cold creep - aluminum wiring 2

[AAlpha]

Hi,
I would like to hook up my new (actually used, but new for me) arc welder in my garage. I need a 50 amp 230 V outlet and am considering Al-wiring since it is about half as expensive as copper wiring. Did some research and found out about the cold creep. Found different opinions, from the suggestion to not use AL to it is fine to use but you have to check the connections every two years.

I am almost at the point where I decided to use the AL wire. Any final suggestions?

Thanks for any input on the subject.

 

[dpc]

The breaker in your panel and the receptacle must both be designed and listed for use with Al wire. The cold creep issue was a major issue back in the 1970s with small 15 A and 20 A branch circuits. The issue was addressed (more or less) and I would not be overly concerned in the larger amperage devices as long as the equipment was new.

Do an Internet search regarding termination of Al wire, follow the instructions, and it should not be a major concern with new equipment.

David Castor

http://www.cvoes.com

 

[rmw]

AA,

As long as you KNOW that you are going to have to retighten the connections periodically, then you can make the decision on that basis. It is those who don't know to check and/or retighten that get in trouble.

Not that I liked them, the Power Company brought their service in with AL wire and I had to get them out every couple of years to take the transformer out so I could retighten their connection.

During a 12 day outage due to a hurricane in our area, I went around and retighened all the AL connections at my home.

Since you know to do it, you shouldn't have any problem.

I hate AL wire, by the way.

rmw

 

[waross]

My understanding is that aluminum wiring is particularly problematic with cyclic loads or overloads. When the conductor is heated, it expands more than the fastening. This results in cold flow or creep. When the conductor cools down it is not as tight as before. This gives aluminum oxide a chance to intrude on the connection. With many cycles the aluminum oxide spreads further into the joint. This reduces the contact area and increases joint resistance and the conductor runs hotter with each load cycle.
If this is a long run and a lot of dollars are involved, consider copper pigtails on the ends of the aluminum conductors. There are a number of crimp type connectors on the market that have enough contact area to connect a copper cable to an aluminum conductor without cold flow issues. Then your connections to the breaker and to the receptacle will be made with copper conductors.

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

 

[ScottyUK]

It's a 50A circuit: it's only a small cable. The cost difference can't be that significant in relation to the overall job. If it was a 500A circuit then 'maybe' aluminium would be worth consideration although personally I would still use copper given any choice.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Skogsgurra]

I was sent to the German Democratic Republic (Eastern Germany) back in the eighties to do commissioning in a steel work. They used Al everywhere. Also for control circuits with 1.5 mm2 wires. It was hell. No two years needed for problems to present themselves. Every week, we had problems.

We finally invented a solution which was based on spring steel that we cut in pieces, bent and put inside the terminal clamp so that contact pressure was maintained. That, combined with some kind of grease that they probably also used for their Kalashnikovs, made the situation somewhat better.

But, given that experience, I would never use Al for control cables and light gauge power cables. For heavier gauge, there are special terminals that work well. No problems there.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[resqcapt19]

The 'retighting" of the connector is something that should never be done. (assuming this mechanical, set screw type connector). You will end up making a problem where there was not one before. Install the connector to the manufacturer's torque spec and you are fine. In the event of an issue with the connection in the future, the recommendation is to cut the end of the wire off and start over.
The following is from an ABB document.

Sidebar: Re-torquing -- A screwy practice

Although "re-torquing" as a way of checking tightness is common in many PM procedures, it violates basic mechanical principles and does more harm than good. A screw has maximum clamping power at a torque value specific to its size, shape, and composition. Exceeding that torque value permanently reduces the clamping power of that screw by reducing its elasticity and deforming it. Loosening and then re-torquing still reduces elasticity, which still means a loss of clamping power. Doing this to a lock washer results in a permanent 50% loss.

What should you do? Use an infrared thermometer to note hot connections. Check their torque. If they have merely worked loose, you can try retightening them. Note which screws were loose, and be sure to give them an IR check at the next PM cycle. If they are loose again, replace them. Finally, don’t forget the "tug test." This checks crimps, as well as screw connections. Don’t do this with the drive online with the process, though, or you may cause some very expensive process disturbances.

 

[resqcapt19]

From the IAEI magazine.

http://www.nmsu.edu/~tdi/pdf-resources/IAEI-1to2-05.pdf

While the article is aimed and the use of fine stranded conductors with standard wire terminations, this part is a general comment.

Over-tighten or Retighten?
Some installers over-tighten or retighten a connection to get fine stranded cable to hold in screw-type terminals. UL standards for connectors require that the terminal be tightened once to the specified torque and there is no retightening specified. Tightening the terminal beyond the specified torque value may cause binding of the threads thereby giving a false torque reading. Both overtightening and retightening of listed connectors and terminals on overcurrent devices and other equipment would appear to violate the provisions of the listing and therefore be a violation of NEC Section 110.3(B).

A quick review of NFPA Standard 70B-2002, Recommended Practice for Electrical Equipment Maintenance, does not find any suggestions that electrical equipment terminals be periodically retorqued. The terminals are to be inspected and examined for signs of looseness or overheating and that situation should be corrected where found. There is a retorquing recommendation for mechanical fasteners on box covers and the like.

 

[waross]

I have found that manufacturers torque recommendations are a perfect solution in a perfect world. Unfortunately we live in the real world.
If anyone wants to get their hands dirty try this:
Install a #6 AWG seven strand cable in a set screw terminal. Use recommended torque.
Give the cable a slight twist in the direction to unlay the stranding.
Drop the cable out of the terminal.
Test two:
Install a new piece of cable and over torque it to a ridiculous extreme.
Give the cable a slight twist and drop it out of the terminal.
The manufacturers torque recommendation is great if the cable is never touched in any way forever after installation.
Don't think of moving the cable so as to be able to measure the current with a clamp on ammeter.
I have encountered many installations where heat damage of one type or another was done as a result of a connection that was probably up to spec when installed but then became loose.
One instance was a 1200 amp switch that was destroyed due to heating that started in the connections.
Recommendations:
1> Tighten the terminal.
2> Twist the cable back and forth as much as possible.
3> Retighten. The screw will now turn easily from 1/4 to 1/2 turn.
Repeat until the screw stays tight.
Exceptions:
Some compacted strand cables will stay tight the first time.
This is for copper. For aluminum conductors in sizes from #14 AWG to about # 8 AWG don't try to put the conductor under a screw type terminal. Use a suitable connector to connect a copper pigtail to the end of the aluminum and connect the copper as you would do normally.
How to connect the pigtail?
There are some excelent connectors on this page for smaller sizes.

http://www.idealindustries.com/products/wire_termination/twist-on/

There appears to be only one connector still listed for cu/al. and it may not be large enough.
The Twister Al/Cu
In years past I have had excellent results using either Wing Nuts or B-Caps with aluminum. Unfortunately these are no longer listed for Al. If you do use them, fill them with a corrosion inhibitor such as Penetrox or No-Alox. After connecting the wire wipe off the excess grease and tape well to prevent the grease from weeping.
For aluminum (and copper connections that may be subject to load cycling) look for:

Live-action, square-wire spring

in the specs.

Avoid connectors with

Fixed, Square-wire spring

for aluminum use.

For larger sizes consider a butt splice. Burndy has a good selection of Cu/Al rated butt spices.
Go here and drill down:

http://www.burndy.com/products/product-by-category/compression/compression-aluminum.aspx

Search on this number for an example:
YSU4W8W
Good luck.

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

 

[stevenal]

The CPSC has issues with the twister.

http://www.inspectapedia.com/aluminum/awtwist.htm

Alumiconn apparently works better using set screws:

http://www.alcopstore.com/

Used with 10-12 solid aluminum and 18-10 stranded or solid copper.

 

[dpc]

Scotty,

I don't know about the rest of the world, but in the US, the cost savings of using Al wire in this size can be dramatic - often 50% cost savings or more for the same current capacity. Virtually all service entrance cable into residential panels in my part of the world is aluminum and most dryer and range circuits are aluminum.


David Castor

http://www.cvoes.com

 

[resqcapt19]

Bill,
Your method of tighting the connecton, wiggling or twisting the wires, and tighting again is a common installation method and one that I normally use. I just wish I could find a written statement from a connector manufacturer or NRTL that tells me this is the correct way to make a mechanical lug termination.

 

[ScottyUK]

Hi dpc,

Material cost is certainly getting higher but our labour rates are higher in whatever country-neutral measurement you use so I can imagine that materials are a greater percentage of total cost in the US. A 50A circuit shouldn't be bigger than 10mm²unless it's a very long run, so in this case I wouldn't expect a large cost difference but certainly on big cables copper is becoming ludicrously expensive.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[racobb]

If it is past the service entrance panel, I always use copper, no question. Not worth the trouble otherwise.

Alan

 

[rmw]

AAlpha,

Here are a couple of interesting reads for you:

http://www.body-borneman.com/Images/pdf/RefSheetsElectricalSafety.pdf

http://copperrod.ca/egy/tecdoc/6.htm

Good luck with your welder.

rmw