Bus bars for 3P4S battery

[sailingharry]

I'm getting close to building an LFP pack for my boat. It will be either 2P4S or 3P4S, using EVE cells around 280Ah. Either studs, or if I get them from 18650 then the two hole version (which I like). Especially for the 3P version, bus bars are a challenge! I'll be fusing the battery with a 400A fuse, but can't realistically charge or discharge over 300A and will rarely see over 200A. My pack configuration for 3P4S (but the 2P4S will be the same layout) is below.

My problem is that I need two LONG bus bars to connect across 6 cells, and a WIDE bus bar to connect 3 cells end to end to 3 more cells, all rated for at least 300A. It's not easy! Almost all of the "300A" or "400A" bars on Ali are 1/8" thick (3mm) and less than 1" wide (20mm), which gives a temp rise of over 30C (54F) at 300A and 65C (110F) at 400A. HOT!

I'm really tempted to DIY the bus bars. Buy copper stock, and drill to match. The long ones, to be 300A, would need to be 3/16x1, and the wide ones are essentially not current limited (1/16 thick would still be good for over 600A!). A quick eBay search says I can get 18" of 1/4 x 1 (need 3) for $35 and 1/8 x 4 x 8 sheet for $21, so for around $125 I can have this.

I'm a bit worried about all that rigidity. Is this going to break terminals off? Or is this the right way to wire this battery up?

 

[DIYrich]

Is this what you are looking for?

https://www.aliexpress.us/item/3256806567952455.html

or

https://www.aliexpress.us/item/3256805888457447.html

 

[HRTKD]

As long as you have the cells in a compression fixture they're not going to move and your cross-column busbar should be OK. If you're really worried about it, there are flexible busbars available, but a custom one to span that gap could be expensive.

I have plenty of copper bar to play with but I still used the cross-column busbar that came with my cells. When I stress test the system, that busbar is one item that I'll be keeping a very close eye on.

 

[HRTKD]

Is this what you are looking for?

https://www.aliexpress.us/item/3256806567952455.html

or

https://www.aliexpress.us/item/3256805888457447.html

That's a good find.

 

[sailingharry]

Is this what you are looking for?

https://www.aliexpress.us/item/3256806567952455.html

or

https://www.aliexpress.us/item/3256805888457447.html

They sort of help. If I stack bus bars on bus bars, sometimes 3 deep, I can make it work. Unfortunately, their 400A rating is more than a little optimistic. They are about 1" x 1/8" (I say about, because they are 2mm thick and1/8 is 3mm), and at 400A a 1" x 1/8" copper bar will be 65C over ambient. 65C is 150F. If you are at 80F ambient, you are looking at bus bar temperatures of 230F. Wow! And since these are 2/3 of that 1/8", the temps are even higher.

This calculation was done on the cross style bus bar. The linear style is much worse. They don't provide width, but I think they are closer to 3/4" wide, making the temps really wild.

The cross style (end-for-end) is actually the easy part, as I can use two or even three. The linear style is the hard one, as not only can't I double them up, but they don't make them in a size to support 6 cells, so I'll have to stack them.

 

[HRTKD]

Instead of stacking busbars I would purchase new ones that are the appropriate size.

Here's a post against stacking of busbars.

Why Not Stack Busbars?

Why it's not a practice to use such busbar spacing insulator arrangement as it's showed in the picture from attachment (one busbar above other)?

www.physicsforums.com

 

[sailingharry]

Instead of stacking busbars I would purchase new ones that are the appropriate size.

Here's a post against stacking of busbars.

Why Not Stack Busbars?

Why it's not a practice to use such busbar spacing insulator arrangement as it's showed in the picture from attachment (one busbar above other)?

www.physicsforums.com

We may be slightly misunderstanding here. When I said "stacked" I'm not talking about increasing capacity (ie, doubling up). I'm talking about the fact that I need a single bus-bar to run 6 cells long, and they only make them up to 4. To get to 6, I'd need to run a 4 cell bar, and then mount a 3 cell on top of it to get to the last 2 cells. In addition, since I need to end-connect a 3 cell wide block to another 3 cell wide block, I'd ideally want some sort of 3x2 bus bar, but I could make do with 2 3-cell long bars, and 3 2-cell wide bars stacked on top of the long bars.

But in either case (not thick enough bars, or not long/wide enough bars), the problem is the same -- no one appears to make a legitimately rated flexible bus bar that works (I've not been able to find a 400A bus bar that is actually suitable for more than 200A). Hence, my thought of building custom rigid bars that are both long/wide enough, and thick enough for the job. To be clear, I would MUCH rather buy flexible ready-made, if they existed.

 

[HRTKD]

Hence, my thought of building custom rigid bars that are both long/wide enough, and thick enough for the job. To be clear, I would MUCH rather buy flexible ready-made, if they existed.

I would have fabricated a custom set already. I have copper bar stock in the garage that would be enough to at least make a prototype.

If you can't find what you want off-the-shelf and you don't want to make it yourself then reconsider paralleling cells in the battery. 4s3p works just fine as long as you get the cable lengths right.

 

[sailingharry]

I'm leaning heavily toward making my own, as nothing ready-made is suitable. I do worry about the rigid bus bars tying 6 cells together, but it doesn't seem to provide real-world disasters that discussions over a beer would indicate.

4S3P is a bit tempting for various reasons, bus bars being high on the list as well as fusing, cable size, etc. However, it does add complexity in BMS and controlling charges and loads. I'll be using an Electrodacus, and 3x (plus 3x shunts, plus....) gets expensive. But then you have to address issues like "does an out-of-standard event open a combined main contactor, or do I put three contactors in?" And from a practical matter, I no longer have a single SOC readout, but 3 (which hopefully will all be within a reasonable tolerance of each other).

 

[Supervstech]

FYI, aluminum bar stock works great and is easy to source.
Clean and oil the surface like you would the cells, and it works great.
Use 1/4" x 2" or even use angle stock.

 

[sailingharry]

FYI, aluminum bar stock works great and is easy to source.
Clean and oil the surface like you would the cells, and it works great.
Use 1/4" x 2" or even use angle stock.

The battery terminals are aluminum. That all by itself validates using aluminum as a bus bar. But damn, being on a sailboat, it is so hard to accept that idea! It's not a scientifically explainable thing -- it's pure gut. LOL. NoOxId Special A should solve any issues with corrosion on the contact surfaces, and corrosion on the rest is just a cosmetic thing. Maybe wrap it in heat shrink to make it pretty.....

 

[Supervstech]

The battery terminals are aluminum. That all by itself validates using aluminum as a bus bar. But damn, being on a sailboat, it is so hard to accept that idea! It's not a scientifically explainable thing -- it's pure gut. LOL. NoOxId Special A should solve any issues with corrosion on the contact surfaces, and corrosion on the rest is just a cosmetic thing. Maybe wrap it in heat shrink to make it pretty.....

41 Feet 2" Black Heat Shrink Tubing Roll 3:1,No Wrinkle Electrical Shrink Tube for Wires,Marine Grade Shrink Tube with Adhesive Lined Waterproof https://a.co/d/9589EG2

 

[Marinepower]

I would make some custom busbars (tinned-copper or AL) but to bridge the 3 cells in paralel only.

You may find making 6 perfectly spaced holes in 1/4inch tinned copper a challenge (and not necessary). Use the 2 -hole bus bars that came with the cells to connect the 3-packs in series. Double them up if you wish.

If there is minor resistance btw. one parallel 3-pack to the next parallel 3-pack in series, that is not a problem and something the BMS can sort with balancing.

I have a 4P4S pack and this is how i approached the same problem. So far so good after a year. I also use Electrodacus.


Also, don't connect the 2nd 3-pack and the 3rd 3-pack with a "wide" bus bar. That connection in your drawing is wrong / sub optimal and will result an unbalanced drain on the far cells of pack 2 and 3 ( by far i mean the cells furthest in distance from the 12volt bat's main pos / main neg terminals) I would only connect the last (furthest) cells to make the connections btw pack 2 and 3. Otherwise all the current will by-pass the more distant pack 2, pack 3 cells.

 

[Marinepower]

See this very long post on current sharing .........https://diysolarforum.com/threads/calculation-of-parallel-string-battery-currents.37937/

 

[tinyt]

The battery terminals are aluminum. That all by itself validates using aluminum as a bus bar. But damn, being on a sailboat, it is so hard to accept that idea! It's not a scientifically explainable thing -- it's pure gut. LOL. NoOxId Special A should solve any issues with corrosion on the contact surfaces, and corrosion on the rest is just a cosmetic thing. Maybe wrap it in heat shrink to make it pretty.....

Maybe after final assembly paint exposed aluminum surfaces of the bus bars? Just an idea.

 

[Q-Dog]

On a boat I would NOT build a huge battery with only one BMS. Build 2 smaller batteries each with their own BMS ... that way you don't lose all battery power if one BMS fails.

 

[sailingharry]

If somehow the BMS fails, and that failure somehow disconnects the battery (strange failure mode...it is more likely that a BMS failure would allow a fault condition to run unchecked), then I would need to manually override the battery switch to reconnect. Obviously, that solution requires that the failure be the BMS, and not the battery itself...

A significant issue with parallel batteries/BMS is that if one somehow fails, you probably are unaware that you are running on only one battery. I've yet to see an alarm system that notifies a user of a disconnected battery.

 

[tinyt]

If somehow the BMS fails, and that failure somehow disconnects the battery (strange failure mode...it is more likely that a BMS failure would allow a fault condition to run unchecked), then I would need to manually override the battery switch to reconnect. Obviously, that solution requires that the failure be the BMS, and not the battery itself...

A significant issue with parallel batteries/BMS is that if one somehow fails, you probably are unaware that you are running on only one battery. I've yet to see an alarm system that notifies a user of a disconnected battery.

If you have a monitoring system for each battery/BMS discharge current, maybe comparison with historical data might help.

 

[Ohms_Cousin]

FYI, aluminum bar stock works great and is easy to source.
Clean and oil the surface like you would the cells, and it works great.
Use 1/4" x 2" or even use angle stock.

If one were to fabricate thier own aluminum bussbars, would cooling fins on the bussbar help to keep bussbar temps down? I am thinking, yes it would - at least to some degree and especially if there is air blowing over the cooling fins. Why would we want to do that? reading Harrys post above, he says that the bussbar dimensions he talks about in copper can hit 230F !! So if we machine our own bussbars out of Aluminum then we can make the connectors, thicker, wider (best) and machine cooling fins in to help dissipate some of that heat being produced.

Is this idea worth pursuing or not of much help at all ? I am really wanting to tune my setup to the enth degree. Every little helps.

 

[Supervstech]

230F is only 110C, so not overly high temp for a bussbar.
I mean, you can boil water on it, so dont lean on it or anything...

 

[robbob2112]

I think I would look up metal fabricators near you and get them to make you a set of custom bars out of copper and the right cross-sectional area to run cooler. I.e. thicker and wider so less resistance meaning less heat. Local places generally will cost you, but they can turn out a quality product in days or hours.

If all else fails they will generally sell copper bar stock in any dimensions you want and you can skip the shipping.

 

[Philipay]

I am building my house battery on my 42 sailboat using 4S 4 group 12v system
Each group has an Individual BMS to regulate with 1/0 awg wire connected to a busbar and my Victron 3000 inverter will connect to the busbar with 4/0 cable.
Lets assume my inverter is at full load 250a. Each group of the battery bank will share the 250a, in my situation each battery group will only draw appox 62.5a each. Thus I can use smaller cable for each group. However im still using 1/0 cable in case 2 of the battery group failure, with only 2 groups of battery sharing the 250a load, I can still keep the power system online. For monitoring I will have my battery BMS (Daly 100) connected to Victron to monitor my system's health.

For high load busbar I suggest this kind of busbar, it is a multilayer busbar which can flex and rated 200a which I believe according its actual thickness.

 

[sailingharry]

You may find making 6 perfectly spaced holes in 1/4inch tinned copper a challenge (and not necessary). Use the 2 -hole bus bars that came with the cells to connect the 3-packs in series. Double them up if you wish.

I have a 4P4S pack and this is how i approached the same problem. So far so good after a year. I also use Electrodacus.

Also, don't connect the 2nd 3-pack and the 3rd 3-pack with a "wide" bus bar. That connection in your drawing is wrong / sub optimal and will result an unbalanced drain on the far cells of pack 2 and 3 ( by far i mean the cells furthest in distance from the 12volt bat's main pos / main neg terminals) I would only connect the last (furthest) cells to make the connections btw pack 2 and 3. Otherwise all the current will by-pass the more distant pack 2, pack 3 cells.

Thanks for the input.

A few questions/comments.

* Making the holes is likely to be easy/fun. I have just come into access to a numerically controlled mill. So I can drill precisely spaced holes!

* The bus bars that come with the cells would be very good for tying the cells end to end, as 3 bars have plenty of capacity. However, they won't go end-to-end, as the batteries are two-hole, and the bus bars are drilled for side-to-side. Rats!

* Great picture. But I'm confused. I don't see any bus bars or connections of any sort. Also, those big honking terminal posts are awesome -- how did you manage that? Finally, I love those snap-on covers, a great safety improvement.

* Your comment on unbalancing the current between packs 2 and 3 make sense. But I am not sure how to do that. Should I make each (2 and 3) have a 3-battery bus bar with a terminal stud in the middle, and then make a 8" #4/0 jumper to jump between them? I do wonder about the real-world implications of unbalance. When my 600 (or 900) Ah bank routinely sees under 10A of draw (that's my daily average), the voltage drop across stuff rated for 40 times that current is extremely small. I would think that the cells would spend .1 hours a day getting out of whack, and 23.9 hours a day self balancing -- and any imbalance of a few Ah would only show at the very top and bottom of the SOC cycle (where the first cell to reach the knee would stop contributing to load or charge, returning naturally to balance).

Thanks again!

I fear about the tax breaks for solar next year, so am intending to make most of my purchases this year -- which is getting really short really fast!

 

[HarryN]

Drilling into copper is a little bit of a learning curve.

The same things that make it soft also make it want to create burs and grab.

The pros that make bus bars punch it rather than drill most of the time.

There are a number of bus bar mfg companies around in the US - just do a search using any browser. You can buy it tin plated ( much better for most uses ) or plain.

The US has a lot of copper mining capacity and China ends up importing material to keep up, so it may or may not make sense to go that path.

 

[sailingharry]

I am building my house battery on my 42 sailboat using 4S 4 group 12v system
Each group has an Individual BMS to regulate with 1/0 awg wire connected to a busbar and my Victron 3000 inverter will connect to the busbar with 4/0 cable.
Lets assume my inverter is at full load 250a. Each group of the battery bank will share the 250a, in my situation each battery group will only draw appox 62.5a each. Thus I can use smaller cable for each group. However im still using 1/0 cable in case 2 of the battery group failure, with only 2 groups of battery sharing the 250a load, I can still keep the power system online. For monitoring I will have my battery BMS (Daly 100) connected to Victron to monitor my system's health.

For high load busbar I suggest this kind of busbar, it is a multilayer busbar which can flex and rated 200a which I believe according its actual thickness.

Parallel batteries with parallel BMS makes for a real coordination problem. When one battery is full, it should call for shutdown of charge systems -- but if other batteries are still charging you don't want to shut it down. This is especially true of the alternator -- if any one battery is nearing a high cell condition, you need to shut down the alternator, but other batteries still need charge. I guess wiring all the controls signals in parallel would solve it -- charge continues until all batteries call for shut down. It also in your case quadruples your BMS cost.

Using FET style BMS is a partial solution, but leaves open the possibility of an alternator dump. I've also not seen any FET style that include advance warning of anything. Advance warning of full is almost pointless, as all charge sources SHOULD shut down at full, but advance warning of a single high cell is a bigger concern. Advance warning of low is truly pointless, as you should aready be aware when your bank is at 20% and start charging. SOC is a bit of a challenge unless all the BMS (or your Victron gear) coordinate to give a single SOC value.

Don't forget that by ABYC E-11, you need Class-T fuses for each battery when paralleling them, and within 7" of each positive terminal. And, you will need the battery cables be fused at each end. Each 1/0 battery cable can only support 285A. If you've fused each with a 200A fuse at the battery, you have a potential short circuit current of 600A, a real problem. Consider a short to ground in a battery cable between the battery and the bus bar. The fuse at that battery will blow, no problem. But the remaining 3 batteries can each supply 200A for a total fault current of 600A. The solution is either a 200A fuse at each end of each battery, or 100A fuses on each battery (limiting fault current to 300A, or not much over 280A).

I would love to find that style bus bar. Unfortunately, they are "Fake rated" at 200A and by dimensions and industry standards are more accurately rated at closer to 100A, and with my current realistically hitting over 200A (I can charge at close to 300A and I have a 3KVA inverter) are way undersized. They also do not have a version allowing end-end connections, only side-by-side, and in any event require stacking 2-3 deep to make the required connections.