Prius Batteries - How to make them work

knightmb said:

Don't get me wrong, I'm glad to see the Lithium tech take off and I'm certain that's where the future lies, but don't count out the old faithful just yet, the NiMH still have a good place in the market, just as you said, tried and true technology now, no surprises about how it will perform or how long will it last.

Click to expand...

I agree. Maybe it's just wishful thinking, but I think that NiMH technlogy will sooner or later get a massive boost because of the insight developing that a decent BMS is worth the effort.

Just overcharging (and turning the excess energy into heat) seems to be the accepted modus operandi for NiMH battery string equalisation. That, or just not using the full capacity, at neither the top or the bottom end of SOC levels.

How unsophisticated (!), but nevertheless, long service life and good performance are possible.

With the progress in computing power it should be possible to record the entire history of each cell in a NiMH string and constantly adjust the amount of trickle-charge balancing that gets shuffled between cells to keep the weak ones happy.

And this would also allow for an early warning system (like on a printer: "Almost out of Cyan") to make planned replacement of weak cells possible before they cause damage to the rest of the pack.

The problem is that this needs individual cell monitoring at least; individual cell balancing would be even better. With individual cell monitoring the charger could limit the equalisation charging to the absolutely necessary minimum and recommend cell replacement if the damage to the majority of good cells outweighs the benefit to the weaker cells.

NiMH are also largely non-toxic and do not tend to explode or vent with flame. That's very good for folks like me who learn this stuff "Hands On"!

Well to sound like a total noob at this what charger are you guys using to charge them? I understand how to connect them in series in parallel but how do you safely charge them and how many cells do you hook up and do at a time? I got a prius battery still sitting in the garage I might as well put it to use before it gets thrown out.

I agree NiMH still has potential... and they have proven to be one of the most tollerant of abuse batteries on the market... They are often times over charged and over discharged in common usage ... and in Hybrids they are usually under used staying above 20% SoC and bellow 80% SoC... of course they under use them as a simple and easy method of avoiding the longevity issues that come up from the complexity of top and bottom capacity usage.

I have more experience with the Gen-I Insight & Civic NiMH Hybrid battery packs than the Prius ones... Although I guess they are comparable ... the Prius ones might have a better wh / kg ratio due to the lighter weight plastic casing of each subpack... I know the Gen-I Insight and Civic 6 cell subpack sticks with cylindrical metal ( D cell Shape ) casing ends up weighing in at 1.085 kg including the bolts, PTC strips, and plastic sheath... given the original 6.5Ah rating of the cylindrical cells that puts them at ~43Wh/Kg .... I know all HEV NiMHs are high power cells ... The Insight for instance pulls bursts up to 100 amp discharge rates ( over 15C ) from its 1999 year ( ~10 year old now ) technology NiMH cells ... and lab tests on the Insight battery packs have shown internal resistance ( including all the connections in a 120 cell pack ) to be as low as 0.36Ohms... which puts individual cells in the range of ~3.0 mOhms ... of course it does vary with temperature as shown in the link above and it also varies with SoC shown here... but still this results in a power density of about ~666W/kg .... and a Volumetric energy density of about ~139Wh/L.

My own personal experience here with my original 2000 Model year Insight Battery pack now has almost 130,000 Miles on it and 9 years... When I got an IMA error code last year ... the dealer offered to replace it under the 10 year , 150,000 mile Honda warranty ... but I declined ... I was curious and I had the means to properly test the pack to determine its condition myself... so I did ... I suspected that given a NREL study of the Insight Gen-I battery pack temperature distribution , that the 120 cells were not all being treated equally ... and that I probably didn't need a new pack so much as maybe one or two new cells.... what I found first was that the SoC in the pack was not balanced from 6 cell subpack stick to 6 cell subpack stick... this made sense because the Insight BCM ( Battery Control Module ) like many HEVs does not charge the cells to 100% ... so it never does the top off over charge trickle charge in order to try and balance them... and it doesn't drain them down to 0% SoC so they won't balance there... and it doesn't have any mechanism for controlling charge to individual subpacks or cells ( the way modern Li packs do )... so it made sense to me that given a varying Internal Resistance with Temperature , and a varying temperature in the pack during operation , and no apparent OEM method to balance the battery pack... So I rebalanced the SoC on my pack and put it back into my car... IMA error code went away ... and still hasn't come back yet almost a year latter... I have also over time tested several other Gen-I Insight and Civic battery packs... and this has shown to be common among all the Gen-I Insights and Civics ... I have traded some data with others who also have done similar tests ... and the results are consistent ... I don't know what the % is ... but it seems like a large % of the IMA error codes that people are getting are just a unbalanced SoC between subpacks... maybe a very small % actually have even 1 or 2 bad cells out of the whole 120 cell pack... this is probably one of the reasons Honda extended the battery warranty to 10 years / 150,000 miles in the U.S.

Or the D cells on my E-Bike... I have abused the heck out of those things for ~5 years now... I've soldered , unsoldered , resoldered the pack into different shapes like 3 times ... slow over charge on every charge cycle ... I know I've pulled some of the cells into voltage reversal several times as I have over discharged the whole 30 D cell pack several times ... I let them bake in the summer sun and heat ... freeze in the winter cold ... I pulled them out a little bit ago just to see how bad they are... They still have ~7Ah of usable capacity... that's crazy levels of tolerance to abuse in my book... No other battery I know of could I treat with a much abuse and still expect them to work at all.... eventually ( but not now ) I will build another battery pack for my E-Bike ... and right now , I think I will still go with NiMH ... although maybe I won't abuse the new pack quiet so much ... :wink:

Overcharge a Li battery... it will kill it very quickly ... depending on the type of Li you might get a fire ... Over discharge or pull into voltage reversal a Li battery and it will kill it very quickly ... Li bake in the summer sun also will kill it... 100%+ DoD cycles ... even A123 cells won't tolerate that kind of abuse as well as NiMH does.

As far as large number of cell pack designs go , NiMH also has the advantage of dying as an electrical short circuit ... a large number of cell series NiMH pack has 1 cell fail , the voltage drops but it keeps on trucking ... taking one for the team so to speak... while Li fails as an open circuit... so a large number of cell series Li Pack has 1 cell fail ... the whole series string turns off like x-mas lights... or the current you force through that one bad cell causes a fire.

Li does have better Wh/kg numbers than NiMH ... that I will give Li ... and Wh/kg has always been the #1 issue for 99% of people when it comes to battery powered anything .... so right off the bat ... Li has a large and significant advantage there... but in my book Li's advantages pretty much stop there... and in just about every other way NiMH continue to have competitive if not superior battery characteristics.

just my 2 bits :wink:

---------------
posted edited to correct typo 0.3 mOhms to correct value of 3.0 mOhms

IamIan said:

The Insight for instance pulls bursts up to 100 amp discharge rates ( over 15C ) from its 1999 year ( ~10 year old now ) technology NiMH cells ... and lab tests on the Insight battery packs have shown internal resistance ( including all the connections in a 120 cell pack ) to be as low as 0.36Ohms... which puts individual cells in the range of ~0.3mOhms ... of course it does vary with temperature as shown in the link above and it also varies with SoC shown here... but still this results in a power density of about ~666W/kg .... and a Volumetric energy density of about ~139Wh/L.

Click to expand...

Does this not make 3mOhm / cell?

Mr. Mik said:

IamIan said:

The Insight for instance pulls bursts up to 100 amp discharge rates ( over 15C ) from its 1999 year ( ~10 year old now ) technology NiMH cells ... and lab tests on the Insight battery packs have shown internal resistance ( including all the connections in a 120 cell pack ) to be as low as 0.36Ohms... which puts individual cells in the range of ~0.3mOhms ... of course it does vary with temperature as shown in the link above and it also varies with SoC shown here... but still this results in a power density of about ~666W/kg .... and a Volumetric energy density of about ~139Wh/L.

Click to expand...

Does this not make 3mOhm / cell?

Click to expand...

yup.... sorry typo ... thanks for catching it... I'll correct... but that 3 mOhms per cell includes all the 120+ interconnects for the whole pack as well... so the individual cells are even less than 3 mOhms.

powermed said:

The performance is quite comparable to A123 cells.

Click to expand...

Just to expand a bit more detail on this.,, and how the ~10 year old NiMH tech batteries compare to modern A123 cells.

( ~10 year old NiMH )
similar to Prius ------------------ Modern today available LiFePO
Gen1 Insight & Civic -------------A123 ( 26650 ) has :
Energy Density:
~139 Wh per L Volume ----------~209 Wh per L of volume
~43 Wh per kg weight -----------~108 Wh per kg of weight
Power Density:
~666 W per kg --------------------~3000 W per kg
~2,150 W per L -------------------~5800 W per L
General Electrical:
less than 3 mOhms per cell ------~8 mOhms per cell
Fails short circuit ------------------Fails open circuit
tollerant of over charge ---------- not tollerant of over charging
tollerant of over discharge ------- not tollerant of over discharge

The best part is that NiMH has also made improvements in the last ~10 years ... for instance modern NiMH cells have Low Self Discharge ( LSD ) and now self discharge slower than Li does... NiMH cells from ~5 years ago had as high as ~276 Wh per L and as high as 90 Wh per kg.

I've been lurking and watching this thread for a while now. Glad to see it is still active. I'm very interested in building a pack using prius cells for my stand-up scooter but need to find two or three others interested in doing a group buy on a whole prius battery. Anyone?

Do yourself a favor and buy used ... not new.

Then test the pack ... pull out the best subpacks / parts to use for your pack... if the rest are good enough use them for something else.

Used Prius battery packs can be had from salvage yards starting at ~$300 for the whole pack.

Testing is easy with any battery analyzer that counts Ah on a discharge cycle.
Charge one subpack up to full all by itself outside of the series string ... then do a full discharge cycle on it... record the Ah counted.... repeat for all of the other subpack... compare the Ah counted from a fully charged discharge ... do not skip the full charge first ... as you do not know the initial balance , or condition of the battery pack when you get it ... so charge it all up before doing your Ah counting discharge cycle.

Do not make the mistake of thinking the voltage of the cells by itself will directly tell you the cells SoC... That works well with LeadAcid and Li... Voltage alone is a very bad / very inaccurate indicator / method for NiMh cells SoC.

If the salvage yard had the battery pack unused for more than 1 month ... NiMH will benefit from a few as in 3 to 5 exercise cycles , before the full SoC to Discharge Ah counting test run.... it will work without the exercise cycles ... but, you might get a couple 100 mAh more out of them if they are exercised a few cycles first.

Of course used is used... and already has some wear on it... and there is a bit more work in testing the subpacks to figure out which are the best among them... but it is only a fraction of the cost of the new packs.

Yeah, no way I'm buying a new pack, definitely used. Thanks for the good advice on cell testing.

Offhand does anyone have the specs for the Prius batteries, like max discharge rate, and max recommended charge rate ?
I just pulled out a few that I have, and amazingly after sitting for a year, they still show 7.2 volts !
I am charging a set of 5 right now with a Futaba CDR5000 charger. I'm watching the temp. as well as voltage, and it still seems to be taking a
charge after the 8.3 volts / stick someone mentioned. My charger can be set to 5mv delta V , I wonder if that would see end of charge ?

Anyways, if these can pull 100 amps or so, maybe I can use some of them to absorb heavy current spikes that may damage
a more sensitive Lifepo pack.

So I finally got to testing my battery packs. They work great. Thanks to Powermed's info I was able to build 4 battery packs. The battery last for about 30mins or so mind you I wasn't really pedaling so it was mostly the battery doing the work. I am thinking of putting two packs together in parallel that way I get more distance out of them. My trip to work has a few big hills and that would drain the battery a bit quicker. Overall I am just really happy that I got them to work.

I have found the most reliable way to charge them is to charge at 4 amps, and cease charge when they *just* start to bulge from the outgas they make internally when they are done. Seems rather crude, but the prismatic cells with the flat plastic sides will bulge at the slightest pressure, unlike a metal cylindrical cell.
I have a 36 volt pack as a replacement for the SLA in a EV global bike, and the 5 prius sticks fit in the frame just perfectly! I will mount a micro switch to sense the bulge, to allow for unattended charging in the future.

lesdit said:

I have found the most reliable way to charge them is to charge at 4 amps, and cease charge when they *just* start to bulge from the outgas they make internally when they are done. Seems rather crude, but the prismatic cells with the flat plastic sides will bulge at the slightest pressure, unlike a metal cylindrical cell.
I have a 36 volt pack as a replacement for the SLA in a EV global bike, and the 5 prius sticks fit in the frame just perfectly! I will mount a micro switch to sense the bulge, to allow for unattended charging in the future.

Click to expand...

good stuff.

Just keep in mind ... the swelling happens after you have already passed 100% SoC ... so by the time you see the swelling pressure deflecting the sides you have already slightly overcharged a NiMH prismatic cell.

But... luckily NiMH is very tollerant of abuse like over charging ... so you should still be able to expect to get a reasonably good service life from them... and as you posted ... if you catch it early you minimize the amount of overcharging stress on the battery.

-------------
sorry was compelled to add a nit picky note too ... personal failing
it isn't out gassing any ... if it was you would be loosing electrolyte ... the swelling happens because the gas formed has no where to go... if you actually do get out gassing that means the cells have at crack or leak in the casing ... which would be a one way trip down hill as it will out gas the electrolyte ... and eventually not have enough left to keep working... I've read about a few people trying to rebuild a vented NiMH cell ... but it generally is far more hassle than it is worth... mostly because they aren't built to have the electrolyte refilled.

IamIan said:

sorry was compelled to add a nit picky note too ... personal failing
it isn't out gassing any ... if it was you would be loosing electrolyte ... the swelling happens because the gas formed has no where to go...

Click to expand...

Do these Prius cells have a safety pressure relief valve on each cell?

Yes, I should have used a different word. Maybe 'Ingassing'
The cool thing about these cells is the plastic is thin, and if you see a fraction of a mm of deflection
you can stop the charge.
There are vent ports, but it seems they have a rather high pressure release point.

i got my prius modules in today.... I started charging them with my smart charger.... As it was stated before these packs dont peak like normal nimh cells. It kept rising and rising. It did eventually get slower and slower. I tried charging at 1A 2A and 4A.

1 A was lack luster and the voltage seem like it plateaued around 8.5 . Never got warm, Never bloated and the voltage Never fell.

2A was better but it started getting warm once the voltage started to plateau around 8.6. It bloated and got warm little but but nothing to be overly concerned. Could be the resistance of the cell... or the cell was about to hit delta and i didnt let it.

At 4A the voltage rose quickly and it sorta plateaued.. the voltage got higher than the 1 and 2 A charge. It rose to 8.68v I terminated charge prematurely because of how high the voltage was. I wasnt willing to go over 1.45v per cell. It never reached delta. The voltage never fell... The cell never got warm. The cell never bloated. It makes me wonder do these cells have a higher charged voltage.

I think my charging tactics are going to be to discharge all the cells in parallel and charge them up using individual chargers while keeping an eye on the voltage and see how long it takes to get to 8.6 or 8.7 and then get a wall timer and set it for however many minutes it took to reach that voltage. There is no true safe way to charge these i think.

Right now im doing another 4A charge on another module. Just to see how high it will actually go. Its at 8.65v not getting warm not bloating not under compression.

At the 4A charge the voltage plateaued at 8.67... at that time i switched from 4A to 2A. Only at that time did the voltage drop and the charge terminate with in 2 minutes.

They don't have a higher or different overall voltage profile from normal NiMH ... but due to the unusually low internal resistance and such ... the same signature events that one sees in other NiMH cells are much harder to see in these HEV NiMH cells ...

keep in mind the high levels of current amps of discharge and charging they were built to take in the HEVs.

This all means that they will produce less heat ... The dV and such indicators will be much smaller than one would see from other NiMH cells even with more Ah of capacity when charged at the same Amp rate.

have fun playing with them ... I know I have with my HEV batteries ... but I'm just a battery ( NiMH ) hobbiest.

Im beginning to understand why they probaly have the 80%-20% SOC on these cells. They probably couldnt detect the DV even if they tried. Ive yet to do any discharge tests on them. Im waiting for a throttle to come in the mail. My clyte controller is having a good time blowing throttles. But probably going to base charging on voltage and time.

Played around with the cells a bit more. I have 9. I realized my voltage was every where on all my packs. I know Im not supposed to parallel charge them but i did. I made some homemade bus bars out of solid copper house electrical wires.and ran a strip along the + and the - wires terminals. I tightened the terminals down to where i had a 7.2v 58.5 AH pack. I put my charger on it and charged it at 4A... i wired my Watts Up and realized Im charging at 3.5A. I charged until my Watts Up meter read 8.3v. I took the bus wire bar off the pack. I check the voltage on all the packs. They all were sitting at 8.29-8.3v. The ohms on the pack i belive were 6.77 to 6.79mohms They are all level they a matched pretty close. I would say if you are charging to a certain voltage even if they are mismatched ... its pretty safe to charge them in paralle like SLAl. They self leveled and I experienced no ill effects no heating no bloating. I will check in 12 hours to see if they have drifted in voltage or internal resistance.

icecube57 said:

I will check in 12 hours to see if they have drifted in voltage or internal resistance.

Click to expand...

How do you measure the IR?

Parallel charging would open up so many possibilities for these cells!
Have you charged to the point that they just start to bulge ? It would be interesting to see if they
all bulge at the same rate also, that would be even stronger evidence that it would be ok to parallel
charge. You aren't afraid to charge to the bloat point, are you? :lol:

I think it's safe to do, but just a tiny amount of deflection. Not like the 1/4" I had once when I forgot about
the charger! ( Cells survived OK even then ).

does anyone know why they bulge? is there material clogging the separator and causing voids to form in the electrolyte or on the surface of the plates from gas evolution? isn't that why the clamping bars are used, to keep the voids from forming initially?

edit: someone made the comparison about the nimh and lifepo4 and said that the lifepo4 doesn't tolerate overcharging. i think that is inaccurate, they are not damaged any more than nimh by overcharging, imo. similarly for overdisharging, because when you reverse a nimh, or lifepo4, both are useless afterwards. jmho

During charge, the positive electrode reaches
full charge before the negative electrode which causes
the evolution of oxygen to begin:
2OH- _____> H2O + 1
2 O2 + 2e-
The oxygen gas diffuses through the separator
to the negative electrode, a process which is facilitated
by the “starved-electrolyte” design and the selection of
an appropriate separator system.
At the negative electrode, the oxygen reacts
with the metal hydride and oxidizes or discharges the
metal hydride to produce water:
2MH + 1
2 O2
_____> 2M + H2O
Thus, the negative electrode does not become fully
charged and pressure does not build up.
The charge current, however, must be controlled
at the end of charge and during overcharge to
limit the generation of oxygen to below the rate of
recombination. Thus, charge control is required to prevent
the build-up of gases and pressure. Duracell recommends
that continuous overcharge not exceed C/300
for optimal performance.

My packs didnt bulge until they sat above 8.4-8.5v for extended periods of time. At that time my charger was still stuck in CC mode and didnt switch into CV. I do hear the electrolyte bubbling during the end of the charge. I think that is a good indicator that the cells are almost done charging.

I checked them again 9 hours later. They voltage has dropped down to 8.09-8.12 the resistance has dropped to 6.47-6.53 mohms. It seems with their loss of charge the resistance goes down. So when charging the current goes to the cell with the lowest resistance until they all become equal and they they all rise together i think. Which further gives credit to the claim that it may be safe to parallel charge.