was limited by cable bell trouble and there was no
failure of the cable within the lead sheath. Assuming
that the dielectric strength of impregnated paper
insulation also varies with the square root of the thick
ness, and using this particular point as our starting
point, then we get the curve shown by B in Fig. 20.
On the same sheet are also shown two points given b y
Weiset in the E. T. Z. for January 15, 1920, also six
points given b y Fernie in Beama for September, 1921,
and also one figure giving results from an Italian cable,
the latter data being obtained from private sources.
On the same sheet are also shown the results of a
large number of dielectric strength tests on various
thicknesses of insulation made for the author on cables
purchased in the last three years. For those thick
nesses of insulation where a large number of tests are
available, these results show a ratio of maximum to
minimum greater than 2 to 1.
In testing the cables purchased under specifications,
it was noted that at first the paper was not applied
smoothly, that is, without wrinkles, and that difficulty
was encountered by some manufacturers in applying
the insulation so that it would pass the bending test.
As these difficulties were brought to the attention of
the manufacturers, it was found that it was not a
serious matter to eliminate the difficulties b y having
the paper tape of the proper thickness and width and
applied with a suitable tension and the right amount
of lap. As these difficulties were eliminated, the
dielectric strength test of the cables'increased, so that
finally for such thicknesses of insulation as are common
in this country for 25 kv. it was found that the voltage
required to cause failure of the cable under the dielectric
strength test was quite beyond the testing facilities
available in practically all of the cable factories. It
would thus appear that by suitable care and perhaps
some additional inspection in the factories, the Ameri
can cable manufacturers should be able to make their
cable of more uniform quality so that it would give
results more nearly corresponding to the available
data on foreign cables.
The American Engineering Standards Committee in
gathering foreign specifications and standards for the
use of its recently appointed Sectional Committee on
Insulated Wires and Cables received an interesting
communication from a Dutch Standards Committee
which includes the following statement: