How To Build a Battery Pack from A123 Cells - Page 2 - RC Groups
Shop our Airplanes Products Drone Products Sales
Thread Tools
Nov 20, 2006, 02:13 PM
Suspended Account
End to end soldering of A123's

Yes A123 strongly advises against it....... but so did every Nicad and Nimh manufacturer, "back in the day".

I'm using SPT's as that gives a lot of convenience to replace a bad cell. While soldering may work to put a A123 pack together, I believe it would be very very hard to break apart two end to end soldered A123 cells without causing a lot of damage. Soldering to tabs should work fine.

Sign up now
to remove ads between posts
Nov 20, 2006, 02:31 PM
Registered User
Originally Posted by asm_
Any one tried end-to-end solder on those cell yet? I wonder how well does those cell hold up in end-to-end soldering heat comparing to the good old GP3300.

Sid solders them....
Nov 21, 2006, 01:25 AM
Use the Force!
LBMiller5's Avatar

Concerns about the Steel Buss Bars

I was a little surprised by the majority of the comments about my battery build thread. Only a couple people had anything positive to say about all the work I put into documenting this build, and most of the responses had to do with the shortcomings of the steel buss bars used in the pack construction.

I guess I am OK with that, if there is that much concern over the buss bars, then I thought I would look into it. I had a 4-cell group of the batteries left over from the build thread that I did, so I soldered a couple pieces of 12 gauge wire to the pack and added a Deans connector. It did not tape up the cells or cover any of the contacts so they would be available to probe with my voltmeter.

To begin the test, I charged up the 4-cell pack to 3.6 volts per cell using my TME Xtrema Charger. I set the charger rate to 2500ma, and the pack took in 1248mah of charge in about 32 minutes. This means that as I received the pack, it was just a hair under half-way charged. While I was charging the pack, I measured the voltage difference from the positive end of one battery to the negative end of the next battery to measure the voltage drop across the buss bar. I put the tip of my probe on the can of the cell for the positive lead and on the button under the buss bar for the negative lead. I got close to the buss bar without actually touching it. My goal here was to measure the voltage drop across the spot welds and the buss bar itself to get the total resistance that the entire assembly added to the cells.

While I was charging the batteries at 2.5 amps, I measured the voltage drop across one of the buss bars and got 0.0016 volts or 1.6 millivolts. Doing the math, that works out to a resistance of the buss bar of 1.6 mv/2.5 amps which equals 0.64 milliohms or 0.00064 ohms. Since this was at a low current and near the threshold of the meter, I figured that this value was a little off, but was probably within 20% of the actual value.

After the battery finished charging, I did another test at higher current. I have a load box that I built for testing battery capacity. I can pull up to about 55 amps of current with this test box by flipping on various combinations of switches that I have connected to 12 volt light bulbs inside.

I hooked up the freshly charged battery to my power meter and then hooked the power meter to the load box. I added load in 2.5 amp increments until I was up to a 35 amp load. I chose 35 amps because this is how much current my power systems in my F-4 and F-18 ParkJets draw. At 35 amps of load, I measured all 3 buss bars in the pack, and all three of them measured the exact same value, 26.7 millivolts. I put my finger on the buss bars after running at 35 amps for about 1 minute, and the bars were a little warm, but not hot at all. Probably about 15 degrees above ambient temperature.

Doing the math 26.7 millivolts of drop at 35 amps gives me a resistance of 0.763 milliohms. This value is 19% higher than the resistance that I measured during the charge cycle. I expect that this is due to 2 factors: first, the original reading was at a low current, and at the very low end of the resolution of my DVM, and second, the resistance got higher when the buss bars heated up a little with the 35 amp current flowing through them.

The distance from the center to center of the cells in the pack is right at 1". To calculate the wire gauge equivilent of these buss bars, I just need to find out what gauge wire has a resistance equal to 0.763 milliohms per inch, or in more common terms, 9.156 milliohms per foot. Checking into my wire gauge charts, the closest size wire that matches that value is 19.5 AWG wire, which has a resistnce of 9.023 milliohms per foot. This is actually quite surprising, I would have never thought that the resistance would be that high!

In the 6-cell pack that I constructed, there are 5 buss bars that connect the pack together. If each buss bar measures 0.763 milliohms, the total resistance of the buss bars is 5 times that value, which equals 3.815 milliohms. That is the equivilent resistance of almost 3 feet of 12 AWG wire!

In actual practice however, when pulling 35 amps of current through that 3.815 milliohms of resistance, it induces a voltage drop of 35 x 3.815 or 133.5 millivolts. Since a 6-cell pack measures 21.6 volts fully charged, a loss of 133 millivolts represents only 0.615% of the voltage, so it reality it is not that big of a deal.

For the ease and simplicity of using the existing tin plated steel buss bars, it costs me less that 1% of the total pack voltage when pulling 35 amps from the pack. In fact, it would take a current draw of 56.6 amps to lose 1% of the pack voltage in the buss bars, and 56.6 amps is equal to a 24.6C discharge rate.

So the bottom line is that if you plan on drawing less than 50 amps out of your A123 packs, and don't mind losing less that 1% of your pack voltage, then the standard steel spot welded buss bars work just fine. If, on the other hand, you want to get that last 1% from your cells, then by all means feel free to add a piece of wire in parallel with the buss bar. Adding a piece of 14 awg wire across the buss bar will drop the voltage loss at 35 amps from .6% to around .2%, and if you really want every last drop, adding a piece of 12 AWG wire will cut your interconnect losses down to around 0.1%.

For the applications that I am using, pulling no more than 35 amps from the packs, the stock buss bars suit me just fine.

Last edited by LBMiller5; Nov 21, 2006 at 01:50 PM.
Nov 21, 2006, 03:41 AM
Registered User
There are hundreds of us out here that usually only read these threads because of all the great information that we gain from them. We hardly ever say thanks. But if wasn't for you guys posting this information we really would be in the dark ages when new technologies emerge. So thanks!
Nov 21, 2006, 03:58 AM
Registered User
Lucien... we all appreciatted your contribution and in fact congratulated you for that well done work! As I stated, I would like you were in my office and not some "individuals" I have lying here...

We just added our own comments on details, no more.

I have also to praise your investigation about the real losses in the steel bars. I knew they are not so big, it's just to me its really easy to solder a small copper braid and so I am more confident on the thing before shrink-covering it. In fact, I rarely use this cells over 35A, and yes, someone useing bir DF stated earlier in this "history" that the original steel tabs were not hot (barely warm) under high loads.

Maybe I am still a bit "intoxicated" from my past days of rc-car racing were you "need" to put big-nice-solid copper bars to be able to show your packs to your mates and receive praise (...and off course have the last mV out from your cells, but that was less important than the praise...).

Thanks again for your trouble.

Nov 21, 2006, 08:36 AM
Registered User
I agree, Josep!

The thread on DeWalt disassembly was particularly helpful. Even folks who are mechanically inclined would be apt to hack the case apart before that thread was posted.

- RD
Nov 21, 2006, 08:43 AM
RC Ace wannabe
Ace in the hole's Avatar
I'll give you two thumbs up and a big thank you too. Your posts are very informative and appreciated.
Nov 21, 2006, 10:35 AM
Registered User
The build thread is good but to actually quantify the issue of the tabs is exceptional!
Kudos Lucien

Nov 21, 2006, 11:27 AM
Use the Force!
LBMiller5's Avatar

Thanks Guys!

I appreciate the thanks and comments. I have been in this hobby for over 30 years, and I enjoy sharing my ideas and techniques with other people. I strongly believe that Electric powered airplanes are the future of the R/C hobby. It just make so much more sense than Glow powered models on many levels.

The big problem is that many people are simply "Afraid" of electronics because they do not understand it. My goal in my threads and posts is to "De-Mystify" the principles of electronics and present them in a way that the average modeler can understand them, and and to replace here-say and rumor with easy to understand facts.

I will continue to present articles like this, as I see the need here on RC Groups. So if anyone out there has a question or a technique that they would like to learn more about, just pose a question here or send me a PM and I will be happy to do my best to answer it.

Now I have to get together my parts and build a balancer that will work at 3.6 volts per cell to balance these new A123 packs.

Back to the soldering station!

Nov 21, 2006, 12:00 PM
Southern Pride
everydayflyer's Avatar
FYI . The Thunder Power Balancers and the Hyperion LBA6 do a very good job of balancing these cells.

The Blinky does not do so well due to the higher voltage needed to get it to start balancing and it's limited current handling abality.

I bet Dan Baldwin's DIY LiPo balancers would do great if adjusted down to 3.60 -3.7 volts range.

Perhaps I should not bet as I do not remember the max. current handling availabe but I seem to remember it being much greater than the other shunting balancers.

Last edited by everydayflyer; Nov 21, 2006 at 12:25 PM.
Nov 21, 2006, 12:32 PM
Use the Force!
LBMiller5's Avatar
I have PM'd Dan and talked to him about that circuit, and I am going to use a variation of his circuit in my 10-cell balancer. Total cost to build the balancer should be around $60-70 for a 10-cell unit if you have to buy all the parts yourself. I have a lot of stuff laying around from old PC power supplies that I have taken apart. The TL431 precision current regulator that Dan uses in his design is a very standard part. Most PC power supplies have 2 of them inside, and some of the better ones have 3 of the TL431 regulators! I have a couple dozen of the TL431's in my parts bin, as well as most of the capacitors and resistors, so I won't have to buy much.

Nov 21, 2006, 01:07 PM
Registered User

Excellent build thread and research. Very interesting about the voltage drops on the weld tabs.

Nov 21, 2006, 02:54 PM
MaxAmps CMO
JAM's Avatar
Lucien your build threads of all types always amaze me.....great work and first class...but any how I just saw an ad on the history channel that might help and may not........Craftsman tools came out with a 20 volt lithium ion powered drill and saw....maybe another source for cells if they use the A123 ones.........I did some checking on sears web site and can find the tools with the packs and chargers, but not the Packs separatly. So a cost is still a mystery......gonna hit E-bat And see if there on the site yet.

If anyone knows of these packs please let us know.

Nov 21, 2006, 03:15 PM
Southern Pride
everydayflyer's Avatar
What might be known.
Nov 21, 2006, 07:30 PM
Registered User
Originally Posted by LBMiller5 next project will be to design and build a custom balancer that can be plugged into the balance connector to keep each of the cells at 3.6 volts during the charge cycle.
Let me know if you'd like to rattle off a few extra balancers to sell!

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion How to build a battery pack? caleberx Batteries and Chargers 0 May 28, 2007 09:12 PM
Sold FS DC9360 battery pack new A123 cells 10sp1 andres.c Aircraft - Electric - Batteries & Chargers (FS/W) 2 Apr 18, 2007 06:58 PM
Learn how to build battery packs squelchF18 Beginner Training Area (Aircraft-Electric) 21 Aug 13, 2004 03:58 PM
Can I build a battery pack from these cells? dmelton Batteries and Chargers 8 Aug 14, 2003 10:56 AM
separate battery pack for 10 cells gliderdude Electric Heli Talk 0 Feb 03, 2002 12:57 AM