Honda Insight Forum banner

1 - 20 of 35 Posts

·
Registered
Joined
·
79 Posts
Discussion Starter #1
I received four GreenTecAuto LTO battery packs a couple weeks ago and I built an aluminum baseplate for mounting seven of the sub-packs into the IPU. My donor car already had it's stock battery pack removed so I used this vehicle for determining the layout. Here's a photo of the baseplate installed within the IPU of my donor vehicle:
Baseplate.JPG

The baseplate rests directly on top of the IPU frame members where the OEM battery pack mounts. The thin aluminum at the front is trimmed 3 inches above the floor so the baseplate does not need to be trimmed in order for it to slide forward. The OEM mounting tabs at the back are cut off flush with the top of the baseplate and the baseplate has been slotted for these tabs. Four holes are drilled into the baseplate for mounting the baseplate to frame members at OEM battery mounting locations. The holes are very slightly oversized for the baseplate mounting-bolts. The baseplate is drilled and tapped for mounting the sub-packs placed into these locations. The baseplate has been cut out on the far left for the battery junction board. The baseplate is also cut out for the two left-front sub-packs. All sub-packs are mounted on top of the baseplate except for the two left-front sub-packs, which are recessed below the baseplate. Spacers are required for all sub-packs except rearward of the two left-front sub-packs. The tabs for these two sub-pacts rest directly on top of the baseplate. A spacer about 3/8" tall is required forward of the left-front sub-packs so that the bottom of these batteries are suspended above the IPU floor by approximately 1/8 inch. Spacers 1" tall are required for the two right-front sub-packs since the LTO battery tabs are 1" above the bottom of the batteries. The two right-rear sub-packs and the left-rear sub-pack require 1 1/4" spacers because these sub-packs rest on top of 1/4" plywood. This plywood allows for the height of the baseplate mounting bolt-heads that rise up within holes drilled into the 1/4" plywood. All spacers are made from a combination of 1" square aluminum tubing and flat aluminum stock. Here's a photo of the LTO batteries along with all the mounting hardware.
Kit.JPG

Batteries and mounting hardware are shown in their respective positions. The rear sub-packs share a single 1 1/4" spacer in the middle. Note the fan has been custom trimmed to make room for the left-rear sub-pack. The outlet at the bottom of the fan was enlarged, for better air circulation, after this photo was taken. These components were taken to the Columbus InsightFest for installation into my working Insight. More to follow.....
 

·
Administrator
Joined
·
10,770 Posts
Good write up Tim keep it coming..
More detail and pics the better for those wanting to follow suit.
 

·
Super Moderator
Joined
·
7,150 Posts
Nice write up Tim. You can avoid the 1/4" plywood at the rear if you use countersunk screws for the rear attachment - not that you were pressed for height at the rear. Tell us how it works as you gain experience, particularly the BCM Replacer part. Did you fit a display with your BCM Replacer and which one?
 

·
Registered
Joined
·
79 Posts
There’s a challenge to building a drop-in baseplate that’s already been drilled and tapped for baseplate mounting as well as battery mounting. Very tight tolerances are required for this 84-cell baseplate. It’s just too bad there’s almost 2 inches of unusable space to the left of the inverter/converter. We could really use that extra space if the inverter converter were only shifted further to the left! Nevertheless 84 LTO cells is very doable. We just ran into a bit of a quandary installing the baseplate in Columbus when it had fit perfectly into my donor vehicle. I have now come to realize that a baseplate, built using precise dimensions for one vehicle, will not necessarily fit properly within the IPU compartment of another vehicle.

There was zero clearance on the right side of the baseplate on my donor vehicle but now there’s a 1/4” inch gap. This gap means the battery-mounting holes, pre-drilled into the baseplate, are now shifted 1/4 inch to the left. The 1/4 inch of extra space for the junction board, between the dc/dc converter, to the left, and the LTO battery, to the right, has suddenly vanished! This was a tighter squeeze than Peter and I bargained for, but we made it work temporarily. I had some extra work regaining just a little bit of this lost space after arriving home from Columbus.

The inside dimensions for all IPU cases are probably very consistent. Likewise, the dimensions of the OEM batteries, and the relative spacing between those holes, are probably also very consistent. The inconsistency is probably in the relative position/rotational angles of OEM batteries relative to IPU cases. This being said, I plan to get some assistance producing a DXF file for an 84-cell baseplate. Hole locations for mounting the baseplate would not be included in the baseplate DXF file since the mounting holes would likely not be in correct locations for whichever vehicle the baseplate were to be installed.

The baseplate is cut from 1/4" X 24” X 24 3/8” aluminum. We may be able to eliminate the extra 3/8”. The spacer for the left rear sub-pack could probably overhang the baseplate by that amount. Spacers could be cut from the cutouts for the junction board and the left-front under-slung battery pack. The appropriate number of spacers could be stacked to give correct heights. Mounting holes could be included in the DXF file for the spacers and also for spacer mountings in the baseplate. The spacer mounting holes would be located to not interfere with any of the battery tab holes.

With so many different battery tab arrangements and uncertainty as which type would be mounted where, locations for the battery tab holes would probably be best saved for final installation. It will take a bit of time before I can publish the DXF file to IC. I need to see about getting a baseplate cut using a CNC water jet machine and then test to make sure it has the proper fit. DXF files and CNC water jet cutting are all new to me!
 

·
Registered
Joined
·
79 Posts
Discussion Starter #7 (Edited)
This is probably the first posting of charge data for an 84-cell LTO pack.
82581

The pack was changed from 183 volts to 202 volts, but within 30 minutes of charging, the voltage had settled back down to 200 volts. The charge data shows significant resistance to increasing voltage at 198 volts. The voltage slope then returned to be more consistent with what it had been earlier. It’s very doubtful this extra energy actually went into useful energy storage. I’d be interested in knowing what happened to this excess energy if not stored in the cells. I will be running the pack through my discharger/car today on my commute back and forth to work. Then run another charge cycle tonight to see if this trend continues.

For those who are wondering what the BCM replacer main screen looks like, this photo was taken soon after charging.
82582


This is the BCM Replacer voltage tap screen.
82587

All voltage taps for this have been tracking in unison during assist and regeneration. I was curious as to whether they would remain together after grid charging and they all still appear to still be dead on.

This is the corresponding OBDC&C screen.
82585

We had been aiming for a 82.5% voltage hack, but the 152V/200V = 76%. Maybe I need to change something in the setup? Peter can weigh in on this.
 

·
Super Moderator
Joined
·
7,150 Posts
This is probably the first posting of charge data for an 84-cell LTO pack.
View attachment 82581
The pack was changed from 183 volts to 202 volts, but within 30 minutes of charging, the voltage had settled back down to 200 volts. The charge data shows significant resistance to increasing voltage at 198 volts. The voltage slope then returned to be more consistent with what it had been earlier.
I've seen this little charging "hump" in every 6 or 8 cell segment that I've cycled on the PL8 charger. First time I saw it I was alarmed, but then recognized it is a characteristic of every cell or group of cells. Not sure of the explanation, perhaps some sort of chemical phase change.
 

·
Administrator
Joined
·
10,770 Posts
Looks like a minor bug in the Replacer code with the BCMD Screen.
If the high V is 14.3 and the low V 14.3, how can the average V be 20.. :unsure:

When your LTO pack is exhausted how low does the OBDIIC&C voltage get?
What is the displayed BCM replacer voltage and the Obdiic&c voltage?

Please set your OBDIIC&C to display the Mdv voltage parameter and report what that says as well as Bvo.
They should be very close.

We may also need to adjust your resting positive recal menu numbers..
Soc perhaps from 80-90%
 

·
Administrator
Joined
·
10,770 Posts
Interesting. Bvo and Mdv are too far apart...
I need to check some stuff.

Please go into BCM Replacer menu 1 and note the value shown for "4 Voltage Hack 0-100% "

Change it up or down by 1 unit and see if the Bvo voltage changes.
If it does try and get it nearer to what Mdv is showing.

Report back. Thanks

This effect might also be due to the resistors we had to use at IF not sure if they were 1% tolerance.
 

·
Registered
Joined
·
6,124 Posts
....The charge data shows significant resistance to increasing voltage at 198 volts. The voltage slope then returned to be more consistent with what it had been earlier. It’s very doubtful this extra energy actually went into useful energy storage. I’d be interested in knowing what happened to this excess energy if not stored in the cells.
So, are you suggesting that the flat portion of the curve reflects a "resistance" to charging, because the voltage isn't increasing? At first I thought you were talking about the 'hump', which starts at about 192V and ends at about 198, but that doesn't seem to be the case. I don't think the flat portion reflects lack of charging - think about NiMH charge curves, which are virtually flat at low current, yet they're charging. There's probably some kind of chemistry change going on, like Jime suggests, that 'simply' results in a slightly different charge curve for that voltage range...

It's interesting - I see a similar hump with my small, yet high power LTO cells - yet it's much less pronounced and at a higher voltage. In general, too, I'm surprised to see these cells charging to such a high state at, what?, only 2.4V (202/84=2.404V)? The nominal for my cells is 2.4V, for yours it's 2.3, but still, I thought yours charged more like mine... At 2.4V and a 1C charge rate, mine would be charged to only about 15%. You're not charging at 1C, but I don't think that's the source of the difference...
 

·
Registered
Joined
·
79 Posts
Discussion Starter #13
I’m also thinking along the lines of you and JIME that there is some kind of chemistry change to explain the flat part of the curve at 198 volts. This trend was repeated again last night during charging. Based on the energy availability while driving, these batteries are storing a lot of energy at only 2.4v/cell, and there is likely a lot of additional capacity above this voltage. I just ordered this energy meter: https://www.amazon.com/DROK-Display-Digital-Multimeter-Voltmeter/dp/B07B4CWKRJ
I thought it might be useful in determining total pack charge and discharge energy under controlled conditions using the Genesis charger/discharger. The Genesis already has all the data required during charging, so this meter could be useful for validation. I don’t recall seeing discharge amps in the Genesis log file which is needed for doing any sort of efficiency calculations.
 

·
Super Moderator
Joined
·
7,150 Posts
I thought it might be useful in determining total pack charge and discharge energy under controlled conditions using the Genesis charger/discharger. The Genesis already has all the data required during charging, so this meter could be useful for validation. I don’t recall seeing discharge amps in the Genesis log file which is needed for doing any sort of efficiency calculations.
The Genesis discharge rate is only about 3 amps so it is even lower than 1C. It will probably yield capacity numbers over the actual level, but it will give you a good reference mark that you can check against over time.

I don't think anyone has figured out what the original spec Ahr of the packs was, but I've seen 21Ah 2.7>1.5V at 1C, so I'm thinking they were originally 22Ah. That was one of the SCiB spec levels.
 

·
Registered
Joined
·
79 Posts
Discussion Starter #15
Interesting. Bvo and Mdv are too far apart...
I need to check some stuff.

Please go into BCM Replacer menu 1 and note the value shown for "4 Voltage Hack 0-100% "

Change it up or down by 1 unit and see if the Bvo voltage changes.
If it does try and get it nearer to what Mdv is showing.

Report back. Thanks

This effect might also be due to the resistors we had to use at IF not sure if they were 1% tolerance.
Changing the voltage hack up or down causes more problems. Raising it above 18 throws a P1576. Lowering it a few units has no affect, but continuing to lower it results in nonsensical Bvo/Mdv voltages. Should I look at the resistors?
 

·
Administrator
Joined
·
10,770 Posts

·
Registered
Joined
·
79 Posts
Discussion Starter #19
Moving between 18, 17, 16, 15, 14, and 13 has no effect on voltage readings, they stay in the range of Bvo 149 and Mdv 161. At 12 and below the voltage readings are nonsensical with reading in the Bvo 157 and Mdv 8 range. For 19 and above, code P1576 occurs. The unequal voltage reading do not seem have any ill effects. If I'd started out displaying Mdv instead of Bvo, then I wouldn't have known anything was amiss.
 

·
Registered
Joined
·
6,124 Posts
I don't think anyone has figured out what the original spec Ahr of the packs was, but I've seen 21Ah 2.7>1.5V at 1C, so I'm thinking they were originally 22Ah. That was one of the SCiB spec levels.
I don't think Toshiba was making the 22Ah cells back when the Fit EV was coming out, could be wrong*, but... I thought they were rated, or at least called, "20Ah" - that's what they're being sold as, aren't they?... The small Toshiba SCiB high power cells are called "2.9Ah" but put out about 3.1 Ah at 1C. The original Insight cells were rated 6.5Ah but were supposed to put out 6.8Ah on average. So, basically, the actual amp-hour count can differ from the nominal rating. I didn't think there was any mystery about the Fit EV cells being the 20Ah ones...

*edit: actually, it looks like they've 'always' been making 20Ah and 23Ah, I don't see any 22Ah ones. hmm... I'm kind of suspicious now. Your supplier calls them 20Ah, a quick search comes up with stuff pointing to 20Ah, which is also consistent with what I recall reading over the last couple years. But, you're pulling 21Ah out of an old, used cell... Despite what 'everyone' says, there's probably a possibility that the Fit EV did come with cells larger than a nominal 20Ah. Were it 23Ah with a potential new discharge of say +5%, which puts us at about 24.1Ah, and say usage and age have dropped capacity to 90% of original, which seems conservative even, we'd be at 21.7Ah... Of course this is all purely 'academic' - it doesn't really matter whether they're called 23Ah or 20Ah, cuz apparently they are the exact same chemistry, and in the Insight LTO context 23 vs 20 makes no functional difference...

edit some more: I'm looking at some degradation info, 3000 cycles 95% capacity, 6000 cycles, 90% capacity, and very gradually falling only very little after that. That's a lot of cycles. It seems completely possible that the Fit EV started with nominally rated 20Ah cells that could actually put out something more like 21.4Ah when new, and have seen maybe 5% degradation by the time they fall into Insight owners' hands, leaving them at about 20.3Ah... In light of this, and since 'everyone' says they're 20Ah, I'd have to go with that, seems more plausible than them being 23Ah all along...
 
1 - 20 of 35 Posts
Top