Copper pipe, 1/2-inch. What is max current capacity?

[Ample]

I imagine that people smash or hammer the copper into inconsistent thicknesses. That also effects the contacts and how well the bolts hold.
I've got a bunch of bus bars to make. I'm thinking copper bar is probably worth the extra cost. Especially considering that it will be easier to stack them in a vice to drill the holes. At least that's what I'm thinking now. I have some time before I have to make them.

It helps if you have the right equipment. I've seen people use hydraulic presses. Since I don't have one of those, I used a sledgehammer and an anvil. The side that was laying on the anvil is nearly perfectly flat.

For sure copper bar would be better but I couldn't resist making 2 bars for what amounts to less than $2 CAD apiece.

I'm not actually recommending it because in the end, it was a bit of process to make. Fun, but some amount of work.

 

[Ample]

I have 4 separate copper pipe busbars on a midsized 24v system with a 3000W multiplus. I periodically check for heat and my busbars have never even gotten warm. That said, I've never loaded my system with more than about 90A and peak average is more like 50A.

I don't overdo things either. I'm a mobile situation and the max I was running was 100A with no problem. Now that I am of the belief that I can use 200+ A, I might use up 125 to 150A with confidence.

 

[Ample]

Interesting idea if one has a lot of undersized pipe. But the pipe then needs to be covered - unlike most wire. It would be almost twice as much work.

A small bag of lugs isn't too expensive. So I likely won't go as far as using them for lugs. I find drilling holes to be a pain. That's why I'm not really recommending people do this. I had to drill 6 or 7 holes for the bolts in the bus bar; and then again for the other one...

 

[berksrunner]

A small bag of lugs isn't too expensive. So I likely won't go as far as using them for lugs. I find drilling holes to be a pain. That's why I'm not really recommending people do this. I had to drill 6 or 7 holes for the bolts in the bus bar; and then again for the other one...

It’s easier with the right tools. Drill press would be ideal, but I didn’t have one when I did my van build. I now have makita‘s most powerful hammer drill that I think would make quick work of it, together with a nice step bit.

 

[svetz]

...I'm getting different results...

Welcome back to math!

I'm not trustworthy pushing buttons on a calculator... so likely that... let's see....

π (15.875/2)^ - π (13.843/2)^ = 47 mm^2 so that matches your number
Next if we calculate the diameter of solid wire with that area: (47 / π)^.5 x 2 = 7.7 mm, or gauge 1 or 211 A
Let's see, your actual busbars are 2.13 x 23.66, so the area is 50 mm^2, the equivalent solid wire diameter = 8 mm, AWG 1.

The nice thing about "flattened" is the increases surface area which increases the heat transfer rate, so theoretically can carry more current if it doesn't increase resistivity. For example, a round wire has 2πrL surface area. But a rectangular one has 2(w x h) + 2(w x L) + 2 (h x L). For example, an 8mm diameter wire has a surface area of 25L, whereas an 2.13x23.66 mm rectangle has 6 sides, so just doing the big ones: 2x(2.13L) + 2x(23.66L) or greater than 51.58L, that is over double the surface area over the wire.

That's the theory anyway... I'd be worried about tossing the 20% safety margin (I know, the wire charts already have a safety margin built in). At least until I confirmed it with testing... keep in mind silly things happen all the time (The power of Murphy makes it so). If you can't run that many amps safely, take how many amps you can run and back calculate the width of a bus bar. Then cut one back to that width for testing. Can't be to safe! Plus, it'll make a fun post to see how well practice meets theory.

 

[Ample]

Welcome back to math!

I'm not trustworthy pushing buttons on a calculator... so likely that... let's see....

π (15.875/2)^ - π (13.843/2)^ = 47 mm^2 so that matches your number
Next if we calculate the diameter of solid wire with that area: (47 / π)^.5 x 2 = 7.7 mm, or gauge 1 or 211 A
Let's see, your actual busbars are 2.13 x 23.66, so the area is 50 mm^2, the equivalent solid wire diameter = 8 mm, AWG 1.

The nice thing about "flattened" is the increases surface area which increases the heat transfer rate, so theoretically can carry more current if it doesn't increase resistivity. For example, a round wire has 2πrL surface area. But a rectangular one has 2(w x h) + 2(w x L) + 2 (h x L). For example, an 8mm diameter wire has a surface area of 25L, whereas an 2.13x23.66 mm rectangle has 6 sides, so just doing the big ones: 2x(2.13L) + 2x(23.66L) or greater than 51.58L, that is over double the surface area over the wire.

That's the theory anyway... I'd be worried about tossing the 20% safety margin (I know, the wire charts already have a safety margin built in). At least until I confirmed it with testing... keep in mind silly things happen all the time (The power of Murphy makes it so). If you can't run that many amps safely, take how many amps you can run and back calculate the width of a bus bar. Then cut one back to that width for testing. Can't be to safe! Plus, it'll make a fun post to see how well practice meets theory.

Awesome! Thanks for verifying my math.

I will test my real-world situation with a heat gun and some big current draws when I get a chance (currently rebuilding my van).

It's good to know that the flat bus bars could transfer more than a wire/cable of a similar amount of copper.

Some more questions, if I may.

1) Does the insulation around a single cable make any difference? The AWG chart we've been using notes 2 values for max current: "in air" and "in jacket". The "in jacket" value is much less at 119A. While I may not often (ever) exceed 200A, I do plan to exceed 125, maybe even hit 150A (my inverter can handle it and now with 3 x 100A-capable batteries, I plan to take advantage). My understanding is that "in jacket" means a bundle of wires in a jacket like a corrugated plastic wire loom.

2) Does it matter the distance between 2 cable lugs on the bus bar? My bus bars have 6 or 7 bolts. I intentionally put the lugs for the 2 highest current devices next to each other on the bus bar (battery bank and 3000W inverter). Will this help? The 2 bolts are less than 1 inch apart.

 

[svetz]

Keep in mind everything I know I learnt off the internet... that is I'm not a "real" expert. Even if it was from a real expert, I'd still say test everything to be safe. For example earlier you mentioned copper pipe was 99% copper...but keep in mind a little goes a long way. For example as we know from transistors.... silicon is an insulator but with just a little doping becomes a conductor.

...It's good to know that the flat bus bars could transfer more than a wire/cable of a similar amount of copper.

Assuming it doesn't change the resistance....not sure about that.

...1) Does the insulation around a single cable make any difference?

Insulation typically restricts the outwards heat flow, and that limits the current. My understanding of jacketed wire is that it is insulated wire in a bundle with other insulated wires that has yet another outer coating. So, more blocking of outward heat flow.

...2) Does it matter the distance between 2 cable lugs on the bus bar?

If that's what I think you're asking first see post #8 in the Battery FAQ and you might also be interested in the discussion in Incrementally adding DC batteries.

 

[GXMnow]

Some wire charts also list the current for bare wire vs insulated wire. Even one layer of insulation holds in more heat than a bare wire in free air, and a flat bar in free air could do even better. As for the pipe, leaving it round or mashing it flat does not change the outside surface area. We just eliminate the inside surface area. Too bad air can't really flow on the inside once the ends are flattened to make terminals. A circle of solid wire would likely have the most even distribution of current through the conductor. A flat bar may have more current in the middle and a bit less towards the edges near the bolt termination. Over a length, it will likely even out though. On DC current it is all about the cross section area and heat dissipation. At 60 Hz AC it does not change much until the conductors are huge, but at RF frequencies, the current starts to flow more on the surface.

The current capacity of a buss bar is not effected by it's length, but the voltage drop and lost heat will increase as the bar between the current source and the loads become further apart. The shorter the bar, the less resistance, so less voltage drop at a given current. I used all the closest positions on my Bay Marine Buss Bars. At only 30 amps, just one lug apart, I see 0.002 volts of drop. If I used the far ends of the 4 position bar, it would be 3 times the drop, maybe a bit more as the lugs are closing some of the gap as well. Obviously, this small drop is not a problem, but at 200 amps, it will add up.

At the same 30 amps, the 8 feet of 2/0 cable, just the negative lead, is dropping 0.029 volts. That was more than I was expecting. My 200 amp rated JK BMS is only dropping 0.025 volts including the two times 4 inches of 2 x #7 awg wires going in and again, back out. With the case on the BMS, I can't measure on the board side of the wires, but this is less loss than expected, so I am happy. I am pretty confident the batter bank will not give me any issues at 150 amps. That is about the most the XW-Pro inverter can pull long term. The surge rating of 12,000 watts could try to pull 240 amps. The charge side tops out at 140 amps. The factory buss bars on my LG Chevy Bolt packs look very small for the current. Between cells, they are almost 2 inches ( about 45 mm) wide by memory, but less than 1/16th inch (1.2 mm or so) thick, so they seem okay for the 125 amps I could pull through them (I have 2 packs in parallel). But the ends of each pack where it comes out to the terminal, seems even thinner, maybe 1 mm, and is only 3/4 inch (20 mm) wide over a longer distance. So far the bars feel cool, but I have not topped 50 amps in any test yet. When I reconfigured the 8S pack into two 4S packs, I had to make my own terminals. I used copper sheet that is about 1.7 mm thick on my calipers. I have it the full height of the buss bar, and it extends out just an inch with the bolt to hold the terminal. I made a clamp from 1/8 inch aluminum angle with a pair of bolts to clamp my copper plate to the factory copper buss bar for the entire height. At 20 amps, I measured the drop from the cell tab to the wire, and found my terminals have a little less drop than the factory end terminals, so I think my connections are sound. Each string is separately fused at 125 amps. Then the pair is fused at just 150 amps.

 

[svetz]

...As for the pipe, leaving it round or mashing it flat does not change the outside surface area...

True! But, that's not what the math was about in #25... it was comparing a 5/8" OD copper pipe smashed flat to solid round #1 wire (roughly the same weight of copper per length theoretically). So, the surface area of the bar is indeed over twice that of the rod, math says its so anyway (unless I messed up).

...The current capacity of a buss bar is not effected by it's length

Technically it is effected ... as resistance increases with length and resistance blocks current flow you need a higher voltage drop to push more current.
We just don't generally think about it that way because yeah... when you measure it you see the volts go down and the amps stay the same because we almost always work with voltage sources. But, if the voltage drop were fixed, the current would drop accordingly with Ohm's law as resistance increased.

 

[Mars]

I looked at the ampacity of copper pipe when I used some copper pipe for bus bars and came to the conclusion that the copper pipe was good for something between 150 and 285 amps. I also realized the copper pipe was not likely to be the limiting factor. The limiting factor would be the connection between the battery terminal and the bus bar.

My estimate of the ampacity is based on flattened copper pipe having dimensions and cross sectional area approximately equal to 1/16" thick bus bar between 3/4" and 1" wide and this ampacity table for copper bus bars:

Electrical: Bus Bar - Table 1: Ampacities

Ampacities in the table below are for bus bars having an emissivity of 0.4.

www.copper.org

 

[fat_old_sun]

Ampacity does not scale linearly with the cross-sectional area at large current magnitudes. The reason is that the magnetic field induced by the current flow pushes the electrons away from the core of the conductor - in other words, the current density (Amps/cm^2) is not the same across the cross section. For this reason, high current conductors are usually made of fine strands of enameled (insulated) copper instead of thick wires or rods.

There's another effect that comes into play when a copper pipe is cold-worked and flattened into busbars. The process introduces defects & dislocations which impede electron flow (higher resistivity). Proper annealing will reduce the resistivity.

 

[Mars]

Ampacity does not scale linearly with the cross-sectional area at large current magnitudes.

That is absolutely true. Ampacity has more to do with the surface area and what is an acceptable temperature rise.

 

[Leo3]

Is it safe to use copper pipe to attach to a shunt like this pic?

 

[grizzzman]

Is it safe to use copper pipe to attach to a shunt like this pic?



View attachment 25775

It works just fine. The "REAL" problem will be corrosion. ( BTDT )

 

[chrisski]

I wouldn’t mind seeing peoples completed pics of these copper busbars to include the lugs and however they are mounted. I thought about doing this, but with tooling up with a 3 LB sledge and an Anvil, or just going to the hydraulic press, Blue seas came out much cheaper.

For them being exposed to air, I haven’t tried it yet, but its supposed to be possible to tin copper with a MAF torch and solder you get from Home Depot. I saw an end picture off a frame ground and it looks like a really good idea.

 

[Hdonly]

True! But, that's not what the math was about in #25... it was comparing a 5/8" OD copper pipe smashed flat to solid round #1 wire (roughly the same weight of copper per length theoretically). So, the surface area of the bar is indeed over twice that of the rod, math says its so anyway (unless I messed up).


Technically it is effected ... as resistance increases with length and resistance blocks current flow you need a higher voltage drop to push more current.
We just don't generally think about it that way because yeah... when you measure it you see the volts go down and the amps stay the same because we almost always work with voltage sources. But, if the voltage drop were fixed, the current would drop accordingly with Ohm's law as resistance increased.

Technically, it is not effected. It may affected though.

 

[dpk1nba]

I imagine that people smash or hammer the copper into inconsistent thicknesses. That also effects the contacts and how well the bolts hold.
I've got a bunch of bus bars to make. I'm thinking copper bar is probably worth the extra cost. Especially considering that it will be easier to stack them in a vice to drill the holes. At least that's what I'm thinking now. I have some time before I have to make them.

even with shipping the copper bar is comparable in price to copper pipe at the home center. - https://www.onlinemetals.com/en/buy/copper-rectangle-bar

 

[sshibly]

even with shipping the copper bar is comparable in price to copper pipe at the home center. - https://www.onlinemetals.com/en/buy/copper-rectangle-bar

shipping usually kills online deals, did anyone order from these guys online? who bad was shipping?
I just ordered off Grainger
.125x.5x72 inch copper stock was $28 for free pick from their store, which is 3 miles from me.

 

[dpk1nba]

shipping usually kills online deals, did anyone order from these guys online? who bad was shipping?
I just ordered off Grainger
.125x.5x72 inch copper stock was $28 for free pick from their store, which is 3 miles from me.

Just put what you would purchase into the shopping cart and go through the checkout process without actually purchasing and it will calculate shipping cost. I bought a 1/4 x 1" x 12" bar (to make (2) 6 inch bus bars) for $26.48 including shipping.
I just ran the same copper plate thru the Grainger website and it came to $33.42 including shipping. ( I don't have a Grainger close by)