The data sheet for the CK727 illustrated on the web page:

http://www.ck722museum.com/page7.html

- shows ratings for the part of

5nA (collector current) and
30nW(collector power dissipation).

Please note the units used.

The only other source for data on this part is from the D.A.T.A.
catalog series, that gives collector dissipation of the CK722 as 4mW.

The beta test for this part involves a static collector bias of 10V
and 4mA to establish minimum hfe of 30. This would require the part to
dissipate 40mW, at least for the duration of the test, unless a curve
tracer was used. Even then, this exceeds the paper collector current
rating by some orders of magnitude.

Similar part numbers in similar packages are either rated at

40 to 100mW,

or 2 to 4mW.

Is it possible that the latter group suffer from practitioners
dithering around the same possible typo, made by the same typist, at
around the same time? The typo seems only to affect recorded ratings
for part numbers

CK721 -4mW
CK722 -4mW
CK725 -4mW
CK727 -4mW
CK790 -2mW
CK791 -2mW
CK793 -2mW
all early Si PNP drift types from Raytheon

CK766 -2mW
CK766A -2mW

both early Ge PNP types also from Raytheon

A facsimile of the D.A.T.A. listing is hosted for these parts by
Datasheet Archive, with the first group of four tabulated on the first
page and first lines of the low power silicon pnp transistor section
and the second group in the same location for low power germanium pnp
transistors. The only parts with lower ratings are those with unstated
(blank) listings.

http://www.datasheetarchive.com/sear...b.x=34&sub.y=3

http://www.datasheetarchive.com/sear...b.x=38&sub.y=5

If this is a typo from the original spec that your sample datasheet
represents, it certainly has gone on for a considerable length of
time.

There should probably be some official notation made, if only for the
sake of museum records, before unprinted reference resources who can
clear it up disappear. I'm sure data for these parts was published and
republished over the years of the part's commercial life.

Anyone with access to other data sources concerning these part numbers
is requested to respond to this news thread or by e-mail to
leggatmagmadotca. I've already contacted Mr Ward for any suplimentary
info to which he may also have access.

Anyone with a copy of the IEEE Spectrum magazine of March '03 is also
asked to review it's contents for more relevent information, and to
report it in a similar manner.

Hpofully there will be a more diffinitive entry available in time for
the new spreadsheet format of bipolar transistor numbers currently in
the works for free distribution on the web.

RL