Honda Insight Forum banner
1 - 12 of 12 Posts

·
Registered
Jack of All Trades
Joined
·
12 Posts
Discussion Starter · #1 ·
Hi Everyone,
I'm sure this has been discussed many times throughout the forums, but I can't quite seem to find the information I am looking for.

I just recently purchased a new 8.0Ah Bumblebee Battery for my insight, but the SOC seems to constantly operate below the halfway mark (usually around 40% on the OBDII C&C). I have even tried driving on the freeway in 4th gear in an effort to charge the battery, as well as grid charging it. It will register a full charge but the level will drop fairly quickly to around 40% after a few 2nd and 3rd gear pulls (while still staying above 140v rest voltage). It also has a difficult time getting above 50% if I drive conservatively and maximize regen while minimizing assist. The interesting thing is I will find the Ah in will exceed the Ah out by a couple Ah when I drive conservatively. The battery voltage also rests pretty high (around 160) and will drop to around (140) while pulling max assist, so to me this would indicate a near full battery pack even though the dash seems to think the battery is half charged.

So what gives? Do we have an Idea on how the factory computer calculates SOC? Is it looking for a rapid change in voltage or some other indicator that doesn't show up easily on a larger pack? Or is it calculating Ah and I am just too heavy footed? What are the indicators that I could read for a charged NiMih pack with the OBDII C&C (other than the SOC). I keep expecting it to do a positive recal but it's only done that a few times.

I'd like to figure this out as I am losing MPG with the car constantly background charging the pack, and I would like to maximize the service life of my new pack; and having the pack constantly being charged if it is nearly full, or operate in a narrow range of charge does not seem healthy. If I could determine the SOC myself I'd reset it with the OBD C&C; But I don't like the idea of the car draining the battery if my new SOC is wrong.
 

·
Premium Member
2001 5S "Turbo"
Joined
·
11,778 Posts
Sometimes I think the SOC has a mind of it's own. I've seen unexpected actions and I've said...NO WAY. Install a CARD (CALPOD) switch and reap it's rewards. IMHO
 

·
Registered
Joined
·
1,395 Posts
It is also possible that there is a weak cell in the pack which is causing the BCM to treat it like an aged, reduced capacity pack. However, Peter has had more experience than I with the aftermarket cells and a BCM issue may be possible but seems unlikely.

The car should not behave any differently with 8.0 Ah cells (except be able to provide assist longe from a full state).

Unfortunately the car provides nothing more than what you already have to determine this. I would contact Eli, as it would be best for him to suggest diagnostics that can be performed (I can't think of any simple ones that can pinpoint the issue.)
 

·
Registered
Jack of All Trades
Joined
·
12 Posts
Discussion Starter · #5 ·
It is also possible that there is a weak cell in the pack which is causing the BCM to treat it like an aged, reduced capacity pack. However, Peter has had more experience than I with the aftermarket cells and a BCM issue may be possible but seems unlikely.

The car should not behave any differently with 8.0 Ah cells (except be able to provide assist longe from a full state).

Unfortunately the car provides nothing more than what you already have to determine this. I would contact Eli, as it would be best for him to suggest diagnostics that can be performed (I can't think of any simple ones that can pinpoint the issue.)
I have also noticed the average pack temperature seems to run on the high side. Even on a cool night (75f) I might see 90f pack temps on the OBDII C&C. I'm thinking there could be an issue with one of the cell temp sensors or one cell is weak and is running hot like you mention. I'll get in touch with Eli and pick his brain on this, what is the best way to contact him?
 

·
Premium Member
2001 5S "Turbo"
Joined
·
11,778 Posts

·
Premium Member
Joined
·
808 Posts
@Eli

I have the exact same issue, 8.0 pack nearly brand new still has 4+ years warranty left. In addition to behaving exactly as you say it also sometimes drops from 50% to 25% in seconds and then recharges. Have no idea what's happening. I PM'd Eli about this several weeks ago but didn't hear back. Luckily I have the warranty but hopefully I won't have to use it.
 

·
Moderator
Joined
·
7,259 Posts
Sorry I missed your PM. I remember reading it now, but must have gotten distracted. Lots going on in the hive! I'll respond here first and then send you a PM.

It seems there are two types of drivers out there - those who are able to keep the battery at 50-100% SoC, and those who keep it at 0-50%.It's definitely better for the battery to keep it at 50-100%, so it would be interesting to get to the bottom of this. I've always been a 50-100% type - I can guarantee that if I drove these vehicles, the battery would fill up and stay there. The above percentages are dash cluster, since that's what most people have. If you have an OBDIIC&C, all the better.

So, let's start from the top. As for the thread title - in the context stated, there is nothing to 'calculate incorrectly'. The SoC percentage scale displayed by the BCM is just based on current in/out of the battery. The scale is 6500mAh / 100, so each percentage point represents 65mAh; 10% is 650mAh, etc. All of the magic happens at the top and bottom end of the usable SoC range - the "hang points".

I'm not sure if I understand your 'depletes quickly' comments - I think there must be a misunderstanding regarding what you were expecting. The battery installed in the car doesn't change the value of a percent of SoC. As stated, BCM SoC is calculated based on current in and out of the pack. So, if you start at 60% and draw 2Ah from the pack you will end up at 30% SoC.. every single time(assuming the pack has 2Ah to give). So then, how long it takes to go from 60-30% SoC depends entirely on current draw during the time in question. The car is capable of drawing the most current in 2nd and 3rd gears, so it makes sense that it would be depleted the quickest there. What was the expected behavior?

The temperature thing is interesting. It definitely isn't what you would expect at first thought given the internal resistance improvement, but I think it makes sense in the context of what is happening here. "My battery is constantly depleted" = "My battery is constantly being used". And it can deliver the power to boot, no throttling here - like there would be with a weak battery. IIRC, this was borne out in the original BetterBattery test data. The new cells ran as hot or even hotter than the OEM pack - because they were able to deliver full power for like 3x as long or whatever it was. So just by the numbers, it seems to "run hot", but the benefits are there. If an OEM pack could deliver the same power over the same period of time, it would be all the hotter.

I think a similar phenomenon is probably behind the feelings of "fast depletion" - it's easy to deplete a battery that can actually deliver lots of current!

The themes behind the (mostly) fixable things that can cause this are drag and engine power. If you have the wrong tires or a dragging brake, for example, do not underestimate how much of an impact this type of thing can have on where the SoC gauge "lives". When you press on the accelerator, except for when you very first start off, it should be pretty easy to let off the gas until assist goes away - but still accelerate. I know that's pretty subjective, but if you find it very difficult to accelerate without using tons of assist, it would be wise to start checking things out. The hardest thing to deal with is an engine that is just plain worn out. Insights with worn out engines will eat IMA batteries for breakfast and the occasional midnight snack. Not saying that is the case here, just trying to give scope.

If I were driving these hypothetical low/wrong tires/bad engine vehicles "for the battery"(as opposed to for mpg), the extent of the drag or lack of power would determine the MPG hit I'd take for that driving style. If I was driving for MPG, it would determine the average speed hit I would take at my chosen MPG... just to maybe help people to see the relationships between all of these things.

Anyway, one thing everyone can do to help keep the battery charged - and I DO highly recommend that everyone do this -

Drive with your headlights on! Even and perhaps especially during the day. Get used to flicking them on whenever you drive and soon it's like putting on your seatbelt. It makes a big difference in the Automatic Background Charge point, and will try to keep the SoC between 65 and 70% which is right where we want it. Anyone having this problem should try that first.

It would be best to make sure you 'start over' from from "actually full", especially if you've been experiencing this problem for a while. Starting with SoC below 65.0%, rev the engine to around 3100RPM - I think that should dial up around 15A of regen. If you have an OBDIIC&C, you should see SoC climb and hang at 72.0% while the battery continues to fill. Hold there until regen stops. At this point BCM SoC should be either 75.0% or 80.0% depending on BCM revision. If you have a grid charger, it would be good to input an additional ~1500mAh at this point, so 4-5 hours if it inputs 350mA.

Then drive(with your headlights on of course) and play the "How high can I maintain SoC while getting the best MPG" game. :D
 

·
Registered
Jack of All Trades
Joined
·
12 Posts
Discussion Starter · #9 ·
Sorry I missed your PM. I remember reading it now, but must have gotten distracted. Lots going on in the hive! I'll respond here first and then send you a PM.

It seems there are two types of drivers out there - those who are able to keep the battery at 50-100% SoC, and those who keep it at 0-50%.It's definitely better for the battery to keep it at 50-100%, so it would be interesting to get to the bottom of this. I've always been a 50-100% type - I can guarantee that if I drove these vehicles, the battery would fill up and stay there. The above percentages are dash cluster, since that's what most people have. If you have an OBDIIC&C, all the better.

So, let's start from the top. As for the thread title - in the context stated, there is nothing to 'calculate incorrectly'. The SoC percentage scale displayed by the BCM is just based on current in/out of the battery. The scale is 6500mAh / 100, so each percentage point represents 65mAh; 10% is 650mAh, etc. All of the magic happens at the top and bottom end of the usable SoC range - the "hang points".

I'm not sure if I understand your 'depletes quickly' comments - I think there must be a misunderstanding regarding what you were expecting. The battery installed in the car doesn't change the value of a percent of SoC. As stated, BCM SoC is calculated based on current in and out of the pack. So, if you start at 60% and draw 2Ah from the pack you will end up at 30% SoC.. every single time(assuming the pack has 2Ah to give). So then, how long it takes to go from 80-40% SoC depends entirely on current draw during the time in question. The car is capable of drawing the most current in 2nd and 3rd gears, so it makes sense that it would be depleted the quickest there. What was the expected behavior?

The temperature thing is interesting. It definitely isn't what you would expect at first thought given the internal resistance improvement, but I think it makes sense in the context of what is happening here. "My battery is constantly depleted" = "My battery is constantly being used". And it can deliver the power to boot, no throttling here - like there would be with a weak battery. IIRC, this was borne out in the original BetterBattery test data. The new cells ran as hot or even hotter than the OEM pack - because they were able to deliver full power for like 3x as long or whatever it was. So just by the numbers, it seems to "run hot", but the benefits are there. If an OEM pack could deliver the same power over the same period of time, it would be all the hotter.

I think a similar phenomenon is probably behind the feelings of "fast depletion" - it's easy to deplete a battery that can actually deliver lots of current!

The themes behind the (mostly) fixable things that can cause this are drag and engine power. If you have the wrong tires or a dragging brake, for example, do not underestimate how much of an impact this type of thing can have on where the SoC gauge "lives". When you press on the accelerator, except for when you very first start off, it should be pretty easy to let off the gas until assist goes away - but still accelerate. I know that's pretty subjective, but if you find it very difficult to accelerate without using tons of assist, it would be wise to start checking things out. The hardest thing to deal with is an engine that is just plain worn out. Insights with worn out engines will eat IMA batteries for breakfast and the occasional midnight snack. Not saying that is the case here, just trying to give scope.

If I were driving these hypothetical low/wrong tires/bad engine vehicles "for the battery"(as opposed to for mpg), the extent of the drag or lack of power would determine the MPG hit I'd take for that driving style. If I was driving for MPG, it would determine the average speed hit I would take at my chosen MPG... just to maybe help people to see the relationships between all of these things.

Anyway, one thing everyone can do to help keep the battery charged - and I DO highly recommend that everyone do this -

Drive with your headlights on! Even and perhaps especially during the day. Get used to flicking them on whenever you drive and soon it's like putting on your seatbelt. It makes a big difference in the Automatic Background Charge point, and will try to keep the SoC between 65 and 70% which is right where we want it. Anyone having this problem should try that first.

It would be best to make sure you 'start over' from from "actually full", especially if you've been experiencing this problem for a while. Rev the engine to around 3100RPM - I think that should dial up around 15A of regen. If you have an OBDIIC&C, you should see SoC climb and hang at 72.0% while the battery continues to fill. Hold there until regen stops. At this point BCM SoC should be either 75.0% or 80.0% depending on BCM revision. If you have a grid charger, it would be good to input an additional ~1500mAh at this point, so 4-5 hours if it inputs 350mA.

Then drive(with your headlights on of course) and play the "How high can I maintain SoC while getting the best MPG" game. :D
Thanks Eli, I wouldn't be that surprised if I drive the car heavier than the average insighter, so to some degree I am expecting some drain. Doing the math I should be able to pull a full 10kw of assist for around 160 seconds before depleting the SOC to 50% (3.25Ah / 70A = 0.046hr = 165 seconds) not taking into account losses from ESR. I'll have to revisit my current measurements but it is possible I am drawing this much over my Los Angeles commute, I live in a very hilly area and the drivers around me are extremely aggressive making the car difficult to drive without relying on assist.

However what I am not expecting is the inability of the insight to replenish the missing charge in the batteries, doing the math again the car will often background charge at 10a on the freeway for more than 20 minutes at a time, which should be enough to replenish all the lost charge in the battery. Thing is though it doesn't seem to replenish the charge at all. I'll watch the battery gauge just sit at 40-45% on the freeway even going down a steep hill or staying in 4th gear. If my math is right I'm expecting to watch the gauge climb at 1bar / 30s with 10a of current flowing into the battery, but it doesn't seem to do that.

I'll see if I can get some more quantitative data as well as try out some of your tips. But from what I have seen the Ah into the battery will exceed the Ah pulled from the battery by a surprising margin. My general observations from driving the car also don't seem to match my calculated assumptions, so it still feels like something unusual is happening here but I could be wrong.
 

·
Moderator
Joined
·
7,259 Posts
Well, so a few things.. Unless you have modded stuff, the Insight draws a peak of 12kW in 2nd and 3rd gears only, and only for ~4 seconds. All other gears are limited to 6.5kW, which is what assist falls back to after the ~4 seconds as well.

SoC from 80% to 50% would be 1950mAh, not 3250mAh

Depleting the dash cluster gauge from 20 bars to 3 bars should yield an output of about 3500mAh. At 6.5kW, that should take around 280 seconds. To fill it back up to 20 bars is going to take more than 3500mAh. Probably around 4000mAh before you get the positive recal to 20 bars.

Input into the battery is always more than the output. Otherwise, you would end up with a dead battery. Plus, it's not 100% efficient. I haven't watched that number much, but I wouldn't be alarmed unless it's like wayy different.

I'm not sure regarding the case of the disappearing SoC .. I mean, if input is going into the battery, SoC will tick up. You can verify this by revving the engine to 3000-3500RPM in neutral....

The trick to avoiding background charge is to never deplete the battery in the first place. Lots of available capacity is a double edged sword in that sense. It's definitely much easier to use the battery up than it is to replenish it.

My recommendation is to rev charge as stated above to reset the baseline to 'full', and then try your hardest to not let SoC drop. Take it to the extreme, to find what is necessary for your area. Once you can drive without SoC dropping at all, find the middle ground. Ideally, you would find some type of routine where SoC gets down to ~60% before it's recharged back up to 80%. Then you get natural balancing on every upswing.

I have exactly such a route here that I can take home from the shop. That's how I did the experiment showing the car CAN balance packs - through the natural characteristics of NiMH batteries.
 
  • Like
Reactions: jeagan

·
Registered
Joined
·
7,957 Posts
Most of this sounds like some IMA computer glitchiness to me. If you're regen-ing for a while and the BAT bar gauge doesn't move - that's a 'glitch' of one sort or another. I've often wondered if the BCM has some sort of check point that would likely fall right about at your 40-45% mark. For example, I usually reset my nominal SoC with OBDIIC&C to 40% at empty (neg recal, which would be 25% without the reset), and then often when I let the pack background charge back up, it will stop charging at about 55%, either that and/or the charge bars disappear, even though it's still background charging.

I've generally thought the BCM must look for an appropriate tap voltage level after a certain amount of charging (~15%) after a neg recal, and if that level is hit it won't charge any more.

But anyway, if you start at the neg recal nominal, 25%, and add 55% minus 40% = 15%, that would put the nominal at 40%, i.e. about where your 40-45% bars are likely to be at...

Most of the 0-50% vs. 50% to 100% behavior Eli's talking about has to do with the "vario-stoichiometry" of NiMH electro chemistry. Basically, the voltage can vary at different charge states - you can have seemingly near-full voltages at low charge states, and low voltages at high charge states, it depends on how the pack is being used (i.e. the size of cycles, actual charge state range of cycles). So, if you stick to the 0-50% range, for instance, the voltages will look more or less like they do in the 50%-100% range. And since the BCM determines 'full' and 'empty' partially based on tap voltage, you can end up with a pack that has 'full voltages' but within a low absolute charge state range...

Have you tried resetting the IMA - pull underdash # 18 fuse, or maybe better, pull the 12V neg cable?

Do you know or can you find out what BCM version you have? When I tried a later model 2005-06 BCM with my older pack, it undercharged. Your sig says 2003, not sure what BCM model that would be. I've generally thought aftermarket packs/cells would have voltage behavior closer to the later year Insight/Civics cells. If that were the case a later year BCM would be more appropriate - but it seems like you'd have the opposite problem than I did: The later model BCM on my older pack undercharged; an early model BCM on a later pack - seems like it would overcharge, or just simply charge more, have a tendency to charge higher...
 

·
Premium Member
Joined
·
808 Posts
Thanks Eli. I'll respond to your PM in a moment. Thanks for outstanding customer service and thank you for providing us with batteries.

So, I'm going to be honest - all this would take the joy out of the car for me. I drive conservatively and get high 60s mpg on average around town. Once in a while I gun it just for fun, but usually I putter along. I don't keep an eye on SOC much - it maintains around 50% though. To think that now I need to try to not let it deplete really takes the joy out of driving. I mean am I enjoying the car or is the car enjoying me? So far so good with my battery, knock on wood. I just hate to think I have to drive it a certain way to keep my battery in the "healthier" zone. Maybe I'm not enough of an oddball to own an Insight after all. ;)
 
  • Like
Reactions: jeagan
1 - 12 of 12 Posts
Top