Tool Battery - Replace or rebuild (partial/full) Opinion?

[gaalcom]

One of my Hitachi 18v batteries (EBM1830) has called it quits. After disassembly I found one cell (Sanyo UR18650W2) has failed. Now, I'm trying to decide -- Buy a $50 new Hitachi OEM pack of unknown age; $30 aftermarket pack of (as always) questionable quality; source a new Sanyo UR18650W2 cell (2pk for $13); or refresh all cells with Samsung 25R 2500mAh ($23 10pk). I'm leaning towards 10 cell replacement with a "supposedly" 1000mAh Samsung upgrade from the 1500mAh Sanyo's. The one downside I see is not gaining the spare electronics that would come with the purchase of a complete pack. I expect there'd be no conflict with the electronics/charging considering the only variable is capacity. Opinions?

*EDIT* -- Nix the $23/10pk Samsung batteries, quite possibly fake. $37/10pk for genuine, still not a leaning-towards option breaker. :-/

Bonus question! ;-)
I was attempting to adapt a wall charger for use with a battery operated motion sensor. The sensor works off 3 AA's in series and lists an input voltage of 5v. It doesn't list amperage but I'm assuming it would be sub 1A. I have way too many homeless wall chargers and found a few with rated output of 5v and from .25mA-1.0A. Thinking this would be a no-brainer I just tapped it in only to find it didn't work as expected. The sensor turned on but didn't operate properly. I found the charger, when attached to the sensor, was outputing nearly 9v and overwhelming the sensor sending into a reset loop. Could it be because the simple wall chargers have no (or weak) regulation? Insights appreciated.

 

[wssix99]

One of my Hitachi 18v batteries (EBM1830) has called it quits. After disassembly I found one cell (Sanyo UR18650W2) has failed. Now, I'm trying to decide -- Buy a $50 new Hitachi OEM pack of unknown age; $30 aftermarket pack of (as always) questionable quality; source a new Sanyo UR18650W2 cell (2pk for $13); or refresh all cells with Samsung 25R 2500mAh ($23 10pk). I'm leaning towards 10 cell replacement with a "supposedly" 1000mAh Samsung upgrade from the 1500mAh Sanyo's. The one downside I see is not gaining the spare electronics that would come with the purchase of a complete pack. I expect there'd be no conflict with the electronics/charging considering the only variable is capacity. Opinions?

There is no problem swapping these cells. The issue I've had in the past is that there is little room and I've always needed a tab welder to weld tabs to the cells. There hasn't been enough room in the packaging to use wires.

Bonus question! ;-)
I was attempting to adapt a wall charger for use with a battery operated motion sensor. The sensor works off 3 AA's in series and lists an input voltage of 5v. It doesn't list amperage but I'm assuming it would be sub 1A. I have way too many homeless wall chargers and found a few with rated output of 5v and from .25mA-1.0A. Thinking this would be a no-brainer I just tapped it in only to find it didn't work as expected. The sensor turned on but didn't operate properly. I found the charger, when attached to the sensor, was outputing nearly 9v and overwhelming the sensor sending into a reset loop. Could it be because the simple wall chargers have no (or weak) regulation? Insights appreciated.

If the charger lists it's output as 5V and you measure 9V, then it's just a bad charger. You should be able to get another charger and give it another shot.

 

[gaalcom]

Much appreciate the reply.

There is no problem swapping these cells. The issue I've had in the past is that there is little room and I've always needed a tab welder to weld tabs to the cells. There hasn't been enough room in the packaging to use wires.

Yeah, don't have the option for tab welding but had looked and determined there's likely enough room for soldering in some 12 or 10 gauge wire (might have to relief a rib or two for the latter). I also have a high watt gun so may attempt tinning and soldering on the original tabs/plates, should be able to minimize heating of batteries.

After sleeping on it I think I'm going to simply replace the single failed cell with the same. For the most part the packs are still quite strong and long lasting. After they wear appreciably will likely do a wholesale cell change.

If the charger lists it's output as 5V and you measure 9V, then it's just a bad charger. You should be able to get another charger and give it another shot.

Odd thing is I tried three different wall chargers for power. Each has an open voltage of between 4.8v and 5.01v but when attached to the motion sensor output raises to between 6.8v and 8.0v.

*EDIT* -- In the meanwhile if I can manage to scrounge some 4.7KΩ to 10KΩ resisters from some of my defunct electronics I'll try a quick-n-dirty voltage divider.

 

[wssix99]

Odd thing is I tried three different wall chargers for power. Each has an open voltage of between 4.8v and 5.01v but when attached to the motion sensor output raises to between 6.8v and 8.0v.

That doesn't make sense. What do you measure when there are batteries in it?

 

[Bretny]

I replaced a cell in a 9yr old M18 pack. It worked once then failed like the time prior. I would guess the balance or charge circuit quit. I wouldn't spend to much time trying to repair for the price of new.

Get your self a flashlight or something that uses 18650 battery, a 18650 charger and harvest the cells. I have done this with a few lithium tool battery and almost never find a cell that s dead.

 

[dave*99]

Much appreciate the reply.



Yeah, don't have the option for tab welding but had looked and determined there's likely enough room for soldering in some 12 or 10 gauge wire (might have to relief a rib or two for the latter). I also have a high watt gun so may attempt tinning and soldering on the original tabs/plates, should be able to minimize heating of batteries.

After sleeping on it I think I'm going to simply replace the single failed cell with the same. For the most part the packs are still quite strong and long lasting. After they wear appreciably will likely do a wholesale cell change.



Odd thing is I tried three different wall chargers for power. Each has an open voltage of between 4.8v and 5.01v but when attached to the motion sensor output raises to between 6.8v and 8.0v.

*EDIT* -- In the meanwhile if I can manage to scrounge some 4.7KΩ to 10KΩ resisters from some of my defunct electronics I'll try a quick-n-dirty voltage divider.

How much current do you wish to draw from the proposed voltage divider?
If you are using resistors in the kiloohms range to make the divider, you can expect to derive milliamps in the output.

 

[gaalcom]

Again, thanks for the replies!

That doesn't make sense. What do you measure when there are batteries in it?

With the three batteries installed (series) it measures a constant 4.76v. This holds true with the On/Off switch in either position. Although I definitely remain curious as to the reason for voltage rise I'll probably throw in the towel and simply use the micro USB power option I overlooked. I just tested and it works properly on 5v/1A USB.

I replaced a cell in a 9yr old M18 pack. It worked once then failed like the time prior. I would guess the balance or charge circuit quit. I wouldn't spend to much time trying to repair for the price of new.

Get your self a flashlight or something that uses 18650 battery, a 18650 charger and harvest the cells. I have done this with a few lithium tool battery and almost never find a cell that s dead.

You offer words of past experience I could probably benefit from.

I did initially give possible circuit board failure some thought. What has me assuming this is not the case would be the fact there are two cells on that circuit -- When disassembled, one cell measured 3.92v while the other was below 1.0v. Using my ability to charge a single cell I attempted to individually test those two cells. The 3.92v cell was within working voltage so a normal charge topped it up to 4.21v which then performed well under load and held a charge. The other cell was placed on a repair charge (.2C) which failed as it never got above 1.5v, within 24hrs of shelf time it drained to less than .8v.

In the end, your line of reasoning is the same that made me decide on sourcing just one or two cells to repair the pack. I was prepared to have nine reclaimed 18650 cells for other uses if I decided to rebuild the pack with all new cells but I really don't want 19 cells laying around if ultimately it ended up being a charge/balance circuit problem with that battery pack.

How much current do you wish to draw from the proposed voltage divider?
If you are using resistors in the kiloohms range to make the divider, you can expect to derive milliamps in the output.

Good catch, something I didn't even consider. It may not be a concern any longer as I'm likely switching to USB power but, for sake of knowledge -- I know it runs off 5v/1A (which is what the mini wall warts are rated at). My musing was -- If the supplies are putting out 7.5v on average I was thinking resisters (voltage divider) of 4.7 and 10 would get me to 5.10v. For the life of me I can't determine end current. Best I can figure is, even using 4.7Ω and 10Ω (vs. KΩ), current would be .5A-.6A. Might that be a good ballpark guestimation on my part?

 

[jkeyser14]

FYI, you can also solder tabs to the cells to keep a flat profile. Use flux and be as brief as possible with your application of heat, you don't want to damage the cell.

 

[dave*99]

Again, thanks for the replies!


With the three batteries installed (series) it measures a constant 4.76v. This holds true with the On/Off switch in either position. Although I definitely remain curious as to the reason for voltage rise I'll probably throw in the towel and simply use the micro USB power option I overlooked. I just tested and it works properly on 5v/1A USB.


You offer words of past experience I could probably benefit from.

I did initially give possible circuit board failure some thought. What has me assuming this is not the case would be the fact there are two cells on that circuit -- When disassembled, one cell measured 3.92v while the other was below 1.0v. Using my ability to charge a single cell I attempted to individually test those two cells. The 3.92v cell was within working voltage so a normal charge topped it up to 4.21v which then performed well under load and held a charge. The other cell was placed on a repair charge (.2C) which failed as it never got above 1.5v, within 24hrs of shelf time it drained to less than .8v.

In the end, your line of reasoning is the same that made me decide on sourcing just one or two cells to repair the pack. I was prepared to have nine reclaimed 18650 cells for other uses if I decided to rebuild the pack with all new cells but I really don't want 19 cells laying around if ultimately it ended up being a charge/balance circuit problem with that battery pack.


Good catch, something I didn't even consider. It may not be a concern any longer as I'm likely switching to USB power but, for sake of knowledge -- I know it runs off 5v/1A (which is what the mini wall warts are rated at). My musing was -- If the supplies are putting out 7.5v on average I was thinking resisters (voltage divider) of 4.7 and 10 would get me to 5.10v. For the life of me I can't determine end current. Best I can figure is, even using 4.7Ω and 10Ω (vs. KΩ), current would be .5A-.6A. Might that be a good ballpark guestimation on my part?

Voltage dividers are a poor choice for this application. I'm glad you chose another path. But just for discussion - if you want 10% regulation from a voltage divider, you need the current flow in the divider resistors to be 10X the load from the batteries.

So to get 100 mA of charging current you need 1.0 A flowing through the resistors. And that will result in a 10% voltage drop in the 5.1V value.

Now this all assumes the battery impedance is 10X the 10 ohm resistor in your divider. The battery is in series with the 4.7 ohm resistor and parallel with the 10 ohm resistor.

But battery impedance changes with charge level. You get the idea.

 

[wssix99]

found a few with rated output of 5v and from .25mA-1.0A.

Just to make sure... Did you check to confirm the output of your charger is DC? A lot of these AC transformers have both AC inputs and outputs. (Not all convert to DC.)

Applying AC to a DC circuit can cause odd things to happen.

 

[Niteman9]

I'd get the new $50 battery. My time is worth more to me.

Sent from my SM-N960U using Tapatalk

 

[yatg]

Cheap wall chargers have poor voltage regulation. You could get an LM7805/L7805 chip and connect between charger and load.

http://www.learningaboutelectronics.com/Articles/What-is-a-LM7805-voltage-regulator

 

[gaalcom]

Much appreciate the many insights and suggestions!

FYI, you can also solder tabs to the cells to keep a flat profile. Use flux and be as brief as possible with your application of heat, you don't want to damage the cell.

I'm fortunate to have a 150/400w gun that would allow for quick heat-minimized tab soldering. I very well may reuse the original tabs/plates.

Voltage dividers are a poor choice for this application. I'm glad you chose another path. But just for discussion - if you want 10% regulation from a voltage divider, you need the current flow in the divider resistors to be 10X the load from the batteries.

So to get 100 mA of charging current you need 1.0 A flowing through the resistors. And that will result in a 10% voltage drop in the 5.1V value.

Now this all assumes the battery impedance is 10X the 10 ohm resistor in your divider. The battery is in series with the 4.7 ohm resistor and parallel with the 10 ohm resistor.

But battery impedance changes with charge level. You get the idea.

Thanks for that detailed explanation. It does seem that considering the questionable quality of these power supplies that use of voltage dividers would have to be trial and error to produce the desired results. The USB route is definitely a simplest and proven option. That said, I like dabbling in uncharted territory so...the IC VR has peaked my interest.

Just to make sure... Did you check to confirm the output of your charger is DC? A lot of these AC transformers have both AC inputs and outputs. (Not all convert to DC.)

Applying AC to a DC circuit can cause odd things to happen.

Just double checked and output is definitely DC.

I'd get the new $50 battery. My time is worth more to me.

My hesitation is this battery pack happens to be the older post style (vs shoe) and, from what I understand, is no longer produced by Hitachi. OEM are probably NOS so shelf life could be well dated. New aftermarket packs of unknown cell type could be a crapshoot. I've seen quite a few capacity reviews of them that routinely measure as little as 55%-60% advertised. Luckily I have some free time. ;-)

Cheap wall chargers have poor voltage regulation. You could get an LM7805/L7805 chip and connect between charger and load.

http://www.learningaboutelectronics.com/Articles/What-is-a-LM7805-voltage-regulator

I'm really liking this suggestion. Although USB power would be the least fuss I'd like to try something new while also finding a use for some of these wall PSU's taking up space. Add to that, ten 7805's for $4 shipped beats the USB chargers and cabling I'd have to buy.

I'll install IC Reg first without the aid of any ceramic caps as the IC will be located very close to the PS. Expect heat shouldn't be an issue either as the input voltage is only slightly above rated output (I do have appropriate heatsinks if I use one of my many 9v-12v PSU's).

 

[Max]

Much appreciate the many insights and suggestions!

I'll install IC Reg first without the aid of any ceramic caps as the IC will be located very close to the PS. Expect heat shouldn't be an issue either as the input voltage is only slightly above rated output (I do have appropriate heatsinks if I use one of my many 9v-12v PSU's).

The short answer is that you should use all of the caps that the manufacturer recommends you to use. That could be as few as two or as many as four.

The long answer is that the voltage regulator includes a very high gain amplifier. If you don’t use the caps, the amp will be sensitive to noise and may be unstable. And if it’s unstable, it may not regulate, may oscillate, etc. The bottom line is that when the VR’s designers designed the part, they were very aware that their customers look at not just the IC cost, but also any associated component cost. So all of the parts the mfg. suggests you add are what they think are the very minimum required...

- Max

 

[gaalcom]

The short answer is that you should use all of the caps that the manufacturer recommends you to use. That could be as few as two or as many as four.

[Snip]

Appreciate the info!

Ah, from my cursory research I was under the impression that their required use could "possibly" be location dependent. A fair number of comments were that if the IC is located very close to the PS filters then caps might be omitted successfully. I guess even if that were the case quality of PS and other components could be a contributing factor. If I end up using an IC Reg (vs USB power) I'll insert the ceramic caps

The kicker here is that, for a guy not dabbling in electronics on a daily/weekly basis, because it's cheaper to purchase bulk packs instead of 2 individual pieces (ex. 7805 IC's $2ea or $4 10pc) over the years I've amassed parts/pieces bins that make me look like an electronics repair shop, sans the expertise. ;-)

 

[Max]

Well, if you get a bulk pack of regulators then get a bulk pack of caps at the same time.

I haven’t looked at that specific regulator, so I don’t know what/where they want caps. There could be just a ceramic cap on the input and output, or they could specify an electrolytic or two as well. If there is a electrolytic on the input and the reg is close to the power supply, then you can skip that electrolytic on the input. But you’d still need the ceramics and an output electrolytic (if specified).

- Max

 

[Max]

OK, finally got off my lazy tail and looked at the data sheet for the TI 7805. Just two caps, and the input one can be deleted if you are close to the power supply. And the output one is not needed for stability - but it does improve transient response. Since you’re just looking at charging a battery (not a big dynamic load) you can skip that cap as well. So after some work you’re back where you wanted to be.

Sorry for any confusion, but not all regulators are as easy to use as the 7805.

BTW, please make sure that you look at the power dropped through the regulator and add a heat sink if needed. You should not be able to hurt the part if you don’t cool it enough, as it has thermal shutoff and it will shut down if needed. Still, you might wonder why the battery didn’t get charged.

Also, part of why the 7805 is easy to use if that it’s got some generous specs. As an example, the dropout voltage is 2V, and the minimum Vcc in is 7.5V. So if your power supply voltage drops with a load on it (and I think it might) then the regulator may drop out of regulation.

- Max

https://www.ti.com/lit/ds/symlink/lm340.pdf

 

[gaalcom]

Very much appreciate that additional due diligence Max!

Going to bite the bullet and order those regulators shortly. As has been said, no damage can be done, it'll simply work or not (will keep an eye on heat and sink if required). If caps are deemed necessary I'll be sure to order in bulk!

You note the minimum Vcc and that might require me to use one of my many other salvaged PS's. Those that I have been trying to use are rated at +5 Vout and test that way open but when loaded they jumped to approx 7v-8v, which is what I was attempting to regulate. I'll probably be better off employing the more appropriate +9 Vout PS's.

Will be sure to update with results.

 

[Max]

Sounds good and have fun!

- Max

 

[gaalcom]

The 7805 IC Regs came in yesterday and I had a moment to deploy them this evening. Attached to 9v PS's they produce a rock-steady 4.97v and 5.00v with absolutely no heat generation. The combination of only 9v IN and what I assume to be negligible current demand made mounting the 7805's without concern. The motion sensors to which they're attached now operate as designed. I thank you all for enlightening me to the wonder of these simple chips.

I'm still searching for a single Sanyo ur18650w2 cell to attempt repair of the battery pack. Don't want to mix cells as I'm not sure the built-in charge/balance circuit can compensate. Samsung INR18650-15M might be a comprable replacement. Only availability for the Sanyo seems to be from China. :-(
.