2016 Spark EV battery size/usable capacity....what the

Adding up the numbers, 14.5 kWh is 78% of the usable battery capacity. So the 100% usable capacity figure should be....18.6 kWh (14.5 / .78 = 18.6)?? The '16 Spark EV's battery is officially 19 kWh....or is it 18.4 kWh?

According to the official GM press release for the '16 Spark EV, it has a 19 kWh battery: http://media.chevrolet.com/media/us/...k-ev/2016.html

But this release says the '15 Spark EV (basically identical to the '16) has an 18.4 kWh battery: http://media.gm.com/media/us/en/chev...2015.tab1.html

So....wtf? Both the '15 and '16 Spark EV have LG Chem batteries, while the '14 Spark had the 21 kWh A123 battery. EPA ratings didn't change for MY16.

Did GM bump the size of the '16 Spark EV battery 0.6 kWh and just didn't bother to update the EPA numbers?

And regardless of whether the '16 Spark has an 18.4 or 19 kWh battery, it seems that the usable capacity is either 100% of the battery or 97%, both ridiculously high!

Usable is about 17.6 kWh, based on my estimation. I've never seen any official number.

The MY15/16 battery capacity is 18.4 kWh AFAIK.

Zoomit said:

Usable is about 17.6 kWh, based on my estimation. I've never seen any official number.

The MY15/16 battery capacity is 18.4 kWh AFAIK.

Click to expand...

I have a 2015 Spark EV. My average calculated battery capacity is 17.64 kWh for 34 charging over the last 3 months. Last year, when the car was new and the weather was much hotter, I averaged 18.68 kWh for 16 charging sessions. I will continue to collect charging session data this year, especially during the same period as last year, in order to see if I really had a drop in capacity or the difference is due to air temperature differences. My 2014 Spark EV is running about 1 kWh higher.

Official GM press release says the '16 Spark EV had a 19 kWh battery? So they upped it 0.6 kWh quietly?

GM said 19 kWh for 2015 as well. It's actually 18.4 kWh for both years.

It's hella confusing since 2014 had 21 kWh (bit more I think) and bit over 19 kWh usable; some said 19.5 kWh. Then GM said 19 kWh for 2015, but it's actually 18.4 kWh and bit over 18 kWh usable. I struggled with this in my blog posts until finally just guesstimating 19 kWh for 2014 and 18 kWh for 2015. Yes, 2015/2016 use almost 100% of battery capacity!

I don't think it's a "bit over 18 kWh usable". If you believe the percentage on the display, it's a little less than 18 kWh usable. The challenge is that we don't drive to the point that the battery quits, so it's hard to know if it'll quit exactly when the display says 100% used.

I ran the battery down just about all the way this time, and here are the numbers:


98% utilization and 18.1 kWh used. If I were to have used that last 2%, kWh used would have been probably either 18.3 or 18.4 kWh. So it definitely seems the '16 Spark uses just about all of its battery capacity, which is 18.4 kWh. It will be interesting how the battery holds up for people hanging onto their Sparks for a while.

bro1999 said:

I ran the battery down just about all the way this time, and here are the numbers:


98% utilization and 18.1 kWh used. If I were to have used that last 2%, kWh used would have been probably either 18.3 or 18.4 kWh. So it definitely seems the '16 Spark uses just about all of its battery capacity, which is 18.4 kWh. It will be interesting how the battery holds up for people hanging onto their Sparks for a while.

Click to expand...

I leased my 2015 Spark EV 2LT at the end of May in 2015. My data for 1000 miles (ODO 1400-2400 miles) covering July and August 2015 and 16 charging sessions gave a calculated average battery capacity of 18.68 kWh. Your number yields 18.47 kWh. Looks like they correlate. My data for 2000 miles (ODO 8410 - 10412 miles) covering Mar, Apr, May and part of June 2016 and 36 charging sessions is 17.66 kWh. I will be looking at my charging data for July and August 2016 to make a same-time-of-year comparison.

MrDRMorgan said:

bro1999 said:

I ran the battery down just about all the way this time, and here are the numbers:


98% utilization and 18.1 kWh used. If I were to have used that last 2%, kWh used would have been probably either 18.3 or 18.4 kWh. So it definitely seems the '16 Spark uses just about all of its battery capacity, which is 18.4 kWh. It will be interesting how the battery holds up for people hanging onto their Sparks for a while.

Click to expand...

I leased my 2015 Spark EV 2LT at the end of May in 2015. My data for 1000 miles (ODO 1400-2400 miles) covering July and August 2015 and 16 charging sessions gave a calculated average battery capacity of 18.68 kWh. Your number yields 18.47 kWh. Looks like they correlate. My data for 2000 miles (ODO 8410 - 10412 miles) covering Mar, Apr, May and part of June 2016 and 36 charging sessions is 17.66 kWh. I will be looking at my charging data for July and August 2016 to make a same-time-of-year comparison.

Click to expand...

I contributed to a kickstarter for the Juiceplug, which will allow me to get detailed charging details for both my Volt and Spark EV. Hopefully my unit ships before too much longer! https://www.kickstarter.com/projects/emw/juiceplug-a-universal-smart-ev-charging-adapter

SparkevBlogspot said:

GM said 19 kWh for 2015 as well. It's actually 18.4 kWh for both years.

It's hella confusing since 2014 had 21 kWh (bit more I think) and bit over 19 kWh usable; some said 19.5 kWh. Then GM said 19 kWh for 2015, but it's actually 18.4 kWh and bit over 18 kWh usable. I struggled with this in my blog posts until finally just guesstimating 19 kWh for 2014 and 18 kWh for 2015. Yes, 2015/2016 use almost 100% of battery capacity!

Click to expand...

Just to mix this up a bit more - My 2014 Spark EV 2LT, which only had 1500 miles on it when I bought it last year, gave me an average battery capacity number of 19.91 kWh for 5 charging sessions in August of last year. So far this year, for Mar, Apr, May and June, I am averaging 18.88 kWh for 20 charging sessions. That is about 1 kWh more in 2016 than my 2015 Spark EV 2LT which is averaging 17.66 kWh for the same time period in 2016. Calculations are made using the info on the car's display. I am using the data to see if I can detect any battery degradation trends.

Because the battery itself is not 100% efficient, what you measure and what's reported by the car should be derated by the battery efficiency. If you assume 95% efficient, 18.1kWh you measured is actually 17.2 kWh stored at the battery.

Even with discharge method, you never know the full capacity without knowing the battery efficiency. Let's say 88 miles range at 5 mi/kWh = 17.6 kWh (62 MPH test by Tony Williams). If you consider battery discharge is 95%, full capacity is 18.5 kWh. Leave a bit for margin of error, and full capacity would be 18.4 kWh. Yeah, SparkEV does seem to use close to 100% of battery.

Without knowing battery efficiency, whatever you measure will always be off by some. I think 95% efficient is reasonable approximation with error of 1% or 2%. Battery efficiency would also change over time (wear) and method (ie, L2 vs DCFC). When I use DCFC, extrapolating "full" only come out to 16.5 kWh, but close to and over 18 kWh with L2, though not sure if that's due to different measuring devices.

Instead of trying to measure absolute capacity, relative capacity is what you're after.

SparkevBlogspot said:

Because the battery itself is not 100% efficient, what you measure and what's reported by the car should be derated by the battery efficiency. If you assume 95% efficient, 18.1kWh you measured is actually 17.2 kWh stored at the battery.

Even with discharge method, you never know the full capacity without knowing the battery efficiency. Let's say 88 miles range at 5 mi/kWh = 17.6 kWh (62 MPH test by Tony Williams). If you consider battery discharge is 95%, full capacity is 18.5 kWh. Leave a bit for margin of error, and full capacity would be 18.4 kWh. Yeah, SparkEV does seem to use close to 100% of battery.

Without knowing battery efficiency, whatever you measure will always be off by some. I think 95% efficient is reasonable approximation with error of 1% or 2%. Battery efficiency would also change over time (wear) and method (ie, L2 vs DCFC). When I use DCFC, extrapolating "full" only come out to 16.5 kWh, but close to and over 18 kWh with L2, though not sure if that's due to different measuring devices.

Instead of trying to measure absolute capacity, relative capacity is what you're after.

Click to expand...

Based off what I've observed with my Volt and the car's kWh used figure (and comparing to Chargepoint usage logs/Onstar charging logs), the Spark's kWh used figure should be fairly accurate too.

So it's probably pretty safe to assume the 18.1 kWh figure I observed was actually pulled out of the battery. So if the '16 Spark's overall battery capacity is actually 18.4 kWh (and not 19), it pretty much uses 100% of the capacity.

That makes me think that the Bolt will use a pretty high % of its battery's total capacity, especially since it will have the latest cells from LG Chem.

Without knowing the battery efficiency, no amount of measurement accuracy will give you absolute scale accurate result. You can measure what happened at the battery terminal, but what's actually in the battery is a mystery and we have to go by what's given in spec. If you put in 18.1 kWh (assuming you know the battery efficiency), what got pulled out will be less than that. As I mentioned, 95% efficiency would be 17.2 kWh that got pulled out.

If you assume 95% each way, it would be even less, but based on Tony Williams range test, I suspect about 95% is both ways. It could actually be even higher, though there will be some variations between cars.

Again, without knowing the battery efficiency, best you can hope for is relative metric. In some cases, you might end up measuing more than 18.4 kWh when charging.

SparkevBlogspot said:

Without knowing the battery efficiency, no amount of measurement accuracy will give you absolute scale accurate result. You can measure what happened at the battery terminal, but what's actually in the battery is a mystery and we have to go by what's given in spec. If you put in 18.1 kWh (assuming you know the battery efficiency), what got pulled out will be less than that. As I mentioned, 95% efficiency would be 17.2 kWh that got pulled out.

If you assume 95% each way, it would be even less, but based on Tony Williams range test, I suspect about 95% is both ways. It could actually be even higher, though there will be some variations between cars.

Again, without knowing the battery efficiency, best you can hope for is relative metric. In some cases, you might end up measuing more than 18.4 kWh when charging.

Click to expand...

I'm gonna see if I can charge my Spark from empty to full at a CP station. There is one a 1/4 mile from my house, but for some reason it is not networked, so no way to pull charging logs.

My Volt's charging losses have been observed between 15-18% (~10.4 kWh reported used by car, ~12.5 kWh used to charge back full) charging on L2, so I would expect the Spark would have similar losses since it probably uses the same onboard charger as my Volt, and has the same TMS system.

How do you factor the energy used by the TMS while charging?
It would be different seasonally.

SparkevBlogspot said:

Without knowing the battery efficiency, no amount of measurement accuracy will give you absolute scale accurate result. You can measure what happened at the battery terminal, but what's actually in the battery is a mystery and we have to go by what's given in spec. If you put in 18.1 kWh (assuming you know the battery efficiency), what got pulled out will be less than that. As I mentioned, 95% efficiency would be 17.2 kWh that got pulled out.

If you assume 95% each way, it would be even less, but based on Tony Williams range test, I suspect about 95% is both ways. It could actually be even higher, though there will be some variations between cars.

Again, without knowing the battery efficiency, best you can hope for is relative metric. In some cases, you might end up measuing more than 18.4 kWh when charging.

Click to expand...

The usable capacity of a battery is invariably taken as the energy available at the battery terminals with a specific load, it is not the chemical energy within the battery. For Lithium Ion batteries it is commonly taken at the 1C rate.

The battery efficiency is thus already factored in and is effectively taken as part of the charging efficiency. You don't need to know it to measure the usable capacity.

For the first year I had my Spark EV I measured the AC energy in and the energy displayed on the energy screen. My average for the year was an AC in to battery out efficiency of 83%. (This includes charger losses, battery losses, TMS consumption and other parasitic losses).

Lithium Ion batteries are commonly ~95% efficient between energy in and energy out. One related figure where they excel is in the coulombic efficiency which is usually better than 99.9% - the A/h out is pretty much the same as the A/h in.
The main energy loss is that the discharge voltage is lower than the terminal voltage during charging.

Lead-Acid and nickel based chemistries are much worse and may only be 70-90% efficient on this basis.

This is the effective capacity of my Spark over 20,000 miles.

kevin

kevin said:

SparkevBlogspot said:

Without knowing the battery efficiency, no amount of measurement accuracy will give you absolute scale accurate result. You can measure what happened at the battery terminal, but what's actually in the battery is a mystery and we have to go by what's given in spec. If you put in 18.1 kWh (assuming you know the battery efficiency), what got pulled out will be less than that. As I mentioned, 95% efficiency would be 17.2 kWh that got pulled out.

If you assume 95% each way, it would be even less, but based on Tony Williams range test, I suspect about 95% is both ways. It could actually be even higher, though there will be some variations between cars.

Again, without knowing the battery efficiency, best you can hope for is relative metric. In some cases, you might end up measuing more than 18.4 kWh when charging.

Click to expand...

The usable capacity of a battery is invariably taken as the energy available at the battery terminals with a specific load, it is not the chemical energy within the battery. For Lithium Ion batteries it is commonly taken at the 1C rate.

The battery efficiency is thus already factored in and is effectively taken as part of the charging efficiency. You don't need to know it to measure the usable capacity.

For the first year I had my Spark EV I measured the AC energy in and the energy displayed on the energy screen. My average for the year was an AC in to battery out efficiency of 83%. (This includes charger losses, battery losses, TMS consumption and other parasitic losses).

Lithium Ion batteries are commonly ~95% efficient between energy in and energy out. One related figure where they excel is in the coulombic efficiency which is usually better than 99.9% - the A/h out is pretty much the same as the A/h in.
The main energy loss is that the discharge voltage is lower than the terminal voltage during charging.

Lead-Acid and nickel based chemistries are much worse and may only be 70-90% efficient on this basis.

This is the effective capacity of my Spark over 20,000 miles.

kevin

Click to expand...

Can I ask how you measured the amount of energy being put into the Spark's battery when recharging? Were you charging 120V or 240?

The '14 Spark has the A123 battery, while the 15/16 Sparks have the a LG Chem battery (same cells as the ones used in 2013+ Volts). I think it is safe to say the LG battery is superior to the A123 one (based off the Gen 1 Volt battery's bulletproof reliability), and noting that the A123 battery has a much larger buffer than the 15/16 HVB, which we've pretty much determined uses almost all of its rated capacity.

The 15/16 Sparks also have a superior TMS setup, with cooling fins integrated in between the battery cells, compared to the 14's much simpler cooling pan setup.

bro1999 said:

kevin said:

SparkevBlogspot said:

Without knowing the battery efficiency, no amount of measurement accuracy will give you absolute scale accurate result. You can measure what happened at the battery terminal, but what's actually in the battery is a mystery and we have to go by what's given in spec. If you put in 18.1 kWh (assuming you know the battery efficiency), what got pulled out will be less than that. As I mentioned, 95% efficiency would be 17.2 kWh that got pulled out.

If you assume 95% each way, it would be even less, but based on Tony Williams range test, I suspect about 95% is both ways. It could actually be even higher, though there will be some variations between cars.

Again, without knowing the battery efficiency, best you can hope for is relative metric. In some cases, you might end up measuing more than 18.4 kWh when charging.

Click to expand...

The usable capacity of a battery is invariably taken as the energy available at the battery terminals with a specific load, it is not the chemical energy within the battery. For Lithium Ion batteries it is commonly taken at the 1C rate.

The battery efficiency is thus already factored in and is effectively taken as part of the charging efficiency. You don't need to know it to measure the usable capacity.

For the first year I had my Spark EV I measured the AC energy in and the energy displayed on the energy screen. My average for the year was an AC in to battery out efficiency of 83%. (This includes charger losses, battery losses, TMS consumption and other parasitic losses).

Lithium Ion batteries are commonly ~95% efficient between energy in and energy out. One related figure where they excel is in the coulombic efficiency which is usually better than 99.9% - the A/h out is pretty much the same as the A/h in.
The main energy loss is that the discharge voltage is lower than the terminal voltage during charging.

Lead-Acid and nickel based chemistries are much worse and may only be 70-90% efficient on this basis.

This is the effective capacity of my Spark over 20,000 miles.

kevin

Click to expand...

Can I ask how you measured the amount of energy being put into the Spark's battery when recharging? Were you charging 120V or 240?

The '14 Spark has the A123 battery, while the 15/16 Sparks have the a LG Chem battery (same cells as the ones used in 2013+ Volts). I think it is safe to say the LG battery is superior to the A123 one (based off the Gen 1 Volt battery's bulletproof reliability), and noting that the A123 battery has a much larger buffer than the 15/16 HVB, which we've pretty much determined uses almost all of its rated capacity.

The 15/16 Sparks also have a superior TMS setup, with cooling fins integrated in between the battery cells, compared to the 14's much simpler cooling pan setup.

Click to expand...

All of this is good information but I think Kevin's graph is what we are trying to understand - how fast does the battery's capacity degrade. I only have 6000 miles on my 2014 but my capacity calculations seem to fit his graph at the start and 5k miles points. I really would like to see the same for the 2015 and 2016 Spark EVs since, as you mentioned, the battery cooling in the 2015 and 2016 Spark EV is better than that in the 2014. Right now, at 10.5k miles, my 2015 Spark is tracking about 1 kWh below my 2014.

MYSTERY SOLVED!! Well, maybe.

I got in contact with one of the moderators on the gm-volt.com forum (he works for GM...probably some kind of engineer), and he told me the 2015 and 2016 Spark EV batteries have 19.44 kWh total energy. He said the press releases of 18.4 and 19 kWh were probably "typos".....so GM never bothered to correct them? Wtf?

He also said under ideal conditions, the battery can exceed 20 kWh under ideal charging conditions.

So assuming the 15/16 Spark batteries have about 18.5 kWh usable, that still means almost 95% of the total battery capacity is utilized.

SparkevBlogspot said:

Can I ask how you measured the amount of energy being put into the Spark's battery when recharging? Were you charging 120V or 240?
...

Click to expand...

I used the energy measured by Chargepoint for L2 (240V) and by a Kill-A-Watt when charging at L1 (120V 12A).

The majority of my charges were at L2 at work, I only did half a dozen at home in the first year.

For the last 6 months I have been using L1 much more often as I moved to a building that didn't have charging points, luckily in the last couple of weeks some have been installed.

I purchased a Bosch EVSE with the $500 discount when I got the car but have not found the need to install it - I can recharge a full 60-70 miles worth of energy overnight at 12A (which I usually set it to in spite of the annoyance that the setting is not persistent) and that is adequate.

kevin