Honda Insight Forum banner

101 - 120 of 154 Posts

·
Moderator
Joined
·
6,916 Posts
Discussion Starter #101
Not on the OEM pack, no. I just got the stuff for another wifi BCM Gauge a couple of days ago, so I'm going to work on building that up and getting it into Insight #3 so we can play with the 10 year old OEM pack. My mountain run route has been closed for the better part of a month due to a fire close by, but I think it should be re-opening soon. I'm anxious to get some more runs in with the BetterBattery too.

The higher amperage data logging is nice though, you can see more than you can even on the BCM Gauge. :) It seems that the car constantly alternates between plus and minus ~0.04-0.25A when nothing else is happening. Kinda interesting.
 

·
Administrator
Joined
·
10,596 Posts
Even the well designed Honda Current sensor setup won't be that accurate at very low currents so i would take anything below about 0.25A with a big pinch of salt.
 

·
Premium Member
Joined
·
2,818 Posts
Even the well designed Honda Current sensor setup won't be that accurate at very low currents so i would take anything below about 0.25A with a big pinch of salt.
If I recall correctly ... Wasn't the IMA in the Gen-1 also supposed to act as an active balancing system for the 3 cylinder ICE?

It might be noise ... but do we know what the level is for that balancing effect?
 

·
Administrator
Joined
·
10,596 Posts
If I recall correctly ... Wasn't the IMA in the Gen-1 also supposed to act as an active balancing system for the 3 cylinder ICE?

It might be noise ... but do we know what the level is for that balancing effect?
Good point could be that easy to forget. :) It's a very small current then if it is.
 

·
Moderator
Joined
·
6,916 Posts
Discussion Starter #106
I think the balancing currents are in the 10-20A short pulses. When idling with the fan and lights on, you can see the first assist then regen led on the MIMA display flash, Same when the A/C is on at idle. So when the 12V system is loaded it looks like stronger balancing pulses may be produced.
Yeah, I have noticed this too on the MIMA display. I don't think I've captured this in the logging though, I'll have to try and do that.
 

·
Administrator
Joined
·
10,596 Posts
Yeah, I have noticed this too on the MIMA display. I don't think I've captured this in the logging though, I'll have to try and do that.
The vibration dampening pulses pulses are probably too quick to be captured by the BCM gauge and they may not be captured by the BCM itself and therefore not in the data stream.

I wonder if the dampening is active when IMA switch is off? Perhaps it manages with the charge in the filter caps, I have never noticed any difference.
 

·
Premium Member
Joined
·
4,389 Posts
you can tap the isolated/amplified and rezeroed current signal off the MIMA card.
The sensor output is analog, so it has infinite resolution. At the micro, 2V = zero current.

there are other battery failure modes to consider.

From my experience, once a cell vents, the capacity is permanently reduced, and the internal resistance goes up due to less effective plate area.
Not recoverable.
Unbalanced cells mostly due to non cycling and infrequent use.
Recoverable

A shorted cell or two will allow the pack to overcharge?
Not recoverable.
All of these could be present in a pack at the same time, what will they look like on the traces.
I will have to dig out my datalogger once the charger project is under control, and can add to the data mix.


The final topping voltage at a 350ma CC seems to be a big variable, which is the better battery, final settling at 165 or 175V?
Ambient temp effect on this voltage?
I used sequential relay subpack multiplexing when I did my subpack monitoring datalogger many years ago. Worked well, but was noisy.
 

·
Premium Member
Joined
·
2,818 Posts
Good point could be that easy to forget. :) It's a very small current then if it is.
I think the balancing currents are in the 10-20A short pulses.
The vibration dampening pulses pulses are probably too quick to be captured by the BCM gauge and they may not be captured by the BCM itself and therefore not in the data stream.
I wonder if it could be a combination???

meaning...

If it is a ~20A pulse that happens fast enough ... the 'average' amps of pulse over a longer interval might look like a much smaller amount of Amps.

for example:
over 100x as long of a sample period a ~20A pulse 'averages' to about ~0.2A.

I guess it might just be dropped completely instead of being averaged over the longer period.... ???
 

·
Premium Member
Joined
·
4,389 Posts
Since the pulses are both regen and assist, and about the same magnitude, they will likely average to zero. If the current sensor signal is filtered in the BCM, the BCM may not even see the pulses, but the MIMA current circuit seems to easily acquire and display them.
Probably not going to be a big error either way.
I will be popping the IMA cover to wire in the grid charger harness, so I may do some playing with a scope.
 

·
Moderator
Joined
·
6,916 Posts
Discussion Starter #111
I've got the wireless BCM Gauge setup in #3. I hope to begin logging data soon on this worn out pack.

The battery was at 152V after sitting for a couple of weeks. I did four negative cable pulls and noted how much was inputted before the positive recal.

#1) 0.42Ah
#2) 0.26Ah
#3) 0.22Ah
#4) 0.18Ah
-----
1.08Ah inputted

Maybe I'll try it some more and go until it does a positive recal instantly. I think there's some life left in this pack, but not much. It's going to be interesting to see how much a grid charge or ten revives it.
 

·
Administrator
Joined
·
10,596 Posts

·
Moderator
Joined
·
6,916 Posts
Discussion Starter #113 (Edited)
The Anatomy of a MIMA Pack Whack

After getting the 2nd Wireless BCM Gauge setup the weekend before last, this past weekend I took the opportunity to switch MIMA over from #1 to #3 so the continued testing of stock, OEM packs could commence.

Due to the revisions of the BCM/MCM in #3, I now believe that this battery was replaced somewhere in the 2005 time frame. I don’t believe it’s an original, 11 year old battery like I first thought. Regardless, this pack is throwing constant P1449-78 “Battery Module Deterioration” errors.

This will begin the data I’ve collected over the last week. First, we’ll develop a baseline. This is the first run I did with the pack. I decided to do a full assist run, as these are easy to keep consistent – peg the joystick and see what happens.

Car #3 Mountain Run #1 - Full Assist - Volts, Temp, State of Charge, Amps

As you can see, the pack did a massive recal after about 1 minute. Note that it plummeted nearly immediately to 120V. It’s also interesting to note that the pack was only able to give me almost full assist of 9.819kW for ~0.25 seconds. The recal happened at 71.3% BCM SoC.

Without thinking about it, I ended up doing half of a pack whack on this first discharge test – I kept discharging the pack even after the recal. So this grouping of data is essentially a test of the pack whack, and will perhaps show how valuable MIMA can be in regards to battery health.

Car #3 Mountain Run #1 - Full Assist - VS. Car #1 Mountain Run #8 - Full Assist - Amps Overlay

This compares the amperage discharge of this first run to what we would expect from a healthy pack. It’s interesting to note the sawtooth pattern that assist takes on, compared to the more flat topped pattern of a healthier pack.

Car #3 Mountain Run #1 - Full Assist - VS. Car #1 Mountain Run #8 - Full Assist – Kilowatts Overlay

Here we have the Kilowatts overlay of this first run compared to a healthier pack.

This is what the amp hour output of the pack looked like:

Car #3 Mountain Run #1 - Amp Hours

The first data point of 0.73Ah is when the recal occurred. This is essentially the capacity of the pack independent of MIMA. After this point, the MIMA Pack Whack begins. You can see that I was able to take an additional 0.24Ah out of the pack after the recal, thanks to MIMA.

Car #3 Mountain Run #1 - Full Assist - VS. Car #1 Mountain Run #8 - Full Assist - Amp Hours

Here we’re comparing the amp hours discharge curve to the relatively healthy stock battery in #1. Lots of room for improvement. :D Again the first data point is when the stock system would have cut assist off.

Perhaps the best overall indicator of pack health will be Kilowatt Hours. This is a measurement of the actual energy the pack puts out. Amp hours don’t tell you the whole story because it’s missing the voltage information.

Car #3 Mountain Run #1 - Full Assist - VS. Car #1 Mountain Run #8 - Full Assist – Kilowatt Hours Overlay

Again, this is being compared to what we believe to be a fairly healthy stock pack at only 1.5 years old.

Now we’ll move on to the Regen portion of this first test. Regen doesn’t seem as interesting overall, but I think it’s an important part of the equation. In my experience so far, the car always seems to throw the IMA light on during the regen portion of the test, but I don’t know if it’s actually looking at the full cycle or what. Note that the P1449-78 error code is supposed to mean 10% or less of 6.5Ah, which would be 0.65Ah. It seems to be more dynamic than this though, as we’ll come to see. It seems to be voltage based, which is what the advanced diagnostic information alludes to in the service manual.

Car #3 Mountain Run #1 - Full Assist - Down - Volts, Temp, State of Charge, Amps

You can see that the pack did the positive recal at 32.1% SoC. This is also when it threw the IMA light(P1449-78 ). The spike of assist in the graph is me testing whether MIMA assist works even with the IMA light – it does. However, regen is odd. It acts like the car is in neutral, and only very slowly notches regen up as I hold the joystick at full regen – it’s not instantaneous like normal and like assist is, which is the reason for the strange regen profile after the positive recal.

Here are the Amp Hours inputted during the run down the mountain.

Car #3 Mountain Run #1 - Full Assist – Down – Amp Hours Input

I was able to input 0.75Ah before the positive recal, a fairly dismal number. However, with MIMA I was able to pump in another 0.46A, for a total of 1.21Ah. That’s a fairly significant input, I think.

Next is the Kilowatt Hours inputted during the run down the mountain.

Car #3 Mountain Run #1 - Full Assist – Down – Kilowatt Hours Input

As you can see, the positive recal happened after only 0.13kWh. I was able to input another 0.08kWh into the pack thanks to MIMA.

That's about it for the first run. If anyone wants to see anything specific or has any questions, just let me know. Over the next few days I will be posting the data from subsequent runs. That's where it really gets interesting....

Tomorrow we will see the effects of the first Pack Whack. Time for sleep, I'll probably be dreaming of spreadsheets...
 

·
Moderator
Joined
·
6,916 Posts
Discussion Starter #115 (Edited)
No, not yet. I wanted to see how the pack behaves with just the car and MIMA influencing it. I do plan on grid charging it and seeing how that ultimately improves things once the improvement stops.

It's obvious that cycling the pack is helping, but I don't know to what end.
 

·
Moderator
Joined
·
6,916 Posts
Discussion Starter #116 (Edited)
The Anatomy of a MIMA Pack Whack #2

Alright, here's the data from Run #2 overlaid with the data from run #1. You can clearly see improvement.

Car #3 Pack Whack Run 1 & Run 2 - Volts, Temp, SoC, Amps Overlay

The voltage profile is better. The recal point improved from 71.3% to 67.0%. Slightly higher full assist output, and a slightly better amperage profile upon commanding full assist.

Car #3 Pack Whack Run 1 & Run 2 - Amp Hours Overlay

A definite improvement. The recal point moved up from 0.72Ah to 1.03Ah. The total amount able to be taken out of the pack with MIMA increased from 0.97Ah to 1.27Ah. That's an improvement of 31%.

Car #3 Pack Whack Run 1 & Run 2 - Kilowatts Output

Here we have the instantaneous kilowatts output profile of each run. Peak output improved from 9.819kW to 9.937kW, an improvement of 1.2%. The amount of time spent at over 9.8kW improved from 0.25 seconds to 0.75 seconds.

Car #3 Pack Whack Run 1 & Run 2 - Kilowatt Hours Overlay

Recal point improved from 0.09kWh outputted to 0.13kWh outputted, and total output improved from 0.12kWh to 0.16kWh - an improvement of 33%.

On to regen!

Car #3 Pack Whack Run 1 & Run 2 - Volts, Temp, State of Charge, Amps

Most of the data in this graph isn't comparable to eachother since the regen rates are so variable. The graphs in the assist part are directly comparable though, as I'm commanding full assist the entire time. You can see that the positive recal point improved from 32.1% to 33.9%.

Car #3 Pack Whack Run 1 & 2 Amp Hours Overlay

Car #3 Pack Whack Run 1 & Run 2 - Kilowatt Hours Overlay

As you can see, only a modest improvement here in amp hours and kilowatt hours as well, an additional input of a mere 0.13Ah and 0.02kWh, respectively. But an improvement none the less.

I actually have four runs completed, and the pack has continued to improve on each run, so I look forward to sharing that data with you all. It's quite interesting, and it will be even more interesting to see to what end the pack whacks improve the pack. It's obviously not hurting though. :)

I have some other data points that will come in the following weeks, including grid charging and the equivalent of SoC resetting with Peter's OBDIIC&C tool. I have some tricks up my sleeve that I think may be the equivalent of grid charging the pack, minus the grid charger.
 

·
Registered
Joined
·
813 Posts
Interesting data Eli.

Note: You should probably update your sig, since MIMA #163P is now in #3.
 

·
Moderator
Joined
·
6,916 Posts
Discussion Starter #118 (Edited)
The Anatomy of a MIMA Pack Whack #3

Car #3 Pack Whack Run 1, 2 & 3 Assist Volts, Temp, State of Charge, Amps Overlay

It's beginning to get a bit crowded, but there's still lots of useful interesting information. Recal point improved from 67% SoC to 63.1% SoC, or 6.2%. Note the significantly improved voltage profile, and the lower current draw for more power as we'll see in the Kilowatts graph.

Car #3 Pack Whack Run 1, 2 & 3 Assist Kilowatts Overlay

A noticeably better overall profile. Peak output increased from 9.937kW to 10.193kW, and increase of 2.6%. Time of output at 9.8kW or more increased from 0.75 seconds in Run #2 to 1.75 seconds in Run #3.

Car #3 Pack Whack Run 1, 2 & 3 Assist Amp Hours Overlay

Here we can see that overall output increased from 1.27Ah to 1.53Ah, an increase of 20%. The recalibration point improved from 1.01Ah to 1.29Ah, or 28%. Perhaps more importantly, you can see that the overall profile of the discharge curve has improved.

Added:

Car #3 Pack Whack Run 1, 2 & 3 Assist Kilowatt Hours

Forgot to include this in the initial post and it's probably the most important metric overall. Overall output increased from 0.16kWh to 0.20kWh, or 25%. It's interesting to note the consistencies. It seems that each run improves by 0.04kWh, and I'm always able to input an additional 0.03kWh after the recal.

Car #3 Pack Whack Run 1, 2 & 3 Closeup Assist Kilowatts Overlay

This is the output profile of the first few seconds of commanding maximum assist. I just thought it was interesting to look at.. lol

Things got a little weird on the 3rd regen run. I decided to try and input a much lower charge to see how the systems reacted. They certainly reacted differently, that's for sure.

Car #3 Pack Whack Run #3 - Regen Volts, Temp, State of Charge, Amps

You're seeing correctly, it took me over an hour to charge the pack up at an average rate of ~3A. Recalibration point went from 33.9% to 60%!

Car #3 Pack Whack Run #3 Regen Amp Hours

A whopping 3.04Ah inputted.

Car #3 Pack Whack Run 1, 2 & 3 Regen Kilowatt Hours

Most of the other graphs look silly with the other runs overlaid because of the time difference, but this one looks OK I guess, though perhaps a little over dramatic. As you can see a huge improvement in overall input, and no IMA light!.

Past history tell us that we can generally get out of a pack what we put into it. Does this mean that I'll be able to discharge ~3Ah on the next run? Did I hit some kind of breakthrough with this pack? Stay tuned to find out. :D
 

·
Administrator
Joined
·
10,596 Posts
Pack Exercise is certainly helping this car nice work. Time for a grid charge soon now to see what happens now it's livened up a bit.
 

·
Registered
Joined
·
813 Posts
Nice work Eli, good research!
 
101 - 120 of 154 Posts
About this Discussion
153 Replies
23 Participants
Mike Dabrowski 2000
Honda Insight Forum
We’re the ultimate Honda Insight forum to talk about Honda’s hybrid car and its fuel economy and specs!
Full Forum Listing
Top