Current flow from battery cell to bus bar

[Bob B]

The discussion came up on another thread about whether or not current flows to the bus bar thru the stud. That discussion was surrounding the use of the stud / flange nut type of connection using a stainless steel stud and nut and an aluminum pad on the battery cell.
I wanted to separate the discussion because the original discussion was being somewhat side tracked.

I have seen it reported a few times that there is no current flow thru the stud .... only from the pad on the cell terminal to the bus bar .... I don't agree with that and would like to see some more discussion about it. I would argue that when a parallel circuit exits, current will flow in all paths. The amount of current flow in each will depend on the amount resistance in that path.

Some things to consider:
1. Is there any way to scientifically determine how much current flows .... or how much resistance there is in each path.
It seems to me that there are so many possible variables it would be hard to determine.
2. How would the contact resistance of the pad and stud connection be determined? It seems to me this could vary wildly depending on how well the surfaces were prepped.
3. How would the contact resistance of the stud threads to terminal be determined?
4. How much difference would the torque on the nut make ... both to the cell pad connection and the thread connection?
5. How much greater resistance would the stud have with that short of a connection point?
6. How would the application of something like Noalox affect the contact resistance. Even though the Noalox will help prevent oxidation, the substance itself seems to be an insulator.
7. How much different would the flow be when there is a ring terminal attached to the top of the bus bar? The main positive and negative will have a lug attached to the top of the bus bar which will add an extra point of contact resistance between the pad and the lug and make current flow thru the stud more likely?
8. How much difference would surface area make on both the pad and stud ... for instance if the stud was screwed in more or less, for instance.
9. How much difference would temperature make?
10. Is that bus bar really flat?
11. Using @HRTKD suggestion .... is the top of the cell terminal pad really flat?

There are probably more variables you engineers can dream up ... kinda hoping there is someone who specializes in these kind of calculations.

My premise is that even if the resistance of the stud to bus bar is 10 times greater that significant current can flow thru the stud. If 10 times greater and current flow of 100A ... the stud would still see 10A .... and it seems to me the stud would carry more current when there is a lug attached.

For those who want to read the comments already made on the original thread about studs, I am posting a link if you haven't seen it ... but please post current flow comments here.

link to proper studs and nuts for eve cells?

Anyone have a US link to high quality proper studs, washers, and nuts that fit the eve 280ah cells? Also any clue to proper torque and seat depth? Mine came with crappy bolts and I want to replace them.

diysolarforum.com

 

[HRTKD]

Thanks Bob. Good questions.

Your point #7 is something I was going to ask, because my balance lead will be positioned between the bus bar and the head of the bolt or nut.

For your point #10, I'll add, "Is the top of the terminal really flat?" I ran into a situation where the end cells in my 8p configuration did not get top balanced and close scrutiny showed that the bus bars from the cells one in from the outermost cells were keeping the bus bars from coming into full contact with the terminals on the outermost cells. I have no problem with chalking this up to user error as I was perhaps too light on tightening the terminal bolts for fear of causing damage.

 

[Luthj]

Its complicated. A bit of reading that came up with a quick search.

https://nvlpubs.nist.gov/nistpubs/jres/5/jresv5n3p757_a2b.pdf

https://www.ewh.ieee.org/soc/cpmt/hongkong/Seminar/Reliability03Apr2007.pdf

Current will always flow through all parallel paths. The split will depend on each paths total resistance, and its kinda complicated when there are multiple components in regular contact.

 

[Bob B]

I'm hoping we can arrive at some sort of SWAG as to what % current may be flowing thru each. I'm not an engineer but had 2 years dedicated electronics training in the Navy and MANY years installing and troubleshooting, so I usually just go with instinct or experience with this type question .... In this case, I know there will be flow thru each .... maybe we just need to consider the perfect scenario for both to get an idea of how the flow splits up.

 

[Bob B]

Wow, the conducting contact area in that 2nd document is very revealing. P8

Edit:
That 2nd document is VERY interesting .... 1/2 way thru you start to think that maybe this whole thing of trying to make a DYI battery is not a good idea ... too many things to go wrong.
He does get into ways to mitigate the risks. It seems the lubricants are VERY effective at mitigating some of the problems that occur over time, temperature fluctuations, vibration, etc.

I am wondering if it might be possible to make a measurement if someone had a micro ohm meter. The first measurement could be taken with the stud insulated from the bus bar and then another measurement with the stud connected normally. With the difference, the resistance of the stud could be determined. (The measurement could actually be done the other way around also. Put and insulator on top of the aluminum pad for the first and then remove it and take a 2nd measurement.)
Anyone have a micro ohm meter?

 

[Ampster]

It is an interesting subject and I would be happy to contribute knowledge or examples to the subject. Even though I discount the current flow through the stud or bolt is not significant. I have observed some phenomena that might be worth discussion as part of this subject.
As background my pack is 2P16S of the popular LF280 cells. I use an Orion BMS that gives me open circuit voltage, live circuit voltage and a calculated value for internal resistance.
I use SS studs that are fixed to the terminal via Threadlocker Red. I am also using different types of buss bars. These two factors may be contributing to some of the issues I have experienced. I top balanced my pack to 3.6 volts but after a few cycles still had some cells that were moving toward 3.65 volts while others were back at 3.33. I did not keep a log of the IR or the open cell voltage but I tried putting a 1/2 Ohm power resistor on those cells. That brought the voltage down but I still had some variation. Eventually I double checked the bolts and on some of those I got an additional 1/4 turn on the nut. That fixed most of the issues.
So my hypothesis is that since the voltage sense wire is on the top of the buss bar that there was some voltage difference between the actual cell voltage and the measured cell voltage. I double checked the BMS voltage with my VMM on the top of the buss so that eliminated any voltage measurement error. That voltage difference could have been because of the higher resistance of the stud with the threadlocker coupled with the lack of tightness of the buss bars. This may or may not be related to the subject at hand but I thought I would throw it out there in case there is a relationship. I know the subject is current but voltage and current have a mathematical relationship with resistance. I have not even worked this out in my head. I am thinking the variety of buss bars could also be contributing to the issue but it has pretty much been resolved now.

 

[Bob B]

In your case, there must have been some bad connections somewhere. Typically, the voltage drop won't show up unless current flow is present .... the more current flow across a resistive connection, the higher the voltage drop would be. How much balance current does the Orion have?

There are so many things that can go wrong with the connections .... this is one area where being meticulous is a major benefit. It also seems that things are not going to stay the same very long and using an IR thermometer on the connection points should be a maintenance item that is done frequently.

I'd sure like to get my hands on some of that composite foam material mentioned on Page 107 of the 2nd document @Luthj posted.

 

[svetz]

My two cents: Ohm's law... Both are conductors, both will carry current. They'll have difference resistances, so different amounts of current.

 

[Luthj]

Honestly there isn't a whole lot to worry about with battery packs. We aren't pushing the conductors to their limits for 12 hours straight. Power transmission is a whole different situation. While we can gain useful data from those applications, their issues are not significant for a dry indoor battery pack operating at 0.5C or less. Also note that in power transmission equipment, it isn't unusual to see long threaded fasteners as primary conductors, for example the ones which penetrate the shell of some transformers. In those cases, when the fastener is the primary conductor, resistance across thread interfaces is of a primary concern.

For most any application building a stationary battery pack, the studs/bolts won't carry an appreciable amount of current. In the end, you simply size the copper/aluminum conductor to carry the rated load. Any conduction by the threaded faster is just extra margin. There isn't any risk of damaging the fastener if the contact area is suitably sized, clean/smooth, and the conductor is also correctly sized.

Now if we are talking spacecraft or aircraft to a certain extent, weight savings my optimizing lugs and terminal connections can be significant, and in those cases each terminal configuration would be tested on a test bed to determine the resistivity of each typically connection. Then it would be optimized to reduce weight.

 

[Bob B]

For most any application building a stationary battery pack, the studs/bolts won't carry an appreciable amount of current.
How do we know that .... especially on the main positive and negative where a lug is present? That seems to be the conventional wisdom, but I'm not sure it's a valid conclusion.

There isn't any risk of damaging the fastener if the contact area is suitably sized, clean/smooth, and the conductor is also correctly sized.
It seems to me that there is a real risk of galvanic corrosion to the stud threads .... and the copper to terminal connection ...due to dissimilar metals and current flow.

I agree that the connections are not as critical as aircraft or power transmission .... I would just like to get some kind of ballpark idea how much current is flowing thru each. Maybe it's worth changing the stud from stainless to some other material if the current flow thru the stud is significant/

Maybe it would be better to use aluminum alloys for everything to prevent degradation over time ... or maybe we also need to be putting Noalox on the stainless stud threads.

 

[Luthj]

I think you are misunderstanding a bit. You don't need to select a fastener for current carrying if your primary conductor is able to carry the full load on its own. If this is the case, the current passing through the fastener, be it wood or pure gold, will never be sufficient to cause damage. The less conductive the fastener, the less current will be diverted through it.

 

[Bob B]

I'm not trying to call current thru the stud either a good or a bad thing. I would just like to get some sort of consensus on what % current flows thru the stud. Although, I guess it could be construed as a bad thing if galvanic corrosion is the result of the current flow.

 

[HRTKD]

A concern with the fastener is that if we use to poor conductor then we're leaving some current on the table. I don't think this is a critical issue, at least not in my situation. I have enough Ah that I'm not going to worry about a 1% loss, if that's what it is.