Trying to create a method to test 18V power tool batteries - help?

21 posts / 0 new
Last post
sac02
sac02's picture
Offline
Last seen: 1 week 4 days ago
Joined: 09/26/2016 - 15:18
Posts: 710
Location: TX
Trying to create a method to test 18V power tool batteries - help?

I know this is a bit off topic from flashlights, but I’ve been trying to come up with a good Budget solution with no success for a while now.

I have a bunch of 18V Milwaukee and Makita battery packs that I need to evaluate their health. Mainly I want to know how they perform (current/voltage) when under at least a moderately high load (25A?). This would be about 450W power.

My first thought was something similar to how I test individual 18650 Li-Ion cells, with a basic resistive load and then compare voltage at rest and voltage/current under load. Then I looked at the price on 500W and 1kW resistors (it took me a while to even find them) and wondered if there was a less expensive method, they were $150 and up. I thought of maybe a fan mounted to a DC motor, but I think motors of that size are kind of expensive too.

Can anyone here help me brainstorm some cost-effective ideas on how to accomplish this?

dekozn
Offline
Last seen: 2 days 9 min ago
Joined: 06/17/2016 - 09:04
Posts: 578
Location: Be

why not use a powertool? connect poles with thick enough cables.

kennybobby
kennybobby's picture
Offline
Last seen: 2 days 21 hours ago
Joined: 05/10/2017 - 09:13
Posts: 102
Location: huntspatch, alabama

You can get some used stove eyes—the heating elements of an electric range, pretty cheap at thrift stores or used appliance shops. i have used them as load banks for many projects—they can take a lot of amps and all they do is heat up…They are low resistance and can be ganged in series or parallel to get the load you desire…flame on!

sac02
sac02's picture
Offline
Last seen: 1 week 4 days ago
Joined: 09/26/2016 - 15:18
Posts: 710
Location: TX

Using a power tool has a couple issues as I see it. running the tool with no load tells me nothing other than yes the battery is capable of turning the tool on at 5% load, so I would have to be making cuts or holes while taking measurements – and even then it is difficult to do repeatedly and accurately (and potentially waste a bunch of wood). I would rather test the battery directly.

  • To test dozens of batteries would be hard on the tool itself, I don’t want to abuse an otherwise healthy cordless tool that I paid for and intent to keep a long time.

A stove burner as a resistance element is a pretty cool idea, I’ll definitely look more into that, thanks.

Edit: research (of my stove) shows the small coils are 24ohms, and the large coils are 40ohms. That’s too high, I’d have to connect a bunch in parallel to get near the kind of currents I was looking for (~20A).

WillyD
Offline
Last seen: 4 hours 43 min ago
Joined: 10/10/2012 - 15:55
Posts: 651
Location: Ohio

Just how many battery packs are you talking here? Sounds like you might have gotten a good deal on some discarded packs. I know you said you don’t want to put the added stress on a power tool, but here’s how I determine if I need to replace one of my DeWalt 18V packs. I charge them up and then test using my circular saw. If they can quickly run a few strong cuts on a 2×4 then I declare them good-to-go. Even if they are down somewhat on power, they are still very useful in a drill or impact driver.

sac02
sac02's picture
Offline
Last seen: 1 week 4 days ago
Joined: 09/26/2016 - 15:18
Posts: 710
Location: TX

About 20 Makita and 10 Milwaukee. I’m not willing to cut 2×4s until the battery dies, and I want to know more than “will it make the first couple cuts?”

I want to be able to quantitatively say with data that these batteries are “like new”, these are “good”, these are moderate, etc.

SawMaster
Offline
Last seen: 1 week 10 hours ago
Joined: 05/03/2015 - 17:18
Posts: 1802
Location: SC USA

You can also use the Nichrome wire in portable mains-powered electric heaters to build high-current resistors. The advantage here is that you can tap the wire anywhere along it’s length to change resistance values Wink It too will get hot in use. Watch your connections for rusting.

Phil

kiriba-ru
kiriba-ru's picture
Offline
Last seen: 29 min 48 sec ago
Joined: 01/17/2016 - 02:34
Posts: 1241
Location: Russia

First way:
Get some useless dc motor and try to put constant load over it. For example, big fan, or water pump, or smth else. There are various of cheap and powerfull 735-775 motors that can carry 20amp load (actually you dont need more than 10Amps). I can give you link for ebay 735 motor that have so big rpm that internal fan produce so big flow and load that I can get max speed with 5Amp power supply.
But if you want to make dead test, you can use car starter motor (I suppose anyone can get used one for cheap).
Best way (how do I do it):
Neverless what do you think about big load tests, there is no way to get original cells work that way when they looks good with small current but fail big current test. So, I prefer disassemble battery case, put crocodile clamps over contacts, make full charge-discharge-charge test with current that your equipment allows to. Then put them aside for are month or near so. This is big problem of old cells and controllers – great self-discharge rate that can easily kill cells. Check voltage just after charging, write it on piece of paper that you can glue to the pack, and then check it one more time a month after. Better use good DMM with 3-4 numbers after point. 0.01V drop is good, 0.1V is bad result.
Also, I practice checking cell balance after first full discharge in real tool. Disassemble case again and check each cell (each 2 cells in 2p5s pack) voltage. 0.02V difference is good, 0.2V difference is bad result.
Using heaters and etc. is bad idea cause load will flow as temperature will grow and it will be difficult to compare results.

Meanwhile in Russia: https://twitter.com/CrazyinRussia

sac02
sac02's picture
Offline
Last seen: 1 week 4 days ago
Joined: 09/26/2016 - 15:18
Posts: 710
Location: TX

Thanks for the input, good points. I was wondering how the load might change with temperature if I used heating elements, and if it would be significant.

A user on another forum mentioned an issue he had when trying to use Makita batteries for a DC powered project – he thinks he made the Makita battery protection logic mad by trying to use the battery not connected to a tool, and his battery would not charge ever again on the Makita charger.

So there are likely some advantages to opening the pack and measuring directly at the battery contacts (and bypass the protection circuit) as you describe.

kiriba-ru
kiriba-ru's picture
Offline
Last seen: 29 min 48 sec ago
Joined: 01/17/2016 - 02:34
Posts: 1241
Location: Russia

sac02 wrote:

A user on another forum mentioned an issue he had when trying to use Makita batteries for a DC powered project – he thinks he made the Makita battery protection logic mad by trying to use the battery not connected to a tool, and his battery would not charge ever again on the Makita charger.

Makita have most simple circuits, and all batteries support old tools without third pin – so circuit could never recognize if there is tool or other load. Problem could be if total voltage is under limit, or voltage difference between cells is too big, or fuse is broken.
20Amp is enormous amount, you will recognize smth wrong when contacts are already melted.

Meanwhile in Russia: https://twitter.com/CrazyinRussia

sac02
sac02's picture
Offline
Last seen: 1 week 4 days ago
Joined: 09/26/2016 - 15:18
Posts: 710
Location: TX

Good info on Makita protection circuit. Do you know anything about Milwaukee protection?

I planned to test using 20A because DC motors can draw large currents, especially near stall. And I thought the capability of the battery would be more obvious at high current. But I am just making assumptions.

What do you think is a proper current to judge the health of a power tool battery pack?

kiriba-ru
kiriba-ru's picture
Offline
Last seen: 29 min 48 sec ago
Joined: 01/17/2016 - 02:34
Posts: 1241
Location: Russia

What do you want to get?
If you need to devide packs you have in two groups – good and bad – you dont need big current.
If you need to have presicion capacity, you can test packs with 5-8A constant current load.
Only grinders and circular saws have motors and electronics and usages when you can have long-time load more than 10Amp. Other tools have smaller motors and non-constant load – average doesnt exceed 5amp (35 minutes of discharging for 3amp pack – not sure you have ever made it faster).

Meanwhile in Russia: https://twitter.com/CrazyinRussia

sac02
sac02's picture
Offline
Last seen: 1 week 4 days ago
Joined: 09/26/2016 - 15:18
Posts: 710
Location: TX
kiriba-ru wrote:
… If you need to devide packs you have in two groups – good and bad – you dont need big current. …

At minimum I want to know this.

kiriba-ru wrote:
… If you need to have presicion capacity, you can test packs with 5-8A constant current load. …

This would be good to know.

kennybobby
kennybobby's picture
Offline
Last seen: 2 days 21 hours ago
Joined: 05/10/2017 - 09:13
Posts: 102
Location: huntspatch, alabama

The control circuits in the lithium power tool packs, especially Milwaukee, have a “feature” to permanently disable the pack if certain limit conditions are exceeded. This is described in the patent that Milwaukee Electric Tool Company owns and for which it receives royalties from all the other tool companies (makita, dewalt, ryobi, etc).

The part numbers of the micro and bms chips used has not been found traceable to anything available from typical electronics parts vendors or found using web searching.

The bms chips are made by O2Micro, a Cayman Island corporation, and no datasheets are found on their website. Can you say offshore tax-haven shell company—sure, sure you can.

So they are doing a great job of keeping their code secret and hidden from those of us who might like to debug this “feature” and remove it from disabled packs that have perfectly good and useful cells with no damage or defect.

There are typically 2 FETs in the packs that must be driven ON in order for the pack to function properly for either charge or discharge. If the micro decides to not drive the FETs, then your pack becomes a brick. The micro will then turn on circuits to drain the cells completely.

Lazy-R-us
Offline
Last seen: 5 hours 38 min ago
Joined: 09/13/2015 - 00:30
Posts: 626
Location: Carolinas, USA

If this is something that you plan on keeping for a while, consider buying the $20 Makita flashlight. You can take it apart and add whatever load you decide to use inside it – keeping in mind any cooling requirements. I have one of these that I used as a housing for another project, it is a great form factor and available at every Home Depot.

I did a search on BangGood for ‘load” and got some results, but since I have not used any of them, I can’t say one is better than another.

Lazy-R-us

dchomak
Offline
Last seen: 7 hours 5 min ago
Joined: 03/17/2012 - 20:10
Posts: 3864
Location: Connecticut

I had already come across this problem and posted my solution some time ago.
Unless one needed to know, they would have no interest in it and would probably have ignored my post.
Here it is.

I needed to measure the capacity of an 18V drill pack, here is how I did it

Lexel
Lexel's picture
Offline
Last seen: 5 hours 18 min ago
Joined: 11/01/2016 - 08:00
Posts: 2885
Location: Germany

Why dont buy some of those resistors and mount them on old CPU heatsink
http://www.ebay.de/itm/100W-Watt-Shell-Power-Aluminum-Housed-Case-Wirewo...

The 6Ohm is cheap and you could mount 4 and change resistance from 1.5 to 24 ohms

[Reviews] Miboxer C4-12, C2-4k+6k, C2, C4 / Astrolux K1, MF01, S42, K01 / BLF Q8 / Kalrus G35, XT11GT / Nitefox UT20 / Sofirn SF36, SP35 / Imalent DM21TW / Wuben I333 / Ravemen PR1200 / CL06 lantern / Xanes headlamp

[Mods] Skilhunt H03 short / Klarus XT12GTS / colorful anodisation

[Sale] TA drivers Narsil or Bistro, XHP35 E2-3C

kotobuki
Offline
Last seen: 7 hours 34 min ago
Joined: 10/17/2016 - 09:41
Posts: 167
Location: US

Hello Sac02,
In case you need some high power dummy loads, try these 200W ceramic heaters (or similars around there), choose 24V for safe:

You can combine them to many configurations as you want …

kennybobby
kennybobby's picture
Offline
Last seen: 2 days 21 hours ago
Joined: 05/10/2017 - 09:13
Posts: 102
Location: huntspatch, alabama

dchomak wrote:

I needed to measure the capacity of an 18V drill pack, here is how I did it

i like it — promote that guy!

sac02
sac02's picture
Offline
Last seen: 1 week 4 days ago
Joined: 09/26/2016 - 15:18
Posts: 710
Location: TX

Lexel wrote:
Why dont buy some of those resistors and mount them on old CPU heatsink
http://www.ebay.de/itm/100W-Watt-Shell-Power-Aluminum-Housed-Case-Wirewo...

The 6Ohm is cheap and you could mount 4 and change resistance from 1.5 to 24 ohms

Do you think the 100W power rating is even close to accurate? (I would mount them on a heatsink) That was my main concern, that even with a heatsink these small resistors are nowhere near 100W capable.

That, and the fact that my first plan was to test at very high current / power.

But if I do not test at 20A, I do not need such high power resistors. A single 4ohm resistor would be 4.5A and only result in 81W, which is significantly less than the 350-450W I was considering. Since I do not necessarily believe the 100W rating, I could easily do two 8ohm in parallel.

Maybe that is the best answer. My assumption of needing to test at 20+A maybe was incorrect and made things too complicated I guess.

Lexel
Lexel's picture
Offline
Last seen: 5 hours 18 min ago
Joined: 11/01/2016 - 08:00
Posts: 2885
Location: Germany

The rating of those resistors it true, its a simple resistance wire coil thermically connected inside an aluminum tube

Just they need active cooling with some sort of heatsink, one with fan is far more effective than those passive big resistors

just imagine how small a CPU core with 100W is, it all depends how you can conduct the heat away from it

[Reviews] Miboxer C4-12, C2-4k+6k, C2, C4 / Astrolux K1, MF01, S42, K01 / BLF Q8 / Kalrus G35, XT11GT / Nitefox UT20 / Sofirn SF36, SP35 / Imalent DM21TW / Wuben I333 / Ravemen PR1200 / CL06 lantern / Xanes headlamp

[Mods] Skilhunt H03 short / Klarus XT12GTS / colorful anodisation

[Sale] TA drivers Narsil or Bistro, XHP35 E2-3C