Welcome to the Countdown to STS-135

In a few days, we will witness the final launch in the space shuttle program. This is a daily series of posts that recount the space program and how I experienced it. If you are new to this blog, start from the bottom (first post) and work up.

Thursday, June 30, 2011

STS-135 T-8 Days - STS-1 1981


Finally, in 1981, nearly six years after the last Apollo flight, the Shuttle was ready for it's maiden flight.  It launched exactly 20 years after the first manned space flight.   I was working in Florida for Stromberg Carlson, at the time, with Mike Shaffer.  We were building computer controlled telephone systems and following the development of the Shuttle.

The first launch was halted a few moments before ignition, when one of the five control computers disagreed with the others.  After launch, there were voting procedures to break the disagreement, but before launch, it was a scrubbed mission.  They found the bug the next day and the launch was rescheduled for the weekend.  The delay permitted us to take the boat and go watch the launch from the intercoastal waterway in Titusville.

Watch this YouTube for a great launch sequence of this first launch.
Remember what I said about the orange external tanks?  Well they painted the first two white, before they figured out they could save important weight by leaving them unpainted. 


Mike and I identified with bugs related to redundant CPUs.  We were implementing the redundancy programs for Stromberg Carlson's PBX, which was the first stored program digitally switched PBX.  It used the Digital PDP LSI-11 CPU with 128K of memory.  The CPU board is shown here.  There is no memory on that board, just the CPU and bus driver chips.  The five large chips comprised the 16 bit CPU.  A real masterpiece of integration at the time.  The address space only supported 64K, but we engineered a paging memory that allowed us to increase the memory to 128K in order to fit all the software needed to control all the phone system features.  We loaded programs from a tape cartridge that took 5 minutes to load the 128K program. There was no Ethernet for computers in these days.  If they needed to communicate, like with our redundant processors, a peripheral was created that would transfer information between them 16 bits at a time, or a 9600 bit per second terminal line could be used to communicate over longer distances.  By comparison, this CPU was less than 1/10,000 as powerful as an iPhone.  The data link was 1/10,000 the speed of a current Ethernet connection.  Yet, we used it to control a 1200 phone PBX, with all the features.

http://en.wikipedia.org/wiki/PDP-11

No comments:

Post a Comment