Should I even Bother?- Question about used batteries and resistances.

[CONSIDERABLE SHOUTING]

I scrap a lot of stuff, and in my possession are tons of laptop lithium cells- mostly 2400-2500mah, rated 1 Amp of discharge cells that have consistent capacities made entirely of 2 brands. Charge to 4.2v no problem on basic opus chargers for info, have similar curves and rates of discharge, but because they're all resales their resistances are all over the map. Lowest are maybe 60miliohms, highest are up to 250. I easily have 100-120 of them, and that resistance has only increased since I salvaged them.

At the same time, I'm trying to start in this hobby properly and I wanted something to make to get started- and I have lots of 20-24v tools, so I thought about building some kind of backpack or carrier to use all these cells in one big, 24v nominal battery that could push 20-21 amps at peak by just having tons in parallel. But with that high of an IR, is there ANY reason to do so at all? From what I'm reading (and still learning), this idea would rapidly die on me because the high resistance cells would rapidly burn out. Or would it be a good idea from the standpoint of learning- and since I have everything sans BMS- to just make it, since 24v is relatively safe?

 

[harrisonpatm]

I scrap a lot of stuff, and in my possession are tons of laptop lithium cells- mostly 2400-2500mah, rated 1 Amp of discharge cells that have consistent capacities made entirely of 2 brands. Charge to 4.2v no problem on basic opus chargers for info, have similar curves and rates of discharge, but because they're all resales their resistances are all over the map. Lowest are maybe 60miliohms, highest are up to 250. I easily have 100-120 of them, and that resistance has only increased since I salvaged them.

At the same time, I'm trying to start in this hobby properly and I wanted something to make to get started- and I have lots of 20-24v tools, so I thought about building some kind of backpack or carrier to use all these cells in one big, 24v nominal battery that could push 20-21 amps at peak by just having tons in parallel. But with that high of an IR, is there ANY reason to do so at all? From what I'm reading (and still learning), this idea would rapidly die on me because the high resistance cells would rapidly burn out. Or would it be a good idea from the standpoint of learning- and since I have everything sans BMS- to just make it, since 24v is relatively safe?

Properly tested, sorted, grouped and used, recycled laptop cells can be used in a stationary powerwall application, such as mine. If resistances are all over the map, that's okay as long as you spec your build to the weakest cell. But as you point out yourself, most of them are rated at 1A discharge, no more. This is not ideal for even a weak e-bike battery. Your backpack 24v battery seems maybe doable, but is it going to be worth it to make a 40p6s battery?

As for you question as to what IR is acceptable, you would have to compare it against the cell make and model spec sheet, to determine how much the IR has risen from age. 250 is way to high for any cell. Some low-quality, but still acceptable, laptop cells could have been originally rated for 60milliohms, and may now be 80. That could be considered acceptable. I personally toss any cells over 70milliohms.

You didn't mention in your first post whether you're testing capacity as well as voltage and IR. If not, you should. A cell rated at 1800mah, now testing at 1600mah, still has 88% of it's lifespan yet. A cell rated at 2600mah, now rated at 2000mah, now only has 76% of its life yet, even though 2000mah is way more than 1600mah. Something else to consider for your project.

You have the right idea that to get any decent use out of them, you'll have to put a ton of them in parallel, which is why stationary storage is the most common route to take.

Hopefully these thoughts will get you started. It's not that your project isn't possible, just you need to take all the factors into consideration. And certainly be careful, it's always a risk dealing with used cells.

 

[harrisonpatm]

Or would it be a good idea from the standpoint of learning- and since I have everything sans BMS

Learning is always a good idea, working with used cells without using a BMS is rarely a good idea.

 

[CONSIDERABLE SHOUTING]

As for you question as to what IR is acceptable, you would have to compare it against the cell make and model spec sheet, to determine how much the IR has risen from age. 250 is way to high for any cell. Some low-quality, but still acceptable, laptop cells could have been originally rated for 60milliohms, and may now be 80. That could be considered acceptable. I personally toss any cells over 70milliohms.

You didn't mention in your first post whether you're testing capacity as well as voltage and IR. If not, you should. A cell rated at 1800mah, now testing at 1600mah, still has 88% of it's lifespan yet. A cell rated at 2600mah, now rated at 2000mah, now only has 76% of its life yet, even though 2000mah is way more than 1600mah. Something else to consider for your project.

You have the right idea that to get any decent use out of them, you'll have to put a ton of them in parallel, which is why stationary storage is the most common route to take.

Hopefully these thoughts will get you started. It's not that your project isn't possible, just you need to take all the factors into consideration. And certainly be careful, it's always a risk dealing with used cells.

Thanks Harrison.

They all actually test at 95-100% of their rated lifespan; I think out of all of them, their variance between all ~120 cells is 100 milliamps or less. IR is off the Opus charger, not a dedicated rig, but average over 3 tests was done. I'd have to search through them all but I believe the highest was 250 milliohms.

I was thinking of using it for the great lot of 20-24v hand tools I have, as well as potentially some 24v brushed bike motors I've picked up for free in the past- but for that little amperage out of the "big pack" idea I'm not sure if it could even power them due to reactive power. I'd probably use a BMS from Battery hookup (since they're extremely cheap and low amperage, but have a great track record) and I've thought about attaching some kind of solar panel and MPPT to it as well, just as a learning exercise too. Does that change your input at all about it?

 

[harrisonpatm]

IR is off the Opus charger, not a dedicated rig

The Opus IR readings can be considered as being random number generation lightly inspired by internal resistance. The fact that you get 3 different readings and take an average should be the first clue that the readings mean nothing. Proper IR testing should give you 3 repeated identical readings.

They all actually test at 95-100% of their rated lifespan

If that's true, that's great, but it's questionable to see capacity readings within 95-100% of the spec sheet, while also seeing IR measurements so damn high. Like I said, you shouldn't trust Opus IR readings, but even so, 60 still seems rather high for a nearly-new cell. Can you share what cells you are getting? For my own curiosity.

I've thought about attaching some kind of solar panel and MPPT to it as well, just as a learning exercise too.

I think that's a great idea, start with some cheapo PWM controller and secondhand solar panel, and it'll give you practical real-world insight into what kind of power can reasonably expected from the sun.

 

[electric_nz]

In my experience a decent laptop cell can handle 1-2C discharge rate with some voltage sag. Many of the early Panasonics etc even had a 3C curve on their spec sheets, however they get warm at that rate.

They are far better than the NimH cells anyway!

 

[CONSIDERABLE SHOUTING]

The Opus IR readings can be considered as being random number generation lightly inspired by internal resistance. The fact that you get 3 different readings and take an average should be the first clue that the readings mean nothing. Proper IR testing should give you 3 repeated identical readings.


If that's true, that's great, but it's questionable to see capacity readings within 95-100% of the spec sheet, while also seeing IR measurements so damn high. Like I said, you shouldn't trust Opus IR readings, but even so, 60 still seems rather high for a nearly-new cell. Can you share what cells you are getting? For my own curiosity.

I can post them when I get home tonight!

How do you test IR then? I know there's that little handheld tester but i'd rather not drop $90 on some new tool I might use a few times.

 

[harrisonpatm]

I can post them when I get home tonight!

How do you test IR then? I know there's that little handheld tester but i'd rather not drop $90 on some new tool I might use a few times.

YR1035, less than $50 on aliexpress. RC3563 is another for a similar price, good results. If you plan on recycling any more cells, it's a good tool to have on the spot. For example, I got lucky awhile ago and saw someone on facebook selling used 40ah prismatic cells for a dirt cheap price. A proper capacity test would have taken hours, but for a spot check, an IR test gives you an immediate idea of the general health of the cell. Ie, it's supposed to be rated at 10milliohms and reads 50? That's a pass for me.

 

[CONSIDERABLE SHOUTING]

Of course I look up the RC3563 and I find this great thread... coulda used it a few years ago.

 

[CONSIDERABLE SHOUTING]

Can you share what cells you are getting? For my own curiosity.

Here's the Spec sheets on the cells Harrison.
Half of the batteries, tested with "lower IR" by the Opus, are these A&S power cells. Mine lack the integral BMS.
The other half are these Samsung ICR -28As. The Opus claimed their IR was "worse".

Per these sheets, it seems like both cells can actually discharge more than 1C but may not have workable charging characteristics with the A&S at 4.23 and the Samsungs at 4.3. I didn't plan on charging anything like these quickly anyways, but that's a hmmmmmm from me.

 

[Dui ni shuo de dui]

I scrap a lot of stuff, and in my possession are tons of laptop lithium cells- mostly 2400-2500mah, rated 1 Amp of discharge cells that have consistent capacities made entirely of 2 brands. Charge to 4.2v no problem on basic opus chargers for info, have similar curves and rates of discharge, but because they're all resales their resistances are all over the map. Lowest are maybe 60miliohms, highest are up to 250. I easily have 100-120 of them, and that resistance has only increased since I salvaged them.

At the same time, I'm trying to start in this hobby properly and I wanted something to make to get started- and I have lots of 20-24v tools, so I thought about building some kind of backpack or carrier to use all these cells in one big, 24v nominal battery that could push 20-21 amps at peak by just having tons in parallel. But with that high of an IR, is there ANY reason to do so at all? From what I'm reading (and still learning), this idea would rapidly die on me because the high resistance cells would rapidly burn out. Or would it be a good idea from the standpoint of learning- and since I have everything sans BMS- to just make it, since 24v is relatively safe?

If the goal is to learn and gain experience, then yes, you should do it.
Not without a BMS though (or at least some very thorough manual monitoring). And not without a good balancing function, because you are likely to have to balance the pack a whole lot if you have a wide range of different internal resistances.

Just don't expect a long lifespan, it's likely this pack will die soon so try to make it easy to work on later for when you'll have to switch a few cells. If the goal is to learn then don't go too crazy on this project, build a simple test pack and then, later, use this experience to make a better one, with fresh cells and more advanced functions.

 

[CONSIDERABLE SHOUTING]

If the goal is to learn and gain experience, then yes, you should do it.

So I don't need to worry about the difference in charging voltages? For the record, I don't ever intend to charge the battery at a rate higher than an amp or two for the entire pack.

 

[harrisonpatm]

So I don't need to worry about the difference in charging voltages? For the record, I don't ever intend to charge the battery at a rate higher than an amp or two for the entire pack.

What difference in charging voltages? Can you explain what you mean by that?

 

[CONSIDERABLE SHOUTING]

What difference in charging voltages? Can you explain what you mean by that?

I my response from thursday with the data sheets, it claims the cells have different charging voltages from manufacturer. I've never seen anyone in guides detail or discuss that as a potential issue, so I'm wondering if I should worry or not. It says the Samsungs are charged at 4.3 but the A&S power at 4.23, and their site is updated; I seem to remember on their original data sheet from a few months ago their cells were only rated for 1C of discharge.

Also I intend to buy that resistance tester soon, once I find a cheap one on eBay. If they test fine, then I'll begin planning the pack in detail.

 

[DogDipstick]

No.

Not when electric car cells are all over junkyard for 1/10 the price.

 

[CONSIDERABLE SHOUTING]

No.

Not when electric car cells are all over junkyard for 1/10 the price.

I would, except:
1. I've got em right here and I'm using them to learn on not-lethal voltages.
2. Those wrecker cells are purchased by a different company who takes the batteries from the yard. No JY around me has cheap, used automotive EV cells and they've refused me in the past (but I fully admit the last time I tried was ~2 years ago). Still tho, few EVs in my region.

 

[harrisonpatm]

It says the Samsungs are charged at 4.3 but the A&S power at 4.23,

I'd just ignore that. There's not really any usable capacity over 4.15v anyway. Yeah, some cells can "safely" go a few millivolts over, but you'll likely want to just be charging to 4.0 or 4.1v max anyway, for safety and longer lifespan.

 

[CONSIDERABLE SHOUTING]

I'd just ignore that. There's not really any usable capacity over 4.15v anyway. Yeah, some cells can "safely" go a few millivolts over, but you'll likely want to just be charging to 4.0 or 4.1v max anyway, for safety and longer lifespan.

Okay, excellent! I'll check to make sure with Batteryhookup to see what voltage their BMSs equalize at still just in case. Being able to do a 6S 30A BMS will be great and should be able to drive most tools and even some of the Brushed DC ebike motors I have.

Plan so far then is- Get that IR tester, check cells IR, recharge them all to test for loss over a week, then build a full wiring diagram for a plan. Should be a good use of these.

 

[harrisonpatm]

Okay, excellent! I'll check to make sure with Batteryhookup to see what voltage their BMSs equalize at still just in case. Being able to do a 6S 30A BMS will be great and should be able to drive most tools and even some of the Brushed DC ebike motors I have.

Plan so far then is- Get that IR tester, check cells IR, recharge them all to test for loss over a week, then build a full wiring diagram for a plan. Should be a good use of these.

sounds like a good plan to start

 

[Stealth_Chopper]

ignore the root cause of the resistance period,

That is the dendrites from the welding and soldering
garbage pile and then MORON { likey {
OK ITWIS

 

[CONSIDERABLE SHOUTING]

I'd just ignore that. There's not really any usable capacity over 4.15v anyway. Yeah, some cells can "safely" go a few millivolts over, but you'll likely want to just be charging to 4.0 or 4.1v max anyway, for safety and longer lifespan.

Oh I realized you might have mistaken the sheets- it claims the charging voltage is supposed to be 4.3 and 4.23 volts per cell. I can't imagine that'll affect too much however, since it's literally tenths of a volt.

 

[CONSIDERABLE SHOUTING]

So here's my current MS Paint plan. Please critique the ever-loving shit out of it, because I haven't worked with electronics at all and this is largely being made out of junk I've gotten secondhand.

Battery will (assuming all 147 cells are usable) be a 6S 24P, which I know can fit into the case I have for it. The BMS will likely be this 30 Amp model from Lithium Battery PCB that other users have had good experiences with in the past, that balances above 4 volts a cell which will be good for longevity (the battery balance leads were not drawn because the BMS could change). The BMS will be connected to two "Rails" from which I plan to attach other modules to the pack to make it modular; I have two other ideas I could add to it, but I'm trying to not have project creep with things like an additional over the shoulder light or phone charger, both of which I could still add (and frankly, would kinda like just because it would look cool and be useful).

The CC/CV solar charge converter will be provided from the AITRIP DC-DC converter to mess with solar and trickle-charge the battery as it's used, and the solar cell I'll have to pick through my stash to find one with an open-circuit voltage less than 30v. Since the converter can only do a max of 4 amps, I think I'll pop a 4A Schottky diode on the negative leg before the panel for protection and have it angled off the back of the pack to act as a sunshade.

The voltimeter is as it sounds- one of the smaller jobbers that shows both amps and volts that I don't have a link for, but should be able to withstand it. The fan and controller is a single probe with an automated function, and I have a 24v DC brushless server fan I can pop into the case- my idea is, that fan will draw in air likely from "below" and exhaust "up", and I'll try to make some method to keep blowing cooler air past the cells when temps reach a certain point as basic temperature control. I'll update later with a link to the controller.

From there, the XT60- single plug out, it'll connect to my tools with a live cable that for safety, will be capable of being easily and quickly disconnected. Since I'm right-handed it'll exit on the right or bottom side. Because of the voltage I'll have to watch the length of the cable to not increase resistance too much.

Finally, fuses and switches- In it's current guise, I think the DC-DC converter can be switched off but I am not sure, and as-is the solar panel can continue to keep the pack "on" as it is. I think the BMS I linked as a on/off as apart of the unit, but if it doesn't I may want to move the on/off switch to between the battery and BMS to cut off power directly, but frankly I'm not sure. My hope is that this project could work with a 24v inverter as needed. Thoughts, opinions?

 

[amberwolf]

Oh I realized you might have mistaken the sheets- it claims the charging voltage is supposed to be 4.3 and 4.23 volts per cell. I can't imagine that'll affect too much however, since it's literally tenths of a volt.

Tenths of a volt can make the difference between overcharging or overdischarging a cell and damaging it or even setting it on fire.

If you haven't ever done so, I recommend going to sites like lygte-info.dk and looking at the battery testing section, where they show the discharge / charge curves for various types of cells and specific brands and models, which will show you directly what voltage corresponds to what capacity at different charge or discharge currents.

 

[Dui ni shuo de dui]

So I don't need to worry about the difference in charging voltages? For the record, I don't ever intend to charge the battery at a rate higher than an amp or two for the entire pack.

Yes and no.
You need to worry to not go over whatever cell is rated at the lowest maximum voltage. Also, you need to make sure the cells are the same technology. For example, you cannot make a mix of li-ion, lifepo4 and ni-cd....
They all need to be the same chemistry.

So, in your example, assuming they are the same chemistry, you have cells rated at 4.23 V max and cells rated at 4.3v max. So you should make sure that your BMS is set up to cut the charge if any cell goes over 4.23V. Their final charge voltage should never go above the maximum rating of the lowest cell (otherwise it can go boom).

In practice I suggest you don't charge over 4.1V, there is almost nothing to be gained going higher (maybe 1-2% more capacity), and a lot to be lost (cells degrade a lot faster at higher voltages).

Also, always keep in mind that you are using bad cells with various internal resistances, so you should really monitor the pack often, not make any power intensive use of it, charge it as slow as possible (and never leave it alone while doing so), make sure you have a proper balancing system that works fast enough, that your BMS is really well configured and basically treat this battery like a small bomb XD

Remember it's for learning, don't plan on using it in a real project. If you have a real project, then get real cells.

 

[CONSIDERABLE SHOUTING]

They all need to be the same chemistry.

So, in your example, assuming they are the same chemistry, you have cells rated at 4.23 V max and cells rated at 4.3v max. So you should make sure that your BMS is set up to cut the charge if any cell goes over 4.23V. Their final charge voltage should never go above the maximum rating of the lowest cell (otherwise it can go boom).

In practice I suggest you don't charge over 4.1V, there is almost nothing to be gained going higher (maybe 1-2% more capacity), and a lot to be lost (cells degrade a lot faster at higher voltages).

Also, always keep in mind that you are using bad cells with various internal resistances, so you should really monitor the pack often, not make any power intensive use of it, charge it as slow as possible (and never leave it alone while doing so), make sure you have a proper balancing system that works fast enough, that your BMS is really well configured and basically treat this battery like a small bomb XD

Remember it's for learning, don't plan on using it in a real project. If you have a real project, then get real cells.

They are all the same chemistry, but I will have to message A&S to see if the cathode is the same as the Samsungs (LMO). They are both Li-ion.
It seems the LTT BMS I linked does balancing above 4 volts, so charging to 4.05-4.1 volts routinely is perfectly fine. All the 24v chargers I have MIGHT be trimmable, but I'll need to yank their plastic cases open for potentiometers to see (most are wall-warts). I don't plan on charging it past 2-3 amps total, so I imagine that I should be able to find a one easily.

I have the RC3563 on order to see if I take a "next" step.