Re: NEC-LIST: Edmond Bruce questions

Hi,

At 06:36 28.12.2003, J. Mc Laughlin wrote:

>I am trying to find out more about Edmond Bruce (the electrical engineer
>not the navel engineer) who co-invented the rhombic antenna (with Friis)
>and did a great deal of antenna work from the late 20s to at least the
>40s.

Bruce was a Bell Labs engineer, who I had until now associated with Karl
Jansky, for Jansky had used a "Bruce" antenna in his famous rotating array
(cf. John Kraus, "Radio-Astronomy", 2nd edition, Cygnus-Quasar books, 1986,
p. 1-5). (I had never given any deep thought about the history of Rhombic
antennas, at least until today).

> Apparently, Bruce filed a patent application for a "diamond-shaped
>antenna" in Feb. 1931 that received a serial number (not patent number)
>of 513,063. The pendency of this application is mentioned as late as
>March 1939, which is a long time for an application to be in
>prosecution.

Looking-up patents using application serial numbers can be a tad tricky.
The US Patent Office recycles serial numbers over time. Their 6 digit
serial number sequences are periodically reset, according to a logic which
I can't quite fathom. The series themselves are nowadays numbered, and
usually prepended to the application serial number with a slash, for
example, 03/123,456. US application numbers from the year 2001 begin with
the "09/" prefix.

The Espacenet service (from the European Patent Office) codes the US serial
number together with the year - the number of the series (eg. : "03/") is
not included. Since you had provided both the year and the serial number,
it was very easy to locate the patents using "US19310513063" as a priority
number. (Priority numbers are the number of applications which are used by
an applicant to obtain a filing date and stake his priority rights
according to the Paris Convention.)

Two results come out with the priority reference thus assembled :

US2285565 , filed 03.02.1931 , and published (issued) 09.06.1942 ;
GB392901 , filed in the UK on 12.11.1931 , based on US priority of
03.02.1931, and granted 12.05.1933 .

The dates are coded DD.MM.YYYY - so 03.02.1931 means Feb. 3, 1931 (and
*NOT* Mar. 2). The patent numbers are also written *without* commas, and
include a country code - patents exist OUTSIDE the USA too :-) , and they
are often coded that way - makes it easier for computers.

The patents may be seen here :
US :
http://l2.espacenet.com/espacenet/bnsviewer?CY=ep&LG=en&DB=EPD&PN=US2285565&ID=US+++2285565A1+I+
GB :
http://l2.espacenet.com/espacenet/bnsviewer?CY=ep&LG=en&DB=EPD&PN=GB392201&ID=GB++++392201A++I+

The Acrobat Reader plugin is required. You must use the navigation buttons
provided at the top - using the browser's "back" and "forward" only brings
trouble. Only one page may be seen (and printed) at a time.

BTL appears to have only filed this particular patent in the US and GB. I
couldn't find parents of this application in other countries - but the
database could be incomplete.

Comparing the GB and US versions of the same application is interesting.
The US patent is assigned to Bell Telephone Laboratories, whereas the
British version belonged to Standard Telephone and Cables. Bruce's address
is given in the British patent as 27 Buena Place, Red Bank, Monmouth
County, NJ. Is there any connection between Bell Labs in NJ, Monmouth
county, and Fort Monmouth ?

The British patent is issued as an "improvement" to an earlier Bruce
patent, identified throughout the description as British Patent No. 353,517
( GB353517 ), filed in 1929 in the US. This earlier patent can be seen here :
http://l2.espacenet.com/espacenet/bnsviewer?CY=fr&LG=fr&DB=EPD&PN=GB353517&ID=GB++++353517A++I+

This earlier application concerns vertical inverted-V antennas. Coming back
to the Rhombic antennas patents, the invention therefore can be said to
reside in the idea that the antennas are laid horizontally (see page 1/4,
right column, lines 89 to 97). The lower side of the Rhombic antenna
(references "3" and "4" in fig. 1) is dubbed "counterpoise". So the logic
would be that the prior art vertical inverted was laid on the side, and
completed by the "counterpoise" - the virtual antenna image from below
ground is raised above ground. But there is something more which isn't
accounted by this "theory" : the far-end terminating resistance "9". The
prior-art had at first glance no such resistance at the far end of the
inverted V (I can't go through all of the 14 pages in detail; the figures
show nothing, nor the passages of the text). There is a small discussion
about this far-end termination on page 2/4, right column, lines 85 to 118,
which is there to improve the front-to-back ratio when the antenna is not
at resonance.

Bruce's US patent follows the same pattern. US1899410 is used as a starting
point instead of GB353517 , but these are in fact the same inverted-V
antenna applications. This earlier application had also been filed in
France ( FR705727 ) and the Netherlands (granted patent NL36493C ). The US
inverted-V patent may be seen here :
http://l2.espacenet.com/espacenet/bnsviewer?CY=fr&LG=fr&DB=EPD&PN=US1899410&ID=US+++1899410A1+I+
. Two other patents are mentioned in the description, US1963723 to Sterba
and US1947247 to Bruce. I don't know whether these references were
introduced with the original filing or during patent prosecution, however
they don't really seem to be related to the invention. There seems to be no
other significant differences between the US and GB documents, other than
the way the claims are drafted, which reflect their (contemporary) styles
and legal thinking.

I have no idea why the prosecution of the US patent had lasted more than
ten years. The British procedure had only taken about 18 months from the
day the application reached the UK patent office.

I had a peek at my 1932 edition of Terman (Frederick Terman, Radio
Engineering, McGraw Hill, 1932). The preface is dated August, 1932 at
Stanford University, so the Rhombic antenna was still pretty hot stuff
then. Under the index entry "Bruce", I find at page 550 the "Bruce
Antenna", which is the antenna used in the Jansky experiment. The qualities
of Bruce antenna is compared with the Beverage and "horizontal diamond"
types. This is how the "Bruce" antenna looked like :
http://l2.espacenet.com/espacenet/bnsviewer?CY=fr&LG=fr&DB=EPD&PN=US1813143&ID=US+++1813143A1+I+
.

At pages 524 to 527 tilted wires are discussed in general, including the
Rhombic (then known as the "tilted wire diamond antenna"). Figure 294 shows
the different versions of the concept. I find it interesting that in this
figure the vertical inverted-V antenna is also illustrated with a
terminating impedance. Terman states in his footnote that "[t]his section
is based largely upon the discussion by E. Bruce, Developments in
Short-wave Directive Antennas, Proc. I.R.E., col. 19, p. 1406, August,
1931". If I am correct in assessment of US1899410 in that it discloses no
far-end termination, then this would mean that Bruce may have realised the
importance of this resistance even for the vertical inverted-V sometimes
between 11.10.1930 and 1931, at the time he was filing for the Rhombic.

Interestingly, the Rhombic antenna may be considered differently by
different eyes. Kraus (John Daniel Kraus, "Antennas", second edition,
McGraw-Hill, 1988) in his chapter on long-wire antennas (chap. 11-16, pp.
502-509) regards the Rhombic antenna as a "double-V type", therefore as a
derivative therefrom, and not of the inverted-V type. The (horizontal)
V-type antenna (discussed on p. 502) is attributed to P.S. Carter of RCA
Communications, whereas the (vertical) inverted-V belongs to Bruce.

Looking up the horizontal V antenna, I see this from Philip Staats Carter
of RCA filed 11.06.1930 .
http://l2.espacenet.com/espacenet/bnsviewer?CY=fr&LG=fr&DB=EPD&PN=US1974387&ID=US+++1974387A1+I+

Looking at other stuff which comes up under "Edmond Bruce", I saw this :
http://l2.espacenet.com/espacenet/bnsviewer?CY=fr&LG=fr&DB=EPD&PN=US1974387&ID=US+++1974387A1+I+
.

There are a few more documents.

In the 1930's, improvement of transoceanic HF telephone circuits was
sought. If I remember well two approaches to combat fading were
investigated (and used) : 1) diversity switching 2) automatically steerable
antenna arrays for tracking the dominant reflection from the ionosphere.
(see "A History of Engineering and Science in the Bell System", 1984, I
think it was in the volume "Communications Science" -
http://isbndb.com/d/book/a_history_of_engineering_and_science_in_the_bell_system_a02.html
). I don't remember which approach won out, I believe both had similar
performances.

Bruce seems to have been later involved in switching technology.

> Does anyone know what became of this application? My speculation is
>that, in spite of the valiant work of his patent attorney (H. S.
>Grover),

On page the US patent's cover page, a Guy T. Morris is identified as being
the attorney. Dunno if they belonged to the same partnership, or if he was
a BTL employee.

>work published prior to 1930 doomed a patent from issuing.

Attorneys (and also examiners) are not always heroes...

By curiosity, what's the reason of your interest in Edmond Bruce ?

Regards,

Alexandre

-- 
The NEC-List mailing list <nec-list_at_gweep.ca>
http://www.gweep.ca/mailman/listinfo.cgi/nec-list