Mach-Lee's Mach-E Cold Weather Testing [Taking Requests]

[Mach-Lee]

Alright, I did cold weather run part 2 last night to compare to post #1. This time I used preconditioning beforehand, and it was somewhat colder, about -9ºF this time instead of -2ºF. Let's post the stats and then we'll get into graphs:

12/21 Trip Summary:

Outdoor temp: -9ºF/-23ºC
Wind: W 3 mph
Trip distance: 89.3 mi
Duration: 82 min
Displayed efficiency: 1.8 mi/kWh
Average speed: 65 MPH

Preconditioning:
Setting: Warm
Initial temp: 34ºF
Energy used: 7.01 kWh
Time: 45 minutes

Starting conditions:
Cabin temp: 61.0ºF
Battery temp: 57.2ºF
Displayed SoC: 80.0%
Displayed Range: 148 mi
kWh to empty: 66.338

Ending conditions:
Cabin temp: 61.3ºF
Battery temp: 33.8ºF
Displayed SoC: 22.5%
Displayed Range: 27 mi
kWh to empty: 16.424

Calculated stats:
Battery used: 57.5%
Energy used (BECM): 49.91 kWh
Calculated efficiency: 1.80 mi/kWh (based on above)
Calculated efficiency: 1.90 (based on wall)
100% range at -9ºF: 156 miles (56% of EPA)
100% energy available: 86.8 kWh

Charge back energy:
51.61 kWh from wall
91.5 % charging efficiency
47.2 kWh estimated to the pack

HVAC set to 68ºF AUTO 1

First I'll talk about cabin comfort. It was colder today, I saw temps down to about -12F while driving. You can see the cabin started warm this time due to preconditioning, and the sensor temp remained relatively flat. Temp stayed around 62ºF the whole time, which was somewhat chilly for me. No changes were made to the HVAC the entire drive (68º AUTO 1). Even though the graph doesn't show it, I was feeling colder as time went on, especially my legs and ankles. I had a jacket and gloves on, but just jeans on my legs which wasn't quite enough IMO. Heater was pegged at full 5.5 kW output almost the whole drive so this is thermal equilibrium folks! The Mach-E cabin sensor was tracking around 64º so it knew it wasn't maintaining the 68º setting.

Fogging was actually better this time, you can see the three humidity spikes when the recirc door closed. The fog during the first recirc event was the worst (maybe 50% coverage) but resolved itself without requiring MAX defrost. There were much less recirc events this time than before (3x vs 6x?), I'm guessing that's because it was colder outside and the strategy is avoiding frosting risk more. I think once the temp gets down to -15ºF the recirc warming events will be completely eliminated and it will use 100% outside air 100% of the time.

There were no recirc events while I was on the interstate, but interestingly it did the second and third events as soon as I got off the exits. Not sure if that was just a timing thing or if it's also speed dependent? I didn't measure the vent temps, but during the times on the interstates when the fresh air was open the air blowing was not hot. It felt like a fan in your room, maybe 75ºF air blowing out? So that would be about an 85ºF temp rise. Which was only able to maintain a 62ºF cabin so things will decrease proportionally with temperature. If it's 10º colder outside (like -18ºF), the cabin will be 10º colder (52ºF) to maintain the same delta. I'm going to call it around -15ºF/-25ºC as the coldest temp you'll want to take a trip in this thing. When it gets below that, I'm driving my ICE car.

Now let's talk about battery temperatures:

Read the color legend for the parameters. HvbT is the mean cell temp, HvbMinT is the lowest cell temp, HvbMaxT is the warmest cell temp, and HvbAve is the average of the min and max. This is from the Car Scanner app BTW. You can see the battery started out at 56ºF (preconditioning heats it to 59ºF). A reminder during preconditioning the battery is heated first, then the cabin (sequentially not simultaneously), so it cooled off a couple degrees while the cabin was heating. See stats for energy. The warmest cells are quite a bit warmer than the mean, I'm guessing those are the cells on the second tier up under the back seat that get warmer.

You can also see the big cooling off trend. Interestingly the pack stayed above freezing the entire 82 min drive, which is good. That means you have about 1.5 hours to drive after preconditioning before worrying about the pack getting too cold, longer if parked. In run #1 the battery temp really didn't change much and stayed around 32ºF.

Now it's time to share my most exciting finding of the drive, which was the Mach-E actually tried to heat the pack while driving! I've never seen that before! Let's see:

The red line is the coolant heater power. A reminder that the Mach-E has a single 5 kW PTC coolant heater that can heat either the cabin, battery, or both depending on the valve settings. You can see the heater is pegged around 5.5 kW for most of the drive, which is its maximum output. Because it's a PTC (positive temperature coefficient) the output will vary some depending on the inlet temperature. When it's colder it outputs more heat. You can also see some spikes when it first turns on, which is a normal characteristic. The recirc cycles show up as slight dips on the heater output because the returning coolant will be warmer (from warmer inlet air) which affects the output.

The green line is the battery coolant inlet temperature, which measures the temp of the coolant going into the pack. This is the parameter (HvbCIT) you want to look at to determine if the battery is being heated or cooled. The temp sensor is in a hose below the frunk, so if there's no coolant flow, the temp will drop and become the same as the ambient temperature. You can see the rapid temp drop from 48º to 20º during the first 15 minutes of driving. But then something happens at 00:17! The diverter valve opened! This means the coolant exiting the heater core is now circulating through the battery before returning to the PTC heater. Once the flow starts, the sensor reading rapidly adjusts to the circulating coolant entering the battery, which is around 48º to start with. That's the amount of heat left in the coolant after the heater core. Just enough to keep the battery above freezing.

Let's see what that heater flow does to the pack temps:

Please note this graph is in ºC now so the blue line makes sense (unit conversion issue in Car Scanner). I put the coolant temp from the previous graph on this graph with the other temps. You can see the point the diverter valve is opened marked "change." The blue line is the battery temp range, which is the difference in temp between the warmest and coldest cell (in ºC). Since there's now flow through the cold plates in the pack, the warmest cells now start cooling down and become closer to the average temp. You can see the blue difference line stabilize and start to fall as the cells even out in temperature. I'm happy to see this since you want all the cells in the pack at a similar temperature. Last, you can see the battery inlet temp trend with the rest of the pack temps, which proves flow through the pack is happening.

On the previous run in similar temps, we got no battery heating or flow whatsoever. So it's kind of a mystery why it decided to work this time. I feel like it has to do with outdoor temp, battery temp, and cabin heating demand. Last time the heater spent the whole drive trying to warm up the cabin, but this time the cabin and the battery were warm to begin with. I previously talked about my concerns with regen on a freezing pack, so seeing battery heating while driving, even slight, is a good thing to keep the pack warmer for longer. The data resolution isn't great, but it looks like the rate of cooling was arrested slightly after the battery was switched into the heating loop.

One thing I need to mention is that switching the battery into the heating loop will dramatically lower the vent temps in the cabin. The battery can drop the return temp of the coolant much lower than the normal return temp of the heater core, so that means the coolant exiting the heater core will be less hot as well. The effect is amplified the colder the pack is, this is why you lose most of your heat for the first few minutes of DC fast charing. The pack will just suck all the heat out of the loop and you won't get it back until the pack temp comes up or the pack is removed from the loop. If the pack is too cold relative to the heater core return temp, the strategy may not want to switch it in for fear of reducing cabin heat output too much.

Last, let's look at some energy numbers for the two runs:

	Run 1 (no Preconditioning)	Run 2 (Preconditioned)
% Battery Used	62.0%	57.5%
Energy from Wall	53.3 kWh	51.6 + 7.0 kWh
Calculated efficiency	1.85 mi/kWh	1.90 mi/kWh
100% Calculated Range	144 mi	156 mi
100% Calculated Energy	78.6 kWh	86.8 kWh
Initial / Final Pack Temp	30ºF / 34ºF	57ºF / 34ºF

Notice the battery % used was less, even though the efficiency was almost identical. A warmer pack has more energy available than a colder pack, and will have a greater range. Keep in mind the increase in range and energy will be greater when the pack is even colder. My garage was about 30ºF but if your car sits outside in worse temps the benefits will increase further.

So in summary, preconditioning seems to have had a couple good effects on the drive. We had a warm cabin from the beginning, less initial fogging, slightly more range, got the battery to heat while driving, and less concerns about regen damaging the pack.

Findings

• Exact same distance driven, about 8ºF colder than previous run
• Cabin started warmer, but got increasingly chilly as time went on
• Cabin temp stable at 62ºF, but could not warm up futher
• Only a few recirc events this time compared to before
• Window fogging was less due to fewer recirc events
• Heat vent temps feel colder than previous run
• Heater as max output virtually whole time just like previous run
• Battery stayed above freezing for a long time (1.5 hr) after preconditioning
• Battery began heating about 15 minutes into run, continued for the remainder
• Regen damage less of a concern due to warmer battery temps
• Energy consumption and efficiency is very similar between the two runs
• Total range is increased with preconditioning by about 8%
• Available pack energy is increased with preconditioning by about 10%
• Increases will be even larger when the pack is colder to begin with

Conclusions

• You will start to run out of heat around 5ºF outside, and by -15ºF you will want to drive a different car
• The 5.5 kW Mach-E heater is inadequate for temps below -13ºF/-25ºC unless trips are short and preconditioning is used beforehand.
• Preconditioning does not affect driving efficiency, and lowers overall efficiency.
• Preconditioning helps increase total range in the winter.
• The energy penalty for preconditioning makes more sense for longer trips (>20 min)
• Preconditioning will lower the risk of lithium plating battery damage from regen by keeping the battery warmer

 

[mkhuffman]

Preconditioning:
Setting: Warm
Initial temp: 34ºF
Energy used: 7.01 kWh
Time: 45 minutes

Did you intentionally limit the preconditioning time to 45 minutes, or is that what the car normally does? I am wondering if a longer preconditioning would have helped.

 

[the golden eel]

Did you intentionally limit the preconditioning time to 45 minutes, or is that what the car normally does? I am wondering if a longer preconditioning would have helped.

The other day my car took roughly 45 minutes to precondition in a 29F garage. So this tracks with what I saw.

 

[sukhoi_584th]

Does the HVAC use the A/C when the vents are set to front defrost and/or front defrost + dashboard (main) vent? I've noticed the indicator light doesn't illuminate but I'm used to vehicles activating the A/C any time a front defrost vent is active.

Assuming A/C is secretly active I'm curious if you would have better cabin temps by running the main vent without front defrost so you don't have to overcome the A/C cooled air.

 

[RedRocker]

I wonder if it would have warmed more if it was set to 72F/Auto-2 vs 68F/Auto-1? I understand the heater was pegged at 5kW, just wondering how recirc works and if it would have recirc'd more to increase cabin temp if the delta between actual/requested temp was larger. Maybe something to try.

I also have noticed my legs do feel colder on longer trips, I attributed it to the seat heater timing out (in auto mode), but maybe there is more to it than that. I run 69F AUTO-1 as my default.

As always great info - thanks for sharing

 

[lightningandmache]

Maybe 32F battery temp is the target, and correctly predicted equilibrium on your previous drive wouldn't require external heat but needed it here and is willing to sacrifice cabin temps to get it.

Why didn't they simply bump the PTC up a bit? This performance would have been obvious to an engineer on a -10F 2hr test drive.

Also, why does the battery only pre-condition to 59F?! Why not push that up? That would be an easy fix they could push OTA now. Cabin stays warmer, battery has more capacity, less regen risk/issues. Having the cells at 100F+ for just minutes before occasional long trips will have no impact on longevity. They could even have a "long trip" mode that enables more pre-heating where it makes sense.

Do you have any recordings of battery temp after a DCFC session in winter?

Such a fantastic thread!

 

[Jiji]

What are you using to capture and display that data?

Some custom software of my own creation. First there is the collection done by a state machine (think Idle, On, Trip, Charging, etc) reading the various OBDII sensors and writing them to the cloud.

The posted screenshot is from a second iPhone app that provides a UI for the data in the cloud. In this case I have an option to graph trips by efficiency and temperature and shade each point by the change in elevation.

 

[mkhuffman]

Also, why does the battery only pre-condition to 59F?! Why not push that up? That would be an easy fix they could push OTA now. Cabin stays warmer, battery has more capacity, less regen risk/issues. Having the cells at 100F+ for just minutes before occasional long trips will have no impact on longevity. They could even have a "long trip" mode that enables more pre-heating where it makes sense.

It does make sense to warm the battery up to at least 70 F. 100 F is probably too hot, but starting at 70 would allow the car to drive a longer distance before hitting 32.

 

[lightningandmache]

It does make sense to warm the battery up to at least 70 F. 100 F is probably too hot, but starting at 70 would allow the car to drive a longer distance before hitting 32.

100F should not be too hot. The cells should get to at least that temp during DCFC sessions already and here we are talking minutes of duration only occasionally. You can see from this post the temps drop within minutes. Give a "long trip" pre-heating checkbox so it is used infrequently, and doesn't needlessly waste energy. Scale it based on ambient. Would have to see what the equilibrium temps look like on longer trips than this post, but this feels like something they could quickly validate and OTA for very significant QoL improvement.

 

[RickMachE]

oh yes, you must be referring to LaRiche Chevrolet dealer in Findlay. I'll be driving through Findlay around 10pm at night. Is the chargepoint charger accessible even after the dealership is closed? I don't want to deviate from course only to find the charger behind a locked gate. Otherwise I'll keep pushing to Lima.
Thanks for all your feedback btw...it's our first EV trip from Canada to FL. I am prepared for an adventure but unfortunately I'll be driving through some severely cold temperatures....something I wasn't counting on. The trip back looks warmer...all the way in the mid 20's.

No, Dave White in Sylvania. LaRiche looked iffy in PlugShare. Wouldn't want to be back of a dealership in the dark either.

 

[RMoore]

Is the battery well insulated from the cabin (or vice versa)? In other words, if there is a large gradient between the two compartment, is there heat flow between them? On a cold day, will a warm battery help the cabin stay warm, and would a cold battery serve as a heat sink, pulling heat from the cabin or is the majority of cabin heat loss on a very cold day a result of flow from the cabin to the outside air (and fresh air coming in that is not adequately warmed)?

 

[Mach-Lee]

Did you intentionally limit the preconditioning time to 45 minutes, or is that what the car normally does? I am wondering if a longer preconditioning would have helped.

No, the car decides when to start preconditioning to meet the targets in time. We have no control over it, I can't make it longer unless maybe bumping up the cabin temp setting to hot.

Does the HVAC use the A/C when the vents are set to front defrost and/or front defrost + dashboard (main) vent? I've noticed the indicator light doesn't illuminate but I'm used to vehicles activating the A/C any time a front defrost vent is active.

Assuming A/C is secretly active I'm curious if you would have better cabin temps by running the main vent without front defrost so you don't have to overcome the A/C cooled air.

The A/C compressor doesn't run when it's less than 32ºF outside (it will not run a freezing evaporator). No combination of buttons will override that.

I wonder if it would have warmed more if it was set to 72F/Auto-2 vs 68F/Auto-1? I understand the heater was pegged at 5kW, just wondering how recirc works and if it would have recirc'd more to increase cabin temp if the delta between actual/requested temp was larger. Maybe something to try.

I also have noticed my legs do feel colder on longer trips, I attributed it to the seat heater timing out (in auto mode), but maybe there is more to it than that. I run 69F AUTO-1 as my default.

As always great info - thanks for sharing

Perhaps, I'll consider testing that. I used 68º because that's what I normally use and any deviations from the normal comfort level would be noticeable to me. Also it's a moderately low setting that should give the heater the most opportunity to keep up. The seat heaters don't timeout, they will ramp down if the seat temp is close to your target climate temp. In AUTO mode the heaters are thermostatically controlled to maintain your set temp rather than running at a constant power.

Maybe 32F battery temp is the target, and correctly predicted equilibrium on your previous drive wouldn't require external heat but needed it here and is willing to sacrifice cabin temps to get it.

Why didn't they simply bump the PTC up a bit? This performance would have been obvious to an engineer on a -10F 2hr test drive.

Also, why does the battery only pre-condition to 59F?! Why not push that up? That would be an easy fix they could push OTA now. Cabin stays warmer, battery has more capacity, less regen risk/issues. Having the cells at 100F+ for just minutes before occasional long trips will have no impact on longevity. They could even have a "long trip" mode that enables more pre-heating where it makes sense.

Do you have any recordings of battery temp after a DCFC session in winter?

The cabin heating always has priority over pack heating, except while DCFC. If the cabin is too cold it will not attempt to heat the pack. Increasing the heater size may created a packaging problem or increased the cost considerably. 5 kW seems like a standard size in the automotive industry so I think that’s why they went with it. Most EVs are designed for California weather so there hasn’t been a big demand for larger heaters until now as they expand into frigid climates. I’m guessing 59°F was chosen as a moderate temperature so as to not waste electricity. They just need to keep the pack above freezing for an hour or two drive until the next charge, and it’s clear from the data 59° accomplishes that in most typical conditions. Maybe in Alaska/Canada they should bump it up to 20°C, but 15°C is fine for most of the US.

I might have a DCFC recording, but it’s not too interesting. It only wants to heat the pack when it’s really cold or it’s a high power EA station. For the smaller DCFC it doesn’t want to do much and will just charge at the limit while the pack is frozen. It’s very inconsistent behavior, and Ford needs to work on it.

Is the battery well insulated from the cabin (or vice versa)? In other words, if there is a large gradient between the two compartment, is there heat flow between them? On a cold day, will a warm battery help the cabin stay warm, and would a cold battery serve as a heat sink, pulling heat from the cabin or is the majority of cabin heat loss on a very cold day a result of flow from the cabin to the outside air (and fresh air coming in that is not adequately warmed)?

I don’t think they’re expressly insulated from each other, the battery top cover is just right under the metal floor pans. If the battery is really warm (like after DCFC) you might feel that in the floor. So having a warm battery might help a little bit, but most of the cabin heat loss is through the windows and via the warm air exiting the cabin through the rear outflow vents.

 

[lightningandmache]

The cabin heating always has priority over pack heating, except while DCFC.

I thought the cabin couldn't reach target temps, wouldn't closing battery loop valve fix this?

I’m guessing 59°F was chosen as a moderate temperature so as to not waste electricity. They just need to keep the pack above freezing for an hour or two drive until the next charge, and it’s clear from the data 59° accomplishes that in most typical conditions.

I think 32F is too low a target. Your drive was also very short, the equilibrium looks like it would have been <32F. They are forgoing winter range for a lot of consumers with this approach. I can understand not wanting to re-work any hardware before a refresh for modest range gain, but bumping pre-conditioning temp is an OTA/validation freebie. Put it in a "long trip" checkbox, so it isn't wasting energy when not needed. This, along with no DCFC on-route pre-conditioning is a head-scratcher. Sure the PTC heater is under-specced for -10F, but in warmer temps it could handle both.

Maybe I'm being dense, but I still don't understand your statement on not valving in the motor/inverter loop which in winter is probably throwing off 1.7kW waste heat. They go oil -> glycol line -> radiator. Cost of an extra valve and a short poly tubing run? Seems like this would help alleviate the undersized PTC heater and cold batteries.

Thanks so much for taking the time to do these posts! Very enlightening.

 

[Av8tor]

@Mach-Lee Can you do a test to see if this Fordpass switch "Allow start when vehicle is unplugged" to see if it would actually precondition the battery when unplugged. I have the climate setting set to Off in the Departure time.

It has the note right under it stating "Remote starting", which we already know is cabin only. Since this only appears in "Depature Times" it would be easy for them to actually have it Precondition when unplugged.

Interesting that it has a Vehicle's Time Zone setting when activated, clearly for folks traveling via the airport and setting this in another timezone.

 

[Mach-Lee]

@Mach-Lee Can you do a test to see if this Fordpass switch "Allow start when vehicle is unplugged" to see if it would actually precondition the battery when unplugged. I have the climate setting set to Off in the Departure time.

It has the note right under it stating "Remote starting", which we already know is cabin only. Since this only appears in "Depature Times" it would be easy for them to actually have it Precondition when unplugged.

I did try that already, unplugged it did NOT heat the battery and strangely heated the cabin AFTER the departure time. So that's really a scheduled remote start feature. As far as I know, you must be plugged into a L2 charger in order to heat the battery during preconditioning. It doesn't make sense to deplete the battery to heat itself, that would likely always be a net loss of range.