| Use of the
Core Loss Tester (Continued)
Theory of LEXSECO core loss testing
Designers of electrical apparatus use data
furnished by electrical steel manufacturers in arriving at the output and performance
characteristics of their machines. These same data items are useful when evaluating the
condition of the iron cores when they are tested.
· The first data item is the core loss in
watts per pound (watt/lb).
· The second value is the ampere turns
per inch (AT/in).
Using the LEXSECO core loss tester with the
built-in computer, we are able to determine with a reasonable degree of certainty whether
a core is good, marginal, or bad. This is accomplished by computing,
from input meter data, the indicated watts per pound and the amp turns per inch. Also,
numerical test values are printed out for good and bad cores so the operator can judge how
good is good and how bad is bad.
Commonly used electrical grades of sheet
steel have watts per pound (manufacturer's ratings) varying over a range of slightly under
1 to as much as about 2 watts per pound. These are Epstein test ratings and will be less
than the actual punched-and-assembled-into-motor-cores values of watts per pound.
Assembly into the electrical machine will
increase these values from one and one-half to two or more times the steel manufacturers'
test values.
Some of the factors contributing to this
increase are:
· Lamination punching or stamping burrs
· Lamination thickness
· Lamination clamping pressure
· Type of insulation coating used on
the steel
· Heat treatment process used on punched laminations
· Lamination assembly method
· Silicon content and hardness of steel
· Heavy welds across the back of core stacking
The watts-per-pound loss is composed
of two components: the hysteresis loss and the eddy current loss.
The hysteresis loss results from the
alternating current frequency reversals moving the molecules of iron to adjust their
polarity. This movement of the molecules consumes energy and is dissipated in the form of
heat.
The eddy current loss results from cross
current flow in the lamination assembly and is higher for the thicker steels. This current
flow also consumes electric energy and is dissipated in the form of heat.
The LEXSECO core loss tester includes a
computer program which permits quick and accurate identification of either a questionable
or bad core.
As soon as the desired test excitation level
has been reached, the actual measured flux, amperes, and watts are input to the computer
for a printout which indicates whether the core is acceptable, marginal, or completely
unacceptable in its present condition. This is followed by a tape printout of the values
of watts per pound and ampere turns per inch.
The magnitude of these values is used by the
computer in selecting which category into which the core falls and may be used to evaluate
the user's course of action in making core repairs or core replacement.
The watts per pound is the principal
criterion of the condition of the test core. The bad areas will show increased temperature
rise as compared with other core areas or, if the fault is general, the entire core will
get very hot.
You must exercise judgment and recognize that
very old (80 percent rather than 92 percent efficient) motors may have a watts per pound
in the range of 7 to 8, yet the stators are large and do NOT have excessive heat.
Conversely, relative late and energy-efficient motors will have watts per pound in the
range of 2 to 3, with some even lower.
When the core is dubbed marginal or bad
because of exceeding limits shown on tape printout, you are alerted to look for special
conditions or to try to do something to lower the watts per pound.
Often, locating and correcting hot spots by
running a knife blade between the teeth laminations at the hot area helps, as does bumping
the ends of the core stacking. Sometimes thin varnish applications will also lower the
watts per pound. The last resort, of course, is restacking or replacing the core.
The ampere turns per inch is a measure of the
magnetizing power to produce a certain flux density in the iron laminations. The lower
ampere turns per inch, the lower the energy and the more efficient the core. However, the
ampere turns per inch values are often badly distorted by welds across the back of core
and through rivets or bolts. You must take this distortion into consideration when
evaluating the quality of the core being tested with the knowledge that high-quality, new
silicone steel normally has an Epstein value of one and one-half to two ampere turns per
inch at 85,000 lines per square inch.
In a normal stacking, you can expect about
three ampere turns per inch, but with rivets and welds the value is closer to 10 ampere
turns per inch.
Hot spots
Locating hot spots is just as important as
the metering test and is a part of the LEXSECO core test procedure.
Upon completion of the computed core loss
test, apply a high-excitation current, approaching two or three times the metered amps or,
if beyond the tester capacity, use the highest amps the tester will put out.
After approximately 1 minute of the high
amperage, a hand-check of the core surfaces will reveal any localized hot spots.
Mark these spots for the necessary repair.
Fused or welded laminations may be cleared by grinding or filing. If there is no evidence
of burning or fusing, bumping or pounding the end of the core stack with a heavy mallet
may break up the shorted laminations so that insulation can be restored between the
individual core steel laminations.
The core is again subjected to the
overcurrent hot spot test to be sure all bad areas have been cleared. If not, the core may
have to be restacked or replaced.
Locating and eliminating these smaller hot
spot areas is important even though the original metered test data may show the core is
either in the good or marginal categories. The tooth area is particularly important
because this portion of the core operates high magnetic flux densities in normal motor
operation and any additional core heating may result in rapid winding insulation
deterioration and failure.
The reason these small hot spot areas may not
throw the computed data into either the marginal or reject categories is that the ratio of
the increased excitation amps and watts to the normal overall amps and watts is very
small. The ratio may only be 40 watts, which is a small amount to measure, but enough to
make a few shorted teeth glow and burn coil insulation.
|
An infrared scanner may be used to monitor the temperature
as an option.