Honda Insight Forum banner

1 - 20 of 76 Posts

·
Super Moderator
Joined
·
8,144 Posts
Discussion Starter #1 (Edited)
I decided to start a new thread on this subject, not because eq1 wasn't giving me plenty of help, but rather so the build/experiment would have a proper title. I owe a debt of gratitude to eq1, mudder and others who have already contributed to the idea of alternative 12V batteries in this original thread:


---------------------------------------------------------------------------------------


I've become interest in trying this approach of using LTO cells to replace the lead acid type 51. I guess everyone seems to have bad experiences with service life on the seldom used PbAcid - me included.

I fortunately, or unfortunately, have an excess of of the LTO packs, some of which have damaged intercell straps because of a particular risk when removing covers. I'm thinking of splitting some of the subpacks to produce either 5 cell or 6 cell starting batteries. I did a little tinkering and I think there might be a decent chance of "isolating" a few cells for this purpose.

I've read through most of your posts on the subject and I understand your concerns. It seems that most of the concerns were answered, but I assume that since you find the NiMH pairs so satisfactory you have lost interest in the Lithiums.
 

·
Registered
Joined
·
1,339 Posts
are used nickel metal hydride batteries to start my car for a short while. I got stranded my cousin came and boosted me off. Went back to flood battery.
 

·
Registered
Joined
·
6,742 Posts
@eq1

....I've read through most of your posts on the subject and I understand your concerns. It seems that most of the concerns were answered, but I assume that since you find the NiMH pairs so satisfactory you have lost interest in the Lithiums.
I didn't lose interest, exactly, but I was working with small cells with tapped terminals that are really crappy, so it was taking too long to assemble a suitable 'package'... I guess I was trying to make a plastic box and kind of petered-out on that part, otherwise, I think I have everything I need... Having a working NiMH 12V - and a virtual endless supply of those - does kind of remove incentive, too...

The Fit cells would be great for a 12V battery - a nice, big one. I think I'd do 5 cells, 13.85V total in use, 2.77/cell (disconnect DVCT wire, which will make the DCDC output voltage max-out at 13.85V, rather than being high when the engine is cold). You're supposed to be able to float the Toshiba cells at 2.7V, I don't think 2.77V would make much difference (a lot of LTO cells actually have 2.8V as the max V) -- provided the cells are balanced and stay reasonably so...

You can buy 'cell balancing extension' wires/harness at Hobbyking really cheap, they have JST connectors I think they're called, male and female, on either end -- like what you plug into your hobby charger cell port. You'd pull one connector off and attach the wires to your cells, so you can then use a 'Battery medic' (cell checker./balancer) to check cell voltages and balance them occasionally...

I guess this has been more or less my plan with my LTO cells. I'll probably do it at some point, maybe soon...
 

·
Super Moderator
Joined
·
8,144 Posts
Discussion Starter #4
Still a bunch of practical mechanical problems with the Fit packs. I was able quickly to separate the two halves, but now the real work begins - splitting off 5 cells. I recall some very early effort by insightbuyer to split off a single cell for testing perposes. He wound up driving a thin paint scraper between cells to separate the cells. We'll see.
 

·
Registered
Joined
·
6,742 Posts
Good luck. Feel free to add anything about your project here if you want... I'd be interested in seeing how you go about it, how it turns out, etc.
 

·
Super Moderator
Joined
·
8,144 Posts
Discussion Starter #6 (Edited)
The Fit cells would be great for a 12V battery - a nice, big one. I think I'd do 5 cells, 13.85V total in use, 2.77/cell (disconnect DVCT wire, which will make the DCDC output voltage max-out at 13.85V, rather than being high when the engine is cold).
Putting the finishing touches on the 5 cell 12V LTO battery. Working on cables and some spacers, etc.

I could not find the DVCT wire in the schematics. Can you give me a page reference in the Electrical Troubleshooting Guide?

BTW, I'm not recommending anyone else actually try this for the moment. There are some safety issues, particularly with splitting off 5 cells, that remain unresolved seems to me.(edit; More later on that. Just don't try it for now.)

LATER: Never mind. I found several threads on the subject. Should have searched first:oops:
 

·
Registered
Joined
·
6,742 Posts
In case anyone else has the same question, here's a link to a post that contains directions to the DVCT wire, and some images that should help: Mod to Adjust DCDC Converter Output Voltage

One can disconnect DVCT at the ECM or at the DCDC connector in the IMA compartment. I suggest carefully pulling the pin out, very easy, rather than cutting the wire.

At the ECM, DVCT is RED/YEL on Connector A (GREY), A7.
At the DCDC, it's the RED/YEL wire.
 

·
Super Moderator
Joined
·
8,144 Posts
Discussion Starter #8
Thanks for the tip on the shotcut wire location. I found it at the DC-DC, which is the most convenient location. I was reading your other refs and learning a lot about the circuit, but I had not found a good "map" to the wire.

Do you think just pulling the pin out creates any lasting danger to the plug grip? I don't have the pin removal tool. I don't mind cutting wires. I've gotten really good at solder splices and heat shrink :)
 

·
Super Moderator
Joined
·
8,144 Posts
Discussion Starter #9
One of the safety concerns I have with using the Fit LTO cells for this purpose is that the enclosure isn't waterproof. The LTO cell outter cases are "hot" and both the top and bottom of the cells are exposed to potential dirty rain water - a fair conductor iirc.

I'm thinking about using a bunch of silicone sealant to seal up the tops and bottoms of the 12V battery I've created, but that raises another question. Is normal silicon sealant a good insulator? Are there better ones like two part exoxy? I'll get some experimental data, but the issue is worrisome.
 

·
Registered
Joined
·
6,742 Posts
Do you think just pulling the pin out creates any lasting danger to the plug grip?
Not sure I understand, "plug grip." I don't see any negative consequence from pulling the pin out. It was real easy when I did it, just a little tab or something to depress with a pin and it slides right out...

One of the safety concerns I have with using the Fit LTO cells for this purpose is that the enclosure isn't waterproof....I'm thinking about using a bunch of silicone sealant to seal up the tops and bottoms of the 12V battery... Is normal silicon sealant a good insulator? Are there better ones like two part epoxy?
I definitely think it'd be good to make sure the enclosure is water tight, or at least mostly-water-deflecting with a drain hole, vent, and the appropriate precautions for keeping the cells apart. But if you do go totally water tight, you need to make sure it is absolutely water tight - because you don't want to have some minor egress for water and have it end up pooling or just making it damp and humid in the box...

I'd have to see what you're doing to understand if silicone would be good. I don't think silicone sticks very well to what I'm guessing is polyethelene plastic, the black plastic around the Fit cells? Silicone is a good 'insulator' against water, what kind of insulator do you want? Electrical insulator? Thermal insulator?
 

·
Super Moderator
Joined
·
8,144 Posts
Discussion Starter #11
Thanks for the tips on the pin. I have some junk plugs from my parts car so I'll experiment from your directions.

I was thinking of using the silicon to completely seal the top especially. If you look at the tops of the LTO packs in this theread:


you can get an idea of just how open to water the tops of the cells and the intercell straps really are. Plus, and this doesn't show well, the plastic wrap forms a perfect little pool on top of the batteries. The bottoms of the packs are also quite open.

I've been thinking that a decent plan might be to use something to seal the top against water, but it must also be a good good electrical insulator, and then leave the bottom relatively open since that will be protected from direct splash by the OEM battery well. I can also create proper spacers and drainage and after all gravity is on my side at the bottom. The plastic at the top does appear to be polyethelene.
 

·
Registered
Joined
·
6,742 Posts
In general the area where the stock battery goes stays dry, at least on the top, so I think what you're saying about protecting the top sounds about right. I've occasionally seen a drop or two of water on my NiMH 12V 'tubes', but that's about it. I think the stock battery box gets a lot of spray on the bottom - but if that's in place and your battery is inside it, it should be protected...

I think silicone would work, but you'd probably have to rough-up the edges of the plastic pieces you're joining, such as with...220 grit sandpaper. I'm pretty sure the silicone won't stick well to smooth, glossy plastic surfaces, it just peels off easily. It partly depends on how much you think the top cover would be disturbed, and/or how permanent you want it. Like if it fit more or less like a loose tupperware cover, and you put a good bead of GE Silicone II around the 'lid' and pressed together, let it cure, I'm sure it'd hold and seal just fine. And you'd still be able to remove it if/when you needed to.

If you wanted a more permanent, robust seal you'd rough the edges and use the silicone, or you'd use a two-part epoxy that says it works with polyethelene (if that's what it is, if it's ABS or PVC I think most epoxies stick to that well, there's just something about polyethelene and polypropylene, I think, that make them hard to 'glue' to). But I wouldn't think you'd want the kind of permanent lid an epoxy would create...

Not sure where being an electrical insulator comes into play, with the top cover... I can't imagine silicone is conductive, but I've never tested it or read that it isn't...

I think you'll need to leave/make a 'breather' hole in the top somewhere, somewhere where water wouldn't get into it, but vapor could escape. IF moisture/water gets underneath the battery, like in the OEM tray, and can then enter the battery box from underneath, such as when it's drying-out and vaporizing, you wouldn't want that to condense on the cells or metal bits, or on the underside of the lid. So a hole in the lid, like underneath an edge or lip, would be needed, to let that vapor escape... I think this is pretty important IF the underside isn't well sealed, basically, if any water vapor can enter the box... If the whole box is perfectly sealed, then there's no problem; if it's only partially sealed/covered, then you need to worry about condensation and making sure there's adequate ventilation...
 

·
Super Moderator
Joined
·
8,144 Posts
Discussion Starter #13
Thanks for the thoughts. All of it is good thought and I'm rolling over in my own mind how to implement much of it.

I do have another more major concern, but for now I'm going to move ahead. The concern - I'm not quite sure that the 5 cell approach is quite right vs. using 6 cells. I've played around with the rather simple math and assumptions in arriving at this guesstimate.
1. In the absence of a BMS intergral to the battery, it seems only sensible to derate the operating range to something like 2.0-2.5V. From my PL8 testing, these old used cells already don't have equalized capacities. For the most part, these capacity differences are small, but they may be significant if an assembly of cell is operated at the limits of the specifications.
2. If one derates to say 2.0-2.5V then the defined operating for a 5 cell assembly is 10.0-12.5V. The operating range of a 6 cell assembly would be 12.0-15.0V.
3. Since cutting the DVCT results in a pegged DC-DC output of 13.85V, the "maximum" charging voltage is well within the "specification" of a cell pack.

This analysis may have holes, but I don't intend that it stop experimental progress. I plan on installing the already built 5 cell pack and doing some testing. I just think it misplaced to do much more than the minimum with the 5 cell pack at the moment. I have to move beyond some of my operational concerns and gather some data - in the dry.
 

·
Super Moderator
Joined
·
8,144 Posts
Discussion Starter #14
Finally found the schematic ref to the DVCT on P32-7 of the ETM :)
 

·
Registered
Joined
·
6,742 Posts
....In the absence of a BMS intergral to the battery, it seems only sensible to derate the operating range to something like 2.0-2.5V... the defined operating for a 5 cell assembly is 10.0-12.5V. The operating range of a 6 cell assembly would be 12.0-15.0V. Since cutting the DVCT results in a pegged DC-DC output of 13.85V, the "maximum" charging voltage is well within the "specification" of a cell pack.
hmm, I think you may be right. It mostly depends on how much capacity you need/want the battery to maintain and what charge state it achieves at such and such voltage. If you did go 6-cell, I think you'd be better off installing a PWM device on DVCT so you can adjust the top voltage, from a range of 13.85V to 14.51V (and a default to 15.13V if DVCT were grounded), rather than disconnecting DVCT...

I see no circumstances where disconnecting DVCT would be necessary; if you leave it as-is, connected, then you'd get a boost of charging when the engine is cold, something like 14.2-14.4V, then it falls to 13.85V. That wouldn't harm anything in a 6-cell config, might be good to have a little extra charging... But, 13.85V seems too low, your battery probably wouldn't be charged enough.

In any event, 15V might be too high for the system - a couple others chimed-in in one of my queries about it and said 15V would start to push electronic components made for a '12V system' beyond their intended limits. For instance, you might get premature burnt-out light bulbs...

I currently have my DCDC output set to I think 14.2V and see no issues. I've had it higher for long-ish durations, never had issues...

The functional range you'll be dealing with in the car will be a low of 12.2V, when the DCDC is in 'low power' mode -- which can happen for longer stretches of time than you'd think, especially now that you'll have a battery whose voltage doesn't sag like a lead acid -- and a high within the range I wrote above, the adjustable range. 12.2V / 6 cells is 2.03V - that seems OK. And on the high side, 13.85/6= 2.31V - seems too low, and 14.51V/6=2.42V - that might work...

So the question is, How much capacity do you get with the Fit cells if charged to about 2.42V? I think the typical CC-CV charge approximates the max charge amount you'd get in the car pretty well; in reality it will usually be maintained at a lower level, maybe by minus 10% on average... Depends on the type of driving you've been doing.

I'd guess with a nominal of 2.3V, a 2.4V charge threshold might get you to 50%? SoC... 2.4V with my small SCiB cells is only like 20% max.* Those have a 2.4V nominal...

Here's a list of the PWM duty cycles on DVCT and the DCDC output voltage you get from it:
100, 13.85V (or DVCT disconnected)
90, 13.85V
80, 13.91
70, 14.02
60, 14.12
50, 14.21
40, 14.30
30, 14.38
20, 14.45
10, 14.51
0, 15.13
(or DVCT grounded)

* I have charge data to 2.5V, 3A charge CC (1C), 300mA at end of CV stage, and ended up at 56% charge state. I don't see a 2.4V CC-CV charge; I do have a 1C charge that shows about 15% charge state at 2.4V, i.e. the nominal for these cells. So, maybe with a nominal of 2.3V your 2.4V charge would be like my 2.5V charge - hitting about 56%?... If that were so you could probably expect to be around 50% charge state most of the time, in-car usage... That would definitely work with your 20Ah cells - you'd have 10Ah of usage. I get by on like maybe 2-4Ah...
 

·
Super Moderator
Joined
·
8,144 Posts
Discussion Starter #16 (Edited)
Thanks for the analysis. It is interesting.

The 5 cell construction is complete and just waiting to put it into my conversion car. I don't see any reason not to give it a try and maybe drive it for a while. No point in doing 90% of the work and not trying. Will need to disconnect the DVCT for this 5 cell experiment. Right?
 

·
Registered
Joined
·
6,742 Posts
Will need to disconnect the DVCT for this 5 cell experiment. Right?
I would. Otherwise your battery will be seeing ~14.2-14.3V / 5 cells= 2.85V per cell, when the engine's cold... I don't know, 2.77V (at 13.85V total), vs. 2.85V -- I kind of doubt it would make much difference, not in the short term.
 

·
Registered
Joined
·
79 Posts
Thanks for the analysis. It is interesting.

The 5 cell construction is complete and just waiting to put it into my conversion car. I don't see any reason not to give it a try and maybe drive it for a while. No point in doing 90% of the work and not trying. Will need to disconnect the DVCT for this 5 cell experiment. Right?
In the past before going a different route I tried to split the 12 packs into individual cells and wasn't able to do it in an easy repeatable manner. How did this go for you and what method did you use? I found the adhesive on the black plastic hard to remove and definitely found the adhesive between cells really hard to split. Any tips?

Also after getting the individual cells did you wrap them? I remember testing the bare aluminum case and terminal to case had a voltage so it seemed like each one needed to be re-wrapped after splitting.
 

·
Registered
Joined
·
6,742 Posts
^ This won't help you, but you might find it an interesting read - where this guy tried to dismantle a Schwinn Tailwind electric bike battery pack, which pack used Toshiba's cells: Schwinn Tailwind Battery Pack Teardown and Analysis

When he got to the point of trying to extract cells, this is what he had to say:
"Unfortunately for my plan, Toshiba also used copious amounts of a very high quality adhesive to stick this pack together. After about an hour, I gave up - I cannot get the inner pack apart beyond this point without destroying it..."

Also after getting the individual cells did you wrap them? I remember testing the bare aluminum case and terminal to case had a voltage so it seemed like each one needed to be re-wrapped after splitting.
Pretty sure they do need to be individually wrapped, or otherwise isolated. It seems really weird to me. I posted a question a long time ago, here at IC, asking what it means to have the terminal-to-case voltages like they are. No one had an answer.

I gather that the electrolyte must be in contact with the case ('can'), as the pos terminal-to-case circuit and the neg terminal-to-case circuit seem to reflect half-reaction voltages, i.e. the potential between pos electrode and electrolyte, and negative electrode and electrolyte, I think... Add them up and you get the cell voltage. As I recall, when you short between terminal and case you actually start to discharge that electrode - or at least the voltage goes down... I don't really understand it well, if at all...
 

·
Super Moderator
Joined
·
8,144 Posts
Discussion Starter #20 (Edited)
Well. I'm going to start with a caution: ATTEMPTING TO SPLIT ONE OF THE FIT LTO SUBPACKS IS EXCEEDINGLY DIFFICULT AND PERHAPS PERSONALLY DANGEROUS. THE PRACTICE CAN ALSO EASILY RESULT IN DAMAGE TO ONE OR MORE CELLS. USE OF METAL TOOLS HAS A HIGH RISK OF SHORTING TWO CELL OUTER CASES, WHICH ARE ELECTRICALLY HOT.

Now, I first cut the plastic around "divide" line with a Dremel rotary saw blade. It is important to stay slightly off the exact divide line between cells so that the metal saw blade only lightly touches the metal outer case of ONE cell.

The spot weld on the stap between the cells must be driled free at one end. I was very careful to use the very minimum drill depth. Only a little drilling is necessary to free the strap. The strap makes a convenient, though somewhat flimsy, terminal.

All the cells are separated by a single layer polyethelene (?) plastic separator. One wants to retain the undamaged separator with the cell(s) to be used and not those to be discarded. As a separation tool, I used a sturdy but thin paint scraper. The tool needs to be sharpened on one side in a bevel manner like a wood chisel. In this way it can be driven with a hammer between the two cells being split and still follow the wall of the discard cell, saving the single layer plastic separator for the usable cell(s). It requires careful attention and some skill. There is a chance that the metal tool will not follow a true course and will contact both cell walls at the same time, creating a short between two cases, and that will I believe cause a positive to negative short of one cell - probably damaging one cell, and potentially creating a fire. One should be prepared to protect self and property. I did the final separation operation on a concrete slab outside my garage , because of risk to property.

Bottom line is that I successfully separated 5 cells, but I don't recommend the practice because of the danger involved. Obviously this account is my personal experience and does not constitute a recommendation to others to do the same.

Later edit: By cutting one end of the appropriate bus bar FIRST, the risk of shorting two cell cases with metal tools is considerably reduced.
 
1 - 20 of 76 Posts
Top