OBDIIC&C Performance Testing

[jime]

OK, after a lot of yelling, bitching and waving of arms I finally learned how to do proper IMA acceleration testing. I have filled my info in the spreadsheet. It looks like this is a really good, strong Honda stock pack from the acceleration numbers. This might be considered some sort of baseline for a new Honda Refurbished IMA battery, but, who knows if they are all this good.(I did start all runs with an 79-81% SOC. Please indulge an old engineer in his belief that the SOC probably matters.)

I tried to do the Ah/wh battery capacity test, but the results didn't come out right and I don't think I fully understand the procedure as yet.

[jime]

It seems to me that the results as reflected in the spread sheet are kinda all over the place. Have we learned anything? I suppose the mods in peter's project car have shown their worth in the data. What are we actually trying to test, the performance of a standard pack? Just curious

Would one for example baseline their own battery with their own test data, and perhaps the battery capacity test, for later comparrison?

[retepsnikrep]

I'm hoping we just build a good base of data for later analysis as all the 50 or so gauge owners perform the tests. We have people like Willie with his turbo car and Matt from Hybrid revolt with his power mods as well to include.

We can see clearly in the data that good packs give much better results than no IMA or bad batteries, and as you say the power mods inc MIMA/IMAC&C & others hacks make a big difference to performance.

[Eli]

SoC doesn't matter much in and of itself, except at the very low end. With the BetterBattery, I could draw 10kW with MIMA and the car would give 11kW+ on it's own even down below 50% SoC.

What matters is the condition of the pack, ie: what the BCM sees on those pesky 10 voltage taps. I think it looks at a lot of things regarding them.. the absolute voltage, voltage difference between taps, the rate of voltage drop and probably other things...

[JohnVirginia]

Originally Posted by Eli

SoC doesn't matter much in and of itself, except at the very low end. With the BetterBattery, I could draw 10kW with MIMA and the car would give 11kW+ on it's own even down below 50% SoC.

What matters is the condition of the pack, ie: what the BCM sees on those pesky 10 voltage taps. I think it looks at a lot of things regarding them.. the absolute voltage, voltage difference between taps, the rate of voltage drop and probably other things...

Pardon this novice intrusion (again), but I keep wondering how much of the car/IMA/operation control is done as digital program/SW/firmware in the BCM, ECM, etc., and how much is just electronic hardware circuits. Doesn't Honda do periodic sw updates (e.g., the one that caused gas mileage to go in the gutter in order to prolong batteries). I see a lot on the forum of hacking the inputs and outputs to the system (FAS, SOC via OBDCC, MIMA, etc.). Is this because most of the operational control is accomplished in the hardware (like past analog systems) and thus hacking the I/O to these circuits is the way to go or is the digital programming Honda has incorporated just unhackable? Other responses have indicated the latter and this will probably be my last attempt to understand.

[retepsnikrep]

We don't have access to the Honda software or source code for any of the modules, neither can we retrieve the data from the modules so we are stuck with the computer side as it is. That leaves hacking the analog side or intercepting and modifying the data communications between modules as out way into the system.

[JohnVirginia]

Thanks. Understood. I'm just surprised there is no known access to the non-volatile memory other than how ever Honda does it with their shop systems and headquarters supplied patches. Good luck in your endeavors with the signals.

[Eli]

Here's an interesting thought -

Let's start a tab in the spreadsheet for acceleration numbers with IMA disabled.

This would give us a better metric to judge pack status by. If adding IMA reduces numbers by 3 seconds, compared to another person at 1 second, that would be a fairly profound indicator.

We might need a 20-40MPH test or something for that though, or we'll have people taking 2 minutes to go from 50-70.

[jime]

Originally Posted by Eli

Here's an interesting thought -

Let's start a tab in the spreadsheet for acceleration numbers with IMA disabled.

This would give us a better metric to judge pack status by. If adding IMA reduces numbers by 3 seconds, compared to another person at 1 second, that would be a fairly profound indicator.

We might need a 20-40MPH test or something for that though, or we'll have people taking 2 minutes to go from 50-70.

It is an interesting thought and would probably add some Insight
I don't think we can change the ranges though, since our meters are already programmed and most of us don't have the ability to change them. A half mile, reasonably flat road may be a challenge for some. Lack of repeatable conditions (i.e. different road section for IMA/non IMA) will badly skew the 50-70 tests.

I think there is reason to suspect that the current test does not provide a true picture of the IMA, since total torque/horsepower is a combination of the IMA and the ICE. I suspect that some ICE's are much better than others, just like battery packs.

How do those of us without MIMA or the clutch switch accomplish the test, just raise the rear carpet and shut off the IMA switch for the short duration testing?

I am going to LA Cajun country in about 10 days. The roads through the rice fields there are absolutely flat and lightly traveled. I would be happy to repeat my test there and avoid the Richmond hills. I'm definitely for it. Just add columns adjacent to the current 30-60 and 50-70 columns.

Good Idea

[Eli]

Yeah, the only way would be to shut the switch off. Kinda a PITA, but I would be willing to do it. That's kinda what I mean though.. what if someone has a very strong engine but a weak battery? Could be pretty telling overall.