Pack capacity, SOC and grid balancing

[mikemo]

I'm curious on how the computer behaves when a pack is grid balanced. It seems to me that as the battery is used the computer is constantly trying to keep it from over or under charging, all the while trying to compute the usable capacity.

My battery was replaced by Honda 10/2009. It is still working fine, but I have noticed that using heavy assist seems to deplete the SOC display more than it used to. I assume that is because the computer thinks the available capacity is less than when it was new.

I will also assume that reduction in capacity is mostly due to imbalance in the cells. If I do a grid-balance and restore the original capacity, won't the computer continue to think that the available capacity of the pack is still at least as low as it was before balancing? If that is the case then I would not see any improvement.

Let's say the computer thinks the SOC is 50%, then you grid charge it. The computer has no idea that you added charge, and when it tries to charge it will detect a full pack and positive re-cal. Won't this reduce the computer's perceived capacity even more???

Should the computers be reset somehow after grid balancing to "start over" and assume the battery is new? Is there a procedure to do that?

BTW, I use the car daily and it still returns 65-80mpg in my commute (depending on lights/traffic). The car just turned 70k miles.

Thanks
Mike M.

[E27006]

The panel gauge for displayed S of C is nothing more than a finger in the wind. You might as well ignore it.

The OBD11 gauge gives the driver a proper picture of the IMA system, Sof C, pack volts, Assist/Regen current, IMA background charging, 12 Volt DC battery background charging.

Example, IMA charging at 5 to 10 amps for several miles, Sof C rising on OBD11, yet zero bars on charge gauge and very little movement on battery gauge

[talonmike]

You can check the ima system with an obd2 reader?


What software are you using?

[AClewell]

IMA is only available on the ODBII C&C.

[mikemo]

Originally Posted by E27006

The panel gauge for displayed S of C is nothing more than a finger in the wind. You might as well ignore it.

Thanks. I understand that, but it really doesn't answer my question of how the computer handles the changes in actuall SOC from grid balancing.

[Cobb]

I never got an answer to the question I posed some time ago about a similar issue. Where is the SOC meter getting its data if the real SOC is available on the ODB2 bus and why doesn't Honda change it?

Its my understanding it guestimates the capacity using a count of amps in and out. Then if it detects too high or low of voltage it does a recal, negative or positive.

As a newer Insight owner its my assumption the system is too slow at monitoring the pack and under regular use it routinely goes too low or high with stop n go driving.

I am guessing this is done on purpose to ensure the battery isnt charged too much or used too much.

What I find neat is that once a recal occurs, the car sems to not use any IMA for a minute, then when it is retroed you hear the blower wind up as we have a variable speed unit.

Sent from my Autoguide iPod touch app

[Eli]

Originally Posted by mikemo

Let's say the computer thinks the SOC is 50%, then you grid charge it. The computer has no idea that you added charge, and when it tries to charge it will detect a full pack and positive re-cal. Won't this reduce the computer's perceived capacity even more???

Should the computers be reset somehow after grid balancing to "start over" and assume the battery is new? Is there a procedure to do that?

No. The computer won't do a negative recal until the pack is actually empty. So if you grid charge at 50% SoC, it will do a positive recal the next time you drive, then you have all of that "extra" capacity(however much that may be) before the pack does a negative recal. Then on the next charge cycle, you'll theoretically be able to put it all back in.

Resetting the systems won't hurt anything, but it's not really necessary. To do this, just pull the negative 12V cable. It will start the background charge, and positive recal once it realizes the battery is full. No different than if it did it at 50% SoC without the reset.

Essentially, a negative/positive recal cycle IS a system reset. It's recalibrating to your new capacity.

[Cobb]

When I grid charged my i2 the charge seemed to past 6-9 miles depending on how I drove, then it went to normal operation. For the first few miles it was rather assist happy and did no regen on braking.

If you are doing this for fuel economy, you should consider a battery maintainer or charger for the 12 volt side. If you do this, this will further your mpg as in that case the charger system will only be on for the first 2 minutes and even then it wont do much as the battery is already charged.

Sent from my Autoguide iPod touch app

[mikemo]

Originally Posted by Eli

No. The computer won't do a negative recal until the pack is actually empty.

I'm not trying to argue, but I still don't understand. I thought that the computer tried to never let the pack get completely full or completely empty. It has to estimate pack SOC because there is no absolute physical signs that the pack is in mid-charge. It integrates energy (joules, watt-hours, AHV) and does its best to keep track of where it "thinks" the SOC is.

Negative recal happens at cell reversal, which means at least one cell is completely empty. When that happens the computer has to readjust its SOC, but also its calculated total pack capacity.

So let's say that the computer thinks that the pack has 5AH capacity as it has aged some. It currently thinks that there is 2.5AH capacity remaining, then you grid-charge. The next time you drive it quickly finds that the pack is completely full (not sure which criteria it used to decide this, temperature or voltage). You are saying that it won't adjust its estimation of total pack capacity down? So how does it know what the total capacity is if it doesn't make assumptions after hitting the limits.

I'm only going on my experience with my car and my knowledge of NiMH batteries. If someone could clarify this for me I would appreciate it.
Mike M.

[retepsnikrep]

The problem is Mike as we don't have access to the BCM software we can only make educated guesses as to what it is doing and what it's parameters and logic may be.

Reading the advanced diagnostics info and other documents on this site relating to BCM/ fault codes may help you understand how the available capacity is calculated and it gives a bit of an insight into some other specs.

IMHO if grid charging you should clear the BCM by pulling the fuse every so often, especially when first grid charging, rather than relying on the reset soc function of the OBDIIC&C for instance, as this will clear the BCM capacity memory and hopefully allow the available useable capacity window to be reset. This useable capacity window and recalibration (Not pos/neg recals) of the soc gauge range is the least well understood part of the BCM operation ASFAIK.