Contact Resistance Test

Just add a Cinemascope infrared imaging camera, and look for any hot spots when the full loads are applied.

Nice paper. It's a shame the OP will not or cannot take the time to study this paper. It answers precisely their question.

Maybe they reject this reality and want to substitute their own.

Maybe they did study the paper and still cannot understand it.

I know. You can lead a horse to water, but you can't change their spots.

all your answers is not helping, we need to make this test before apply full power, and if we make this test for all line then we will include resistance for busbar it self plus the resistance for joints, but we need to check only the resistance for the joints only, and if we make this test for joints by joints it will take one year to finish, which not acceptable at all, so whats your opinion ??

Thank you

There are 2 aspects to the question your are asking, the physical aspects and the moral aspects. The physical are simple enough that much elaboration should not be necessary for a qualified individual.

The statement that they don't want to apply full power before this test is complete brings up the moral questions. If it CAN blow up when full power is applied and someone is saying they cannot properly test it, it seems that a huge amount of liability, moral and financial, will come in.

It also seems that A LOT of individual joint tests can be done in a year. If the project is that big they are not done and the project is completed and does not work because of a high joint resistance problem somewhere, if it does not blow up and make it obvious where the problem is, finding the problem may be very difficult and time consuming - expensive. -- JHF

Is there sufficient time to attend the funerals of the observers that will not have survived the experiment (rhetorical question - NNTR)?

Is there sufficient time to hire additional personnel to carry out the testing, thereby shortening he delivery time (rhetorical question - NNTR)?

Is the outcome of this experiment going on YouTube, please?

VOM when the power is off.

Rami good afternoon any electric device or equipment do you want to test it in order to detect any probable defect .you should subject your doing to real test

this is electric don't say i don't want the job takes a long time. if you ignore such like this important step maybe cost you too much .

wish to you finish your job without problems.

With a magic ground bond tester that subtracts the resistance of each bus span from the total resistance accross all the joints.

Apply a reliable, stable DC current source capable of providing maximum rated voltage and current to each bus bar while taking voltage drop readings across each joint using a high resolution VOM to record the readings as they are taken.

After the voltage drop across each and every joint/connection has been recorded, compare the results to identify if and where there are any high resistance connections.

From your post it appears that the installation is being performed now.

1. Your buss bar installation contractor should be testing the integrity of every joint after the connection is made and give you a report of every test!
2. If the joint integrity test is not being performed, then someone went cheap and the final test will be a "power on" test.
3. Let the person who specified "no testing required" flip the power switch for the first time!
4. Stand back and video the results! Post on YOUTUBE and hope it goes viral!

Do a search on the term "Ductor test". Ductor was (is?) the brand name of an instrument made by Megger designed specifically for this purpose, essentially a micro-ohmmeter. The values to look for vary by the type of bus material and connection method, but a Ductor used to come with a book that listed typical values one could expect. When I used one in making MCCs with tin plated copper bus bars, we were always looking for 10 micro-ohms or less across each joint. But I can't tell you if that is appropriate for your situation, you will have to finish your own research on that.

I don't know of a test that would be of value for getting around testing each joint. I suppose if you determined the total number of joints in a series string and measured the total micro-ohmic resistance of the entire string, you could calculate what the total series resistance should be, including the natural resistance of the total length of bus bar material and if your value is higher, it indicates a problem. But that would not tell you which joint is the problem, or if it is more than one, etc. etc. That could actually lead to taking LONGER than just having the right instrument handy when assembling the bus joints, testing it the moment the joint is done, and move on. What are you adding, 30 seconds per joint? Why are you trying to shave off those seconds?

It might be possible to use a Time Domain Reflectometer as a "go-no go" test.

This type of meter if adaptable, would indicate the distance(s) from the point of applied signal to the high impedance connection point(s).

There is a myriad of this type of meter available from Fluke, Tektronic, Toshiba, Biddle, ?

A call to any technical support group of your preferred vendor would be a good place to start.

The contact resistance between two joined bus bars can generate heat and power loss when there is large current. To measure a relatively low resistance, you need to use the 4 point method. Two probes supply a constant current through the contact and the other two are connected to a voltmeter to measure the voltage drop. This allows you to measure a low resistance without consideration of the probe contact and lead resistance.

http://en.wikipedia.org/wiki/Four-terminal_sensing

What test manner is this and how much current will pass through the bus bar and what

a battery could hold on for a time while you are busy to look for a drop voltage at points (2,3) or others on bus bar i am sure you will see the huge spark immediately if you close this circuit and the voltmeter will not read.any thing .

The point is to run a measured current through the bus bar via terminals 1 and 4. Across each junction (2 and 3) measure the voltage. For example, if the current (1 and 4) is 10A, then if the voltage (2 and 3) is 0.5 volt, then the junction between 2 and 3 is 0.05 ohms.

If you instead measured across 2 and 3 with an ohmmeter, you would be measuring the junction resistance plus the resistance of the leads and the connections at 2 and 3. The added resistance is unknown and may well be larger than the junction resistance that you want to measure.

The "Four point method" avoids this problem because almost no current flows through the voltmeter.

I wonder what the drawings specify?

Rami, if the the joints are performed in a local industry acceptable manner for the busbar size and there are no connection errors eg shorts, then passing current through the bars replete with joints and using thermal imaging will identify any loose connections. That is what you are looking for? You don't have to do this at Imax.

The resistance check doesn't test the design it tests the assembly.

If everything gets hot then the design is flawed or you have exceeded the design Imax.

Curiously....Cu or Al busbars?

I suggest you measure the resistance of the complete bus with any instrument that provides the required accuracy and sensitivity. That measurement should be the benchmark for subsequent assemblies. Any assembly that shows a reduced resistance should become the new benchmark, and any assembly that shows increased resistance should be viewed as suspicious and tagged for detailed study.

An LM317, a resistor, and a heat sink mount will provide you with a usable current source. ( I like 200mA )

Measure the voltage drop across each connection to find poor ones. You cannot isolate a bad contact without individually testing it.

O.P. has not said anything about isolating or testing individual joints. If I understood his question correctly, he wants to verify the resistance of a bus assembly. This can be done by measuring individual joints, or by summing them. Also, most of the bus assemblies we test are powered by at least 100 amps. Highly unlikely that an accurate measurement could be made at 200 mA.

The fact that he has not said anything about testing individual joints does not mean that he does not have to do it. If on testing the whole system he finds a fault, what do you propose that he does to solve the problem?

I propose the use of PETN detonation cord to remove the faulty metal. Then replace the missing conductor with a new conductor. This now makes two junctions where there were once one.

What lame, boring technique would you use? Would you follow the engineering documentation? How quaint.

In an attempt to provide a constructive answer to the OP who clearly has a problem, either self inflicted or caused by a third party, I have already detailed my suggested lame, boring, technique in post 22. I designed busbar systems for many years so I was the person generating all than quaint engineering documentation. Yes, I would follow the documentation and not get myself into the mess that the OP now finds himself in. So I would not have posted the problem and had to suffer a barrage of snide remarks and sarcasm. I am on CR4 to try to help, obviously one of us is in the wrong place.

There is such a pleasingly old fashioned feeling about following the engineering documentation that was agreed upon in the contract. It also feels great when the engineering document I labored over to make sure it was clear, concise and out of necessity boring as hell to write gets followed. Don't you agree.

This statement leads me to the critical unspoken questions for our OP.

Does the engineering documentation for this project stipulate testing of joints as they are assembled? If it does then why didn't QA catch this mistake early? If it does not then who pays for this engineering change after the fact?

I almost forgot. Your technique in #22 is a very good, after the fact, testing approach. I've given you a belated GA. When I've used a similar technique after the fact I've rarely finished this kind of testing. One of two scenarios quickly happen. Either a systemic flaw in the assembly technique quickly becomes apparent. This causes the whole assembly to be dismantled and then properly installed. Or the time keepers of the project insist the random check just performed with only passing results is sufficient. Turn it ON.

Without knowledge of O.P's detailed requirements, it is difficult to come up with detailed recommendations. I would suggest that the bus assemblies be sorted into as many bins as required, and that the customer select from the low resistance bins. If higher resistance bins are encountered, they can be reworked or disposed of as warranted. The objective, as I understand it, is to speed the testing process and this is an accepted method.