Saturday, September 10, 2016

DeWalt 20V Max 3.0Ah Battery Pack Teardown & Analysis

It's been a little bit since I've torn apart a new battery pack!  The last new-to-me pack I pulled apart was a 26v BionX battery (which, I'd add, I rebuilt to nearly twice the stock capacity by filling all the space with cells).  And I've got this cute little DeWalt 20V MAX battery pack (model DCB200, 3.0Ah)  that's just not behaving right.  It would charge, but then only show one LED on the status bar.  I got it for $6 at a pawn shop when I asked for defective batteries.

Well, I've got a dead battery on my bench - that means that it's time to tear it apart!

And you know you want to see what's inside!

Read on for an awful lot of photos inside this solidly built battery pack.

Specs & Warnings

On the back, I've got specs about my DCB200 Battery Pack (Made in Korea).  It's a 20V (Max) pack, 3.0Ah, and claims to be a 60Wh pack.  Type 2, whatever that means.

I'm going to grumble a bit here.  They call the pack a 20V pack, claim it as 3.0Ah, and then say "60Wh."  This is the math, but it's not a 20V pack by anyone else's standards.  This, in normal terms, is a 18.5V pack (5S pack, 3.7V/cell).  To call it a 20V pack requires a nominal voltage of 4V/cell, which is just weird (and not what they have).  Anyway, just my thoughts.  I'm clearly not a tool pack marketing guy.

Amusingly, the warnings here don't say anything about not disassembling the pack, so I'm good!

Except, I apparently did charge it while damaged.  Spoiler alert!

Pack Interface & Voltages

The pack has two significant interfaces: The power interface, and the "How charged is it?" interface.  

The "How charged is it?" interface is simple enough.  Press button, receive bacon LED status lights.  Zero, dead.  Three, fully charged.

The electrical interface is a bit more interesting - and, really, quite a bit of fun!

The leftmost slot is a double connector for the positive terminal.  This is a full height terminal (two contacts).  The rightmost slot is the same thing for the negative terminal.  These always appear to have the full pack voltage on them.

In the middle are some other terminals of interest.  From left to right, top to bottom: TH, ID, C1, C2, C3, C4.

A tiny bit of experimentation with packs demonstrates that C1-C4 are actually the individual cell bank voltages, conveniently brought out for my use!

If you look at the charger side, the pin that makes contact with the "ID" pin is longer by a tiny bit than the others (the top right of the center 6 pins).  This pin seems to be always at low potential, and there's a bit of a voltage floating on the charger pins when there's no battery.  I suspect pulling the ID pin low brings the charger circuitry online, but I don't care to actually mess about with my charger since I use it.

I grabbed some voltages from a good pack of mine (I've got 2 good 20V Max batteries).  Everything is measured relative to B-, with the relative voltages from the previous reading in parentheses.  My pack is fully charged (20.43V, 3 LEDs, solid charger light).

C1: 4.08V (4.08V)
C2: 8.17V (4.09V)
C3: 12.26V (4.09V)
C4: 16.34V (4.08V)
B+: 20.43V (4.09V)

This is a healthy pack.  The per-cell voltages are almost identical (within 0.01v), which means a well balanced pack in good condition.  Conveniently, this also means that the cell group voltages are externally accessible.  This is great news for testing a pack!

Also of note is that the cells are charged to 4.1V instead of the more typical 4.2V.  This is great news for longevity - sitting at 100% state of charge (which, for these cells, is 4.2V) is bad for calendar life.  Charging almost all the way (but not quite to 100%) means they'll last longer on the shelf when charged - and most people store tool batteries charged.

From this "dead" pack right off the charger?  Let's see!

C1: 4.08V (4.08V)
C2: 4.10V (0.02V) (uh oh)
C3: 8.19V (4.09V)
C4: 12.19V (4.00V)
B+: 16.37V (4.18V)

I did double check this, thinking I'd made a mistake in transcribing numbers.  The C1-C2 reading is 22.9mV.  This isn't good.  That's a stone dead cell bank, if it's accurate.  And the rest isn't very pretty either.  You'd expect to see a nicely balanced pack like mine, not a voltage horror show like this.  If the voltages are accurate, there's something seriously wrong with this pack.

Disassembling the Pack

Pulling the pack apart involves removing 4 Torx screws with center pins.  These, obviously, don't work with regular Torx head bits.  I hear a rumor that things like center pins mean they're a "Security Bit" intended to keep people out.

Fortunately, my shop has something called a "100 Piece Security Bit Set."  Any good electronics shop does.  So, bring your "security bits trying to keep me out."  Hint: I brought it.  It's a T10 size Torx security screw, if you're interested.  One of which is very much included in this 100 piece set.  It's not even very creative...

Pack Insides

With the "Security Screws" out, the top comes off cleanly, exposing the pack guts and a really neat little spring for the pack release button.  Sproing!

From this angle, the "double height" B+ and B- terminals are visible, as is the state-of-charge indicator.

Interestingly, it looks like the B+ and B- terminals are a different material than the cell-group contacts and the TH/ID contact.

The B+/B- terminals look like copper to me, with the rest being brass or something along those lines.

That makes sense, though.  The B+/B- terminals are the ones carrying the power, while the rest shouldn't ever be transmitting any significant amount of current.

Looking down from the top, there are wires running around to each side of the pack.  These are almost certainly the voltage sense/balance wires.  It's interesting that they're bare, but it's fine inside a sealed pack.  They're custom bent to fit, and fit in perfectly.

The red and black wire on the left are going to the LED indicator.  Based on the lack of any other wires, it's likely just a basic voltmeter that lights up the LEDs.

One thing of note here: The battery is always connected to the terminals.  There is no way that the BMS can cut off current if the pack voltage is low.  It's up to the tool to determine the cutoff point and refuse to work below that.

This also means that if you stick something in the terminals, you can get power out.  Don't leave it running and drain the pack, but this would be a really easy pack to repurpose should you care to do so.  There's literally nothing to do but jam metal blades into the B- and B+ terminals.

Removing the Battery Pack

There's no real trick to removing the battery pack from the bottom part of the case.  It just pulls out.  However, it's in there very tightly.  Start with one end, pry (carefully - don't short the pack out), and it comes out.

The wire between the battery pack and the output terminals is quite substantial, and is secured with an awful lot of solder.  This is very clearly designed as a high power pack!  Look at the size of the lug going into the positive terminal...

After a bit more prying, the pack is out.  There are 10 cells (2P5S pack), and they're held together with rather substantial interconnect strips.  The pack claims to be 3.0Ah, which means each cell should be 1500mAh.  Are they?  We'll soon find out!

The LED board is secured to the battery pack with some sort of goop.  It's not going anywhere.

Cells and Interconnects

One of the cells is very conveniently oriented so I can read the model number.

The model number is LGDAHB41865.  This works out to be a LG HB4 cell - which is a 1500mA cell rated for a 30A discharge!  The battery chemistry is NMC (Lithium Nickel Manganese Cobalt Oxide or LiNiMnCoO2).  This is one of the newer chemistries, which is nice to see.  They're in every way better than the early lithium ion chemistries.

This is a solid, solid choice for a power tool pack - and not a cheap choice either!  DeWalt didn't cheap out here with last year's discount cells.  People have tested this cell (it's popular for vaping as well), and it's a legitimate 30A cell.

With batteries, one can generally get a lot of energy (watt-hours) or power (watts), but not both.  These are hardcore power cells - a single cell at 30A is putting out over 100W!  The low capacity (1500mAh - below half of what a high energy cell can do) is the price you pay for a cell with that kind of power output.

The observant reader may note that the cell in the middle here, with the model clearly visible, looks a bit off.  And, you'd be right.  This cell is the 20mV group - even after charging.

The plastic wrapper on that cell is incredibly brittle compared to the other cells.  It seems like it shorted out and overheated, or overheated and shorted out.  I have no idea why, but it's quite dead.  Not knowing the pack history, I don't know if it was a cell defect, or if something happened to it.  But these two cells aren't even nailed to their perch.  They are dead as granite.  What likely happened is that one of the cells shorted internally, and the other one dumped its current through as well.  Exciting - and not in the good way.

The interconnects here are thick, massive, and well spot welded.  I'm not sure if this is a machine built pack or if it was built by humans, but I'm leaning towards "humans" - the placement of the "H" interconnects isn't identical between the left cluster and the center cluster, and the rightmost strip is somewhat off center.  I don't see why one would configure a machine to do that.

The shorted cell shows some corrosion and discoloration on the ends (the second pair down, left side and right side).  Perhaps it overheated enough to damage the coating on the case?  I'm not sure.  This is the negative end, and it doesn't look like it's burned through (I scraped the corrosion off and it looked normal), so... still no idea what happened.  Nothing good, that's for sure.  The top left cell also looks a bit off (the discoloration on the positive terminal doesn't exist on any other cells).

Skipping ahead slightly, I measured the thickness of the interconnect strip used.  It's a hair under 0.012" - so 0.30mm, or twice as thick as the common 0.15mm nickel strip used in ebike battery packs.  Again, a beefy, beefy pack.  For what it's worth, I can't spot weld this stuff - too thick.  It takes some serious amps to spot weld 0.30mm strip.

Getting to the BMS

Next, I want to get the BMS off - that lets me get deeper into the pack, and figure out what's going on with the BMS board (not that it's much of a management board).

The solder connections are substantial.  This pack really is designed to carry a lot of power.  Lots of solder, lots of thermal wicking from the nickel strip... I'm going to need a bigger iron than I normally use!

Fortunately, I have such a beast, warming up nicely in this picture.  This is a Portasol 125W butane iron, available for around $70 on eBay.  There are times for a precision temperature controlled iron, and times for a hot running beast of an iron.  With the butane flow turned up, this thing is the second.  Hot, massive, and will desolder almost anything I come across.

Also, a solder sucker for some of these solder globs.  Because why not?

This unit took more heat than is usual to desolder stuff.  Lots of solder, lots of thermal wicking, and lead free solder leads to needing a serious amount of heat.  Light the iron up, let it get good and hot, get some solder on the tip, and go.  There's no way I could have popped the main power wires free with my normal bench iron.

A bit of fiddling, and the BMS is clear!  The wire routing for the balance wires is really well done.  I'm very impressed with this pack so far.

The BMS, free and clear!

BMS Board

I refer to this board as a BMS, but I think it's probably better to call it a balancing board, or a balancer board.  And I'm not entirely sure I understand its point.

The back of the board is entirely passive components.  Some resistors, a few capacitors, and a diode.  Plus some hefty solder joints to hold the prongs in place.

Looking down into the gap between the prongs and the board, there's what appears to be a small IC in there, and a few more passive components.

The bottom of the board shows another small IC down to the left, and what looks an awful lot like a thermistor in the top right.  After removing the conformal coating, I measured 10.68kΩ at 21.6C - so probably a standard 10k/25C thermistor.  For those unfamiliar, a thermistor is just a resistor that changes resistance with temperature.  It's an easy way to monitor pack temperature.  Unfortunately, this one is really only monitoring pack temperature - not cell temperature.

It's hard to see in the photos, but this board has a nice conformal coating on everything.  A conformal coating helps keep water off the board, and generally improves reliability in rough environments - which tool packs will certainly see.  There's basically no excuse to not have one in 2016.

I played with it a bit, and it certainly seems to be a balancer, but I don't understand why it exists.  Whatever functionality it allows for can be done externally on the charger.  The charger has pins for all of the parallel groups, and an external balancing charger is more reusable than a single pack balancer.  On top of it, this BMS can't even protect the pack from an overdischarge or any sort of fault condition.  It might be able to alert the tool to an imbalance condition, but the charger could also refuse to charge the pack in that state.  So I'm a bit lost as to what, exactly, this board does.  Given what it can't do, I'm having a hard time figuring out why DeWalt even put it in.

LED Status Board

The LED status board has a sheet of plastic over the top - great for helping keep the environment out.

Beyond that, the switch, when pressed, applies pack voltage to a system that lights the proper LED based on the voltage.  That's all it is - a basic voltmeter.

So, of course, I set about testing it to find out what the voltages involved are.

Results from the testing:

1 LED: >15.3V (3.06V/cell)
2 LED: >18V (3.6V/cell)
3 LED: >19V (3.8V/cell)

Sane enough, though one LED means you really should be charging it now.  There's no reason to run a power cell below 3V, and 1 LED is almost there.

I did confirm that the BMS will balance cells.  With voltage on the positive and negative leads, the balance leads have proportional voltage across them as one would expect for a battery balancing system.  So, cool!  Weird, because the charger could do it just as easily, but still neat.

Rebuild or Keep Going?

At this point, I have a decision to make.  I can either stop pulling things apart at this point and rebuild the pack, or I can keep going.

I make the decision to rebuild or not based on two major factors: Rarity and cost.


How rare is the pack I'm dealing with?  For some of the packs I've rebuilt, they're rare, and either very difficult or impossible to obtain a replacement for.  If that's the case, rebuilding it makes a ton of sense.

How hard is it to get a DeWalt 20V Max pack?  Well... there are are over 800 listings on eBay - so not rare at all.  You can buy them in nearly endless quantities at your local hardware store.


Cost is my other main consideration.  If I can rebuild the pack radically cheaper than I can buy a new one, I'm likely to rebuild it.  If I can't - then there's no real point in the labor to rebuild it.

For me to buy 10 of these cells, I'm spending basically as much as a new pack.  BatteryBro has LG HB4s for $4.57/ea - in quantities of 50.  eBay isn't any better, at around $5/cell.

I can buy the 3.0Ah battery packs for $40-$50 shipped, all day long.  You're better off buying these packs to extract the HB4s than you are rebuilding them!

The Decision

This pack is not rare.  And to obtain similar 30A rated batteries, I'll spend more on cells in small quantities as I will on a brand new pack.  Plus, I don't have a spot welder that can deal with the thick interconnects.  DeWalt sells these things in insane quantities, and it shows in the pricing.

So, there's no reason to rebuild it.  I keep going!

Oh, and if you are trying to rebuild one?  Make sure you put a high amperage cell in it.  Don't go "Hey, I can double the capacity with some 3200mAh cells!" - they won't like being subjected to the power demands of a tool pack.  And those 9000mAh cells on eBay?  Lies.  Total, complete, bald faced lies.  They're recycled junk wrapped in a meaningless number.

Pulling the BMS Apart

This means pulling the BMS apart to see what's on it - as much as I can.

The balance leads are easy enough.  They pull off with a bit of fiddling.

They're just standard header pin sizes, and they seat in a regular connector.  Slick!

That nice green thermistor is right there too.

The leads going to the LED panel are nicely secured - they're much more solidly connected than what I'm used to seeing in some of my teardowns.  Excellent!

Removing the connectors is a lot harder than I expected.  I tried to desolder them, and concluded that they're just too large and radiate too much heat.  I gave up after a minute or so of trying to free them with my big iron, and went to Plan B: Pliers and wire cutters.  Which work, though that approach pulls the pads up with the main contacts.  The heat bubbling in the lower right corner is me.

There's nothing much on the front either.  Only one IC (and I can't read the model through the conformal coating, so I have no idea what it does).

Is Your Pack Bad?

If you're here because you've got one that's acting up and you want to see if it's bad or not, I suggest doing what I did way up towards the top and measuring the voltages across the pins.  You'll need a thin wire to probe the contacts - though a pair of paperclips will work if you're careful.  Put the black lead of your voltmeter on the B- pin, and probe C1-C4 and B+ - you should see a steady increase in voltage as you go from terminal to terminal.  If one of them is significantly less (like my 0.02V jump), you've got a dead cell bank.

It's worth pointing out that this pack will happily spotweld any short circuit it sees.  So don't touch your paperclips to each other, or you will likely find them glowing within, oh, a second.  Probably less.

Keeping Your Packs Alive

Lithium batteries are easy enough to keep alive, but there are a few things to know if you want them to last as long as possible.

Don't charge them when they're below freezing.  You can use them in the winter - that's fine.  They'll be a bit sluggish on output, but you won't damage them.  Charging them when cold, especially with the fast chargers that are common for tool batteries, will permanently reduce their capacity.  Let them warm up before charging them.  You also don't want them to bake.  Sitting in the trunk of a black car in Phoenix in the summer will also kill them early.

Don't let them sit empty.  That's a great way to kill cells.  Always charge them after use.

And, if you're storing them for a long period of time, don't let them sit completely full.  It's less of a problem with this pack (as it never charges to "full" by the cell chemistry), but charge them, run a tool for a few minutes, then store them if it's going to be a few months.

Otherwise, just use them.  Lithium cells don't have a memory effect, so charge them when you get a chance, and have fun!

Final Thoughts

This is a solidly built pack that will do its job nicely.

The cells are solid 30A performers, which means this is a 60A pack (two cells in parallel add amperages).  The 0.3mm nickel strip will handle plenty of amps, especially over the short distances involved here.  The main wiring will handle lots of power as well.  This is a very nicely built pack, and it's entirely suited to being a power source for small tools.

Fully charged, at 20+V, that means it's an honest 1200W power source (for, oh, three minutes until drained).

The only real complaints I have are that the BMS is really just a balancing board, and the thermistor is more or less useless.  With the battery leads connected straight to the output terminals, there's no way for the pack to cut off output - it's up to the tool to not kill the battery pack.  This is probably fine, since it's unlikely that most people will use the pack for other purposes.  But, if you do, don't be stupid.

The thermistor hanging out on the BMS board also means that it's more or less useless for tracking issues.  All it will show you is a lagging reading of the temperature of the cells near it.  It'll provide a general idea, but nothing nearly so useful as a set of thermistors on the cells to find if one is running hot and suggest the user back off.

But from an end user perspective, this is an incredibly nice tool pack.  It's solidly built, is using great cells, and is capable of handling a ton of power.  Use it as intended, don't run it below 1 LED if you can help it, and it should keep you going for a long time!


  1. Excellent report! The information you provided was very useful... I'm pondering using banks of these for a homebrew mutant e-bike and your info helped me along quite a bit.

    1. That was certainly one of the things I was thinking about with the teardown. I'd love to see what you end up with!

  2. Hi Russell - you may be able to help me! I've built up an e-bike for my 9yo daughter - using XXS Yeti frame from 2007 and a 27v Bionx kit (Canbus) that used to power my wife's bike: I've removed the battery pack and control board from the casing and have been successfully been using (or rather my daughter) the set up; however the battery mounting is pretty crude.. and it's too big to fit inside the main frame triangle - I've now purchased a Milwaukee 28v drill/charger and battery set; my aim is to use the Milwaukee power packs for the bionx system: they are smaller, fit neatly inside the frame and assuming I can make this work, can be swapped out simply using the original tool fittings' quick release mount..(each pack is 3Ah, and extra packs are easily carried) - my question is whether I can tie up all the wiring from the original Bionx battery pack (given there are multiple wires heading to the circuit board - including I assume temp monitoring...) and substitute wiring from the tool mount - hope this makes sense... Is this feasible? thanks Carl - -great site BTW

    1. Carl -

      It will probably work, assuming the Milwaukee pack is a 7S lithium pack. You may run into early cutoff issues if the voltages don't match, but the early BionX BMSs are pretty tolerant.

      Depending on the space available, I could also build you a custom 7S pack of the same chemistry BionX uses.

  3. This comment has been removed by the author.

  4. I Just picked up a drill (only the drill) with the intention of running it off of a rc lipo pack but I cant get the drill to chooch with just a 10k thermistor on the th pin as well as power to the two 20v in lines. Any advice?

    1. Try hooking the 10K thermistor between the TH pin, and the Positive 20v.
      Most companies hooks it to Negative

    2. Eric - if you wish to emulate the pack, assuming there's no special signaling going on, try running a 10k thermistor between TH and B+, and an 800Ω resistor between ID and B-. That should replicate the signaling for a Type 2 pack.

    3. Hello Russell,
      I tried this (putting 800Ω resistor) between ID and B-, no luck. Any thoughts ?

    4. No idea at this point, sorry. Let me know if you figure it out! I just run mine off the legit DeWalt packs, so I haven't been too interested in poking around at alternatives.

    5. No worries mate, I have looking around for weeks now with no luck on how to crack the ID pin......, I will definitely share the info when I find something.

    6. the c3 pin on the drill needs to see a voltage of 10v approx in order to run.

  5. Excellent write up. The only thing I'd add is that individual cell group voltages are obtainable by using the terminals in series:

    - Group #1: B+ (pos) to C4 (neg; DW numbered them back asswards)
    - Group #2: C4 (pos) to C3 (neg)
    - Group #3: C3 (pos) to C2 (neg)
    - Group #4: C2 (pos) to C1 (neg)
    - Group #5: C1 (pos) to B- (neg)

    I'm sure you (Syonyk) get it, but for anyone who doesn't, polarity is relative to where you take the measurement from (e.g. C4 is negative relative to B+, but positive relative to C3). Path of least resistance, etc. Cheers.

    1. So my question is.. when building a charger.. do I connect any voltages to any other pons besides B+and B-?

    2. Very late reply but, if it is a CC/CV charger, no. The little "BMS" should do the balance charging for you!!! Please monitor the first few cycles to make sure it is balancing sufficiently! And regularly check that the pack is not drifting out of balance. But these packs seem well suited to repurposing, weird honestly.

  6. Thanks for this outstanding informative article, as the PCB from my battery pack is coated with white goop I was really glad to find these details about the PCB.

    I am about to build a 2.0 Slim pack from scratch and just have some 4.0Ah Packs as a reference. Now i need to find out more about de ID Pin, i guess there is a resistor between ID and B-. I measured it on my 4.0 Packs and got a value of 360Ohm. It would be from great help to find out how they vary between the different 18V models. If you still own the PCB I would be really thankful if time allows you a quick measurement of you ID to B- resistance on the 3.0 battery.

    Anyway thanks a lot for all you efforts,

    1. FWIW, my two 4Ah packs measure 318 and 317 ohms between ID and B-. They're 2014 date coded (2014-18-47 4-3), made in Malaysia and have green Samsung 20R cells.

    2. Oh, I just found out i hat a brain fart writing the first comment. I meant 316 Ohms. I further got to measure 2 14,4V DeWalt Packs with 1,5Ah capacity. They also measured 316 Ohms. Seems like they specify just the course generation or charging current handling ability with this ID Pin, I guess i try 316 Ohms then for the 2.0

      Thanks for taking the time,

    3. No problem, buddy. If you can get back with what cells are in your pack, the date code and country of manufacture, that'd be great. Do yours also have the 20R cells? Cheers.

    4. My DeWalt 18V 4.0 packs contain Sanyo cells, i suppose UR18650RX but i can't confirm that - they are sitting snug inside and i don't want to destroy the battery. Datecode is 2014 05-47 2-2 made in Japan.

      In the selfmade 2.0 I'll be using Samsung SDI 20R (if it works out :D).

    5. It seems that the smaller packs are around 300-350Ω between ID and B-, and the higher capacity packs are 800Ω. This appears to correspond to the "Type 1/Type 2" pack description.

    6. Thanks for the additional information and again for the detailed article. I eventually did build the 2.0 from ground up with no balancing (charger does take care, I measured it before making the pack ;) ) and 316 Ohm on the ID connector. Works perfectly fine, charger accepts the battery and takes care of the balancing.

  7. Thank you for this uninformative post. I have a bunch of 20v max dewalt tools, but occasionally I dont need the portability and would like to have more run time. I have been planning on 3d printing a connector and making a board that can provide the 20V. Amperage was one thing I was not sure about, and apparantly my board will need to provide 30A max, which is not a trivial amount of current. Im sure most of my tools dont draw that all the time, but the peak of 30A still needs to be provided by the board I make if thats what those cells can provide. Your stuff on the ID pin are also going to be useful.

    1. Informative** Sorry, I didn't try to be sarcastic or anything. I really enjoyed the post. Autocorrect changed that and I dont see how to edit on my phone...

    2. Check out my 6.0Ah teardown and 1.3Ah teardowns for more info on ID and TH.

    3. I know this post is over a year after your original, Salvador Razo. However, I think that the maximum current needed from an AC/DC converter to power the Dewalt 20V max tools in a similar way to the battery packs would be 60 Amps or 1200 W. My reasoning for this is that there are 10 cells, each cell is 4.1 V nominally charged. Hence, 5 cells are needed to reach nominal voltage. So, my thinking is that there are 5 groups of cells, with each group having 2 batteries in parallel (2 times the 30 A current rating). The 5 groups are then in series to achieve the desired 20+ volts. Hence, the specs to replicate the battery performance would be 1200 W, 20 V @ 60 Amps. My thoughts are that this would be more than a simple board project. In fact, I did something similar for a Power8Workshop, and it worked reasonably well. It involved a large transformer (step down AC voltage to ~20 V DC), high power rectifier diode module (rectify the AC voltage to DC voltage), and a rather large and high power rated capacitor (to help handle voltage spikes common to electrical motors). I am not sure how much these components would cost, as I salvaged all of them. it fit into a standard computer power supply case, but was certainly over just a few pounds in weight. Hence, I don't think its a "board" but an assembly. Someone seems to do this for Dewalt and Makita power tools ( Apparently, its not cheap ($250), which may be reasonable considering the cost of components and heavy gauge electrical wire needed to handle 1200 W DC. I was just looking to do the same thing, which is why I posted my thoughts on the matter.

  8. Very nice job, much appreciated

  9. Such a methodical teardown. Great job! The LED trigger voltages are helpful, since I'd like to rig these battery packs to provide power to some older 18V power tools that I only use sparingly. As long as I keep an eye on the indicator LEDs, I shouldn't ever run the cells into that undesirable undervoltage scenario that causes permanent damage to the cells. I think. Lol. Thanks again!

    1. I'd generally agree - if you keep them at once LED or above, you'll be fine!

    2. I know this is an old post, but Dewalt makes an adaptor for using 20v batteries on 18v tools

  10. Are you aware of DeWalt's new 20/40/60/120(?)V FLEXVOLT line, where the batteries change configuration based on what they're plugged into? That's a teardown I'd like to see, but I don' t have the $$$ to subsidize a purchase. They'll supposedly work on the 20V tools like you already have.

    I sure would like to see a teardown of their adapter (DCA1820) to use the current 20V MAX packs with my 18V tools. They seem overpriced for what they might provide.

    On my 18V stuff, only the charger connects to the thermistor in the batteries. But at least in the XRP ones I've torn apart, it's taped to one of the cells instead of just in a random spot inside the pack case.

    PS - keep up the good work - you're helping me decide what direction to go as my 18V NiCd DeWalt tool packs die out. Thanks!

    1. 1. Flexvolt won't do 40V, only 20V and 60V. Some Flexvolt-specific tools put two packs in series to yield 120V.
      2. The DCA1820 *is* overpriced, but still cheaper than buying new tools. Unless you have a 3D printer or machine shop, making your own may not be more cost-effective.
      3. A DCA1820 teardown will definitely involve some tearing, as the guts are potted in epoxy.
      4. At the risk of stating the obvious, the adapter has no thermistor terminal, just + and -.

    2. I'm aware of the FlexVolt line, but haven't had anyone donate some batteries to pull apart yet (and I have enough 20V Max batteries to last me for many years at this point).

      If you'd like me to rip some of those things apart, the contact form is over there and I accept donations of battery packs! :)

    3. I bricked at Flexvolt battery by trying to see how it works. Here's what I know...

      There's a bar on the battery that the Flexvolt tools depress that is part (not all) of what is needed to put it into series (60V).

      Flexvolt tools short C1 and C3 (no resistance using an ohm meter)

      When you depress the bar AND short C1 and C3 you get 60 volts vs. 20V. But, you also brick the battery so it won't work in any tool or the charger.

      The Flexvolt tools use the C4 connection for something as well as the thermistor connection. I didn't have either of those connected when I bricked my battery. It would still switch between 20 and 60V output but it would not work in any of the tools.

    4. Also, the Flexvolt batteries charge in parallel mode (20V)

    5. I bricked at Flexvolt battery by trying to see how it works.

      You are a motherfuckin' baller. My hat's off to you, sir.

    6. Per the maker of this 3D-printed Flexvolt adapter, the battery is bricked only if you connect C1 and C3 *before* you engage the series/parallel switch. It sounds like he might've bricked one (or more) in trying to figure this out. I can't confirm the findings yet, but it seems promising!

  11. Thank you so much for this write up! I was able to fix my brand new battery that has never worked. I know I should have taken it back but it was a gift. My battery was reading full charge but it would not work in the tool. I found out that I had no voltage in the c3 pin. It was a bad connection at the plug. I soldered a wire from the bank to the board. Works like a charm now. Thank you again!

    1. Awesome! I'm really excited to hear that, because that's the sort of thing I've been hoping people will be able to do with the information I provide. Repair things and keep them out of the landfills. There's no way that the store would have tried to repair it - they would have just disposed of an otherwise perfectly functional pack because it needed a few minutes with a soldering iron.

  12. Thanks for such an interesting article.
    Do you know how to measure the output voltage of the charger? I recently replaced a capacitor with one that supports 220v and I would like to verify that the output voltage on each pin is as expected. But since it does not work when the battery is not there, I do not know how to measure it.

  13. Thank you for this post but i need your help. I recently got a DCB205 20v 5ah and have issues with the fuel gauge. Upon initial charging, red light on charger does not stop blinking (fuel gauge 1 bar). Removed battery and checked voltage and it was full capacity (fuel gauge 1 bar). Used it on a drill and it works. What could be the issue? Is there a way to check fuel gauge if it is working properly? How to do this? Thanks!

  14. Merci trop frt
    Grace à vous j'ai sauvé une machine dont la batterie était morte et pour laquelle je n'avais pas de chargeur !!!!!!
    Saloperie de TH et C3

    1. Rough translation for Russell and Friends:

      "Thanks so much
      Thanks to you, I saved a tool with a dead battery that I didn't have a charger for!
      TH and C3 bullshit

  15. I just salvaged two packs thanks to this post. They have been in a discharged state for a long time- one since late 2014. The Dewalt charger wouldn't do anything so I put a 12V 3A charger on the +/- for about 15 minutes then transferred the packs over to the Dewalt charger and it worked. I will try to run them down once to see how much life they have and report back...

  16. Thanks for the teardown. I referenced it to possibly salvage two old packs that would not take a charge from the Dewalt charger. By connecting a 12v 3A charger for about 15 minutes to get some charge in them, both packs were charged by the Dewalt charger. Now to find out how long they will run... One has been discharged since 2014.

  17. I'm considering using 3 of these packs in series to power a motor through a DC-DC converter. Is there anything in the BMS or elsewhere that would prevent wiring them in series to get 60V? We're hoping to discharge them as fully as possible in 25 minutes (>6A discharge rate) - it's a 25 minute race with 180WH maximum allowed battery capacity... Are these battery packs a good option or are you aware of better choices? Thanks!

    1. They'll work fine in that configuration. There's no BMS to speak of.

      That said, the 60Wh rating on these is a bit optimistic. They're really closer to 55Wh packs. You might be able to do a bit better if you find something else.

  18. Hi Russell,
    Awesome writeup!
    Were you ever able to read what IC was used on the BMS / interface board (the one under conformal coating)?
    Thanks a bunch!

  19. Strange about your state of charge indicator, all my packs will read about 17 volts at 1 bar, 18 at 2 bars, and 19 volts at 3 bars. All 3 of mine seem to drop down to 1 bar pretty quick and always have, and still run for hours like that before dying. I've never tested the point at which it actually goes to 0 bars since I don't let them get to that point if I can help it, it's definitely close to 15 volts at that point because if it works at all it will just be for a moment. Maybe they changed the cutoff points to make people charge them sooner instead of running them completely dead?? (they might sell more packs, who knows) Though I've got one pack that does act up, occasionally it will only light 2 of 3 bars when full, and if you press harder it will light up all 3, and has since stopped altogether (not sure if it was related, but same pack seemed to show a dead cell at the same time, and upon checking it again, seems to fixed itself (maybe the internal balancer takes a while if there is a big difference between cells??) reading 3.9 volts on every cell though the charge indicator still doesn't work at all, guess I won't take it apart just yet... I'm hoping it was due to being loose on the charger and only charging some of the cells, sometimes the pack doesn't get seated all the way in to the charger and won't charge it all the way, seems to be a common problem, and possibly why they don't rely on the charger to get the balancing right?

  20. Hi Russell, the pcb is coated with peters conformal coating material.

  21. Great post, very informative. I have a question that hopefully Russell or one of the other experts on this blog can help me out with.

    I have a set of Yardwork tools (chainsaw, pole-saw, hedge trimmer, blower and weed eater) that use a proprietary 20v 9ah battery. The battery died and I can't find a replacement anywhere. I tried dissembling the battery pack to see if I could replace individual cells but unfortunately this thing has one large monolithic battery in it. The only number I could find on the internal battery didn't show up in any searches so I think I'm going to have to be creative if I want to keep using these tools.

    If I built a custom battery mount with the old Yardworks battery casing and an old 20V Dewalt tool (which I do have on hand) is there any reason why I couldn't tie the +/- 20V pins from the Dewalt tools battery mount to the +/- 20V Yardworks battery casing pins and supply the juice to the tools with my Dewalt batteries?

    I guess what I'm not 100% sure of is if there's any signaling passed from Dewalt tools to Dewalt battery packs for proper operation. From some of the other posts I've read it seems like this should work, but I'd really hate to fry a battery to find out the hard way it doesn't.

    Thanks in advance for any advice.

    1. There's no signaling to get the pack to power on. The tools may be picky about other packs, but you can absolutely use a DeWalt 20V pack on another device by jamming two spades in the positive and negative terminals. Low voltage cutoff is on you, of course.

      You can see this sort of setup in my egg frying post:

    2. Thanks for the info Russell. Love the egg frying post.

      I was thinking I might try and rig up a low voltage cutoff circuit, but for the amount I use the garden tools, it's probably not worth the effort. I'll just make a conscious effort to keep and eye on the LED status indicators more frequently.

      It's a shame they didn't add this to the BMS in the first place.

  22. I am looking to use my 20v/60v dewalt packs to power 12V LEDs by using a junk charger as a base for the battery. Do I need to worry about the ID pin doing something to the battery seeing as how it's on a charger base and not in a tool?

    1. I'm not sure how the 20V/60V packs work - I believe they default to the 20V mode, so you should be able to draw off them, but they rather substantially exceed the voltage on a 12V LED strip.

  23. Hi Russell.
    I found your taredown very informative.
    I'm currently trying to put a gps tracker in one if my battery packs.This means removing a couple of cells to make room.
    The gps tracker I purchased runs off 12v. would it be possible to still use the pack at 18v and somehow run the tracker at 12v ?
    I would very much appreciate your help on this.

    1. You'd have to strip out a full row of cells to keep the configuration sane. And then probably would need a buck converter. I'm not sure why you'd do this, really...

    2. If you're committed to this idea, which Russell is being charitable in not outright calling stupid, you might consider swapping the guts from a smaller pack into one of your existing "premium" cases (so, 3, 4 or 5 Ah). This assumes you require the battery to appear untouched. Come to think of it, you'd have a bit more room if you put 5-cell guts into one of those newfangled 6 Ah packs with 10 fatty 21700 cells like Russell played with a while back.

      If you don't give a shit what it looks like, you could probably pay someone with a 3D printer to make you an extended battery case that will hold the full complement of ten cells AND your GPS fuckery. Of course, I presume this won't do for your spying if whoever you're spying on uses the tools. I assume you are because the unit you bought is obviously not giving you real-time updates via cell data, which means you're not doing this to prevent theft of your tools.

      Good luck...I guess. Ouf.

  24. Anyone out there have any idea what cells are used in the newer 5Ah packs?

    1. Nope. Want to ship me one and I can find out?

  25. The charger doesn't allow the pack to be charge if it is to cold, the thermistor could use for this purpose only !?

  26. Excellent report!

    Quick Question:

    Are over discharge protection built-in to the battery itself? Or is that being monitored by the tools themselves?

    1. The over-discharge protection is built into the tool. There is no way for the battery to shut off discharge current.

  27. Not sure my post got published?

    My question about if over discharge protection is built-in to the battery itself or is being monitored by the tools themselves?

    1. Sorry, I have comment moderation on due to spam issues (lots of people trying to sell Best Solar in Kondwaku or somewhere totally irrelevant to my blog). I have to approve non-spam comments on older posts.

  28. I thought the blog post was excellent, but the comments section also have alot of great info.

    I found this blog while trying to research a way to make my 20v dewalt tool run off 110v ac. The tool itself has 4 pins which make contact w the battery B+/- and two inner pins TH and C3.

    What i was hoping to do is seat a battery and charger together as my power source. and just jump those same the 4 pins over to the tool. Any reason that wont work? Like the battery knows its on a charger and wont output?

    I did something kinda similar to this with my dewalt corded/20v vacuum. I modified vac to charge a battery while plugged into 110v but still use the battery as power source when unplugged. Only thing i had to do was dremel out a way to swap the battery terminal. The charger board/guts fit fine.

    1. It should work fine - C3 won't be an issue, but you might need to interpose TH somewhat. I don't know what the tool/charger side of that looks like to know if it would be dragged out of range by two devices trying to read it.

      There's no way for the battery to shut off output - the output terminals are wired quite firmly to the battery terminals, with nothing in the way that could terminate current.

  29. I appreciate the work done here and the information provided. This gives me the ability to use these Dewalt batteries for other purposes around my home. These batteries will be useful with the inexpensive DC-DC step-up and step-down boards available from China to run other DC devices.

    For those of you on a budget, you can sometimes find cheap cordless tools at second-hand stores. People donate these because replacement batteries are too expensive, so they just buy a new tool set. Years ago my wife bought a nice Ridgid 3/8" drill and 1/4" hex impact driver set with charger, bag, and three batteries that didn't work, all for $10. I took a battery apart and used a $22 switching power supply (24 V at 15 A) to power the impact driver (I adjusted the supply to 18.5 V). Though it's a now corded driver, I don't mind because I keep it on my shop bench (it also runs the drill). It's a nice way to continue to use the drill and driver, which are well-made and didn't have to get tossed out.

  30. my 20V battery is out of balance. the 3 lower cells are 3.6ish when the upper 2 reach 4.2 and it quits charging. I dont think think the DCB115 does balance charging. even when I leave it on over night, the voltages never equalize.
    There is one more pin called ID. The little chip could be an eeprom. or it could be a voltage monitoring chip that works when the battery is in a tool.
    Either way, by batteries are only reaching 2 bars on when the test button is pressed... my 5A packs are missing a lot of watts....

    1. Try using a load resistor of some variety and bringing the upper cells back down to 3.7V or so and see if it improves.

      But if the pack is out of balance and you don't know why, it's likely some weak cells.

  31. Russell,
    I travel a lot and need to take DeWalt 20 v batteries with me to power such tools as a cordless drill, battery powered light, etc. I also have to carry a laptop and have an extended battery, but it is finicky. I noticed that my wall adapter shows a 19.5V 3.33Ah output for 65Wh. I was thinking about ripping the bottom off an old Dewalt Tool and Frankenstining it to a laptop adapter cord. I know I can bring the voltage down to 19.5 fairly easily. What I don't know is if the Ah can be pushed up a bit to get to 3.33Ah and 65Wh.

    1. Your laptop charger isn't rated in amp-hours (Ah) - it's rated in amps (A).

      The DeWalt packs will supply 3A easily, but the packs will be below 19.5V when drained. That's 3.9V/cell, which is still nearly fully charged. You'd want a buck/boost converter with a configurable low voltage cutoff, and then it will probably work fine, unless the laptop and charger have some sort of communications among themselves.

  32. R1, R3, R4, blow up, jump then as a fuse

  33. My DeWALT DCB181 18V 1.5Ah 27Wh has 9k68 on TH--B+ and 316R on ID--B-

  34. Hey, I've got an old battery and want to use it with a 20V DC power supply. After I take the batteries out of the pack, will it be possible to fool the circuits into thinking that there is a battery present so that the tool works?

    I'm guessing a few resistors to get the required 4, 8, 12, and 16 volts for the C1 C2 C3 and C4 terminals?

  35. Hey, I'm planning on using an old battery case to make an AC adapter through a transformer.

    I'm confident I've got enough juice, but unsure what will happen when I disconnect the batteries.

    I'm guessing I need to do something like set up resistors to whisper 4, 8, 12 and 16 volts into C1-4's ears. Is that right or is there more I will need?

    Best Regards,


    1. Sorry, I don't understand what you're trying to do enough to be able to offer advice.

    2. Hi, Dwayne. If you're looking to power a tool (or whatever) with an AC adapter, you need not fuck with the C1-C4 terminals at all on the tool end, which are there for the charger to balance the cell groups. If it won't work with just B+ and B-, try emulating the "TH" and "ID" terminals; "TH" is a standard NTC 10k-at-25-degrees-Celsius thermistor, and "ID" is a resistor whose value corresponds to DeWalt's specified charge rate for the pack (e.g. 1 kOhm for the 21700-celled DCB230, ~800 Ohm for DCB204, ~316 Ohms for the wee DCB203; IOW, more resistance means a faster charge rate up to the charger's limit). The newer DW brusless motored tools *may* need one or both of those terminals to work, but brushed ones ("20V Max" or any older ni-cad era tools) absolutely don't; proof of the last one is that DeWalt's so-called 18V-to-20V adapter (which in no way changes the voltage, of course) only spits out B+ and B-, no other terminals.

      Hope that helps, comrade. Please feel free to reply with your results! Cheers.

  36. Thanks, this is really useful.

    I tried wiring the AC direct to the board (with the balance board still in place). B+ and B- reporting 19.8 volts

    Using a brushless driver, it would connect so that the light on the driver would go on, but not operate the drill.

    I suppose its possible I burnt out one of the components or the board when soldering the main terminals with new wires, but I can't really see the heat transfer being high enough.

    I'll check the components you identified, try adjusting the voltage and get back to you! I also have one cordless that I can try.

    Thanks again!

  37. Thanks, I sourced a spare (empty) battery pack and started again. ( My first battery pack had no voltage indicator, whereas this one has it built-in. It's a handy addition! No second guessing whether power is being delivered to the pack.

    Same result - ie, enough to power the lights on Brushless driver and Brush Drill, but not enough to power it.

    On looking at the terminals in the tools, the all have B+, B-, TH and C3 terminals. I should have realized this earlier. (Flexvolt uses way more)

    Tim Medders noted above that the packs wont work if C3 is disconnected. I experimented slightly before removing the batteries and note that cutting the connection to C4 didn't stop anything, and I assumed it therefore wouldn't matter to cut the others. Shoulda checked the tools! Still a strange decision by DeWalt to check for dead cells 1-3?

    I'm now surmising that I need to give the correct voltage to the C3 terminal. Looks like that will be at least 11.1 (3.7v per cell cut off) and up to 12.6. Anyone have any idea what resistor I should use? The current pulled through must be minimal, but not sure how to calculate. Alternatively I guess I just drop a few bucks on a step down converter or regulator - your thoughts appreciated!

    I should also be able to use the same solution to fix my first battery pack, so I'll have two - luckily I included two outputs on my AC to DC transformer!


    1. They might be checking for weird balance issues.

      A note: TH is a 10k NTC thermistor going to the /positlve/ pack terminal - it's weird.

      Try it with TH connected, then see if you need C3 as well. They can't be pulling much power from it or it would unbalance the pack, so I'd just try it with some 10kΩ class resistors first, then modify as needed once you get it working.

  38. Thanks Russell, after thinking about it a bit more, and quite a bit of experimenting, I got it working with a buck an adjustable buck transformer feeding into the C3 probe.

    Adjusted the buck to about 12V (I think 11.1 is the minimum, and 12.8 the maximum) and it is working really well. I haven't tested under a huge load yet, but it is powering the 20V Max Brushless tools and Brushed. It needs the TH connected also.

    A friend has already asked me to build one for him also and I'll do a build video and post it here!

    I didn't have many of the parts, and tended to go bigger than what might be necessary, but total cost about $120 and now have mains powered DeWalt tools (I have 7 battery tools).


    1. Awesome! Glad you were able to work it out! Any chance you want to send me some details of what you've done, and/or photos? I'd be happy to do a writeup!

  39. The small chip on the board in the battery is for OVP only, and only if the charger is monitoring the ID pin to detect it. It's a BQ771803 or possibly CW1051 depending on the OVP style board in the pack. I've just started poking around in these, but didn't see this info here.

    Some mostly complete schematics of the OVP and LED boards I've messed with so far for those interested. Working off and on for one for the DCB118 charger now.

    Vertical style OVP board, 4 layer PCB, (uses the BQ771803) -

    Flat style OVP board (seems to use a CW1051) -

    LED board -

    1. Hey, awesome! Thanks for sharing those details! Certainly more detail than I've put into my teardowns...

    2. Welcome. It's mainly for S&G, but figured may as well post it somewhere it could be useful. These were simple, now the 18v Ridgid packs, those things are a mess by comparison.

    3. My schematic so far of the DCB118 20v Max Charger, for any interested or curious.

  40. I don't own a 118, but I appreciate the work that went into the schematic. I downloaded it as a work of art. Thank you.

    1. Welcome. The thing is a little more 'busy' than it really needs to be, but that's because I left all of the jumper wires and jumper Resistors in it so I could redo the board virtually in the PCB software that I use here. That's how I make sure I didn't miss an connections and the schematic is as accurate as I can get it, by actually remaking the board based off the original.

      Here's some board scans and screenshots of a bit of that process, for any interested.

    2. RDC, you're insane, you know that. Ping me via the contact form if you want to chat - I enjoy getting to know insane people better, especially the kind who are good with PCB layout software!

  41. Welcome, that wasn't really too much of a chore though being a single sided board. The "BMS" in the DCB606 FlexVolt was a little more fun to do, as it was encased in a couple different materials to get at it first, and then being a 4 layer board.

    Not the greatest screenshots as the detail gets lost, but you can see the mess a schematic made from a board usually is, then after it's cleaned up and the PCB 'remade' to check things.

    Sent you a message also.

  42. I can’t believe I found you guys. Just today I wired a AC/DC 20 amp transformer to my 20v Dewalt driver but all I use was B+/B- terminals.
    The LED light would come on momentarily but no sign of power to motor. I presume I need to hook resistors as stated by Dwayn Baraka to the C3 and TH terminals? Any more specifics would be greatly appreciated. I am running a 40’ 12 ga extension cord from the transformer and hardwiring to battery shell. Can anyone foresee any other problems I might encounter? Thanks everyone in advance. I will post pictures of transformer, storage box and wiring if I get it working correctly.

    1. Sounds like that should work!

      This is definitely a weird little backwaters of the internet. I'm quite surprised by just how much work some other people have put into reverse engineering this sort of pack...

    2. I home there's quite a bit more than just a transformer in that box?

      It depends on the tool. I've only had the DCD780 apart here, and some of them use the TH pin while others don't have it, but it needs to have voltage on C3 and TH to work if it has both of those pins in the tool.

      For the TH pin, put a 12k Resistor from B+ to TH and you're done there. A 10k will also work, but may throttle the tool a bit as that would be the same as the pack being installed and a bit warm. If you don't have a 12k, a 10k and then two 1k in series is the same thing, 10 + 1 + 1 = 12.

      For that C3 pin, you need 3/5 of the pack voltage there, around 12v, which can also be done with a couple of Resistors to make a voltage divider off of the B+ and B- lines to get that 12v.

      There are plenty of Voltage Divider Calculators online to work that out, just make sure you get the divider ends connected to the B+ and B- correctly and then the middle of it will go to C3, and do check it with a meter, couple times, before plugging it into the tool.

      Plug in any 3 values here to find out the last one you need -

      Source Voltage is the battery pack voltage, ~20v
      Output Voltage is the 12v you want.
      R1 = for example sake here, say you have a 10k Resistor.

      Plug all of those in and hit Calculate, and you'll get 15k for the R2 value.

      C3 is not a current source, at least not in the DCD780 anyway, so you should be able to get away with 10k/15K there, or 1k/1.5k, anything that gives you that 3/5 value on C3.

      Likewise there if you don't have any 15k and only have 10k and 1k you can do a 10k + five 1k in series to get the 15k, or use three 10k to make the 15k up, series/parallel, oh what fun. Two 10k in parallel make 5k, then you do the third 10k + those two 10k in parallel that are 5k and there you are at 15k again. ;)

    3. "C3 is not a current source" should be, C3 doesn't seem to need any real current for that pin.

  43. I can’t believe I found this blog and you guys. I hooked up a 110v/20v 20 amp converter to my 20v Dewalt driver using a 40’ 12 ga extension cord. I thought I just needed power to the B+/B- terminals. I see Dwayne Baraka and Russell Graves mentions I need some voltage to C3 and TH. That’s about all I know LOL. Any details would be greatly appreciated.

  44. Rob - I abandoned trying to find the right resistor combination, after having the same problem as you. It might be as simple as that (although I doubt it), but it was easier to order a battery 'blank' with related PCBs and components whispering "I'm a DeWalt battery" to the tools and rewire that with a buck transformer to make the tools work.

    Battery pack: $16

    The PCBs and components in the blank battery pack, along with the buck transformer allow you to send the relevant
    voltages to the terminals.

    I plan to make a video about how to do what I have done (transformer with two leads that will power two tools simultaneously) over the holiday break. I'll post it when I have it done, and the owner of this blog has graciously undertaken to post a new thread about it.

    Sorry I am not more immediately giving you a fix, but I'd rather be comprehensive than pithy!


  45. Oh yeah, and I think I initally used 14ga, but 12ga should be enough and is what I will use for the video...

  46. Thank you for all the work you put into sharing this info. I'm building a portable kit for ice fishing using a Garmin Echomap + 9sv and a Panoptics livesope transducer (live sonar). Hoping to power it using Dewalt 20vlt / 5 Amp hour battery platform since that is what I use to run the drill for the ice auger already. I have installed a Drok step down module to bring my voltage down to 17.5 vlts as the maximum range on the head unit is 18vlts. My question for you is at what low voltage should I swap out batteries as to not permanently damage them? I can configure the Garmin head unit to warn me and shut down at specified voltages. What would you recommend?

    1. The thing about setting the voltage cutoff in the Garmin is that your buck converter's output voltage is, I'm assuming, constant at 17.5V (a synonym for buck converter is "voltage regulator"). IOW, the Garmin will never see a voltage decrease, so can't shut down on that basis. You'll need another approach.

      If you're able to have a voltmeter (either a digital multimeter or hobby-type mini voltmeter+LCD) on your battery while it's being drawn down, you could take it to 3.0V/cell (so, 15V total from B+ to B-) pretty safely, or 3.3V for longest life/utmost caution. Failing that, I think Russell Graves lays out somewhere on the blog the voltages that correspond to the lights on the 3-bar battery indicator; IIRC, when it drops from "two bars" to "one bar", it's 15.x volts (I don't remember the tenths). If that's all you have to go by, defo stop at "one bar". "Zero bars" is 2.xV/cell...and that's bad! I dunno how warm an ice fishing hut typically is, but closer to room temp is better, and I know you guys have propane heaters nowadays, so...yeah! :-)

      All that said, I've recently had this thought of building a Frankenpack for the DW "slidey battery" platform (I still hate calling it "20 Volt"). So, in a nutshell, neither the DW tools nor chargers give a flying fuck what the capacity (amp-hours) of a pack is. This being the case, you could make a single pack by parallelling as many packs as you wanted -- let's say four 5Ah packs, as those are the cheapest per-amp-hour in Canada right now -- in order to get an 18V "slidey battery" with 20Ah capacity. The key for ease of use would be to leave the top pack's battery board intact, to connect to a DW charger like it ever did (except perhaps with the charger slid over the pack upside-down as a hat, rather than the usual way).

      This honestly may not be the way to go if keeping the battery warm is an issue, but then, a custom battery case could be insulated! Also, with a large-by-huge battery, there's the option of adding a tiny heating pad that turns on and off at temperatures chosen by you to keep your electrons nice and toasty...but not toasted!

      I did something similar last summer to make an ebike pack, buying 4 Ryobi 9Ah packs at Home De Pot (Cdn.$100 apiece = 33% cheaper per-Ah than DW, which is otherwise always the cheapest), removing the BMSes and making a 2-bricks parallel/2-bricks-in-series configuration (6P10S, cellularly,, 3P10S2P?). In my case, I'm contemplating actually decoupling it back to two 18V packs, adding DeWalt battery board/charger interfaces and charging as I outline above. The board doesn't give a fuck who made the cells, or even their capacity. Neither does the charger, except for the resistance on the ID terminal -- which you may want to bodge to full current with a 1MOhm resistor anyway if you have a large pack. Even without the DW board installed, I do in fact use two DeWalt "slidey" chargers to charge my Frankenpack.

      I have to connectorize it though, so as to avoid any more "glowing orange wire" incidents. On the bright side, I learned a lesson about using sketchy "temporary" wiring in conjunction with lithium-ion batteries!

      P.S. I certainly wouldn't suggest anyone try this with anything they care about, but...input voltages are sometimes understated on things. For example, I recently converted a Zoom recorder that uses two AA batts (~3.3V) to a single 18650 (~4.1V), and it works a treat. Again, such things are experimental, so don't try unless you can really afford to lose a thing or the money to replace it. Needless to say, I'd be less positive about the experience if I had cooked the thing! lol

      Happy New Year, electro-weirdos one and all!

    2. Joe - depending on the buck converter, things may actually work as Unknown hopes. Most of the buck converters I've met will just go to passthrough if the input voltage is lower than the regulated voltage. So, if you have a battery ranging from 15V to 20V, and regulate to 17.5V, anything above 17.5V will be regulated down, but anything lower will be just passed through - since the buck topology can't boost the voltage, there's nothing else it can do (except shut down - some nicer ones might do that, most cheap ones won't).

      If that's how the converter works, then you could still use the head unit to shut the system down.

      I'd set the shutdown around 15.0V - there's just no good reason to run the cells lower than 3.0V/cell. All you do is trash longevity for a tiny bit more energy.

    3. Sincere thanks for the correction, Russell. That makes perfect sense, and is surely more welcome as news than my ignorant scaremongering. If you can choose any voltage <17.5V tho, why not 16.5V (3.3V/cell)? Perhaps that's overly cautious, but >90% of the pack's energy is released between 3.6V and 4.2V anyway, so not much incentive to go a lot further.

    4. I generally consider 3.0V a good compromise between capacity available and cycle life. You could set it higher if you wanted, but especially in the cold, the batteries will sag a bit on the output side, so...

      It really doesn't matter. Just don't run the cells down to 2.5V as a matter of course, and they'll be fine.

  47. You don't want to run the pack down any lower than 15v really, 3v per cell.

    What converter are you using exactly? If it's purely a step-down type, it will die before the battery hits the 15v mark, as you're not going to get 17.5v from anything under 18v or so. If it's a buck/boost type of converter, then it would drain the battery down well past the 15v and the Garmin would never know because the regulator will try and maintain that 17.5v for as long as it can. Unless it is connected directly to the battery voltage to monitor it?

  48. I've seen low voltage recommendations from 2.7 to 3.1 V per cell. Anyone know about the particular cells DeWalt uses? Or indeed what the Tools shut off at? Seems that might be the safe limit!

    1. The cells are typically cycled to 2.5V for capacity testing, though I don't know where the tool actually shuts off. I swap packs when mine start feeling saggy, which is typically around 1 LED, but I don't carry a voltmeter with me when I'm working on stuff. I just carry spare packs.

  49. The cells are most likely Samsung, though you'd have to open the pack and see exactly which ones they were.

    Just a quick check here, and the drill I have cuts out around 16.5v, or 3.3v per cell, but I'll guarantee that's dipping down lower than that before the tool is cutting off. I just have other projects on the bench currently to get deeper into setting all of that up to measure it more accurately, but sticking with anything from 3.3v to 3v minimum per cell would fine.

    The DeWalt pack indicator will show 1 LED from around 17.8v (3.56v per cell) on down until around 15v (3v per cell) or so. The FlexVolt indicator will show 1 LED down to around 12v, but at that point the tool has long since cutoff.

  50. Hi Russell,
    Many thanks for your in depth explanation. Very welll done and informative.
    I attempted 3 dewalt batteries this morning; 1 x original 4amp/h and 2 x copies 4/h. Igot to say that the build quality of the copies is fantastic and it look lik ethe moudling are identical. Any way the copies only needed the thermistor to be resoldered. Both had come adrift using my multitool and I think that it was just due to vibration. The good thing is that the thermistors are located on one of the cells. Bit fiddly but hey ho all done and repaired. The dewalt battery had been dropped and the connection to the 4th pin on the tool was missing. I stripped off the whole thing and reconnected it and once again success. Wouldn't have tried this without your blog so ManY Thanks

  51. Well guys thats to much for me:)
    I have question if I can replace cells at Dewalt battery with some similar cells without problems.
    Lets say 18V 5Ah bat. has LGHE2 2500 mah cells so can I replace it eg with Samsung 18650 30Q or BSM will allowe to work just exactly same battery.I replaced my old Hitachi 1.5ah batteries with Samsung 20Q and 20R but there wasnt any board and its workiing nicely over 2 years now.Regards

    1. BMS doesn't care about capacity. Happy spot welding! :-)

    2. well any better alternative cell to HB4 at this time as I need to rebuild 5ah battery,I am not sure if Samsung 30Q will make it?)

  52. Incredible awesome article! I was searchjing for one to make a "big battery" for the leaf blower or the mowder, there is an adapter to make the battery huge: that I'm going to be printing and I have a lot of 18650 that Used for another project.

  53. Hi Russell, thanks for the great explanation. I'm stumped with something, maybe you or someone can help me figure it out: due to missing wiring inside the house, I decided to run the doorbell off a battery pack (DCB207, DeWalt 20Vmax, similar to yours). The circuit is open except when pressing the doorbell, and it works fine. But the pack gets discharged to zero in a few days, even when the bell isn't pressed at all in that period. I checked for possible leaks but current was zero (down to the microamps setting) when the doorbell button is not pressed. I'm using the B+ and B- to feed the doorbell solenoids, with the doorbell button as a normally open switch. I took out the LED that permanently illuminated the button to avoid drains. But still, the battery discharges to zero. When I leave the same pack on the drill, it lasts for a very long time, it doesn't self discharge.
    Any hints would be appreciated.

  54. Do any of you know how to get a replacement clip for the BMS board? It's the clip from the battery to the drill.

  55. Hello Russell,

    The info you shared here are so helpful, I really learned a lot from it, thanks a lot!

    I have a question you may know the answer. For the Dewalt drill using 20V Max battery, if I have other DC power supply(enough for the drill, e.g. 20v 10A), do you think the drill will work if I just connect the B+ and B- on the drill(ignore the other pins)?



Comments on older posts are moderated due to spam issues. If you don't see your comment immediately, and you weren't just spamming me with some irrelevant comment and a link to whatever site you're trying to SEO, your comment should show up relatively soon. If you're trying to use my blog for your SEO purposes, your comments will never show up, so don't waste your time.