satchel prefect

Well-Known Member
Joined
Sep 2, 2022
Threads
9
Messages
157
Reaction score
269
Location
South Florida, USA
Vehicles
Crusty old PHEV
Country flag
To investigate the effect of low ambient temperature on battery capacity (and by extension, range) I have been gathering battery temperature and capacity data using Car Scanner. With some readings from my own car and some from @21FE , I have some interesting results to share. It's not micrometer perfect, and it is largely based on my car (with a lower temp reading provided by 21FE), but probably close enough to support what many have been reporting anecdotally, including some of us who have been surprised and alarmed by number shown on the car's "Guess-o-meter".

TLDR: Winter range is lower not only because you're using more power to drive and heat, but also because you lose battery capacity

The data suggests that the ER battery will have around 77kwh when ambient temperature is around 13F, and will gain about 2.6kwh per 10F until it reaches full capacity.

I found the strongest correlation between HVB Min (F) and estimated capacity. HVB Min tended to be consistently 7F above ambient when ambient was below 50F, provided the car was parked and not charging or preconditioning for a few hours beforehand. Let's consider a chart showing what this implies--again, at least for my car:

Ford Mustang Mach-E Winter Range Loss in Mach-E -- a more detailed look with some measurements 1674877082293


When ambient temp was 50 or below, HVB Min temp was consistently about 7F warmer than ambient temp. This was true both for 21FE and myself, and I was fortunate enough to take readings with temperatures ranging from the low 40s to the mid 80s over a 3 week span. So, for example:
  • At 38F ambient, HVB would be approximately 45F. At that temperature, battery capacity is around 83.6kwh, which is a 9.1kwh reduction from max, leaving 90% available. My personal consumption in that temperature range is 2.9mi/kwh, yielding an actual driving range of 242 compared with the 330-340 I usually see in tropical South Florida.
The numbers in the white above are based on the linear model below, derived using the Car Scanner data. The numbers in the grey cells are based on interpolating and extrapolating my normal and cold weather mi/kwh readings (at 80F and 50F). For anyone wanting to try this out for their own car with the ER battery, simply use the Est. Capacity numbers and multiply them by your observed or expected mi/kwh numbers to estimate or sanity check the range on the Guess-o-meter. Keep in mind that this is based on a limited data set, but the relationship seems strong enough that it should land us in the ballpark. Also remember the 7 degree shift if ambient is below 50F. (In the warmer temperatures--it appears to be a bit more complicated, as it seems the car regulates temperature, which would include cooling it. And it seems to do so with some type of logic that I couldn't infer from the data on hand. Luckily once you get over 60F, the effects are mild enough that hopefully no one was worried anyhow.)

Here is a chart of the data, along with the transfer function and R-squared value:

Ford Mustang Mach-E Winter Range Loss in Mach-E -- a more detailed look with some measurements 1674880181391


Think of this as just a fun exercise. As someone mentioned earlier, there's of course a lot more to the picture than just temperature. But it seems temperature alone is not a terrible predictor of battery capacity. This may not in fact be a linear relationship across its entire range. In fact, we know it tapers off, as no one would expect to have 105kwh capacity at 130F for instance. But with a 95% r-squared value, I was content to roll with it for the temperature range most of us are likely to encounter.

Even if these numbers are off, hopefully they help paint a picture of why winter range can drop so dramatically. Let's say you have only 84% of your battery capacity and 65% of your running efficiency when it's below 20F, and you're heating the cabin, seats, windshield, and steering wheel. Under those conditions, your range could very well work out to (.84 x .65)= 55% of your mild weather range.

New owners: The good news is, as many members have said repeatedly, the car is probably fine, and you probably got the right battery. The range will revert to a more attractive number as the temperature goes up. (It's important to also consider that the range displayed on the Guess-o-meter may not respond immediately to nicer weather because it seems Ford tends to be very conservative and has some type of smoothing built into their calculation.)

If you have data points you'd like me to plug in, I need HVB Min, HVB SOC, and Energy, and whether you have the ER or SR battery. For consistency, this should be captured as soon as possible after starting, with a 4-5 hour downtime beforehand with no charging, driving, or preconditioning. Car Scanner shows these readings on this screen:

Ford Mustang Mach-E Winter Range Loss in Mach-E -- a more detailed look with some measurements 1674879973098


Thanks all.
 

hawkeye3point1

Well-Known Member
Joined
Nov 13, 2021
Threads
0
Messages
481
Reaction score
453
Location
NH
Vehicles
Space White ER RWD, Born on 12 Aug. '21
Country flag
The data suggests that the ER battery will have around 77kwh when ambient temperature is around 13F, and will gain about 2.6kwh per 10F until it reaches full capacity.
Nice work, appreciate the effort. My back of the envelope calculations make your numbers look good. I calculate ER HVB cap. @ 100% to be closer to 70 kWh at 34F (HVB @ 51%, energy @ 35.15 kWh). That assumes the BMS energy measurement is fairly linear of course.

I found the strongest correlation between HVB Min (F) and estimated capacity. HVB Min tended to be consistently 7F above ambient when ambient was below 50F, provided the car was parked and not charging or preconditioning for a few hours beforehand.
Interesting, I have found that HVB temp. tracks with ambient generally speaking. January has been quite mild in the Northeast, but my HVB is still usually 32F or lower in the AM.
 

Nklem

Well-Known Member
First Name
Norm
Joined
May 20, 2021
Threads
105
Messages
1,318
Reaction score
1,688
Location
Coast of Maine
Vehicles
Subaru Solterra
Occupation
Mechanical Engineer
Country flag
I am so glad you documented this. I do not know why Ford's battery loses so much capacity in the cold. Yes I know that all EV batteries lose capacity in the cold. A buddy and I did a comparison a month ago at 100% SOC. The 2022 Mach E ER was 78kWh available at 32F (100% SOC, from 91 kWh). My Hyundai Ioniq 5 was 72.3 KWh (from 75 kWh) at the same. Basically a 3.5% loss (or 96.5% available from the Hyundai) at 28F. Stone cold, no preconditioning. I have done a few Winter trips and confirmed that is my available Hyundai battery capacity. I also did many Mach E trips to confirm that reduced capacity as well. Why the difference? Is Ford protecting the battery where Hyundai is not? This is why my Ioniq 5, has the same or slightly higher Winter range over my Mach E ER with a battery that is 17.5% smaller. I wonder if Ford Calculates the inefficient resistance heating capacity to maintain the battery at its warmer State (like 39F), and makes that power unavailable for the drive per the BMS? Does Hyundai do or not do the same but assumes the Heat Pump efficiency to Maintain it. On a cold trip, certainly some of the Mach E energy goes to the battery. With My Hyundai, I have a several shorter and one single 400 mile road trip, in 20F-30F weather and not a kWh went to the battery for temperature management.

My only real disappointment from Ford was the displayed mi/kWh (in Winter) is not "real" on the display. It is based on the "reduced battery capacity" (but when you recharge, you put the full kWh back into the battery, and I would love someone to confirm that again), so it is actually a true lower than displayed mi/kWh on the trips meter in Winter. I could never get real KWH in (minus Charge losses) VS Real Miles out to match the screen (in Winter) unless I took the "credit" for the de-rated battery capacity found in the BMS. The difference was about .3 mi/kWH real vs display which is significant. I tell every Mach E (And Lighnting) owner you cannot calc your winter Range from Battery Percentage and mi/kWH. You have to believe the GOM (which is actually pretty darned close) or apply the adjustment factor from the BMS.

The Hyundai comes out dead-on every time displayed vs true calculation, charged in to miles out.

Inquiring minds want to know.
 
Last edited:

Mach-Lee

Well-Known Member
First Name
Lee
Joined
Jul 16, 2021
Threads
204
Messages
7,732
Reaction score
15,318
Location
Wisconsin
Vehicles
2022 Mach-E Premium AWD
Occupation
Sci/Eng
Country flag
Keep in mind this data is a direct result of the BMS energy estimation, which itself is a computer program applying a temperature curve to the energy estimate. In reality, the loss of capacity in lithium batteries with colder temps is not linear, it should be a curve that gets much steeper at colder temps:

Ford Mustang Mach-E Winter Range Loss in Mach-E -- a more detailed look with some measurements Screen Shot 2023-01-29 at 2.18.00 AM


Your data basically falls between 0ÂşC and 25ÂşC on this graph. There should be a slight curvature the data, it's hard to tell. It's possible whoever wrote the BECM software just applied a linear fit, but you would have to test at much colder battery temps to find out definitively. Using a linear fit instead a curve will create increasing error at cold temps.

In my experience, the BECM energy estimate can be inaccurate, especially when the battery changes temps. The real driving range will be different than estimated.

The loss of capacity at 0ÂşC will be somewhere between 10-20% depending on the exact battery chemistry and the rate of discharge. Looks like you got about 13% which is ballpark.

In summary, the capacity/temp curve could probably use a little bit more tweaking for maximum accuracy.

Also which BECM software were you running? Post 3.6.2 update?
 

Mach-Lee

Well-Known Member
First Name
Lee
Joined
Jul 16, 2021
Threads
204
Messages
7,732
Reaction score
15,318
Location
Wisconsin
Vehicles
2022 Mach-E Premium AWD
Occupation
Sci/Eng
Country flag
I am so glad you documented this. I do not know why Ford's battery loses so much capacity in the cold. Yes I know that all EV batteries lose capacity in the cold. A buddy and I did a comparison a month ago at 100% SOC. The 2022 Mach E ER was 78kWh available at 32F (100% SOC, from 91 kWh). My Hyundai Ioniq 5 was 72.3 KWh (from 75 kWh) at the same. Basically a 3.5% loss (or 96.5% available from the Hyundai) at 28F. Stone cold, no preconditioning. I have done a few Winter trips and confirmed that is my available Hyundai battery capacity. I also did many Mach E trips to confirm that reduced capacity as well. Why the difference? Is Ford protecting the battery where Hyundai is not? This is why my Ioniq 5, has the same or slightly higher Winter range over my Mach E ER with a battery that is 17.5% smaller. I wonder if Ford Calculates the inefficient resistance heating capacity to maintain the battery at its warmer State (like 39F), and makes that power unavailable for the drive per the BMS? Does Hyundai do or not do the same but assumes the Heat Pump efficiency to Maintain it. On a cold trip, certainly some of the Mach E energy goes to the battery. With My Hyundai, I have a several shorter and one single 400 mile road trip, in 20F-30F weather and not a kWh went to the battery for temperature management.
Have you looked at your Ioniq 5 battery temp using a scan tool? Does it keep the battery warm while parked? Mach-E doesn't heat unless plugged in.

Before 3.6.2 the GOM was too conservative in winter temps, now it's too generous IMO. So I actually do use the % and my expected mi/kWh to estimate the range, and that's usually pretty accurate.

With different software versions the GOM can be all over the place, so I don't even look at it much anymore.
 


Nklem

Well-Known Member
First Name
Norm
Joined
May 20, 2021
Threads
105
Messages
1,318
Reaction score
1,688
Location
Coast of Maine
Vehicles
Subaru Solterra
Occupation
Mechanical Engineer
Country flag
Have you looked at your Ioniq 5 battery temp using a scan tool? Does it keep the battery warm while parked? Mach-E doesn't heat unless plugged in.

Before 3.6.2 the GOM was too conservative in winter temps, now it's too generous IMO. So I actually do use the % and my expected mi/kWh to estimate the range, and that's usually pretty accurate.

With different software versions the GOM can be all over the place, so I don't even look at it much anymore.
No. From what I have scanned to date, I have seen no battery heating down to 19F (driving screen shot Attached). We have not had anything cooler except 2-3 overnights and I have not left the I5 plugged in, yet. I Hope for just a few colder (hopefully super cold) days in February for more testing. I know what the old Mach E BMS would do overnight (I think that has changed with the new software).

Ford Mustang Mach-E Winter Range Loss in Mach-E -- a more detailed look with some measurements 6159932E-5E5C-483E-88EB-DD14D124665C
 

Nklem

Well-Known Member
First Name
Norm
Joined
May 20, 2021
Threads
105
Messages
1,318
Reaction score
1,688
Location
Coast of Maine
Vehicles
Subaru Solterra
Occupation
Mechanical Engineer
Country flag
No. From what I have scanned to date, I have seen no battery heating down to 19F (driving screen shot Attached). We have not had anything cooler except 2-3 overnights and I have not left the I5 plugged in, yet. I Hope for just a few colder (hopefully super cold) days in February for more testing. I know what the old Mach E BMS would do overnight (I think that has changed with the new software).

Ford Mustang Mach-E Winter Range Loss in Mach-E -- a more detailed look with some measurements 6159932E-5E5C-483E-88EB-DD14D124665C
I guess I should not have wished for it. 7F Wed Night and -16F forecasted for Friday night. It looks like I will have to plug in and monitor the battery treatment and kWh it draws as well as charge to 100% and monitor the capacity loss due to cold. I will also be doing a -10 to 75F Cabin Heating test against a Mach E Friday night or Saturday AM for Youtube.
 

kltye

Well-Known Member
Joined
May 21, 2021
Threads
17
Messages
880
Reaction score
1,381
Location
Chicago
Vehicles
IB MME Premium RWD
Country flag
How does having a large top and bottom buffer factor into temperature calculations? We're basing these numbers on the 88/91 kWh available numbers, rather than the gross 99kWh the pack has. Can the BMS "hide" some of this loss by minimizing the top buffer in super cold temperatures? Does doing this shorten the pack's cycle life?
 

SpaceEVDriver

Well-Known Member
Joined
Aug 26, 2021
Threads
59
Messages
2,279
Reaction score
4,001
Location
Arizona
Vehicles
Ground-based: CA Route 1 AWD, ER
Occupation
Planetary Science
Country flag
Here are some similar (though not identical) data from a single test.

Last night, after a full day of cold soaking, I unplugged the car, started it up, and let it sit for ~25 minutes with a nearby DCFC destination plugged into the onboard navigation system.

The battery's SOC decreased from about 86% to about 83% while energy to empty increased from about 68 kWh to about 74 kWh (with an initial dip). Battery temperature increased from about 34 F to about 63 F.

Ford Mustang Mach-E Winter Range Loss in Mach-E -- a more detailed look with some measurements Screenshot_20230131-084244
 

awp0

Well-Known Member
First Name
Aaron
Joined
Jul 23, 2022
Threads
11
Messages
775
Reaction score
932
Location
boston, ma
Vehicles
MME Premium AWD ER
Country flag
My only real disappointment from Ford was the displayed mi/kWh (in Winter) is not "real" on the display. It is based on the "reduced battery capacity" (but when you recharge, you put the full kWh back into the battery, and I would love someone to confirm that again), so it is actually a true lower than displayed mi/kWh on the trips meter in Winter.
Are you talking about the trip computer which reports the current trip's miles/kWh efficiency? It's not actually miles/kWh? This is a new one for me.
 

Nklem

Well-Known Member
First Name
Norm
Joined
May 20, 2021
Threads
105
Messages
1,318
Reaction score
1,688
Location
Coast of Maine
Vehicles
Subaru Solterra
Occupation
Mechanical Engineer
Country flag
Are you talking about the trip computer which reports the current trip's miles/kWh efficiency? It's not actually miles/kWh? This is a new one for me.
Correct. In the Winter it’s actually less than displayed. Here is a shot of a 22 Mach E ER at 100% Displayed charge. 91 kWh battery is only 78 kWh at this state and your Mi/kWh on the trips is calculated from this reduced number. Also see that 100% displayed is only 96.2%. So, in the winter, full battery capacity (91,88, 68 or 71) *displayed %*displayed Mi/kWh is not your real range. The GOM takes all of these factors and estimates it for you. Test it yourself sometime. Charge to 100%, drive in the cold for 100 miles or so, and level 2 recharge to 100% with a unit that tells you how much energy it delivered. Deduct 10% from that for charge losses and you will be about 10 additional percent (or more) less than the mi/kWh displayed. I struggled with this for months until I got Carscanner and saw what is happening in the BMS. If you factor in the reduced battery capacity it matches the trips perfectly.

71DDA6CF-537E-4206-BD4A-3004CE51431A.png
 

Nklem

Well-Known Member
First Name
Norm
Joined
May 20, 2021
Threads
105
Messages
1,318
Reaction score
1,688
Location
Coast of Maine
Vehicles
Subaru Solterra
Occupation
Mechanical Engineer
Country flag

awp0

Well-Known Member
First Name
Aaron
Joined
Jul 23, 2022
Threads
11
Messages
775
Reaction score
932
Location
boston, ma
Vehicles
MME Premium AWD ER
Country flag
Correct. In the Winter it’s actually less than displayed. Here is a shot of a 22 Mach E ER at 100% Displayed charge. 91 kWh battery is only 78 kWh at this state and your Mi/kWh on the trips is calculated from this reduced number. Also see that 100% displayed is only 96.2%. So, in the winter, full battery capacity (91,88, 68 or 71) *displayed %*displayed Mi/kWh is not your real range. The GOM takes all of these factors and estimates it for you. Test it yourself sometime. Charge to 100%, drive in the cold for 100 miles or so, and level 2 recharge to 100% with a unit that tells you how much energy it delivered. Deduct 10% from that for charge losses and you will be about 10 additional percent (or more) less than the mi/kWh displayed. I struggled with this for months until I got Carscanner and saw what is happening in the BMS. If you factor in the reduced battery capacity it matches the trips perfectly.

71DDA6CF-537E-4206-BD4A-3004CE51431A.png
Thanks, I have Car Scanner, though I've only scratched the surface.

Question: I would have assumed the difference between "HVB SOC" and "HVB SOC Display" is related to the battery buffer. Yes/no?

I suppose this also means the FordPass trip logs are wrong? They tell you the mileage and the absolute number of kWh used, and in my experience they're the same as the trip computer.

Anyway, I'm curious and I'll try what you're suggesting on my next longer trip. Since I have Car Scanner, I should be able to understand exactly how many kWh were used, and compute the actual miles/kWh and compare with what the car's trip computer tells me. Assuming you're right, the whole thing would be pretty disappointing IMHO. I expect a kWh to be a kWh. Not some scaled kWh thing that's proportional to the total capacity.
 

Nklem

Well-Known Member
First Name
Norm
Joined
May 20, 2021
Threads
105
Messages
1,318
Reaction score
1,688
Location
Coast of Maine
Vehicles
Subaru Solterra
Occupation
Mechanical Engineer
Country flag
Thanks, I have Car Scanner, though I've only scratched the surface.

Question: I would have assumed the difference between "HVB SOC" and "HVB SOC Display" is related to the battery buffer. Yes/no?

I suppose this also means the FordPass trip logs are wrong? They tell you the mileage and the absolute number of kWh used, and in my experience they're the same as the trip computer.

Anyway, I'm curious and I'll try what you're suggesting on my next longer trip. Since I have Car Scanner, I should be able to understand exactly how many kWh were used, and compute the actual miles/kWh and compare with what the car's trip computer tells me. Assuming you're right, the whole thing would be pretty disappointing IMHO. I expect a kWh to be a kWh. Not some scaled kWh thing that's proportional to the total capacity.
I was disappointed at first. If I was paying for power (mine is free from work) it would have bothered me more. I am sure Tesla does the same. Knowing the car was reporting more efficiency than in reality was a surprise. once I figured it out, it was not the end of the world. That’s part of the reason for the low Mach E winter range. But all EVs lose winter range and Ford does actually tell you with the GOM estimates. I found my GOM was pretty accurate , within 10 miles during the winter….What’s 10%….
Sponsored

 
Last edited:
 




Top