I think that the tank end of the anchor had some limited side to side motion and the fitting at the other end, with the spring, allow some degree of rotational movement, allowing horizontal as well as vertical motion of the anchor. The advantage of the Sullivan anchor is that “swelling” or changes in circumference due to temperature changes are much better compensated that with tank bands, which essentially cannot compensate for that at all and are stretched and loosened every time the tank swells.

Regards

Bruce

Bruce F. Smith            

Auburn, AL

"Some days you are the bug, some days you are the windshield."

On Jan 17, 2019, at 3:07 PM, Tony Thompson <tony@...> wrote:

David Thompson wrote:

I am curious why you characterized them as "Gulf-style tank anchors?"

An article in the June 1934 Railway Mechanical Engineer, p.198, attributes the design to Gulf general superintendent J.C. Sullivan, and includes a small drawing.

https://imgur.com/a/3vsAshm

   In the article, they are referred to as "Sullivan" anchors. With these anchors riveted to the tank, it seems to me that they defeat the whole idea of tank bands, namely to allow the tank to move horizontally as it expands or contracts, anchored only at the center anchor.

Tony Thompson             Editor, Signature Press, Berkeley, CA
2906 Forest Ave., Berkeley, CA 94705         www.signaturepress.com
(510) 540-6538; e-mail, tony@...
Publishers of books on railroad history