How do I determine busbar junction resistance?

[matejcek]

I am looking for advice on determining 'tyupical' busbar junction resistances. My applications run up to 50kADC. Usually minor losses aren't critical to me, but sometimes I need to be able to estimate what's going to happen every 12 feet when I bolt all that copper together. Any advice?

Paul Matejcek

 

[electricuwe]

The copper development assosiation (UK)

http://www.cda.org.uk

publishes a publication "copper for busbars", which contains information on that topic.

For a good connection (suitable area, flat, free of oxide and bolted with hardware giving the right force in the whole temperature range) the loss is usually negliable.

 

[matejcek]

electricuwe,

Thanks for the lead. I didn't find that document on the US CDA page. I briefly browsed the article on joints at the UK site, and it looks as though it will be helpful.

Regards,

Paul Matejcek

 

[busbar]

Maybe I misunderstood the question, but measuring joint resistance requires a Kelvin connection. First, a DC current is passed through the joint with one pair of leads. [Microohmmeter are available with up to a 500-ampere test current, so the leads may need to be fat.] Then, the voltage drop is measured with a second set of leads [with small AWG] connected to a (milli)voltmeter. Then: R=V/I, making sure the decimal point is in the right place. Simply, if precisely 1 ampere is passed, the each measured millivolt equates to 1 milliohm.

 

[matejcek]

busbar,

Thanks -- I understand Ohm's Law, and how to get the data empirically. My questions was about predicting the resistance. Based on what I have learned today, it seems not to matter, because a well-made overlap joint (the sort I'm using) has a lower resistance than a similar length of bus.

Thanks again,

Paul Matejcek

 

[Traian]

Very simple!

Run 100Amps DC from a current generator and read the voltage drop on the contact. The higher the temperature the higher the drop voltage.

That means resistance is not constant!!!

Traian

 

[electricpete]

Paul - I agree that measurement is not relevant since we now understand your interest is in predicting junction resistance.

But I wanted to mention that Busbar's point was not simply a statement of ohm's law. He was explaining the 4-wire resistance measurement principle which almost entirely eliminates errors from lead resistance and probe contact resistance. Simple resistance measurements with 2-lead multimeter/ohmmeter are not nearly as accurate when measuring low resistances. Also a good discussion by Traian.

 

[matejcek]

Thanks to all who have replied. What I have subsequently learned probably should have been intuitively obvious: a properly made overlap joint has less resistance per unit length than the bus it joins. A well-made joint is 'better' than a continuous run of bus.

Paul