Wirenet Image Band
wirenet.org mobile image band

WAI Forums


× Nonferrous topics such as copper and aluminum, annealing, etc. go here.

Diameter variation on rod mill finish wire

6 years 5 months ago #1451 by Archived Forum Admin
Does anybody have experience or gathered data what you can expect on a diameter variation annealed Cu wire between 10 - 14 gage? According to EN-DIN 13602 Cu wire round can have a tolerance ≤ +/-1% of the OD in a diameter range from 0.40 to 5.00 mm (.0157 - .200”). Is there any North American Standard that specifies the tolerance?

The a/m questions were coming up by swapping from weight to resistance measurement to specify the diameter and to maintain uL minimum wire size. The process is for the so-called T-wire, rod mill & annealer – PVC & Nylon extrusion – cooling and take-up, all in line at 6000 to 7000 fpm. We found that the diameter in the range of 10 – 14 ga varies up to 1 mill (25 µm) over a section of one foot.
We have checked the die situation, made sure we don’t have any galling or die blocking at the finish size. Experimented with different wire tension between the drawing machine and the resistance annealer. All without any significant improvements.

Hard wire test shown variation up to .4 mills, when annealing with elongation in a range of 24 – 28% elongation variation is amplified all the way to .8 – 1 mill variation of the wire OD. If the annealing is further increased to 30 – 32% the variation is decreasing below .4 mills, and this is across the a/m wire range 10 – 14 ga. Does anybody have an explanation why higher annealed wire has less diameter variation? Just to clarify, with the higher anneal we have of course more stretch looses in the annealer and possible down stream the line. If talking about variation we mean the variation of a given length of wire not the losses between the finish die and the finish wire size at the take up. We try to understand why we have this variation and what can we expect from the process.

Please Log in to join the conversation.

6 years 5 months ago #1452 by Archived Forum Admin
Hello again,

Re Swapping From Weight To Resistance Measurement

Because the American Wire Gage just gives the nominal diameter, what we tried to do at one of my former companies was to be accurate to the fourth decimal place on all finished wire. (eg) +0.0000X/ -0.00000. In other words, we would never be under the nominal. (In fact, I have watched solid, insulated wire stripped (unmanned) at very high speed and then redrawn because it was undersized. It was quite remarkable to watch the plastic peel off at say 1500 feet per minute. Indeed steel guider plates were needed to direct the insulation to the plant floor. There was also a tremendous static charge built up in the pile of stripped insulation so one had to be very careful to ground it.)

Statistically, we then measured the oversize (negative variance) as a quality loss (The price of non-conformance or PONC.) and worked very hard to ensure that we were not giving away any metal in an non controlled fashion. (eg) Knowing that there were statistical diameter losses at bare wire stranding and that some statistical oversize was required to compensate for it.) Ongoing cable dissection audits are a fundamental tool in this regard as plastic give away can also be a very serious problem. If you want to learn more about quality losses, I suggest you go to a library and carefully read "Quality is Free" by Philip B. Crosby.

We never used the European concept of cross-section resistance measurement as a referee test because (a) it was not appropriate and (b) we would not be under size. Moreover there was a trace of silver in our copper rod and that meant our copper conductivity was approximately 101% so we were always confident our customers were getting the very best conductor.

In today's world, some building wire plants have not yet learned how effectively control tension along their primary tandem extrusion lines. This is something that the telephone wire manufacturers had to learn decades ago on their high speed tandem extrusion lines. High wire line speeds mandate precise control and much measurement to guarantee a quality product. Quality in this sense means "conformance to the requirements".

Kindest Regards,
Peter J. Stewart-Hay
Stewart-Hay Associates

Please Log in to join the conversation.

6 years 5 months ago #1453 by Archived Forum Admin

In magnet wire you sometimes used resistance and even weight to determine diameter. But this was only done when you were dealing with a very small diameter. The reason being is that you can easily take a sample of 100 to 300 feet of 45 - 52 awg and weigh it and then calculate the diameter. It is kind of hard to do the same with 100 - 300 feet of 14 awg. Or even rod.

Generally speaking, a fraction of an inch on large diameter wire could make a big difference in calculating diameter. For example say a piece of wire rod had 4 feet of length per pound. If your sample was exactly 4 feet long which would be hard to work with as rod, so you might cut a 12 inch long sample but even a mistake as small as 11-15/16" would be calculate to a much different diameter than say 12-1/16" sample.

Even if your sample was 12.00001 long, it would not tell you anything about concentricity. When dealing with 50 awg concentricity is not that big a deal.

Please Log in to join the conversation.

Time to create page: 0.373 seconds



Contact us

The Wire Association Int.

71 Bradley Road, Suite 9

Madison, CT 06433-2662

P: (203) 453-2777