nimh cells

[1000kw-hunter]

I have 1000!! of nimh cells---->10kwh for free, if healthy!
And more to come.
What are the min and max v for them?
How to test them?

I think max v is 1.8v and min v is 1v???
Is there a bms needed for them? or i can just program a bms? (daly) for a 48V system?

And what is that soppy smell when i open a pack? i think one of the cells was leaking?
All the cells where testing 1.2v btw
I will hook them up and charge them to 1.8v???discharge them to 1.0v????
To find out there SOH
What are the parameters of nimh cells?
I aspect before end of this year a 5000 total!

Lot of questions, most important one...max and min voltage...charge to? and discharge to?
What is the minimum SOH to keep a nimh cell?

When i buy lifepo4 cells 10kwh i pay 2300 euro.
oke with a cycle life of 6000.

If i remember correctly nimh cells have a "memory" but you could erase that memory or prolong there cycle life...
But how to erase and how many cycle they can have?

More questions than answers.

Half of the batch.....

Yes 9ah and 11ah basterds at 1.2v


More questions than answers.

Thanks in advance for your help and info, with best regards Igor

 

[italianuser]

Oh my!
After a three year lithium addiction it's not easy to change, ehm not all. I bought a bunch of AA and AAA NiMH batteries last year which arrived fully charged at 1.65V (not the most expensive cells, must say). After 6 months they were flat and at the first recharge I had to throw them all in the bin because they went to 0V. So I was quite annoyed and went to search the charge curve to try and charge them myself, and it's a bit of a mess.

They don't have a fixed charging voltage and detecting a full charge is based on a delta-V that happens at a certain point (and which also is temperature dependant). "Delta-V" is a voltage dent in the curve.

"The coulometric charging efficiency of nickel metal hydride batteries is typically 66%, meaning that you must put 150 amp hours into the battery for every 100 amp hours you get out. The faster you charge the worse this gets" - https://www.powerstream.com/NiMH.htm

But BU says this: "It is difficult, if not impossible, to slow charge a NiMH battery. At a C rate of 0.1C to 0.3C, the voltage and temperature profiles do not exhibit defined characteristics to trigger full-charge detection, and the charger must depend on a timer" - https://batteryuniversity.com/article/bu-408-charging-nickel-metal-hydride

What I understood is that my maybe-5-year-old charger maybe wasn't good any more, or maybe the cells weren't good, I really don't know. Too much uncertainty for me. I think you need a good quality charger as a first thing.

On my side I'm going to buy lithium AA and AAA from now on... sorry! I'll be happy if someone can give you more positive info than me!

 

[1000kw-hunter]

It is difficult, if not impossible, to slow charge a NiMH battery. At a C rate of 0.1C to 0.3C, the voltage and temperature profiles do not exhibit defined characteristics to trigger full-charge detection, and the charger must depend on a timer

Yes, thank you, from my memory:
If they are full the charging must stop.
They are getting lazy when slow charged and slow discharged, once in a while the must be waking up and not with a cup of coffee but with a bucket of ice cold water(not litterly!)

Also very important:
They are all self discharges, every month you must charge them to avoid 0v cells aka dead ones.
Keep a minimum of 1v.

They must be charged at 1.6v per cel, so after the delta (thus after the full charge) they can settle down to 1.4v.
Discharge to max 0.9v per cel.

To use them in a ess system is very difficult yes, but hey i am getting for free a 1000! cells.
If i can succeed i can have approx 5000 to 8000 cells a year for free.
I really think this is worth to do some effort.

They must be charged at cc and not cv.
This might take a home built bms?

Back then i never experimented with higher then 30v per pack for my model vehicles
Maybe it is a clever idea to not exceed 24v packs?
It will also be easier to maintain those packs when they need to wake up?
I also think they wont need a balancer, maybe a v cut off per cell at 1.45v?

Yes this is a different ball park and they are very good for ballast under the solar panels.(win win)
This weekend i will do some research and report back.

Thoughts and comments are appreciated, on my thinking out loud, facts and my memory.

With best regards Igor

 

[1000kw-hunter]

On my side I'm going to buy lithium AA and AAA from now on

why not lifepo4?

I'll be happy if someone can give you more positive info than me!

I am very happy with the info provided, it was refreshing for the grey mass up there.

 

[italianuser]

why not lifepo4?

I meant I'm going to buy lithium AA and AAA instead of rechargeable NiMH AA and AAA. I only use them for remote controls and my infrared PIR boxes and some other stuff. They are mainly commercial devices and just have enough space for AA or AAA batteries otherwise I would have used those pouch Li-Ion batteries got from cellular phones, I have a box full of them.

But couldn't you sell all of these NiMH you get and buy lithium ones? Unless you have a good plan to use NiMH, I see you're very enthusiastic about them! I'm too addicted now to change... LOL

 

[1000kw-hunter]

NIMH continu:
The min and max v without delta V
0.9v to 1.35v
Max dis/charge: 0.2c
6 to 8M ohm IR when full charge aka 1.35 after delta cooling down.

pro's:
They are self balancing up to max 24v,BUT:
They need a heat sensor that will cut off above 40C, if not applied it will self damage beyond repair.
Memory effect is minimum, BUT
They will get lazy...once in 6 months they need the wake up call.
Estimated cycle life >500 up to 1000.
Estimated self life 5 to 10 years(reports found of 15 years and going strong)
Even after venting they still work.
After discharge to 0v they can be revived, several times(going to test this several times.)
Very safe chemistry, they will boil of some moister when to hot or over charged.
Will not burn or explode in normal (ab)use nor with short circuit.
Will test to set the safe parameters.

Lifespan can be prolonged if having the good equipment, like h2 and o2 gas injection.
But this is a paper a read and i did not go further into this idea, sounded to expensive

Con's:
Very hard to detect a bad cell once in a powerwall, almost impossible.
Keep an eye on the total v per cell, if one stays behind with charging, time to check that pack.
Hard to solder on, need a 250w 240v solder iron.
24v max system voltage.
Very hard to get those cells out of there housings...


My idea for a powerwall, 24v x 2 to make 48v
Connected to get min 2.7(2.8v) to max 4.3(4.2v) and get myself a li ion bms x 2 for 24v

All is still in experimental state!

Lets continu...

All those cells where solo charged to 1.35v with a cc/cv ali charger, checked for ir 6-8m ohm.

The final packs will get a different format: 3 times 16 cells in a 3p16s pack to create a single cell of 144Ah on 3.6v nominal.

I hope this was appreciated, the data and info giving, i will continu with building, thanks

With best regards Igor

 

[DG98]

Thanks for continuing to experiment and document your results. I find it interesting. NiMH might not be the usual chemistry of choice around here, but it seems that cells are plentiful and there might be something to be learned. Please keep it up.

Cheers, John

 

[1000kw-hunter]

Found the paper:

Increasing NiMH battery cycle life with oxygen

Nickel-Metal Hydride batteries (NiMH) have long cycle life. But in the end corrosion of the metal hydride is detrimental for life expectancy. Corrosio…

www.sciencedirect.com

 

[1000kw-hunter]

What a work to get them out, its more work than laptop or li ion bicycle batts.

Anyway I have not verified the ir but i have some suspicion about the good the bad and the ugly cells.
All the packs had at least one or two 0v cells, those are still in the mix.
As you can see the color differences on the top of the cells.
Here are some close ups with explanations of my thoughts of what happend to them:

As you can see the color of the middle white insulation ring, from left to right:
The white ones are not heated up, the middle ones had some temp raise the right ones where clearly to hot, but did not vented!
V was good on all of them, did not checked the IR.
This chemistry will get to hot with a over charge and kill itself.
If one cell fail the charge will go into the other ones, it is to much so it will heat up and start to boil.
This is why heat resistors are so important in those packs.
They tend to fail, so those packs are going to be mine.

This threesome is to make clear what venting is on those cells.
Note the color on the right bottem cells white insulation ring..
That one had a instant over charge and no time to heat up.
The left one had time to heat up with a over charge(before or after i cant tell)
Cell at the top is for comparishment.

Again the black sticky stuff is comming out, last one heated up to much(before or after)

One at the top is suprisingly 7m ohm(6 to 8 m ohm max for IR) and had 1.28v.
His story? a wet housing, probably humid shed when charged.

Some bright light thought it was a safe idea to put soapy water into the housings, probally to make them fire proof or something.....just like throwing li ion bicycle batts into a bucket of water to prevent fire....


With best regards Igor

 

[1000kw-hunter]

Very sorry, i was to busy to give you a propper update...
The most important update is that i changed the format of my cell's.

With this "model" i can mimic li ion.
0.9v to max 1.35v per nimh cell, times 3 would be 2.7v min and 4.05v max.
37.8v total minimum, and 56.7v as max for the system.
When charged up to 4.2v they give around 80Ah when charged to 4.05v they give around 75Ah, but in lazy state!!!
The wires should handle easy 15A.

Remember:
Two times per year charge to 1.5v per cell, to get them out of there lazy state.: with some single cells it worked, but the whole setup?

To incorporate this in my lifepo4 system i have a small problem.
All the settings are max 56.00v incl in the chargers.
3x1.5=4.5v x 14 = 63v!!! to wake them up
But i think daly will cut the charging off for the lifepo4, so i will let one solar charger 2 times a year go to 63v, to top them up

Oke this was the theory for the next experiment.
Lets get back to today.......

I did some research and i found no paper video or blog of someone making a powerwal @48v in nimh.....only 24v.
and someone who toke apart his car battery (i thought a prius?) for energy storage also nimh.

I think i am the first one to do a 48v system with those d cells......cross your fingers

Panik cut off in case of problems.

Balancer and bms are connected.
Lets see what is going to happen.
Tomorow i will connect some leads from the system to this setup to charge.
No worries, all the cables are rated for 60v 5A max for testing.
There will be no more A for charging and discharging, the bms is also set.

With best regards Igor

 

[1000kw-hunter]

small update, i connected the new guy to the main system, while mine wife was cooking and the boiler for the shower was warming up....
lifepo4 was on 100% before all of this, now the nimh is giving power to the whole system.(150w, but oke)
Tomorow around 10.00 the lifepo4 will be full as usual...i am really curios about the nimh...what they are going to do.
All the bmses are balancing all the packs together to the same voltage...now how nice is that?
Tomorow i want to see a charge in to the new guy the nimh pack....to full 100%
I will keep you posted

 

[1000kw-hunter]

Regarding size.......
280ah 48v lifepo4 is 1meter20 and 20cm high....~14kwh
80ah 48v nimh is 1meter20 and 40cm high....~4kwh
So with lifepo4 you can pack 3.5 times more in a smaller space

Li ion is about 50% of the size regarding to nimh for the same 4kwh
Even smaller, stack 30 li ion cells in 1.5 4x5 cell holder and flip it on the long side.
So its big, huge and heavy
Charge and discharghe is max 0.2C
80ah * 0.2 = 16ah max for nimh
280ah lifepo4 you can discharge charge at 1C 280ah

I set my float charge from the victron higher for the v needed for the nimh.
As suspected when the others are "full" at a lower system v the bmses will cut them off

So far so good, happy charging at 2.5 to 5A, the bms is set to 4.05v cell cut off and 56.7v sum high volt cutoff.
But you really need room for them.

For a 290ah nimh cell it would be 7cm thick 67cm tall and 24cm wide 120cm long, this is incl the wood for the frame.
But without sheets (5 or 6mm) to give them more stiffness and dont allow them to fall out..(covers)
You will need 100! of those bicycle batteries, with some luck you can get 15 healthy cells out per battery.
Matrix would be 6x6 cells (36 cells @ 1.2v) times 3 to get your 4.05v.
Lets see if i can get 100 of them.....just for fun

First lets see what the watt meter is going to tell me after they are full charged.

With best regards Igor

Ps i also write down data for myself as a memo, but also for people that are really going to copying this experiment.
I hope that it will be useful for a other DIY er, thats why so much ( not ) needed and (not) important info.

 

[1000kw-hunter]

Small update.
The pack that was connected to the system, i disconnected after it was 100% at 56v
After 10/11 weeks itself discharched to 51v, this is common for NIMH.
also i left the balancer and the bms connected.
Now a Daly bms will shut itself off if there is no activity and the standby mode is less than 10mah.
Same for the Heltec balancer?


Assembled cell testing. (3 times a nimh cell = total 2.7v to 4.05v)
The ones i tested all have the same thing in common.
After 4.05v the A charging drops, but when over 4.10/4.15 the amps are going up, till 4.25.
So cut off will stay at 4.05v @ 1.35 per cell.

Discharge testing is in progress now, will keep you updated.
What i would like to know is do they discharge as one? or at the end will be one section (out of 3) lower or higher in v?

 

[1000kw-hunter]

follow up on the update and some new.
I have accumulated a lot more of them, to much to keep playing around.
So lets get serious with them.

13-8
i connected the NiMh battery again to the powerwall, they are really playing very nice.
I have no decent numbers on this pack, but it plays a fair game.
This battery is ! only ! tested on IR, nothing more nothing less and zero v cells duh
All the cells are used are between 4 and 8 ohm.

15-8
A new battery is in the making, but this time i take my time for it.
Like the previous cells i tested on IR and "paired" the cells into packs.
All the packs that are assembled (or 4 ohm or 5 or 6 ect per pack), i charged them full, just below there delta, 1.35v

7-10
All the packs were measured on v again with the yr meter, all went down between 1.295v (lowest) and 1.310v (highest) from 1.350v

Atm i am building a decent tester for NiMh cells, a bit tricky if i may add, max charge and max discharge is 1a an hour.
More will damage them.
I will also keep on track what the cells are doing after "salvation/liberation" out of the pack: some leaked, some blow out there gaskets some where to hot.
It is showing on the top of the cell, look at the photos in the beginning of this thread.
So the initial segregating of the cells before charging them up, will be a nice process.

Goals for now to tweak the process:

1 to find out what the cells are worth after the blow their gasket or leaking.
2 find out if the cells that where to hot in their e-bike pack, if they are still good(minimum 50% soh)
3 finding out if the delta charging curve is really needed to get some more mAh out of them(that answer is no btw)
4 when is it needed to wake them up? after some period of time? .....after some cycles?
5 is a bit of a side track: how far can i go on the parameters of those cells.
6 what is going to be there added value? especially regarding there size..
(7 at last; optimum IR, SOH, charge discharge and min/max v)

Regarding goal 3, after delta they will use extra power to balance themself and dissipate the extra power into heat. and that's is why most (95% i dare to say) e-bike packs fail.
Also extra power is used/needed to keep them "awake".

Fun fact: the youngest cells i have are "just" 10 years of age....

Enough rambling for now.

With best regards Igor

 

[1000kw-hunter]

Oke lets get back to serios testing of the (obsolete) NiMH chemistry.

That one pack i have in service could be a lucky shot.
All the cells where at 4 to max 6 Mohm, all cells were 1.1v and above, when retrieved and put into service/use within 2 months.
The cells i am playing with now are almost 7 months sitting still and were doing nothing! and with IR not tested yet.
I wanted to step in the unknown. IR will be tested.

Now for the cells that are 0v and or low capacity.
There are apparently two methods:
1: give them for several seconds a short burst to get those crystals to dissolve again/ waking them up.
Start a charge/discharge cycle.
This one is known be me that it will work, but the aftermath i forgot in time.
1a: The short burst can be done with some same cells but then full. Or a equal power supply.
1b: "tack" a single cell with 3 times their v----> 4 to 6 volt.( no more than 5 - 10 seconds)


2: start a trickle charge right away, aka 50 to 100mAh per cell for a day or so and stay under the delta v! aka 1.3
If they are all happy at 1.300 volt discharge them in two steps.
Step 1 discharge to 0.900V with no more than 200mAh draw.
Step 2 discharge further to 0.400v with no more than 100mAh draw.
Remember that i am playing with 10A cells originally, so with 1A cells divide all the draws accordingly!!!
Now go back to the trickle charge, double up the charge, if you see that the cell is getting hungry, its good to go.
You will probably also see some cells at the first stage that they dont want to have so much, maybe not even 10mAh.
You have to learn them again that they can do more.
The memory effect in NiMH is what they did last time, that is why you need them to wake up.
How i am going to do this in my powerwall? i think! deplete the whole battery in the summer in those two stages (0.9v and 0.4v) set the charge v in my bms to 1.450v per cell( i use a li ion bms so that would be up to 4.35V).

After the second trickle charge, charge them up to 1.35 or even 1.4v to trigger the delta, but with this i will report back!


Hopefully the crystals will take a step back as well. Once they formed there is no way back?
But i am not to that point jet. They will not last forever.
First let me see if i can get decent capacity out of them, then we will see about age.

I will try both methods, currently i am doing number two, and this will take a couple of days.
(0v up to 1.25v with low capacity)
There is also a third methode. will try that one as well-----> freeze them overnight.

Its really funny how different chemistries are look a like, just like those other old chemistries like zinc bromine, you have to deplete them completely after 4 cycles to get full capacity back.

Regarding that some call NiMH obsolete, i dont think they are, they have their limits in what you can do with them.
Don't forget their maintenance.
No wonder li ion toke over so quickly.

I really hope this time-consuming testing will lead to some decent results, for all i know i am back in the roman ages, lemon/patato and some zinc plus copper plate.

I hope testing will be done end of this week and i can give some nice results.

With best regards Igor

 

[1000kw-hunter]

Thanks Korishan, in my country the proper saying would be:
At your service!

Thanks

 

[1000kw-hunter]

after some testing two things are pretty obnoxious....
the number/percent of self-discards at a day and the level (lost power) a day, incl keep up the charge when almost full
I believe the level of charging is optimum for storage is max at 1.25v per cell and not as stated earlier at 1.35(delta level.
This means also 10 to 15% less storage, a healthy 10ah cell can do max 9ah of storage....in a setup like we have.

But you have to wake them up every 6 months.
Next couple of test will reveal this on SOC what is needed for them to get the max out.

In the mean time i have a total of 3000 cells waiting, 27x14 are connected, they play nice but they discharge.
I think i have to lower the overall max pack voltage.
I disconnected that pack from my ess btw to see what will happen and adjust the max overall pack andcell voltage.
Minimum is set at 40v 0.9v per cell 2.7v per pack, works perfect.
Math tells us 37.8v, i set a min pack 40v 2.85v per cell
A pack with NiMh is 9p3s makes 2.7v minimal a cell is 0.9v minimal.
max 9p3s is 1.35 per cel, is 4.05 max per pack is 56.7v max per battery
And in here is i think my mistake.
My max should be with 14p 56, and that is beyond what NiMh can do on cell level.
So 15s or 16s i will try, lets see if it will fit in the lifepo range instead the li ion range...
Or a different bms....

I will keep you updated.

 

[1000kw-hunter]

After a lot of experimenting i found two things, that are coming back, again and again.
The overal voltage (nominal 48v) is to high for this nimh. NiMh in cars can go up to 360V dc, not with bicycle packs.
The first two (packs) in the battery will not sustain/survive this high voltage, i had them replaced and broke down again.

For optimum testing the Nimh cells, I charge them up to 1.410 volt, I let them sit for a week so they will settle on 1.35ish volt.
Leave them longer in their idle state is not worth the effort.
Only a bad cell will drop more after this time of sitting.
Bad cells also cannot hold so much energy.
I am separating the cells after testing in groups of 1a each.
To make a separation with IR at forehand is a waste of time, IR is al over the place.
I think this is due to the v level? I will test this.

I will go back to a 24v system for my NiMh cells and add some sort of dc dc step up to obtain 48v nominal, lets see if that will do the trick i want them to do.
Something like a 48v charger? or a dc dc step up? If you have any idea, please share.

What i also noticed is that if some cells test "bad" you must test them again, meaning charge all the way up and discharge all the way down.
The SOH will rise big time.
Some 40% SOH will bounch up to 90-95% SOH.

After each cell is tested i put them away with 0.85 up to 0.920 v in them, after three months you must retest them on v again, the ones that dropped more in v discard them, they are useless.
Mostly they are cells that tested under 60%soh.

Charging them up under the delta, meaning charging them up to 1.35v will not get them full.
After the 1.300v the load that they will take is less then 100mah per cell and some not even 50mah.
But they really need that 1.410v, go over this will result in energie loss and will not result in extra capacity.
Charging lower then 1.390v will result in less storage.

All was tested and measured with a yr1035 meter, to get propper and constant accurate readings.

With best regards Igor

Pics will be added of the tester.

 

[DG98]

Why do you think the cells are failing when you put more in series, to get to 48v? That's certainly inconvenient since so much is built around a 48-52v nominal standard. If you cannot get to 48v in series, then I suspect you would realize much greater efficiencies by buying/using 24v inverters, as opposed to suffering the losses of a 24-48v DC step-up -- but I'm no electrical engineer, just my guess! I also realize, of course, that probably doesn't fit your global plan.

Cheers, John

 

[Boron]

hi Igor @100kwh-hunter Your work is epic and contributes enormously to our community knowledge. Its also on a par with the legendary Pete from https://www.youtube.com/@HBPowerwall/videos in OZ. He spent 7 solid years along a similar path trying to recycle old free Lion cells from Laptops. You must surely know this work.
He had to spend a huge amount of time testing thousands of cells individually for SOH before he would commit them to a power wall.

It seems that NiMH cells just wont play nicely like this and stay within acceptable limits to use in a power wall. Once inside the structure its impossible to detect cells that are drifting out of limits (and must be replaced).

I think you concluded elsewhere that NiMH is just not suitable in strings for power walls at any price and the stored energy density, wiring complexity (thousand+ joints) and consequent weight are also a negative factors

Yet these cells can perform for years in small cordless power tools for example. They get used intermittently and recharged very crudely (remove battery pack from charger when warm is all the warning of charge complete you get)

I have a hedge trimmer still going strong after 6 years and a screwdriver after 10 years (its a NiCad)

BY CONTRAST - Lion WARNING HERE TO ALL I recently had a spontaneous Lion battery fire on a cordless tool bought new from a reputable name. It was only by sheer luck I didnt burn my house down. The tool was dormant not on charge in my desk drawer. See pix