Any recommendations for li-Ion batteries?

Hi Gene,

I would look for batteries used in a commercial application. If you buy the OEM batteries for a product and then salvage them you are likely to get good quality as the manufacturer has his name and reputation on the line. They also have better resources for finding quality manufacturers and ensuring better quality control. Otherwise it is a jungle out there as far as I can tell. The race is on to see who can deliver the cheapest product with the biggest lie (mAh) printed on the side.

Just out of curiosity what were the parameters of your test to determine the mAh of the batteries. I want to test a few Li-ion that I have salvaged from a computer power pack but I haven't had time. Not having much experience I would benefit from your insights.

John

Hi Gene,

For cylindrical Li-Ion, I've used cells from Farnell, and they are great, but the price may not suit all hobby needs (I don't know if you're looking for a commercial project or not). For flat LiPo for hobby projects, Olimex have LiPo batteries at great prices, and they are a reputable seller. However shipping may be expensive to the US. Otherwise, I've bought from Seeedstudio before too, a 2AH LiPo I think.. Anyway, it worked great. I think both Olimex and Seeedstudio are reputable vendors for one-off batteries for hobby use, but for traceable parts then Farnell/Newark have a decent range. As John says, there are plenty of ultra-cheap battery suppliers that are not truthfully advertised, and could also be risky to use.

The ultrafire name reminded me of a school project..

We had to build a toy for less-abled children, so we basically made a radio controlled car (using off-the-shelf remote control) and the thing was heavy - it had a brazed steel chassis, steel threaded rods for axles, and a huge motor and the usual radio control NiCd pack.. and a Herbie shell vacuum-moulded.

We adapted the remote control, by wiring in a joystick that could be clamped to a wheelchair.

Anyway, the kids loved it, because it went fast and would bruise the ankles of any teacher in the way : )

Our teacher was quite proud of us, and took the car home one weekend to show it to his own boys. But it came back as a

blackened charred mess, still smelling of smoke : ) It turns out the wire I used wasn't particularly flame tolerant, and when it caught fire it was all a load of plastic kindling : )

A friend still had the photos from the local paper (they printed anything.. not much happened in the area!) and shared them with me the other day..

John,

I am working on a battery charge/power pack.  I have instrumented the charger side with the ability to measure the current supplied to the battery during the charging phase.  I log the data and then compute the total charge delivered to the battery.  This is done after completely (safely) draining the battery.  The charge delivered is roughly equal to the cells capacity.  After fully charging the battery I then drain the battery into a into a load (drawing 500mA), while again logging the data (unfortunately I can't log the battery current).  I compute the battery current during draining, based on a fix load and fixed voltage (5V boost) and the ratio of battery voltage to boost voltage.  I apply a rough correction for the efficiency of the boost regulator.  The charge in does correlate nicely with the charge in.

I was first tipped off that there was a problem as it was only taking about an hour to fully charge the battery. Given that I am using a constant current of 500mA, up until the battery reaches 4.2V then a constant voltage (diminishing the current until < 50mA), a 1200mAh battery should take around 6-7 hours.  After several runs, it be came apparent what was happening.  After doing a little searching (which I should have done before I bought the batteries), I found several articles about the UltraFire batteries being way over rated in terms of capacity.

I like your idea about salvaging batteries.  When I was working with a company that used li-Ion batteries in their products, we did buy the very best that we could find.  Unfortunately, I have built my charger based on these AA (14500) batteries, and most commercial devices have direct soldered in cells, or wired in cells (not a AA battery holder).  I got a like cute, using the battery holder (for quick disconnect and replacement).  I will continue looking.

Thanks,

Gene

genebren

We have a guy in another club that was in charge of battery packs for mobile radios.

He had some very interesting tales.

Basically he knew everything on the Battery University site

http://batteryuniversity.com/

I recall one talk he gave :-

Charging any of the Lixx batteries takes 2 x Capacity in terms of time.

You can limit the fast charge and then have a shorter float, or faster charge the first stage and a longer float, but either way it's the same.

Also it is extremely important to establish the maximum battery voltage and NEVER exceed it.

He warned about unknown cells and if in doubt err on the side of lower float voltage ... he even advocated pulling the pack apart to see what was stamped on the cells.

There was also something about life (number of recharges) v capacity, and lowering the voltage even 0.1v gave something like 100 or 1000 more times the life (I can't recall sorry)

He described a charged Li xx as a Hand Grenade with the pin pulled.

Mark

Hi Gene,

You got me inspired to do what I have been intending to do. Here is a 18500 battery that I salvaged from a computer battery pack

I had charged this battery fully about a week ago to 4.2 volts and in the mean time it had settled back to 4.1 volts. I hooked it up to the Process Duration Timer and set the target voltage for 2.5 Volts. I set a load of 2 amps on the electronic load and started the Process Timer.

31 minutes later the voltage of the battery crossed below the 2.5 Volt Target and stopped the clock. This is about as close to the 1000 mAh ratting on the battery as one can get. Frankly I was quit surprised as this battery came from a pack with 2 failed cells and many hours of use and many recharges.

Sorry about the light flare on the timer, it says 00:31:10. The DUT voltage is back to 3.02 volts on the display as I have cut the load and it has recovered from the 2.5 volt stop point. The other thing that impressed me was how fast the voltage fell off in the last 30 seconds that the load was on the battery. It was like it held voltage pretty well and then just ran out of gas. I know that this isn't the size of battery that you want to use so this may no help much.

What about trying Batteries Plus and getting a guarantee of accurate labeling for the mAh.

John

Hi Shabaz,

What a nice project and memory. Now I understand why the project that you sent me had all the Teflon covered wires in it.

John

Hi John,

That is pretty impressive that this well used battery is so close to it's specified capacity.  The drop-off in voltage at the end of charge is pretty steep, but normal.  Here is a graph of one of the batteries that I have been testing.

So, how do you like that KUNKIN electronic load?  I have been in the market for a solid load with a rating up to 10A.  Any advice on loads?

Thanks for the data.  It is nice to see a quality battery at work!

Gene

Hi Shabaz,

I checked my local Newark, but I could not find any 14500 (AA) Li-Ion batteries listed.  This is just a for fun project, so I don't have any cost targets (except to keep my wife happy by not spending too much).  I will check around on Olimex and Seeedstudio.

Thanks!

Gene

Hi Gene,

The Kunkin is OK for the money. My only irritation is that it doesn't work well at low load settings. For example if I want to put a 10 mA load on something then it is unstable. For the type of experiments that I do most of the time it works well and if I need to go to less than 50 mA I use a load that I built that covers the low current situations better. your graph is an excellent portrayal of what I saw when the battery discharged.

John