cleaning buss connections 1

[sparky970]

We have a rectifier where I work that produces 100,000 amps DC at about 330V. It has copper splice plates that mate to aluminum buss off the load side of the rectifier. Last week, we found one of these connections in excess of 520 degrees F. After about 12 hrs of finding the problem, management finally decided to shut down the plant to repair the problem. When we disconnected the splice plates they were very black on the mating surfaces. We ended up using scotchbrite pads, fine sandpaper, and HP Cleaner/Degreaser, which took some time to get them clean. We were about 75% done with the cleaning, when the Maintenance Manager stopped by. He wanted to know what was taking so long and said we should be using grinders with wire wheels or the wire cups to clean them. I told him I didn't think this was a good idea because of the grooves and valleys this would create and that it is important to have a flat smooth surface to make the best connection. He scoffed and made comments to other employees that I was f-ing him to death by taking so long. Any thoughts about using a wire wheel to prep the surface.

 

[davidbeach]

Stick to your guns. It got hot in the first place because it had less than the required contact area/pressure. Now, if you can groove both sides so that all the grooves fit together and you have more contact area you will be even better off than perfectly flat. On the other hand, if you groove one up/down and the other lengthwise you will have very little contact area and really bad heating problems. Best surface unless you can control the machining of mating grooves is as flat as possible.

 

[itsmoked]

Good on you. I will say that that copper oxidation is extremely hard and I'd have considered a hand power sander coarse paper, followed by fine sand paper.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[edison123]

How about some mild acid cleaning (followed by water cleaning) ? Using steel wires would surely pit the soft copper and the contact area would only get reduced.

 

[LionelHutz]

Beyond what the others posted, I also don't think you want iron particles left in the joint that can lead to future corrosion.

I hope you used some type of electrical joint compound suitable for copper to aluminum connections. Also, take a good look at the hardware holding the joint together. Not having enough bolts will lead to a connection problem.

 

[sparky970]

When we raassembled the connections, we did use a light coating of Penetrox. The actual copper to aluminum connection is about 8" x 8" square and each has 6 bolts connecting them.

 

[alehman]

I would recommend Belleville washers on the joint bolts to allow for differential thermal expansion.

This seems kind of small for 100kA. U.L. for switchboards is 200 amps per square inch of contact area. I realize this is not a U.L. switchboard.

 

[edison123]

I agree.

64 sq. inch for 100 KA looks to be small. Is the bus force cooled ?

 

[sparky970]

Sorry, to clarify, there are actually multiple connections. There are 8 connections of 64 sq. inches on the positive and 8 connections on the negative side of the buss. I surely wish management would listen to my recommendations. We operate a chlor-alkali plant with electrolyser cells which have thousands of copper to copper connections between cells which are all cleaned with wire wheels. In my opinion, it is only a matter of time.

 

[alehman]

Do you know if this joint had been wire-brushed previously?

 

[edison123]

8 of 64 sq.inches. That's more like it.

BTW, you see the copper blackening on the mating surfaces or on the exposed surfaces ? If it is latter, it is probably due to the chemical environment and nothing to worry about (unless it was eating away the copper)

 

[Zogzog]

"BTW, you see the copper blackening on the mating surfaces or on the exposed surfaces ? If it is latter, it is probably due to the chemical environment and nothing to worry about (unless it was eating away the copper) "

That can develop into whiskers that can cause a fault.

 

[sparky970]

I'm am not sure how the contractor that oringinally assembled the connections cleaned them, but the exposed surfaces were definately black. This is not the first time this has happened. We also had other connections with the same issue. The last time, it actually melted a big chunk of the aluminum buss and we had a contractor fill in the voids with a MIG welder and re-flatten the surface. By the way we are using grade 5 bolts and bellvilles.

 

[edison123]

zogzog - what are them whiskers ?

sparky - 100 KA Al bus and the "management" decides to MIG weld that ? Aluminum is notoriously hard to weld properly due to porosity.

Wouldn't it have been better to replace that Al bus with copper and get rid of the root cause of overheating and failures ?

 

[LionelHutz]

Isn't it either the tin plating or the silver plating that develops into wiskers?

Are you using the correct size of Bellville washer? They should be thick heavy things and not about the same thickness as a standard washer. You may also need to stack a few to get the right spring force. More then once I've seen bellville washers specified and then these thin washers supplied that are wrongly speced and basically useless in the application. The bellville basically needs to be capable of the same spring force as the tension on the bolt when torqued.

You may also want to look for a different joint compound that can better protect in that application.

 

[waross]

Yes aluminum is hard to weld. However there are welders that have been trained to MIG weld aluminum bus joints. Some manufactures hold training courses for aluminum welders so that they can depend on the field welds. This work is often X-rayed to prove the absence of porosity. Most contractors doing substation or power station work have factory trained aluminum welders on staff or on call.
A large contractor may have dozens or hundreds of good steel welders available. Only the very best will be chosen to take the factory aluminum welding courses.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[stevenal]

After cleaning, you should tin plate the copper contact surface using solder, flux, a torch, and a rag.

Make sure you use non-gritted anti-oxidant.

Agree, make sure you spec the Bellvilles. Simply ask for a Bellville and you will get a washer that flattens out after you torque it, and stays flat.

 

[NAZ55]

Did I hear connection between copper and aluminum?
Where's jghrist when you need him? I thought it's always a mess to connect copper to aluminum unless you have some sort of bimetallic connector as a transition plate. jghrist had a good article on that.

Bottom line aluminum and copper don't go together! use transition.

 

[alehman]

Make sure stevenal's tin plating is perfectly smooth and flat or it will interfere with proper contact between the surfaces.

 

[stevenal]

Some like the transition plates. Others consider adding a second connection surface to be a good way of doubling the chance of having a poor connection.