Reducing cell-to-cell contact resistance

[Battery Guy]

Hello Everyone

I was wondering what people are doing to reduce cell-to-cell contact resistance in large series strings.

For example, I have 8AA->2D holder from fivemega that holds 8 PowerGenix AA cells. When fully charged, each cell has an impedance of 0.014 ohms, which should give me 0.11 ohms on the pack, but the fully charged pack has an impedance of 0.15 ohms. So I am picking up an extra 0.040 ohms of contact resistance, and losing a full 1 W of power with a 5 amp draw.

I am curious to hear and see your solutions to this problem.

Cheers!
Battery Guy

 

[VidPro]

roll them? that is the deoxidation method i apply first, spin them around grinding the contact points together like grinding valves.
if that does not do anything then the parts and pieces themselves would have to be replaced or shunted.
There was a pictorial on shunting a Mag Spring with flexable copper wick wire , reducing the resistances by lots. People applying fat copper and direct soldering battery to battery. Many different designs for the whole package resistance have existed ,but the cell itself just to the next cell? what could you do, silver coat it?

when pushing mass amps via a high discharge or even a short, you see quick how many molecules of metal are actually making contact. in high discharge with poor contact the battery itself (and holders) will heat up at that point, so the quantity of the metal to metal contact is such that even with great conductive metals the quantity of the metals connected makes a difference. spread out the contact somehow?? silver conduction pen?

prime "point" conduction areas to make more contact, emeryboard grinding , while you can make a rough surface with sanding , the peaks of the rough surface will bend onto the tightly pressed item onto it. so when you squish the battery into the rough ground contacts the molecular mountains will flatten into the battery contact areas (so to speak).

i often use Silicon spray to keep contact points from being oxidised, while silicon oils are not conductive , it seems to be better than getting a layer of metal oxides (which are not conductive) between things. stubborn battery connection points like Mini-mags i have improved by cleaning then applying lightweight high solvent silicon spray for long term maintance of the contact points.

those "pro" gold contact sprays are used by some, i always assumed they were just de-oxidiser and silicon oil anyways.

When analising why they use Gold for connections when silver is a better conductor (and cheaper) i noticed that gold Squishes quite a bit, so it is not always about the conduction of the metal as it can also be how squishy it is because you could squish the metal into having more molecules connect. Try impacting it with amalgum

Reducing the number of cell units. use 1-3 Big Fat cell items, or higher voltage cell items, i have a great dislike for Series anyway due to reverse charge situations during deep discharges, so i prefer not to use wads of batteries and prefer to use big ones instead.
or the opposite
Increase the voltage even more and reduce the amperage, its the amps that gets ya .

 

[Battery Guy]

When analising why they use Gold for connections when silver is a better conductor (and cheaper) i noticed that gold Squishes quite a bit, so it is not always about the conduction of the metal as it can also be how squishy it is because you could squish the metal into having more molecules connect. Try impacting it with amalgum

The use of gold as a contact material has more to do with the fact that it does not form a native oxide layer, unlike silver and copper.

Reducing the number of cell units. use 1-3 Big Fat cell items, or higher voltage cell items, i have a great dislike for Series anyway due to reverse charge situations during deep discharges, so i prefer not to use wads of batteries and prefer to use big ones instead.

I agree that reducing the number of cells in the series string helps, so going with higher voltage lithium-ion cells is an answer. For my personal, larger flashlight mods, I intend to use lithium-ions in part to limit the number of cells. But for the mods that I give to family and friends as gifts, I prefer not to use lithium-ions because they are not "beginner friendly" cells. Few people want to hear my hour long lecture on the care and maintenance of their flashlight batteries when they get a gift. I can't understand why. :shrug:

Getting back to the topic at hand, I was thinking about using a small dab or smear of silver epoxy at the battery contact points. With respect to Deoxit or other similar products, has anyone measured a reduction in contact resistance?

Cheers,
Battery Guy

 

[MrGman]

I have to ask how it is you are measuring battery impedance?

I am guessing you would be making your battery connections of the pack "permanent" with silver epoxy. If you have a way of testing how that changes impedance of terminals with say a 3 mil thick layer silver epoxy that would be interesting to find out on a test sample prior to committing the full battery pack to it.

High current flow through dissimilar metals has a way of pulling them apart. Your silver epoxy may actually cause a more resistive joint in time with the current flow then just leaving it alone. If you have Tin plating on the battery contacts it may disassociate the Tin from the base metal with current flow. It may disassociate the silver epoxy bond from the surface metal. Silver epoxy is a not a conductivity cure all.

 

[Battery Guy]

I have to ask how it is you are measuring battery impedance?

The impedance measurements were made using a Hioki series 3550 HiTester, which does a 1kHz impedance measurement.

I am guessing you would be making your battery connections of the pack "permanent" with silver epoxy. If you have a way of testing how that changes impedance of terminals with say a 3 mil thick layer silver epoxy that would be interesting to find out on a test sample prior to committing the full battery pack to it.

I would prefer to not make the connections permanent. At least, I would like to see how far I can reduce the contact resistance without soldering or spot welding the contacts. I was thinking that a small amount of silver epoxy could be removed if needed relatively easily.

High current flow through dissimilar metals has a way of pulling them apart. Your silver epoxy may actually cause a more resistive joint in time with the current flow then just leaving it alone. If you have Tin plating on the battery contacts it may disassociate the Tin from the base metal with current flow. It may disassociate the silver epoxy bond from the surface metal. Silver epoxy is a not a conductivity cure all.

Silver epoxy was just a thought, and frankly not something I really want to mess with if there are better or equivalent solutions.

I see many mods with 10+ AA cells in series drawing 5+ amps. I have seen several clever solutions to reduce the resistance of tail springs, switches, etc..., but I can't seem to find any good threads on reducing the resistance of the contacts between cells.

Cheers,
Battery Guy

 

[Justin Case]

I don't see conductive epoxies or even compounds as viable. Their conductivity depends on the silver particle/flake concentration being at the percolation threshold so that there is a continuous path through the medium. Why would you expect that to be such a great electrical conductive path?

I think you'd be better off using a dedicated, hardwired battery pack if the resistance is really an issue.

True, you are wasting 1W of power. But that's spread out over all of the contacts, so I doubt that the additional resistive heating will really be an issue. And the voltage drop from the contact resistance is only 0.2V. If you haven't done other low resistance mods to your Mag yet, then IMO you are worried about second order issues. You probably have more parasitic resistance in the tail spring, switch tower spring, and switch contacts. What are you trying to run? Sometimes a little parasitic resistance is a good thing, helping to avoid instaflashing a filament.

 

[Mr Happy]

I see many mods with 10+ AA cells in series drawing 5+ amps. I have seen several clever solutions to reduce the resistance of tail springs, switches, etc..., but I can't seem to find any good threads on reducing the resistance of the contacts between cells.

My guesstimation would be that contact resistance is reduced by:

1. Larger contact area
2. Reduced surface oxidation

In many applications it seems this is achieved by putting a layer of soft metal like tin, solder or lead on the contact surfaces and applying pressure. The pressure has the effect both of deforming the soft metal producing a larger contact surface, and also of breaking the oxidized surface to expose clean metal underneath.

So one experiment you might try is tinning the battery contacts. Another possibility is to use some of the silver plate compound used to coat the tracks and contact surfaces on circuit boards (like this, for instance: http://www.cool-amp.com/).

 

[Battery Guy]

I don't see conductive epoxies or even compounds as viable. Their conductivity depends on the silver particle/flake concentration being at the percolation threshold so that there is a continuous path through the medium. Why would you expect that to be such a great electrical conductive path?

I am attempting to solicit ideas, and just threw out the first that came to my mind. I never said that it was a good one or that I expected it to work. I was just trying to "prime the pump", so to speak.

I think you'd be better off using a dedicated, hardwired battery pack if the resistance is really an issue.

How do people generally hardwire packs together? Solder? Spot welding with nickel tabs? Sorry if this seems like a dumb question, but it seems like it would be difficult to hardwire a string of cells end-to-end. Obviously people do this, so I would like to know the method they use.

True, you are wasting 1W of power. But that's spread out over all of the contacts, so I doubt that the additional resistive heating will really be an issue.

Dude, that is 1W of precious power that is not going into making beautifully blinding light fly out of the end of my flashlight!

And you call yourself a "flashaholic"? For shame :tsk:

I kid, of course.

And the voltage drop from the contact resistance is only 0.2V. If you haven't done other low resistance mods to your Mag yet, then IMO you are worried about second order issues. You probably have more parasitic resistance in the tail spring, switch tower spring, and switch contacts.

Perhaps true, but there is already a lot of great posts and solutions to those "first order" issues, so I think I have them under control.

Sometimes a little parasitic resistance is a good thing, helping to avoid instaflashing a filament.

Personally, I would prefer to drive as much resistance out of the system as possible. Later, if I need some resistance to prevent instaflashing, I can add it in a controlled manner.

 

[core]

For example, I have 8AA->2D holder from fivemega that holds 8 PowerGenix AA cells.

for the mods that I give to family and friends as gifts, I prefer not to use lithium-ions because they are not "beginner friendly" cells.

Well, I gotta say it since no one else did:

Giving someone a device with 8 (!!) NiZn cells in series is certainly not "beginner friendly" either.

 

[Battery Guy]

Well, I gotta say it since no one else did:

Giving someone a device with 8 (!!) NiZn cells in series is certainly not "beginner friendly" either.

Point well taken! :thumbsup:

I only used the NiZn example because I had recently made those measurements and had the numbers fresh in my head.

Cheers,
Battery Guy

 

[Justin Case]

I am attempting to solicit ideas, and just threw out the first that came to my mind. I never said that it was a good one or that I expected it to work. I was just trying to "prime the pump", so to speak.

You stated you were thinking of using a silver epoxy as a possible solution, and I gave you my assessment of that approach. I take it that you didn't want any assessment of using silver epoxy because you were just "priming the pump" with random thoughts? I missed the mind reading classes in college.

How do people generally hardwire packs together? Solder? Spot welding with nickel tabs? Sorry if this seems like a dumb question, but it seems like it would be difficult to hardwire a string of cells end-to-end. Obviously people do this, so I would like to know the method they use.

Google capacitive discharge (spot) welding. There is a lot of information on the Internet.

Dude, that is 1W of precious power that is not going into making beautifully blinding light fly out of the end of my flashlight!

What bulb are you trying to drive? Do you really think that you will even see an output difference due to the 0.2V drop from your battery pack? I tend to doubt it.

 

[cckw]

Caig De-oxit on the contact surfaces would be worth while in the big picture. I have not done measurements like you are doing, but with use you are going to develope connection issues that can be avoided with a little deoxit every now and then. (every charge or 2 or 3)

Don't just spray and leave it, spray a Q tip or paper towel so you are wiping away all residue from previous chafing wear

 

[45/70]

The use of gold as a contact material has more to do with the fact that it does not form a native oxide layer, unlike silver and copper.

I always thought gold was used because it looks better. And, as Vid pointed out, silver is a better conductor. Also, as I recall, silver oxide is still a better conductor than copper. At least I'm sure it's better than copper oxide.

Interesting discussion. With the exception of my welded and soldered packs, I think I'll just stick with "rubbing 'em on my pants leg before installation" routine.

Dave

 

[Vortus]

For button top, maybe have some small silver washers made to increase the contact.

 

[Battery Guy]

I missed the mind reading classes in college.

Too bad. Really useful class that was. But yet, I sense that you don't believe me.

Google capacitive discharge (spot) welding. There is a lot of information on the Internet.

Will do. Thanks!:thumbsup:

What bulb are you trying to drive? Do you really think that you will even see an output difference due to the 0.2V drop from your battery pack? I tend to doubt it.

I don't have have a specific bulb in mind. I'm simply trying to drive resistance out of the system.

Let's take a setup that I currently have with 7 AA NiZn cells in an FM holder driving an 1185. I am losing approximately 0.5W due to contact resistance in the holder. Now, that is only about 1.5% of the total power draw on the battery at 3.5A.

But lets say you were running a 2S 4P configuration with a similar wattage bulb at 6V. The current would double to 7A, and the power lost through the contacts would be a factor of four higher, or 2W. Now about 6% of the total power draw is wasted in the contacts.

Small losses I know, and you correctly labelled them as second order. But if you can identify a few second order losses that are additive, then sometimes impact of the improvement can be of a first order magnitude.

Cheers,
Battery Guy

 

[Bullzeyebill]

All right guys, you can drop the factious comments to each other. It won't serve any useful purpose, and it can only get worse.

Bill

 

[bshanahan14rulz]

What about those metal mesh grounding pads you see in electronics?

Maybe a wad of copper desoldering braid?

 

[WDG]

I think I'll just stick with "rubbing 'em on my pants leg before installation" routine.

I figured I was the only one who did that! :huh: Of course, the good old pink eraser's always good for taking off contact oxidation, too.

IIRC, there used to be something called Tweek that was for audio connections. I don't know if that would help, but it might. My quick googling isn't really turning it up, so I may have the name wrong. :thinking:

 

[MrGman]

Too bad. Really useful class that was. But yet, I sense that you don't believe me.



Will do. Thanks!:thumbsup:



I don't have have a specific bulb in mind. I'm simply trying to drive resistance out of the system.

Let's take a setup that I currently have with 7 AA NiZn cells in an FM holder driving an 1185. I am losing approximately 0.5W due to contact resistance in the holder. Now, that is only about 1.5% of the total power draw on the battery at 3.5A.

But lets say you were running a 2S 4P configuration with a similar wattage bulb at 6V. The current would double to 7A, and the power lost through the contacts would be a factor of four higher, or 2W. Now about 6% of the total power draw is wasted in the contacts.

Small losses I know, and you correctly labelled them as second order. But if you can identify a few second order losses that are additive, then sometimes impact of the improvement can be of a first order magnitude.

Cheers,
Battery Guy

If you put batteries in parallel contact resistance doesn't go up. The current distribution through the sets of cells in parallel goes down and therefore the voltage drop and I squared R losses for any set of contacts should not be any higher than before. You have a misconception.

The basic answer to your question of high current multicell battery packs really already has a standard answer. They either have welded or soldered battery tabs and not mechanical contacts. More typically the tabs are welded to minimize this issue.

Putting any other metal in series with these metals unless its liquid mercury will really only add more resistance across more mechanical interfaces that don't have ideal surface contacts to begin with. If you could plate the contacts you already have with pure Gold by melting the Gold into the surface contact metal in a fairly generous quantity, you would have some marginal improvement of the mechanical contacts. But of course at a considerable cost.

The general idea of putting more batteries in series to deliver the same amount of total power to the original load is that the series current goes down and therefore wasted drop across things like contacts becomes a meaningless second order effect.

24V at 2 amps as opposed to 6V at 8A still delivers 48 watts as a simplified example. But 2 amps across 0.200 ohms of total contact resistance (for this example) is a lot less than possibly 8 amps across 0.2 ohms. The first is 0.4V and 0.8 watts, the second is 1.6V and 12.8 watts which would be ridiculous and you wouldn't have 48 watts to the load any more. But in reality the second one may have the total parallel contact resistance of 0.050 ohms and not 0.2 ohms and thus again 0.4V drop and ony 3.2 watts loss. Still better to go with higher voltage and lower current if the load is something you can adjust to optimize for that type of source conditions.

Otherwise the reality would be as in most laptop computers or high current RC car battery packs for example to weld all the contacts together as well as have some type of parallel battery configuration (more so in laptops than RC cars). Just an example.

 

[HarryN]

Hi, just throwing in some more ideas and comments.

- I tried to measure a before / after with CAIG deoxit, in particular on a tailcap spring. The main problem was that the tailcap I was interested in had remarkably bad corrosion problems and the measurements were unstable. I can say for certain that it went down substantially.

I don't have an ability to measure on just a battery.

CAIG usually recommends first to "clean" then to "stabilize" so perhaps if both steps are used it will help. (guessing here, but that is what they do for a living)

- I had some NiMH packs made from sub Cs by cheapbatterypacks.com. Using their recommendations, they believe that the battery to battery contact resistance is better if the packs are end to end soldered vs welded tabs, and made a good argument as to why. I just paid the small premium and had them solder them up. :shrug:

- I have used silver filled conductive epoxy before. They usually take a rather long bake to obtain their ultimate properties, and even then it usually is not as good as solder ( by a lot) These keep getting better though.