Data Logging With the OEM Pack

[retepsnikrep]

I know you will be very keen to install the new cells but it may be worth waiting a week to just get a bit more data from the oem pack inc when it has been gridcharged.

Your increasing battery performance seems to tally with others who have packwhacked

You can't really expect any more regen (ah in) on your downhill journey unless the terrain has changed.

Attached is v3.21 which should monitor the fan voltage and output it as a value between 0-255 in the last column on right when logging. You will have to determine what value = what speed, then add a formula in spreadsheet to turn the value into something more meaningful.

[Eli]

I agree Peter. The weekend is almost here, so I could spend all day one day gathering data.. It takes about an hour per trip, so I could do it 4-6 times.

I'm cycling the pack from full to empty and back to full again. It stops allowing me to input regen just as I get home. That's why I'm doing the two runs as separate entities. If you'll notice, they are "true" runs. No regen during climb, and no assist during descent. Of course I've flubbed up a couple of times, and the trip home has stop lights and thus start cycles, but these are negligable over the course of the run. I guess I should post more about my methodology, sorry if I wasn't clear.

But anyway, shouldn't any improvement in output be matched by an overall similar increase in input? But maybe not. I guess that's assuming that the BCM is a logical device, and it is clearly not! Maybe it will manifest itself as another "stuck SoC" episode?

I wonder if this is like an "internal" recal? The gauge was one bar from full during this time, and I inputted another 0.75Ah before it stopped regen.

It would be really cool to have a dash SoC meter bar output in the spreadsheet? Is that possible? lol.. It just eliminates another variable.. Instead of having to remember how the gauge was behaving, there's a snapshot of it right there?

Edit: I was looking over Run #2 Up, and I noticed that the negative recal happens at the exact time I do an ~81A, 10.5kW burst. Pack voltage momentarily dropped below 120V. I bet this triggered a subpack pair dropout and hence the SoC drop. I guess it was a recal? They're hard to see with only 3 bars though. I wonder how much more I could have gotten if I wouldn't have done that little burst?

The gauge hung for a long time at 3 bars this last run up, much longer than in Run #2 Up.

[retepsnikrep]

Originally Posted by Eli

It would be really cool to have a dash SoC meter bar output in the spreadsheet? Is that possible?

Although the soc data is in the bcm data stream (we use it) the gauge is driven by the MCM not the BCM. The OEM bar display is probably based around the percentages sent to the MCM by the BCM but we haven't established that relationship bewteen the % and the bars. You could pick up the soc gauge bars data but would need another BCM gauge type device on the METSCII lines that's not practicable in the timescale. Close observation of the bars and % might allow you to roughly make a corelation between them. Again others could help with this!

Originally Posted by Eli

The gauge hung for a long time at 3 bars this last run up, much longer than in Run #2 Up.

That's good in my book and indicates remaining capacity is increasing which you then used before it finally dropped to zero!

I have uploaded another gauge version untested in my prev post with the fan hack. A 0.1uf cap across the 10k resistor might also help to stabalise the voltage readings from the fan supply.

Gentler discharge will always allow more AH to be drawn due to peukerts law. Try to keep your assist/regen levels consistent between runs.

[Eli]

Yeah, I'm aware of that. Ideally, I would like to get all types of data points in.. from heavy assist to light assist. The picture is starting to appear, but it's far from complete.. Like I said, I need months worth of data really, but that won't be possible - with my pack at least. I'll eventually get a gauge and data logging setup into Insight #2.

I would need to set up a camera or something pointed at the dash, then try and match the timestamps up? It's virtually impossible to keep watch on all the data.. dash cluster, bcm gauge.. and the road ahead. Dangerous, in fact.... lol.

[JimIsbell]

Originally Posted by Eli

It seems the cycling is helping? I've gotta get more of this data... no grid charge tonight. I'll try to get another run or maybe two in tomorrow.... I wish I would have been doing this for the last few months.

There is no doubt in my mind that cycling helps. I started 3 years ago with a 9 year old battery that held something over 1.5 AH of charge. It RECALed 3 or more times a week. I have done nothing but grid charge it and refine my driving habits and I have a near "New" acting battery today that cycles between 20% and 75% without generating a RECAL more than once a month, at most, if I really discharge it. In normal driving I will never see a RECAL and the battery is now 12 years old. I dont have MIMA up yet, too many other projects, so all forced charging was by grid charging.

Edit....well there were a few times when I force charged it coming down out of the Siera Madres on 30 mile grades...LOL

[Eli]

It's just strange to me because like I said, I considered my pack to be healthy at only 1.5 years old, and no obvious anomalous behavior.

If I am gaining capacity by cycling the pack, it appears that cycling - not just regular use - is far more important than I had thought. The last few months, my pack has lived between empty and half on the SoC gauge, with an occasional top off on the weekends. This is from using MIMA on my commute to/from work. It appears this practice of using the battery in the lower range of the dash SoC may be detrimental?

[JimIsbell]

Be aware that your "New" battery was not NEW. It was refurbished by Honda.

[Eli]

The sticks in the warranty replacement batteries are New in the sense that they have never been used before. However, they are NOS - New Old Stock - likely manufactured years ago.

The case and junction boards are re-used. That's why they call them "refurbished".

[JimIsbell]

So the NiMh battery is not harmed by sitting on a shelf?? If true, that is good news. Now we no longer have to use a trickle charger when letting a car sit for long periods of time.

[MN Driver]

Originally Posted by JimIsbell

So the NiMh battery is not harmed by sitting on a shelf?? If true, that is good news. Now we no longer have to use a trickle charger when letting a car sit for long periods of time.

Not quite, Honda ships the batteries quickly and the dealers get them installed quickly, they have a date printed on them to when they are supposed to be installed by. There was a recent thread with pictures from a guy from New Zealand where this was discussed with the battery he received. They get charged before they leave the facility and it isn't long before they end up in the car and back to the customer. I'd make an assumption that Honda cycles the sticks a few times prior(probably through the BCM taps, which is convenient now that Honda's 2006+ model year cars use 12 cell double sticks) to delivery to both kill off the memory from sitting, weed out lame sticks, and know with confidence the pack they are sending out is good.