"How good is good enough" is a fair question, and for an answer, when designing production line equipmet for an auto company the specification is simple: "The system must not fail at any time under any condition". So it makes sense to go for a bit more than just good enough.
Most of the time the customers are able to understand that most of the time.
I held back from a more descriptive explanation as this circuit is currently going through the patent process and the panel of patent attorneys would probably frown on a public description at this time.
Basically it checks both the voltage and currents, in this case checking one is not definitive enough, then it shuts things down if things look askew. And it periodically tries a restart just in case the fault condition goes away. Ret is hush-hush for now.
When I was back in college we were donated what is now an antique video graphics machine called a "Harry"
In 1985, Quantel released the "Harry" effects compositing system/non-linear editor. The Harry was designed to render special effects in non-real time to the video recorded on its built-in hard disk array (much like most computer based non-linear editing systems today). The hard disk array used drives made by Fujitsu, and were connected to the Harry using a proprietary parallel interface, much like a modern-day RAID array. Technically, it was the first all-digital non-linear editing system, since it could also do editing of the video that was recorded on the Harry. Due to technical constraints of the time, the Harry could only record 80 seconds of video, albeit encoded in full broadcast-quality, uncompressed D1-style 8-bit CCIR 601 format. This aside, the Harry was quite an advanced machine, and the only system like it for its time.
When we were fitting it not only were the HDDs a three man lift each, but the mainframe had a PSU which mated to it with 3/4in copper rods for the DC. It looked like some giant Swiss mains connector. Along with the Klystron PSUs that I used to handle I have to declare that I am a wimp when it comes to HT and high current, I stay away.
This interesting story certainly is quite entertaining, no doubt. But what I would like would be more of an explanation as to just how the fix fixed the product and was able to have the product still providing the functions that the customers expected it to do. So what we have is a great setup for the story and then a very short description of the wonderful fix that somehow saves the day. A better description of how that fix kept the thing going would be quite in order.
@Janine, you are absolutely right. Never any time to think. Never any time to do it right the first time, but always time to do it over... Thinking always pays off!
As an RF power guy, connectors and packaging are almost always my biggest headaches. They cost too much, they take up too much room, they don't have enough current capacity, they break...etc, etc, etc. Mundane, boring, unsexy areas like this are where advances could have a real impact. Given time :-)
Great story George! You have a real knack for humorous writing. I love the line: "When a boiler explosion starts sounding like a good thing, you know you're in a pretty dark place." But seriously, you do raise a good point. Thanks to taking your dog for a walk, you thought of a solution. I think one of the fundamental problems of our work culture today is that we are constantly needing to be more "productive" and have no TIME to think of the good ideas. Or even the mediocre ideas that might lead to good ones...
NASA's Orion Flight Software Production Systems Manager Darrel G. Raines joins Planet Analog Editor Steve Taranovich and Embedded.com Editor Max Maxfield to talk about embedded flight software used in Orion Spacecraft, part of NASA's Mars mission. Live radio show and live chat. Get your questions ready.
Brought to you by