperature charging strategy would require a fairly difficult empirical experiment, where you measured how many kWh were required to add, say, 10 kWh to the battery, and doing successive charge runs at different ambient temperatures and different charge rates.
     
  15. markwj

    markwj Moderator, Asia Pacific

    Joined:
    Apr 10, 2011
    Messages:
    3,656
    Location:
    Hong Kong
    I suspect that the roadster has different 'tolerances' for battery temperature, depending on whether it is charging or not.

    At 30celcius, when charging the aircon is on pretty much continuously. Once charging has completed, the aircon turns off within a few minutes. But, just sitting there plugged in all night, I don't see/hear the aircon coming on.

    The original report was that charging was using 2kW from the socket to put 1kW in the power pack at 13amp 220v. I'm not sure it is that bad, but it does seem to be about 1.5:1 at 13amp 220v, from my own experiments.

    I've sent the logs to Tom and Scott, to see what they come back with. It will be interesting to see.
     
  16. tomsax

    tomsax Member

    Joined:
    Jun 19, 2008
    Messages:
    867
    Location:
    Sammamish, WA
    I took at look at Mark's log file and calculated Wh per Range Mode %, then adapted the results from my blog to compare. The results:

    Hot-Weather-Charging-Stats.png

    I didn't have data for a cool weather charge at 240V/12A so I threw in a 120V/16A charge.

    Mark's 32A charge required 3.3% more energy per range mode percent then the 70A charge, and the 13A charge required 20% more than the 70A. Mark explains that the temperature was dropping across the days from the 70A to 32A to 13A, so if anything this understates the effect of lowering the current level in hot weather.

    Mark's energy use is about 34% above mine in cool weather. I also charge at about 238V instead of his 220V, so that's also presumably helping my energy efficiency.

    My penalty for 16A vs. 70A is only 14%, but I would expect it to get noticeably worse at 12A. For me, charging at 120V/16A, which should be equivalent to 240V/8A, the efficiency penalty is 40% compared to 240V/70A.
     
  17. markwj

    markwj Moderator, Asia Pacific

    Joined:
    Apr 10, 2011
    Messages:
    3,656
    Location:
    Hong Kong
    Thanks Tom. I gave a copy of the logs to Scott, and hopefully he can get ambient temperature readings from them. A 20% hit is not too bad. The original numbers suggested at something more like 50%->100%!

    Are you sure your figures are wall-to-battery, not just charger->battery?

    The corresponding figures from the car screens I got for the three charges are:

    • 7.4kWh usage. Charge was 50mins at 70amp, using 8kWh.
    • 10.0kWh usage. Charge was 2hours at 32amp, using 12kWh.
    • 9.2kWh usage. Charge was 4hours 25mins at 13amp, using 15kWh.
     
  18. markwj

    markwj Moderator, Asia Pacific

    Joined:
    Apr 10, 2011
    Messages:
    3,656
    Location:
    Hong Kong
    Did some work with Tom on this, trying to check logs vs energy usage screens on the car.

    There are two screens that show usage: "Trip" and "Energy Use History".

    I did a 40.4km run and monitored all the screens. I am running the 'latest' firmware (installed just a few weeks ago, containing the fix for the J1772 adaptor). I was not using air conditioning, heat or cabin fans (although other fans were most likely on).

    The Trip screen showed 40.4km, 53minutes drive time, and 5.81kWh net energy used (resulting in 145Wh/km).
    The Energy Use History screen showed Net Energy Used Today 7.5kWh and Regenerated Today 1.6kWh.

    Both Tom and I have seen that going down a hill, the Trip screen Net Energy Used number gets smaller. The resolution (0.1kWh) on Energy Use History is not enough to see this affect.

    5.81kWh + 1.6kWh = 7.41kWh (~ 7.5kWh).

    It seems that the "Trip" screen Net Energy Use figure is the energy taken out of the battery minus the energy put back in by regen. It is also seems that the Energy Use History screen is not as accurate, and the Net Energy Used Today figure does not include regeneration.

    These figures look ok to confirm energy use. I still need to confirm that the trip meter also records aircon usage (which is simple to do - just run the car idle, with the aircon full blast, and hopefully the energy use number will increase at a faster rate than without aircon).

    On the charging side, we still can't match-up the charge duration and energy figures shown on the charge history screen with the logs. They are way off (but Tom's previous tests in cold climates had them very close to what a meter on the wall showed). I'm going to get (and install) a meter on the HPC so we can record actual usage and compare against what the car is telling us, because the figures at the moment just don't make sense.

    Ongoing...
     
  19. tetsuos

    tetsuos HK #488 (2.5)

    Joined:
    Apr 10, 2011
    Messages:
    32
    Location:
    Hong Kong
    I have a meter which records usage in my carpark area, and since the only significant usage is charging my Roadster, it should be relatively accurate. I will take a look at the usage tomorrow to see what's going on.

    This morning, I discovered that (according to Charge History), I used 16 kWh last night, and I didn't even drive my Roadster yesterday! Crazy...
     
  20. tetsuos

    tetsuos HK #488 (2.5)

    Joined:
    Apr 10, 2011
    Messages:
    32
    Location:
    Hong Kong
    Confirmed this morning: 16 kWh on the charge usage screen that pops up when activating the car, 16 kWh on the charge history, and 16.1 kWh on my analog electricity meter in the carport. The 0.1 kWh is most likely from the mosquito zapper, intercom, and few exterior lights on the same circuit.

    Verdict: charge history is very accurate.

    My trip range yesterday was 44 km, equating to 363 Wh/km. Efficiency on the trip calculator shows 185 Wh/km. That's approximately 100% more power consumed for the charge cycle at ~220V/13A on my mobile charger.
     

Share This Page