Budget Battery Capacity Readout

[AnthonyW]

I just took a look at the Electric Powertrain print out from the dealer. It shows a Charging Voltage Target of 4091.6mV.
I saved a previously published EP report from this forum which showed a Charging Voltage Target of 4093.8mV.

My battery capacity is 49.9Ah, the battery capacity on the downloaded report (Autel) was 53.8Ah (VIN 9537) Could there be a relationship between the Charging Voltage Target and battery capacity? Or is there some variation between the readings from iHDS, Autel and CarScanner?

lincomatic, what is your battery capacity?

I am very excited about this developement. I baby my battery and rarely charge it above 85% or below 45%. When I had my capacity tested after the first 18 months I was at either 54.6 or 54.8. I am interested in seeing what it is now.
To answer your question, somewhere in the documentation it states that he battery capacity test should be done at the nominal voltage. For the NMO battery that is 3.7 volts (or 3700mV). Therefore the entire pack voltage should be around 311 volts. Also Honda designed the pack's optimal operating temp range to be between 25c to 35c (77f to 95f). You should make sure your pack is in the temp range when testing. I don't see that we have a PID for battery pack temp yet. From my tests with the Scanguage a few years ago it takes the temperature of the battery (in non-extreme temperature environments) a max of 6 to 8 hours to reach ambient. If your garage is in that temp range (probably not at this time of year) then you could test after an overnight. From my test the ambient temp (and the temp of the battery) needed both start out atleast at 70f and maybe possibly the batter temp would end up around 77 degrees after driving it close to empty.

 

[MrFixit]

To answer your question, somewhere in the documentation it states that he battery capacity test should be done at the nominal voltage. For the NMO battery that is 3.7 volts (or 3700mV).

Just keep in mind that this is NOT performing a test. It is simply reading out a value that Honda maintains on a running basis. It is probably heavily averaged, and may have 'adjustments' to account for temperature, etc... We will never know what is actually behind the number, but it IS the number that they apparently base their warranty on, so it is very relevant.

I wouldn't recommend trying to over-interpret this by attempting to read it out with specific conditions in mind. Better to just keep checking it periodically (every month?) to look for trends, etc. If there is some kind of trend that happens seasonally (temperature) maybe it will show up. I don't think you will be able to see anything on a short-term basis, or while trying to apply specific conditions.

 

[DucRider]

I baby my battery and rarely charge it above 85% or below 45%.

Pretty extreme to only utilize 5 kWh (30%) of a 17 kWh battery. There is no evidence that this will help your battery long term, but also unlikely to permanently harm it.

 

[AnthonyW]

Pretty extreme to only utilize 5 kWh (30%) of a 17 kWh battery. There is no evidence that this will help your battery long term, but also unlikely to permanently harm it.

Not extreme if that covered my daily commute. While it may be true that there is no evidence saying this helps the Clarity battery there are mountains of evidence that this greatly extends the life of a NMC lithium battery which is what our battery is. Therefore not a stretch to believe this would extend the life of the Clarity battery.

 

[WolfpackCL]

if you guys are really interested in CAN messages I recommend checking out comma.ai - they created a self driving dash cam and custom CAN message reading tool along with an improved OBD reader called a “panda”. there is extensive CAN decoding work done for a large selection of cars and a community maintained honda clarity port. at the very least come to the discord channel and ask your questions, you may save yourself a lot of time https://discord.gg/bCUZya4s

@MrFixit

 

[Eva Farkas]

Hi everyone!

Let me begin with MANY thanks for all the work being done to help our Clarity community get this elusive, yet vital data. @lincomatic and @MrFixit - I see you and thank you!

Okay, a quick (stupid) question, followed by a long, long explanation of my circumstances. Hopefully that will help add to the info about Clarity battery decline after MANY MANY kms.

Question: Should I (or is it optimal) to run the Battery Capacity Test with a full charge? Or can I run it at any time?

I have purchased the Vgate BLE 4 OBD2 Adapter (arriving today!) and I'll download both Carscanner and Torque. I will post my findings as soon as I can.

Background:
2018 Honda Clarity (Touring) - purchased Feb 2018
Avg. monthly usage - 4000+ km (I drove Uber, now doing Doordash/SkipTheDishes)
Current mileage - 134,000 km
Extended warranty purchased (to 160K km) - I expect to hit this number by end of June 2021!!!

Usage pattern - approx. 150-200 km per day, short delivery trips, mostly city driving
Mode most often used - HV mode (almost exclusively, due to a lack of charging availability) - I try to keep the power in the "blue EV" zone when driving in HV
Charging pattern - almost no charging (explanation below)

We live in an older condo (in Toronto, Canada). When we bought the car (after years of wanting a hybrid or electric, but not buying one because of no charging in our building), I put together a 30 pg proposal to our Board of Directors on why/how/when they should install chargers. I also asked for a regular outlet in the interim (better than nothing!). I got neither. I'm still fighting the fight! Sigh.

So, that means I have been driving this car 40-50K per year with only occasional charging "in the wild". Generally because that means I have to sit in the car to charge (especially now, with no malls open), and often I'm doing this while waiting between delivery orders. So my charging is in short stints (very sub-optimal, I know).

History of battery capacity (as displayed in vehicle, by range):

Feb 2018 - initially we were getting great capacity, even with it being the middle of winter. 75-80 kms on a full charge was the norm. This continued well through the summer and into the following winter.

post-2019 winter - The range dropped dramatically after the "2nd" winter. This followed the software updates that corrected range, so I assumed it would even out over time. It hasn't. Range was more like 60-65 kms on a full charge

2020 - present - Our range for the past year + is never more than 52 km on a full charge.

When driving HV (which is almost always for me), it generally hangs around 25 km, even with using Regen braking a huge percent of the time. I am gentle on the pedal, glide as much as possible, try to keep heat and a/c to reasonable levels (though I'm working in the car, so I do keep myself comfortable!). I do on occasion use the charging mode while driving, but from an economics pov, that makes no sense. It's cheaper to buy the gas to drive the distance. Mostly I do it because I like to keep a "baseline" of 20 km range so that it adds to my overall efficiency (and the angry bees are buzzing the whole time I drive). I work long hours and just don't have the energy to sit and charge for 2+ hrs after each shift. Oh if I could only get a damn regular outlet at home! ARGH!

Now the real kicker......we are leasing this car (with original plan to buy at the end of lease). We prepaid kms to 160K (knowing we'd use at least that much with my gig work). Our buyout is about $10K (CDN), but if I keep up my current driving pattern, we'll probably be 28K over our lease, and that would cost us $4500 in km penalties. So yes, of course, buy the car at that point.....

But what if the battery is NOT under warranty replacement within the next few months (as per the Battery Capacity readout), maybe just shy of it.....and I go out of warranty.

Then the buyout becomes a much different financial consideration (factoring in the possibility of a new battery needed in a year/two/three)?

So, as you can see, the Battery Capacity readout is more than just an intellectual consideration for me, especially in the next few months. I need to know NOW, what my status is. And also, I guess, if I'm borderline, how do I make the battery "worse" so I can get that warranty replacement before I hit 160K in the next 4 months.

Sorry for the extremely long post guys..........

 

[MrFixit]

Question: Should I (or is it optimal) to run the Battery Capacity Test with a full charge? Or can I run it at any time?

As I have stated before...
This battery capacity readout is NOT a "test".

The Clarity does something continually in the background, likely an integral part of the Battery Management System (BMS) to track battery capacity. This is undoubtedly a highly averaged value over a variety of conditions experienced during normal operation.

As such, don't think of this as a "test" where you can "tinker" with the conditions and maybe influence the outcome.

It is just a readout of a very complicated process that Honda Engineering mechanized as part of their background Battery Management process.

 

[Hoon]

Question: Should I (or is it optimal) to run the Battery Capacity Test with a full charge? Or can I run it at any time?

I think you should at least do one or two full charges and discharges so the battery gauge is calibrated. Then you will get more realistic values of what the car thinks your battery has.

It might actually be good that your battery wasn't cycled too much, and there is a bottom guardband that prevents you from over-discharging the batteries. Since you are driving the car frequently, the battery voltage is not dropping any further like a car sitting in a dealer lot. One of the advantages your battery might have is that it wasn't at high voltage levels so it would have less aging from high voltage levels.

 

[MrFixit]

History of battery capacity (as displayed in vehicle, by range):

Let me also point out that your extremely complicated scenario is precisely why we felt it was so important to expose this battery capacity measurement. We have seen repeatedly in this forum that many complain about range, but there are SO MANY variables to the range estimator and with range performance that it is just not a good way to determine whether your battery is responsible. People are quick to blame the battery when it could be any one of a dozen factors, some you might not even think of.

To me, this is the the way to know how 'good' your battery is. Honda is hanging their hat on this. Look at the battery capacity... It WILL show some decline from the original 55 Ah. If that is enough to explain what you perceive as a range loss, then you have the answer. If it is not, then forget about the battery and look elsewhere.

Going forward inputs from the Clarity community will be invaluable a we learn more about the battery behavior over time.

 

[MrFixit]

I think you should at least do one or two full charges and discharges so the battery gauge is calibrated.

I will respectfully disagree with this. Don't try to "out-engineer" the Honda battery engineers, and guess what may or may not affect this number and what you might do to get a "better" answer. All of that is 100% guesswork and doesn't help with any understanding.

Take it at face value under the conditions that you operate the vehicle in. Read it out periodically (monthly?) - it is a long-term kind of a thing.
We will learn over the long haul how this battery system behaves.

 

[Hoon]

I will respectfully disagree with this. Don't try to "out-engineer" the Honda battery engineers, and guess what may or may not affect this number and what you might do to get a "better" answer. All of that is 100% guesswork and doesn't help with any understanding.

Take it at face value under the conditions that you operate the vehicle in. Read it out periodically (monthly?) - it is a long-term kind of a thing.
We will learn over the long haul how this battery system behaves.

You don't think there's any battery calibration involved? If you took what the voltage levels were at 100% a year ago and keep your battery all the time between 20%-30%, then that 100% reference from a year ago is outdated. I don't think there can be an accurate simulation on the aging as the battery is chemistry and with so many variables to change things. I don't see this as trying to out-engineer anyone. I don't know if charging to 100% just once would be enough data for the BMS to guess more correctly, that's on the Honda engineers.

 

[MrFixit]

I don't see this as trying to out-engineer anyone

Do what you want to do.
I am going to drive my car just as I normally do and keep an eye on this parameter over the long-term.

 

[Eva Farkas]

As I have stated before...
This battery capacity readout is NOT a "test".

The Clarity does something continually in the background, likely an integral part of the Battery Management System (BMS) to track battery capacity. This is undoubtedly a highly averaged value over a variety of conditions experienced during normal operation.

As such, don't think of this as a "test" where you can "tinker" with the conditions and maybe influence the outcome.

It is just a readout of a very complicated process that Honda Engineering mechanized as part of their background Battery Management process.

Yes, I am clear that it's a readout, not a test. I was inaccurate in the language in my post! (sheepish grin)

As to the "tinker", I suspected there wasn't much that could be done to sway things up or down, but hey, it doesn't hurt to ask? LOL As it stands, I don't know my current reading, so I may be well within the ideal range, I could be on the "edge" of a warranty swap, or I could even be actually in swap territory now! I won't know until I do my first reading (hopefully today or tomorrow!).

Thanks again for your help. I'll no doubt have more questions and postings on the topic going forward. Cheers.

 

[Eva Farkas]

You don't think there's any battery calibration involved? If you took what the voltage levels were at 100% a year ago and keep your battery all the time between 20%-30%, then that 100% reference from a year ago is outdated. I don't think there can be an accurate simulation on the aging as the battery is chemistry and with so many variables to change things. I don't see this as trying to out-engineer anyone. I don't know if charging to 100% just once would be enough data for the BMS to guess more correctly, that's on the Honda engineers.

So given there there is no way to no for SURE if calibration is a factor in the Battery Capacity readout....

I'm going to do a reading straightaway when I get my Vgate BLE (without any charging prior). I will post my results here.

Then, when I have the opportunity, I will do a full charge/discharge/full charge and complete another reading. And again, I will post my results.

It will be interesting to see if there is any substantial difference in the readings (and other values I assume the Vgate BLE will produce using Carscanner and Torque).

I'm on the geeky side, but perhaps not quite as immersed in things as some on this forum (some of the tech talk here is "greek" to me - LOL). So forgive me if I'm not as precise in my terminology etc.

 

[lincomatic]

I agree w/ @MrFixit. Testing Battery Capacity doesn't seem to be affected by charge level, deep cycling, etc. I've been testing it for the past week or more, at various SOC, and it always returns exactly the same result: 52.78 Ah. My car is driven about 24mi/day on weekdays, and varying amounts on weekends. I charge it up to full every night.

As for another post I saw, positing that the cells can somehow be "reconditioned" by deep cycling them, I think what's really happening with the cars that show improvements is that battery packs were initially very out of balance, and doing the full discharge/charge cycles triggered the BMSs to balance the cells more evenly. Lithium ion cells don't have the kind of memory effect that we used to see with older battery chemistries, such as NiCD

 

[Hoon]

Just watched a virtual seminar from ST Micro and thought I'd share.

This slide showing 2 mechanisms of battery wear (cycling & calendar aging).




Slide showing how BMS controls the depth of discharge to provide more consistent usable range to the end-user (at the expense of more wear) until the end-of-life. There's no definition of "end-of-life", kind of like "lifetime fluid".
But at least now that we can read out the actual capacity, that part is transparent.

 

[Gary1]

First off I wanted to thank MrFxit and lincomatic for there efforts in figuring out how to decode this. I'm guessing this took a significant amount of time do but it's very much appreciated.

Just to share I have a 2018 that is 31 months old, with a little over 31k miles. We drive it 92% on EV and the Battery Capacity is showing 50.1Ah.

 

[lincomatic]

Just watched a virtual seminar from ST Micro and thought I'd share.

Slide showing how BMS controls the depth of discharge to provide more consistent usable range to the end-user (at the expense of more wear) until the end-of-life. There's no definition of "end-of-life", kind of like "lifetime fluid".
But at least now that we can read out the actual capacity, that part is transparent.

Interesting presentation, but the BMS behavior is really up to the implementer. I'm guessing the algorithm they're showing is how one of their IC's behaves, perhaps for small consumer electronic devices. My 2001 Nissan Leaf actually does the opposite. It keeps the depth of discharge fixed throughout the life of the battery. As the battery has been aging, the SOC at full charge has been going down (100% SOC was originally 100% on the "fuel gauge," now it's only 94%), and it the regen is severely limited almost to the point of being useless, further limiting its range.

 

[Hoon]

Interesting presentation, but the BMS behavior is really up to the implementer. I'm guessing the algorithm they're showing is how one of their IC's behaves, perhaps for small consumer electronic devices. My 2001 Nissan Leaf actually does the opposite. It keeps the depth of discharge fixed throughout the life of the battery. As the battery has been aging, the SOC at full charge has been going down (100% SOC was originally 100% on the "fuel gauge," now it's only 94%), and it the regen is severely limited almost to the point of being useless, further limiting its range.

The ST Micro presentation was on a EV topic, so I'm pretty sure they are talking about cars there. True that the behavior would depend on the implementation.

On your 2001 Nissan Leaf, could it be that the depth of discharge you are seeing is actually another layer? Just like what we see on Clarity's battery level (also shown in HondaLink app) as 100% is actually somewhere between 90%-95% or maybe even lower?

I wasn't recommending full-charge to recondition the cells, it was just for more accurate readings since @Eva Farkas sounded like the car was never fully charged for 3 years or any time recently. I would guess that it will actually wear the battery a little more since it is a deep charge and discharge, I didn't imply doing this to revive the cells. I don't know. It would be interesting to see how it reads in current state and then reading it again after one time of full charge. I would bet that you will get a more realistic reading after a recent full charge-discharge cycle which will feed that data to the BMS.

About babying the battery, all this may not matter if it's only a few % difference after few years but at least it's not that much of a hassle for me so I'm okay following through.

 

[JohnT]

First off I wanted to thank MrFxit and lincomatic for there efforts in figuring out how to decode this. I'm guessing this took a significant amount of time do but it's very much appreciated.

Just to share I have a 2018 that is 31 months old, with a little over 31k miles. We drive it 92% on EV and the Battery Capacity is showing 50.1Ah.

interesting - our 2018 is 34 months old about 36K miles, at least 92% EV and showing 50.3 -