[IamIan]

Just a few thoughts on M cell size NiMH batteries in relation to the Gen-1 Honda Insight.

The M cell size is the size one full step larger than the F ...

AAAA ( the size in '9v' batteries )
AAA
AA
A
C
D
F
M

Like the other sizes there can be slight variation between different manufactures even in the same size and same chemistry... different Capacity , size details , weights, power rating, etc.

The sample of 2 cells I picked up a few years ago to test out as a potential candidate for future use in my Gen-1 Insight were.

~400grams each
~42mm Outside Diameter
~91mm Long
Rated at 26Ah

I pulled them out the other day ... to consider the pros and cons vs the A123 20Ah rated pouch cells I bought more recently ( with the group buy we had here in InsightCentral ).

In theory they could be a near drop in replacement ... would still need a SoC fooling device similar to Peter's ... sense it was discovered by others here in Insight Central that OEM the car will cap you to a limited amount of useable battery capacity no matter how much you actually have.

They are significantly larger than the D cell size ... but due to all the extra space open space in a OEM Battery box one should be able to nest 120 cells into that space with little modification... which on that front would put them near the the A123 in terms of space occupied in the vehicle.

Charging is simpler with NiMH chemistry because they are more tolerant of abuse with things like over charging and over discharging... the OEM had rated these M cells to sustain 3A over charge for 26 days without cell rupter or venting ... just heating up ... and as long as the heat produced was dissipated they would not be toasted or significantly harmed from just an event ... of course such events on a regular basis could potentially reduce the usable cycle life into the 300 to 500 cycle range ... but not a death sentence like such a 26 day 3 Amp over charge would be to a A123 20 Ah pouch cell... slower rates at even lower amps will over charge less , are less stressful , and have less negative effects ... and of course like Panasonic did for Honda in the Insight ... narrowing the SoC / DoD usage band closer to the middle ... 90-10 , 80-20 , 70-30 ... etc ... will greatly increase the cycle life of the cells.

They are also very tolerant of over discharging and even the occasional cell reversal ... although of course best to avoid if one can.

So one could potentially save a bit of money $ and complexity in a far simpler electronics for a M cell based battery pack.

Also there is very slight total capacity benefit ... the 120 M 26Ah cells vs the 50 A123 20Ah cells ... the NiMH M cells would have about ~17% more total capacity in that same vehicle space.

Those are all good things ... but there are significant cons as well.

The best pricing I know of for NiMH M cells like the 26 Ah kind I got ... would cost about 3x as much for the 120 M 26Ah cells compared to 50 of the A123 20Ah pouch cells... that's a major negative.

Another is that while the volume used in the vehicle may be virtually the same between the two options ... the NiMH 26Ah M cell's lower energy density would mean that it would weight almost 2x as much ( ~25 more kg ) ... compared to the 50 A123 20Ah pouch cells... which is another major negative for such a option.

As far as I know ... none of the Low Self Discharge manufactures of NiMH are yet producing in the M size ... largest I've seen is the D size ... which means that such a 120 cell build will have significantly more self discharge than the A123 cells ... this is another negative as long as the LSD-NiMH are not used ... LSD-NiMH would be fairly similar self discharge compared to the A123.

Both are able to have cycle efficiencies in the mid to upper 90% range ... but to use a trickle over charging method to balance the NiMH will waste a lot of energy and can bring that down a great deal... of course using a bleeding resistor based BMS to top end balance the A123 would be no better than the NiMH top end wasted heat either.... but there is a few slight difference to consider there as well.

There is also something to be said about the difference in Peukert effects ... meaning ... the faster one discharges a given battery cell ... the less % of Ah or Wh you get from it ... the chemical reactions are slower than the electronics ... you could let the fast discharged cell rest for a while and come back to get a bit more out and so forth to narrow the gap ... but the effect is due to the chemical reactions dispersion in the medium in the cell being much slower than the rate of the electronics pulling energy from the cell.

Tests of the M cells showed that at fairly lower Amp rates ~10A ... they still maintained more total usable pack energy than a A123 pack would have had.

However there are many people who have tested the A123 20Ah Pouch cells at very high amp rates ... and shown very little Peukert usable energy loss ... Although to be fair I haven't done it yet ... eventually on my list to do is to run a higher amp series of tests on the M cells to see how well they hold up ... but until I do that ... not all NiMH cells , but most NiMH cells have significant Peukert effects from higher power rates ... which is something to consider in a vehicle application with the potential for higher power rates.

So while there are some pros ... over all I think the cons would out weigh the pros for most people ... ~2x weight , and ~3x $ ... seeming to be the 2 biggest cons of the 26Ah M cell.

 

[kaypee171]

How many amps can they take? That has been one of the big issues with finding suitable NiMH batteries. 100 amp discharges and 50 amp charges are a tall order. I would like to say cost is the next big obstacle, but I imagine plug-n-playability is second. While the M batteries would be far better than the A123s, you would still at the very least need to bore out the holes for their diameter. For now, Bumblebee batteries and the like are the best option, but if they can take the charge/discharge rates, I see some potential here, as the cost of alterations to make a lithium battery feasible dwarfs the cost of the batteries right now unless one is capable of a DIY.

What M did you use, and where did you get them?

 

[kaypee171]

Found them, I think. They can handle about half of the discharge we need, and about one tenth of the charge rate we need. Now, if we could get the company that makes the batteries used in the Bumblebee batteries to adapt the chemistry to F or M sizes, we would really be on to something.

 

[IamIan]

How many amps can they take? That has been one of the big issues with finding suitable NiMH batteries. 100 amp discharges and 50 amp charges are a tall order.

I agree the power level is the big issue with finding suitable batteries ... NiMH or otherwise ... and of the two ... ~50A charge rates is harder than 100Amp discharge rates.

Sure a given battery might take it a few times ... but it needs to take it hundreds ... preferably thousands of times to have a reasonable life expectancy in use.

There is one saving grace ... the closer to the middle and further from away from the full 100% DoD window the easier it is to do higher power rates ... and the Insight stays mostly in the 20-80% range.

If the seller is to be believed I was originally told ( a few years ago ) these M cells will do ... 3C charge and discharge rate for the full 100% DoD window for a 80% retained capacity after 500 cycles ... 3C for a 26Ah cell is ~78Amps ... they also quoted a max rapid rate of 10C ... for 26Ah cell that's ~260Amps ... I think those are overly optimistic rates ... but I suspect they will do the 50A Charge and 100A discharge rates ... But I have not verified it with tersting yet.

Knowing we will stay closer to the middle 20-80% DoD ... they should be fine with it ... they did not seem to warm up any more than the OEM NiMH 6.5Ah D cells did when I did tests on them at the lower 10A rate ... if anything the M cells stayed much cooler on the discharges than the OEM D cells did... on par with the A123 for low level heating on discharge ... on charging the M cells only started to heat up toward the top of the charge window ... above 70% SoC ... although they didn't seem to get significantly hotter than the OEM D cells did ... they did get warmer during charge than the A123 did.

I've previously tested these NiMH M cells at a 10A rate ... They still put out a bit over 26 Ah ... the A123 -20Ah Pouch cells I tested at that rate were giving under ~19 Ah ... testing the NiMH M cells at a higher rate ... somewhere around 60 to 120 amps is on my to do list ... I just haven't got around to it yet.

I got a bit side tracked with other projects... Story of my life.

I would like to say cost is the next big obstacle, but I imagine plug-n-playability is second. While the M batteries would be far better than the A123s, you would still at the very least need to bore out the holes for their diameter.

Depends on the person of course ... for nearly the same Wh of capacity the 50x A123 -20Ah pouch cells are lighter and cheaper for the batteries than NiMH M cells would be.

The NiMH M cell , might be a simpler option , but it is heavier and more expensive... and has a higher self discharge rate.

The A123 and other Lithium options need additional cost and complexity for more sophisticated BMS is a negative for them ... but that balances out a bit with the more expensive and heavier NiMH alternative for a comparable amount of capacity... no free lunch as they say.

It has been well tested that unless you have a bad cell ... the A123 can take the charge and discharge rates without issue... and although they might look to be ok with it at lower power levels ... the M cells have not yet proven themselves at higher rates... although the OEM makes some bold claims.

For now, Bumblebee batteries and the like are the best option, but if they can take the charge/discharge rates, I see some potential here, as the cost of alterations to make a lithium battery feasible dwarfs the cost of the batteries right now unless one is capable of a DIY.

I think anything with more capacity than what Bumblebee offers will require a modified Battery case / box ... and very likely some type of alterations to the OEM BMS.

I agree ... Unless it is a DIY , any other new cell option will cost more than what Bumblebee offers... only if you are doing DIY or reusing old used batteries do I think you'll have a chance to beat what he is currently offering.

What M did you use, and where did you get them?

M is the size ... like a D cell ... or a C cell ... or a AA.... etc.

Although there are some variations around the base size for any of these ... the M size just tells you the generic approximate size of the cell.

Same thing in Lithium if you say 18650 ... that's just a standard size for the cell .. there will be some minor variation around it ... but fairly close to that standardized size.

I got these 26Ah NiMH M cells from HardingEnergy.com a few years ago ... before they changed their sales policies ... I only bought 2 cells at that time for testing ... I don't know what their new sales policy is ... but I suspect it will involve a minimum order quantity ... but you can still buy and find NiMH M cells around other sites from time to time ... batteryspace.com consistently has either the 20Ah or the 26Ah version of the NiMH M cell available... although it might not be exactly the same as the one Harding Offers ... so I would recommend testing to verify performance.

 

[Eli]

Ah, as usual, great minds..

I picked it up several months back to play with. Naturally, I haven't had a chance to do any serious tests on it. I suspect it likely has too high an internal resistance to be used at full assist/regen rates, but would be interesting at lower currents.

I should set the single cell test rig up again and see what it'll do at some higher rates. The problem is going to be consistency - these cells are almost certainly hand made. Same problems overall as the hand made King Kong cells Mike and I tested. They might not even have real metal plates etc.

 

[kaypee171]

I figured there would be some fabrication, but NiMH has a big advantage there. It seems like NiMH would be a case of modifying the battery case a bit-perhaps as simple as boring out the holes. Lithium would be a new case, BMS, and a ton of foolers. As Peter once quoted on an old thread, he could convert your car to lithium for you, if you could get your car to him along with £7000. Even at $40 a pop for the M's, I bet it can be done cheaper than that. My gut still tells me a normal battery (or Bumblebee) plus a design similar to Mike's V buck is the simplest, most cost effective way to add capacity.

 

[retepsnikrep]

The Lithium works well as we have seen with the right cells (A123 20ah), fits in the ipu compartment and weighs about the same as the stock pack for ~18ah of usable capacity.

The problem with more/bigger Nimh is the wh/kg is a lot worse so they weigh three times as much We really want extra capacity without the weight/space penalty.

Having driven the A123 car for three years and given loads of people test drives I would not put a bigger capacity lithium pack in unless it could. A) Fit in the oem battery space. B) Weigh the same or less than what I already have.

A heavy lump at the back does effect the handling of the car, and I could notice the weight of the 40ah cells I had in the citrus car bootwell of my first conversion. You could not push that car so far into bends and even with matiz rear springs it was working hard at the back.

Fitting heavier/more/different cells to a slug like the HCH1 Civic is easier (technically), and it can handle the boot/rear suspension load and you have quite a bit of space in the back for packs/cells.

I doubt there will ever be a reasonably priced G1 off the shelf plug in conversion offering greater capacity using Lithium or bigger Nimh cells. As soon as you start re-engineering the battery case and electronics the cost spirals.

IMHO the cheapest and simplest way to add extra capacity is a single flat layer buddy/parallel stock or 8ah nimh pack on the rear shelf area.(With appropriate safeties!! Separate grid chargers, big diodes etc) In a buddy pack configuration you could use cheaper lesser spec but higher capacity cells, so perhaps the yabo/sucre sticks would be fine for that as it won't be exposed to the same current demands if set up correctly.

 

[IamIan]

Ah, as usual, great minds..

I picked it up several months back to play with. Naturally, I haven't had a chance to do any serious tests on it. I suspect it likely has too high an internal resistance to be used at full assist/regen rates, but would be interesting at lower currents.

I'm pretty much in the same boat ... Picked up the two I have for testing ... should at the very least , be interesting ... Although they did fine at lower rates ~10Amps ... gave a estimated V/I of about ~3mOhms ... but higher Amps rates are different ... proof is in the pudding , as they say ... I suspect Harding ( the seller in my case ) was stretching things with the 3C and 10C they were telling me ... but I'll know better once I beat the crap out of the ones I have at higher rates.

I should set the single cell test rig up again and see what it'll do at some higher rates. The problem is going to be consistency - these cells are almost certainly hand made. Same problems overall as the hand made King Kong cells Mike and I tested. They might not even have real metal plates etc.

I agree many questions would need to be answered and ironed out , for them to have a chance of being viable ... and even if all of that happens ... the pros and cons would still have to be weighed out , to see if they are better than some of the other options... paying more money for a heavier battery pack needs a significant benefit to offset those 2 strikes against it.

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

I figured there would be some fabrication, but NiMH has a big advantage there. It seems like NiMH would be a case of modifying the battery case a bit-perhaps as simple as boring out the holes. Lithium would be a new case, BMS, and a ton of foolers. As Peter once quoted on an old thread, he could convert your car to lithium for you, if you could get your car to him along with £7000. Even at $40 a pop for the M's, I bet it can be done cheaper than that. My gut still tells me a normal battery (or Bumblebee) plus a design similar to Mike's V buck is the simplest, most cost effective way to add capacity.

Anything over about 8Ah will need a fooler of some kind in order to make use of the additional capacity ... some other people who used a double OEM pack ran into the OEM BCM limits on usable capacity... and that was with about ~12Ah ... not even as high as ~26Ah or so.

Hard to tell where the break even point is ... less BMS needed vs more $ the cells themselves ... and the additional ~25kg of weight that a 120 M cell pack would weigh compared to a 50 A123 Cell pack ... for about the same usable Wh of energy capacity.

And it is not yet known for certain ... needs more testing ... to know if these M cells will take the 50A charge and 100A discharge rates for a reasonable number of cycle life... they gave good indications at 10A ... but 10A is a long way from proof at 100A.

The booster 2nd battery is definitely a potential option ... it will add a weight penalty ... although Mike did 300 Lbs for several years ... it trades the foolers and BMS for a power control electronics for the interface between the primary and secondary booster battery.

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

The problem with more/bigger Nimh is the wh/kg is a lot worse so they weigh three times as much We really want extra capacity without the weight/space penalty.

Their is a weight penalty ... the M cells Harding sold me would weigh about ~25kg more for a 120 M cell pack than the 50 A123 cell pack ... at least at the battery level ... not counting the BMS ... but the BMS is way less than ~25kg ... so the 50 A123 cell pack will still be lighter.

I think the specs are ... ~80wh/kg for the M cell vs ~120wh/kg for the A123.

Having driven the A123 car for three years and given loads of people test drives

Tried and true ... tested and known ... etc ... always has an edge.
We know it can work reliably.

I doubt there will ever be a reasonably priced G1 off the shelf plug in conversion offering greater capacity using Lithium or bigger Nimh cells. As soon as you start re-engineering the battery case and electronics the cost spirals.

Agreed.

 

[kaypee171]

Don't get me wrong, I would dearly love to have a lithium-equipped Insight. I just don't think it will ever be viable for people like me whose electrical skill is limited to installing a stereo or replacing an alternator. I was pretty proud of myself that I built my own dumb grid charger even though I was just copying another's design. Now my goals are to pay off my car, buy one of Mike's smart chargers and buy a Bumblebee battery, in that order. I will keep an eye on re-involt, as they imply they will have a lithium battery ready for the HCH1 soon, although I think they have been saying that for a while. Otherwise, I will probably set up a buddy pack.

That does make me think of a previous thought. I wouldn't mind it if we could limit the draw the IMA puts on the batteries if it meant more capacity. Of course, I might feel differently if I had a battery that could pull it off.

 

[IamIan]

That does make me think of a previous thought. I wouldn't mind it if we could limit the draw the IMA puts on the batteries if it meant more capacity. Of course, I might feel differently if I had a battery that could pull it off.

MIMA-L the DIY version will let you limit the IMA amps for charge or discharge ... it is a bit of a project ... so if you are not the type yourself ... that would either be a big step for a initial type of project ... or you could check around to see if you can find someone to build it for you.