OBD II Data for HVB

[larryh]

The numbers provided are only gross approximations based on my best guesses from the graphs provided in the papers.  I had to make many assumptions to come up with them.  Tweaking the numbers I used isn't going to provide any useful answers.  The actual computation is very complex, solving a set of differential equations using numerical analysis.   The author's in the papers already did this much more precise calculation incorporating additional factors such as temperature and battery cycling.   They determined charging at a faster rate caused less degradation.  I have no way of verifying their more precise analysis.  The point of my post is simply to illustrate that even though faster charging causes faster degradation, this degradation occurs for much shorter period of time resulting in less overall degradation than slower charging (that was the point made in the papers).

 

The calculations were 0.22e-8 Ohms/s * 6 hours * 3600 s/hour = 4.8e-5 Ohms and 0.63e-8 Ohms/s * 2 hours * 3600 s/hour = 4.5e-5 Ohms/s.  The difference between the two is going to be small and as I mentioned, I don't have anyway to improve the accuracy of these calculations and come up with the correct answer for the Energi's HVB. 

Edited June 7, 2016 at 09:56 AM by larryh

[larryh]

Note that in the simple analysis in the previous posts, the SOC and temperature difference effects on degradation were not taken into account. The analysis only considered the effect of an increased charging rate on degradation. The amount of degradation due solely to charging at a high rate for 2 hours vs. charging at a slower rate for 6 hours is small. Too close to call to be certain which is better.

 

However, because Level 1 Charging starts four hours earlier than Level 2 Charging, at any given time, the SOC of the battery will be higher with level 1 Charging. After 4 hours of Level 1 Charging, the HVB SOC be at 73%. For Level 2 Charging, we will just be starting to charge the battery (SOC will still be at 20% assuming the HVB SOC started out at 20% SOC). Higher SOC means higher degradation. As a result of the higher SOC for the Level 1 Charging profile, the increase in internal resistance due to degradation of the battery will be about 1.8e-5 Ohms greater with Level 1 Charging vs. Level 2 Charging if you go through the math.

 

We also need to take into account the effect of the temperature differences in the two charging profiles on degradation. With Level 1 Charging, the temperature of the HVB will increase at approximately a constant rate for the entire six hours. With Level 2 Charging, the HVB will continue to cool down for the first four hours, and then the temperature will rise at an accelerated rate for the last two hours exceeding the final temperature reached during Level 1 Charging by perhaps up to 5 F (being generous). I don’t have any data on internal resistance degradation vs. temperature so I can’t do a thorough analysis. However, I can easily compute the average temperature for the two charging profiles. It is going to be lower with level 2 Charging, simply because the battery is cooling down perhaps a degree during the first four hours and then begins rising only during the last 2 hours. During most of that time, the HVB temperature will be less with Level 2 Charging than Level 1 Charging except during the final hour (and then it will only be a few degrees higher). So most likely Level 2 Charging again will result in less degradation.

[flyingcheesehead]

You can also set the value charge windows to indicate a high electric rate to avoid charging right after you plug in at work (provided you at at work long enough that you can delay charging).  My first priority though would be to make sure that the car is not in the hot sun.  The can easily cause the HVB temperature to rise above 100 F.  

 

It's a nicely shaded spot - Under a tree, with some trees to the east that further shade it in the morning and the building to the west shading it in the late afternoon. Unfortunately, with the problems I've had with the L1 EVSE, I can't really delay it and still get charged - I have to babysit it and often don't leave with a full charge anyway. I guess that's not so unfortunate as I'd thought. ;)

[rbort]

At the end of the day, I still prefer to use L1 charging at home or on the road when I'm not in any rush (at a hotel or a friend's house), and L2 charging when I need to on the road to recharge for the next leg of my trip.  I still believe slower battery charging is better for the pack than faster, but its not that much of a difference just like time spent at higher charge levels before 100%, meaning 30, 50, 70% are still way better than sitting at 100% all the time.  Usually with Lipo batteries they want you to leave them at 50-60% for long term storage, Li-ion is similar.

 

Interesting part seems to be though, when I get home and the HVB is still warm from driving and it low (into the Hybrid) I like to raise it off the bottom a bit as I've seen it crash lower in the past if left as is and next time you go start the car the engine just starts as the HVB is too dead (below 1/4 on the hybrid portion).  This usually happens in the winter time.  

 

This time around I charged it up after returning from Vermont sunday night on L1 until it was at 12% according to the log, but the next day the battery is now at 21%.   The opposite happened, most likely due to the hotter weather the next day and having the battery voltage rise or fall after some rest due to temp swings.

 

-=>Raja.

Edited June 7, 2016 at 06:29 PM by rbort

[bdginmo]

I've noticed the phantom discharge while letting the car sit during work as well. If I run it down to hybrid mode and then park it with about 50% on the hybrid battery meter sometimes it'll show 25% when I start it up to head back home. Like you I figured this was due to the difference in battery temperature causing a different SoC to be reported.

 

I wish there was a way to tell the computer that you want to charge the battery to a certain percent and then halt. That might be useful when you know you're going to store the car for several days or more without driving and you want the battery to be between a certain range to prevent excessive degradation.

Edited June 7, 2016 at 07:25 PM by bdginmo

[larryh]

There are two reasons why the SOC of the HVB changes from the value it was when you turned off the car. 

 

1.  The SOC of the HVB is determined by the equilibrium voltage of the HVB.  The battery needs to rest for an hour or so before coming to equilibrium.  Then the voltage can be measured and the HVB SOC can be computed accurately.  When the car is running, the voltage of HVB fluctuates wildly and cannot be used to accurately determine HVB SOC.  Instead, the car has to estimate the SOC based on the amount of energy it believes is in the HVB.  Unfortunately, the car cannot precisely determine the amount of energy in the HVB either (but it provides a better estimate than voltage).  So the SOC estimate that it computes while the car is running is going to be a bit off.  You will see this estimate when you park the car at work.  After the HVB has rested for a while, it can make a voltage measurement and compute an accurate SOC.  Thus when you leave work, you see the corrected/accurate SOC.

 

2.  When the battery cools down, it radiates energy.  This energy is no longer available to propel the car.  The difference in energy in the HVB from when you park the car to when you leave work is the amount of energy that has been radiated.  This is reflected as a lower HVB SOC. 

 

Ford needs to provide better software to manage the HVB.  If I could tell the car the SOC I wanted the HVB charged to each morning, then I wouldn't have to charge to 100%.  I only need about 66% SOC to get to work and back.  That would help reduce degradation of the HVB.  In another paper, they compare battery life with optimized charging plus charging only to the SOC which is needed for the day's driving vs. just plugging in at night and charging to 100%.  The battery life in the later scenario was 5.5 years.  In the optimized partial charging scenario, battery life was 12.8 years.  A very significant difference.  They could at least use optimized charging where the car figures out the optimal rates and times to charge rather than simply charging at the maximum rate during the least cost time windows for electricity. 

Edited June 8, 2016 at 12:02 AM by larryh

[jdbob]

Ford needs to provide better software to manage the HVB.  If I could tell the car the SOC I wanted the HVB charged to each morning, then I wouldn't have to charge to 100%.  I only need about 66% SOC to get to work and back.  That would help reduce degradation of the HVB. 

 

I don't expect Ford to do anything. I wonder if it were possible to do something at the EVSE end. I thought you had a graph that showed the charging current tapering off when getting full, but didn't find it. I'm thinking that a EVSE could be setup so that it will charge up to say 80% (just in case you need the car at an unexpected time) based on the taper and then remove the pilot signal to stop charging. At some later time (like 30 or 40 minutes before you expect to leave) it could re-enable the pilot to resume charging to 100%.

 

I almost always use the car to go into town shopping around 1:30-1:40PM on Monday and Thursday, so that's what I set the "GO Times" to. I also often go into town on Saturday, but not at any specific time, in that case I use Remote Start to precondition the cabin. But of course I might make an unexpected trip at any time so don't want to have just whatever was left in the battery the last time I came home, but 80% would do me except for winter.

[bdginmo]

Great information larryh. That makes perfect sense. You bring up a good point that there are legitimate everyday use cases to having a feature that allows you to specify a charge level.

[larryh]

The internal charger in the car controls the charging rate--it can charge at any rate it pleases.  The pilot signal to the EVSE tells the EVSE the maximum amount of power that the charger will draw.  The duty cycle of the pilot signal that the car sends to the EVSE is 50%, indicating that it will draw up to 1/2 the maximum power that a Level 2 EVSE can provide.  The car does taper off charging power during the last five to ten minutes of charging.  The internal charger will also charge the HVB at a slower rate when the HVB is extremely cold, and probably when it is extremely hot.  When Value Charge is selected, they could program the internal charger to use an optimal charging profile, varying the charge rate as necessary, and giving preference to the lowest cost electric rate windows, to minimize degradation to the HVB, just as long as the car is fully charged by the next GO time. 

 

In another paper, they claim a Lithium-ion battery should not be operated at a level lower than ~25% SOC (This includes Hybrid Mode in our car.  When MyFord Touch states the SOC is 0%, the actual SOC of the HVB is around 20%.  In Hybrid Mode, the battery SOC ranges between 15% and 20%). 

 

"If lithium ion cells are discharged or operated at a level lower than ~25% SOC, their efficiency and performance is degraded, plus significant heating and aging will occur".

 

So if you have a long commute, it might be advisable to stay out of Hybrid Mode until the end of your trip. Also, you probably don't want to leave the HVB SOC at less than 25% SOC for too long (This corresponds to about 5% as reported by MyFord Touch). 

Edited June 8, 2016 at 10:07 AM by larryh

[jdbob]

The internal charger in the car controls the charging rate--it can charge at any rate it pleases.  The pilot signal to the EVSE tells the EVSE the maximum amount of power that the charger will draw.  The duty cycle of the pilot signal that the car sends to the EVSE is 50%, indicating that it will draw up to 1/2 the maximum power that a Level 2 EVSE can provide.  The car does taper off charging power during the last five to ten minutes of charging. 

 

Not even close Larry. The EVSE generates the pilot signal, not the car. The pilot signal lets the car the maximum current it can draw from the EVSE. For instance, 24A is represented as 40%.

 

If the charging power only tapers off during the last five to ten minutes then it wouldn't be practical for the EVSE to use that to guess the battery state of charge. I was thinking about a Tesla, where the power starts to taper off much much earlier.

[rbort]

Charging level tapers off in the last few minutes because the Li-ion cells have reached maximum voltage.  If the charge level was to stay up, they would go overvoltage.  Once they reach max programmed charge voltage (in the case of the Cmax I believe its 4.10v), the current (charge level) keeps dropping while pegged at 4.10v until it drops to just about nothing, which means the batteries are full.

 

-=>Raja.

[Hybridbear]

The charge rate for the level 1 charger is about 0.15 C and for the Level 2 Charger is about 0.4 C.  Both are between 0 and 1C.  Hence charging using a Level 2 Charger degrades the battery less than using a Level 1 Charger.  For the battery in their study, the optimal charge rate is 1C, which is approximately the full Level 2 charge rate of 6.6 kW. 

I don't know any BEV or PHEV that allows charging anywhere near 1 C other than via DC fast charging. The Focus Electric, for example, has a 23 kWh HVB and charges at around 6.6 kW. This is 0.29 C. The second-generation Volt has an 18.4 kWh HVB that only charges at 3.6 kW. That is less than 0.2 C. Fast charging a Nissan Leaf at 50 kW with its 30 kWh battery is 1.67 C. Tesla has been slowly increasing their Supercharger power levels, with some Supercharger locations being able to deliver up to 135 kW of power. Charging a 70 kWh Model S at 135 kW is 1.93 C, charging a 90 kWh Model S at 135 kW is 1.5 C. The Model S can charge via L2 at a maximum rate of approximately 19 kW (80 amps * 240 V), which is a very low C rate for the size of their HVBs. It's interesting that they say that 1 C is the optimal charging rate.

 

When driving the Energi can charge at a maximum rate of 35 kW which is about 5 C. When driving in the mountains we saw sustained charging rates around 15 kW when descending grades. That approximate 2 C charging rate caused very rapid increases in HVB temperature. The Focus Electric can charge the HVB at up to 60 kW when driving, although I've never seen the charging from regen exceed around 40 kW in my driving.

 

Ford should also use this information to minimize HVB degradation. The user should be able to plug in the car and then the car could determine the optimal charge profile to minimize electricity cost and simultaneously minimize HVB degradation, ensuring the car is fully charged by the next GO time.

 

The battery is very expensive. Each time you charge the HVB, it degrades a little more. There is a cost associated with this degradation. Eventually, you are going to have to replace the battery when it degrades too much. I would rather pay a little more for electricity if it saves me from having to replace the battery. It may be better to delay charging to a time when electric rates are more expensive if it significantly reduces degradation of the HVB. It may cost me an extra $1 to charge at the higher rates, but the additional degradation caused by charging in lower cost electric rate windows may be $10. The user should be able to plug in the car at any time and the car will take care of the rest to minimize my overall cost. If properly implemented, this could significantly increase the HVB lifetime by years.

This is a great idea. I've often wished that instead of the MFM value charge profile starting at the same time each morning that it would instead start at whatever time is needed in order to be done charging shortly before the Go Time. I've tried to limit the time at a full charge by delaying the start of charging until 3:00 am as well. However, some days the Focus Electric only needs 30-50 minutes to charge & other days it needs 3 hours. I'd like to tell it to not start charging until closer to 5:00 am, but then the car would think that it couldn't complete charging before my Go Time on some days so it would charge earlier. So, my compromise is 3:00 am every day.

 

I also don't have a time-of-use electrical plan, so I haven't set up a value charge profile, nor do I bother with waiting to plug the car in until I need to. Also, sometimes I drive it on weekends and sometimes I don't, so when I plug it in Friday night at home it'll charge up and stay at 100% all weekend. Luckily, I live in Wisconsin so the battery also never gets particularly hot.

Larry & I are both in MN & the HVB will routinely exceed 90 F and even 100 F even in our relatively cool weather this far north.

 

In another paper, they claim a Lithium-ion battery should not be operated at a level lower than ~25% SOC (This includes Hybrid Mode in our car.  When MyFord Touch states the SOC is 0%, the actual SOC of the HVB is around 20%.  In Hybrid Mode, the battery SOC ranges between 15% and 20%). 

 

"If lithium ion cells are discharged or operated at a level lower than ~25% SOC, their efficiency and performance is degraded, plus significant heating and aging will occur".

25%, wow! The Focus Electric goes down to about 7% absolute SOC at the very bottom end and charges up to about 89-91% absolute SOC when fully charged. In the winter I get down below 15% absolute SOC at least once per week, if not multiple days per week. In the summer I rarely go below 25% absolute SOC, I rarely go below about 35% absolute SOC in the summer. I have seen measurable capacity loss in the Focus Electric in the 2+ years and 20,000 miles that I've driven it. The Fusion Energi rarely got that low, probably less than 12 times per year, and those days were usually grouped together when taking it on long trips.

[flyingcheesehead]

Larry,

 

Any thoughts as to charging efficiency as the battery degrades? I heard a person make the assertion that as the battery degrades the charging efficiency goes down. For example, if the available capacity was 6.0 kWh and it took 9.0 kWh to charge it, that when that battery gets down to, say, 5.0 kWh of usable capacity, it will take MORE than 7.5 kWh to charge it. I can't find anything either way. Any truth to this?