crack detection of steel bars

[giri]

We manufacture cold finished steel bars of carbon and free-cutting
steel grades.

I have spent a lot of time researching this on the net, but could not
locate an answer.

We use a magnoflux-crack detector for locating cracks in the bars.
However, when checking free-cutting steel, error creeps in due to
presence of streaks.

Is there any other more efficient method,equipment which will enable
quick and accurate detection of cracks? Is Eddy-current testing or UT
suitable, cost effective and practical.

[jsan...@swales.com]

Eddy current finds discontinuities in the surface. It measures
instantaneous changes in diameter well, and some surface flaws, such
as scabs or laps that lift off the surface, are well diagnosed with
this process. It is not quite so good at finding seams, if that is
what you mean by streaks. I don't know if there are manufacturers of
these systems these days. I worked with it in the 70s. The two types
are rotating coil and differential coil. For faster rolling
processes, the differential coil method is best. There are two coils
set about an inch ar two apart, and one signal is subtracted from the
other. If the signal is simjilar (no flaws), the output is near zero.
Otherwise, the bells and lights go off.

Jack

[Filip Spruyt]

Maybe it's possible to use ultrasonic devices to detect cracks in the bars
you produce, it is very suitable for detecting cracks on thick bars. This
method tells you about all cracks, on the surface and in the center of the
bar.

When you have smaller bars or wires you can use eddy current testing, this
method however only detects cracks that appear on the surface. An other
disadvantage of this method is that you detect resistance-variations, and
that you are not sure wether it was a crack or something else. The
advantage of this system however is that you know how deep the crack is. If
you want to proceed with this type I can give you a suplier.
In contrast to what jsanders wrote however, the diameter changes of your bar
doesn't have any effect on your measurement if you use the right equipment.

"giri" <wad...@sify.com> schreef in bericht
news:655f66ac.02052...@posting.google.com...

[Edward D. Vojcak]

Giri,

It's not a question of "efficient" methods - but EFFECTIVE methods.
Testing a cold finished bar is part of the production operation and as
in all production each test method has limitations, limitations that
must be understood and observed.

I've worked with cold finished products for about 15 years and IMHO
you will not find a economically feasible answer - except to purchase
better quality hot roll steel bar - make the steel right the first
time before you add value to it by cold finishing.

Advanced methods like x-ray (tomography), or thermal (heat the surface
with infrared and observe how it cools - revealing flaws) are out of
consideration since most inspection equipment manufacturers see the
cold finished bar market too small for the effort.

Eddy current (roto-bar) and magnetic (differential coil) took and
saturated the bar market a long time ago - but have high (and
undocumented) type II beta errors (not rejecting bad material and
calling it good).

Besides - the steel market is a commodity market, and your management
most likely "know" that they get paid to ship steel and not get paid
for rejecting it!

Ed Vojcak P.E.

wad...@sify.com (giri) wrote in message news:<655f66ac.02052...@posting.google.com>...

[JFMACR]

Are you looking for surface or internal cracks/defects.

Jim MacRae

[Prokhorenko Sci.Groups]

Try to test a Level of an Acoustic Emissions radiated During Cooling of your
steel bars.
(compare Differences of this Level)

(I use low-cost AE-inventory constructed in Kiev University of Civil Aviations,
but produced from PAC is not so bad too).

Sergiy Prokhorenko