Battery discharge curve analysis

[Barkuti]

Hi!

Fellows, I'm in the process of restoring an old Ni-CD drill/screwdriver battery pack.

Jokes aside

, the batteries are to be replaced by, you know, a marriage of LiitoKala 26650-50A cells.
Well, since there are no known 20A discharge curves for these cells (besides the stuff available at 高一致性3.7V 5000mA镍钴锰酸锂三元锂电池 生产厂家｜东莞力朗电池科技有限公司), I've come up with some extrapolation figures which I'd like to share in order to see if this makes enough sense.

The material presented below is a discharge curve done through a 300W Turnigy Reaktor hobby charger and the voltage was monitored right at the cell contacts using a Junsi Powerlogger 6S. Source: EasyB's #118 post at “LiitoKala 26650-50A 5000mah 26650 - great first impressions” BLF thread.

First of all, the curves seem to start a bit below where it should, in my opinion, for a battery with no more than 23mΩ of internal resistance. At least 35mV, as I see it. So the testing setup may had some stray resistance, doesn't it? A minor detail, of course.
Well, the key here for me is to attain a good idea of the expected capacity at 20A discharge rate. The 20A curve should have a voltage offset of current delta times Rbatt, so (20 - 5) × 0.023Ω ≈ 0.35V, no more.
With this figure in mind, I thus see close to 4.9Ah of capacity at 20A rate with 2.8V cutoff, and almost 4.5Ah if we step up the cutoff to 2.9V, which is the figure for the BMS I bought (tad high, doesn't it?

).
Well, that's all for now folks.
I'll be glad to hear from you.

Cheers

 

[zoiDman]

Looks like an Interesting Project. But one Question...

Why did you decide to Rebuild it with 26650 Batteries?

 

[Barkuti]

zoiDman, the main reason is I have these batteries in storage unused, plus I am not going to continue with the more complicated than expected project I had them in mind for. I have no formal money intake atm, so spending money on stuff which isn't guaranteed to at least give me $om€thing back is of no service to me.
The drill belongs to a regarded friend who (€20) funded the restoration.
He's gonna get impressed with the runtime of the overhauled thing, I guess.

Cheers

 

[zoiDman]

zoiDman, the main reason is I have these batteries in storage unused, plus I am not going to continue with the more complicated than expected project I had them in mind for. I have no formal money intake atm, so spending money on stuff which isn't guaranteed to at least give me $om€thing back is of no service to me.
The drill belongs to a regarded friend who (€20) funded the restoration.
He's gonna get impressed with the runtime of the overhauled thing, I guess.

Cheers

Gotcha.

Let us know how things work out.

 

[DaveP]

I have a Black and Decker 18v drill with two batteries. One works and the other gets rejected by the charger. Guys on the tool forums go to Harbor Freight and buy HF 18V drill batteries for $12 and $18 and strip the batteries out to install in the Black and Decker case.

I plan to do that when I get the urge ... probably when the last remaining battery gets rejected by the charger!

Barkuti, if you replace nicads with lithium, what do you use to charge the battery pack once they are connected together? NiCads and Lithium are going to need a charger that can recognize the cells.

 

[Barkuti]

This inexpensive CC/CV module: Free Shipping 5A constant current LED driver module battery charging constant voltage DC DC power module-in Integrated Circuits from Electronic Components & Supplies on Aliexpress.com | Alibaba Group
Check out this video about it:

​

Another potential problem is the wall wart which came with the tool: “24V 500mA”. Freakin' lies of course. Measured ≈19.7V no-load, and 11.21V when loaded with 37.5Ω, which means less than 299mA and God knows how is the voltage waveform

as I don't have an oscilloscope to check it out. LoL!
I already told the owner of the tool to take a peek at his stash of power bricks, just in case.
The CC/CV module is to be fit inside the battery cradle, which still has the stock charging PCB: a lame board with a red led, a big power resistor and maybe a diode. Tomorrow I'll post a pic of it for fun if you will.
I opened a thread on BLF about this long ago: 24 chinese volts drill battery pack overhauling…
Have fun!

Cheers

 

[DaveP]

I opened a thread on BLF about this long ago: 24 chinese volts drill battery pack overhauling…
Have fun!

It's time for my 18v Black and Decker cordless drill to get new NiCad battery packs. Of the ones that came with it a few years ago, one flashes on the charger to indicate a bad pack and the other is still going with shorter charge life.

I read on one of the power tool forums that people are buying the cheap Harbor Freight battery packs and shifting the prewired cells to the B&D housing. Ever heard of that? It would surely be cheaper and the wiring is already done.

 

[Barkuti]

Drill overhauling finished. Owner is quite happy with the result, reporting higher performance than stock when new.
The only “qualm” is the protection trips a bit early, it is a 15A continuous (34A peak) BMS. Gonna fix it stacking 2 or 3 additional Si4324DY mosfets.

Cheers

 

[Barkuti]

WoW! Just told my friend I can overhaul the other battpack of its tool into an 3S2P array consisting in LiitoKala 26650-50A + LG HB6 marriages.
It would be nearly €36 between the cells and two 20A BMSs in parallel. Performance?
Go look a video of a Radical SR8 or a Caterham R620 on youtube… sort of.

Cheers

 

[Barkuti]

Lil update on this, fellow spectators.
Finally found out the drill problem was due to some kind of stray resistance on one of the cells. I haven't thoroughly inspected the stuff, but that is the only reason the BMS could be low-voltage (over-discharge) tripping. No surprise to me here, because I am nearly sure one of the (over aluminium) solder joints is bad. It just works because of the hot glue keeping it all together, nothing more.
Additionally, since the INR26650-50A has been reviewed and I see it has trouble over 20A, I am going to solder a couple of TrustFire “Flame” IMR14500s in parallel with the Liitos which, at about ≈81mΩ of internal resistance, will boost torque by at least a 25%.
I'll be using Rose's Metal to join the batteries together, an under 100° fusion temp alloy. It can also serve as overheating fuse, LoL!
More on this once I receive the alloy and my new 80-450° 110W “climate controlled” soldering iron.

Cheers my fellows

 

[Barkuti]

Final update (for the sake of completeness)

HKJ recently reviewed my aforementioned TrustFire batteries: http://lygte-info.dk/review/batteries2012/TrustFire IMR14500 700mAh (Red-gold) UK.html
They were rated “bad” because of a consistency issue, but they live up to their specced amp delivery.
I did a binning test with my pair, tad accidentally discharging them in series down to (!) 4.34V. A few seconds after their voltages were 2.59/2.57V, so I can presume my pair is well matched.
Seems they could be used for vaping at ≈20+W safely in some sort of pencil style minimod.

Layer of AO4430 MOSFETs over the SY4324s, reducing the combined RDS(ON) below 2.5mΩ, thus setting the overcurrent limit at 60+A.

Some pictures taken during the overhauling reassembly:

Middle point. 3x 0.1mm sheet over the LiitoKalas.

“Copper is mine, copper divine.”
Usage of Rose's metal together with aggressive acidic flux (Goot Super Soldering Flux) makes soldering batteries together super-easy.

Ready to be closed battery tray.

Massive wiring up-down to guarantee no less than 25A continuous.

Lamentable stainless steel contact sheets upgraded with proper 0.1mm (≈3 ounces) copper sheet.

Since the drill has a low rotational speed (≈600 rpm) available torque is massive. No problems trouncing wood with a 10mm bit.

Cheers fellows