Protecting aluminum wire not in use



When I was learning about crimping lugs on control circuit wiring (many years ago...:-), we learned to only use the 'calibrated' crimping tools (ratcheting type that won't release until proper compression is made). The reason for that was that the lug material, when pressed against the copper wire, would actually form a type of 'cold weld' between the two metals. Some of the copper would actually diffuse into the lug metal and vice versa. Completely excludes air/moisture from the surfaces, preventing corrosion.
I suspect that these same 'compression crimps' have to be installed with a suitable tool (on large wire, we used hydraulic crimpers). The pressure may be enough to form the same type of 'cold weld' between wire and crimp casing. You can't get that kind of 'cold weld' joint with a screw connector.
daestrom P.S. This is one reason why the Wal-Mart / Radio-Shack special "complete crimping kit", with it's plier like 'universal crimp tool' was banned in all cat 1E wiring systems in favor of the 'oh so much better but expensive' Amp crimping tool :-)
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
daestrom wrote:

True, but probably better than the solder joint these same people would make if they knew which end of the iron to hold. These lugs are usually used on a screw terminal. So the lug may better than bare wire, even with a poor crimp tool. It's possible to do a bad job with the best tools.
Wire nuts seem to work without even crimping. In California, low voltage wires are supposed to be soldered, but I don't know anybody doing that.
--
Virg Wall, P.E.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
| Wire nuts seem to work without even crimping. In California, low | voltage wires are supposed to be soldered, but I don't know anybody | doing that.
Having seen wire nuts actually fail, I'm more inclined to go with either the screw lugs or the compression crimps.
--
|---------------------------------------/----------------------------------|
| Phil Howard KA9WGN (ka9wgn.ham.org) / Do not send to the address below |
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

To get UL listing on equipment we built, we were required to use positive stop crimpers as you describe. In large sizes they are available in hydraulic (either hand pump, or air over oil powered pump, much nicer) or a double leverage mechanism similar to bolt cutters. The manual ones had handles 5 feet long for the larger wire sizes.

I sure hope so! Did they also ban the insulated crimp terminals? Did they also require a cal?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Yes, the crimping tools all had to be 'calibrated'. Actually, it's more a 'cal-check', as there is no adjustment, just a go/no-go test. The recommended method was use the tool to install a proper lug on a brand new piece of wire, then put the lug/wire in a tensioning device. Measure how much tension it takes to pull the wire out of the lug. If the tension was too low, the tool failed (Amp didn't provide any adjustment to control the jaw closure).
daestrom
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Tue, 24 Oct 2006 20:50:57 GMT, "daestrom"

I don't see why that can't happen in a properly torqued terminal lug. If these lugs are so bad (AL/AL) I would expect to see services bursting into flames all over the country. Gerald may be having a problem in Barrow Ak but I still don't understand why splicing dissimilar metals would make it any better.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Tue, 24 Oct 2006 21:36:43 -0400 snipped-for-privacy@aol.com wrote: | On Tue, 24 Oct 2006 20:50:57 GMT, "daestrom"
| |>When I was learning about crimping lugs on control circuit wiring (many |>years ago...:-), we learned to only use the 'calibrated' crimping tools |>(ratcheting type that won't release until proper compression is made). The |>reason for that was that the lug material, when pressed against the copper |>wire, would actually form a type of 'cold weld' between the two metals. |>Some of the copper would actually diffuse into the lug metal and vice versa. | | | I don't see why that can't happen in a properly torqued terminal lug. | If these lugs are so bad (AL/AL) I would expect to see services | bursting into flames all over the country. Gerald may be having a | problem in Barrow Ak but I still don't understand why splicing | dissimilar metals would make it any better.
As was describe, the compression squeezes out all the air gap and thus prevents any oxidation of the aluminum.
--
|---------------------------------------/----------------------------------|
| Phil Howard KA9WGN (ka9wgn.ham.org) / Do not send to the address below |
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 25 Oct 2006 03:47:27 GMT, snipped-for-privacy@ipal.net wrote:

The contention was that expansion and contraction fropm the seasonal 150f swing would loosen up these splices. That should allow "flow" when it got squeezed and the entrance of air when it opened up. We had a guy from Alcan at an IAEI meeting with lab data that showed an aluminum wire in a typical tin over aluminum lug actually performed better than copper in a series of heat/cool cycles. ... but he is an aluminum salesman.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Well, I'm not the one that said they were 'bad'. As far as thermal cycling, I agree, it seems Al/Al shouldn't be much of a problem. But screw terminals depend a lot on maintaining compression. Even without thermal- cycling, Al can 'creep' when under a constant stress. The solution is to arrange so that the stress is below the point where inelastic creep comes into play. The compression lug puts a very high stress on the conductor while it's being crimped, but not that high a stress after you remove the tool. It doesn't need to maintain a high stress because the two metal parts are 'bonded' by the cold weld process. With screw or other mechanical fasteners, the stress isn't high enough to get this bonding, so the they have to maintain a high mechanical stress (high enough to cause creep over a long time).
Surely you've opened up service panels and found you could 'get another 1/8 to 1/4 turn' on all the terminals. It's not because the last guy was limp-wristed, the metal has 'creeped' and loosened by that much over several years. The old problems with Al wiring under Cu terminals was worse by a long shot, but it still happens even with Al/Cu terminals today.
At least that's how I learned it :-) But I'm not a metallurgist (don't even play one on 'net :-). Perhaps someone has more info??
daestrom
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:
| At least that's how I learned it :-) But I'm not a metallurgist (don't even | play one on 'net :-). Perhaps someone has more info??
Back when I was in junior high school, my parents were building a new house. I collected the wire scraps the electrician left behind. Among them was about 2 feet of what I now know to be Al USE cable. I tore it all apart and extracted all the individual aluminum wires. That's when I discovered they were no good for soldering.
But one thing I did try was twisting them. I had a few inches of one of the aluminum strands and bent about an inch on each end 90 degrees to make it easy to twist. I was curious how much I'd have to twist the aluminum to make it break. I expected to see the whole length between the bends gradually twist all at once. But that didn't happen. Instead, the twist took place at some point, usually, but not always, closer to one end. As I continued twisting, the was a boundary between straight wire and twisted wire, and then another boundary back to straight wire. The part where it was twisted then grew along the length, usually, but not always, just in one direction. There was a definite pitch to the twist that was constant. Eventually I would have the entire distance between the two bends twisted. At that point it was much harder to twist, but quickly broke somewhere when I forced it. On about the 3rd or 4th try at this I decided reverse the twist. I noticed that this produced a reverse twist zone of wire that propogated back over the forward twist as I continued reverse twisting it. Eventually I had reversed the whole wire. In one case I made the wire twist forward then reverse then repeating many times and it didn't break. In another case I managed to get the wire to have several alternating (between twist direction) zones along the length, each about 3mm long. I also measured a very slight drop in distance between the bends after twisting as compared to a straight wire. There did appear to be some kind of compession taking place. I could never get a wire to untwist back to straight.
Eventually I ran out of little aluminum wires.
I suspect this may be related to aluminum's ability to flow. Maybe it just isn't as dense in this form as a wire as it otherwise could be.
Maybe I should look for a phase state diagram for aluminum and see just what allotropes it might have under various pressures in the solid state.
--
|---------------------------------------/----------------------------------|
| Phil Howard KA9WGN (ka9wgn.ham.org) / Do not send to the address below |
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

<snip story of experimenting with Al>

"Work hardening" also comes to mind. Repeatedly flexing a metal tends to move tiny defects along until they are 'stopped' by bumping into another defect. Eventually, you end up with relatively defect-free crystals with the defects all 'pushed' towards a grain boundary. Because the crystals are much stronger, the metal becomes harder to 'work' (losses malleability). Continued working then creates fractures that reduce the area available for strength, and the whole thing breaks with lower force (but higher stress within the material).
Think of a soft iron wire coat hanger. We've all flexed them back and forth repeatedly until it cracks/breaks. That's 'work hardening' in action. The exact point of bending shifts back and forth a bit as each 'bit' gets harder. But a crack soon develops, so it's easier to bend (lower cross-section). Eventually even modest effort will break the hardened metal where cracks have reduced the cross-sectional area.
Or trying to straighten a bent wire. The place that was 'bent' is actually harder (less malleable) than the unbent wire (defects migrated to grain boundaries). That's why when you straighten it, it always has a little 'knuckle' that is much harder to get out.
That's for iron/low-carbon steels, I'm not sure Al will 'work harden' the same way.
daestrom
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:
| Or trying to straighten a bent wire. The place that was 'bent' is actually | harder (less malleable) than the unbent wire (defects migrated to grain | boundaries). That's why when you straighten it, it always has a little | 'knuckle' that is much harder to get out.
Yes, those bends are always annoying. And adjusting a loop to go under a screw is always a pain.
| That's for iron/low-carbon steels, I'm not sure Al will 'work harden' the | same way.
The twisted parts were harder to bend. Since they never could untwist, I don't know if it was the twist itself, or the effect of being twisted, that caused that.
--
|---------------------------------------/----------------------------------|
| Phil Howard KA9WGN (ka9wgn.ham.org) / Do not send to the address below |
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
| Check on this link: | http://www.tnb.com/contractor/docs/colorkeyed_spec.pdf | | Look on page E49 to E51 for splices. The connectors are fairly compact and | they have insulating covers. | | If you look on page E70, they have terminals that are meant to go into | copper only connections for putting on the ends of aluminum wire.
The splices on E67 look interesting.
--
|---------------------------------------/----------------------------------|
| Phil Howard KA9WGN (ka9wgn.ham.org) / Do not send to the address below |
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@electrician2.com wrote:

Correction: That should have been from 1988 to 1996.
The fact is we have very cold temperatures in interior Alaska and there is a problem with aluminum connections loosening even with Cu/Al rated terminals. Also, UL and other testing laboratories do not test for extreme cold temperatures. Likewise, Carlon for instance, does not manufacture cold temperature rated PVC products because as I was told at an IAEI Section meeting the market simply is not there. So in cold climates we learn that aluminum is a liability, XHHW insulation works while T insulation becomes brittle and breaks, Al/CU rated terminals do not stand up when used with aluminum, and PVC conduit expands so much that an expansion joint is required for as little as a five foot run per code.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 19 Oct 2006 18:00:56 -0700, snipped-for-privacy@electrician2.com wrote:

Again I was wrong Sorry. I am a lower 48 guy and I prefer the 27th parallel or lower. I have been to Fairbanks and I understand you have a 140-150 degree swing. In most of the US you will always see aluminum on service conductors up to the service point and usually all the way to the service disconnect. I am surprised you have a problem with aluminum in lugs since they are usually aluminum too. They are telling us an alumnium lug actualy tests better with aluminum wire than copper.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@electrician2.com wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
js5895 wrote:

Aluminum is quite reactive. A "clean" aluminum surface will fairly rapidly oxidize to a self-limiting layer. It should be assumed a wire has an oxide surface layer. The last lug installation instructions I saw said to wire bursh the wire and apply antioxide paste. My understanding is that for any but the smallest gauges, the setscrew or crimp process deliberately deforms the wire to break the oxide layer. I don't think there is any point to trying to protect the wire from oxidation.
bud--
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

.
That is why any listed "AL" lug or splicing device does rude things to that oxide layer. If you strip the required amount of fresh conductor you are as good as the first run off a new spool of wire.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.