We looked at the electronics, and we thought, "Boy, number one, we don't have the time or money to build or qualify anything. The full design, build, qualification cycle for space hardware runs 3-1/2 to 5 years, and it is very expensive."
We thought, in any case, we don't have to do that. There are all kinds of black boxes that fit and fill all of these functions that are currently built for the United States Air Force and for NASA and kept on he shelf by companies like Motorola, Aydin-Vector, regular suppliers. When I say on the shelf, what they kepe on the shelf is not the black box. What they keep on the shelf is the blueprints, and you call them up and order it, and they build it. They are happy to do that, because they know that when NASA comes and orders it, never ever does NASSA come and say we would like one of those please. They say we would like one of those, but, of course, we would like 16 channels instead of 14. We would like pulse-code modulations that are frequency modulation. We would like these 17 other changes. And that is what is called in the aerospace world as making it up on the changes.
So we knew that we would be able, if we intelligently selected and integrated existing black boxes without making any modifications, to put together a GNC system and electronics very reliably and without having to build or anything.
One other part to this is that if Motorola builds one of those black boxes for the Air Force, they have an extraordinary documentation system. They have to literally be able to tell you not only where every part and piece came from, but also when it was tested, and what the test results were for each resistor, each capacitor, each integrated circuit. The paper trail, which are not black but usually gold colored on the outside and cost about $40-100,000 a piece, following them around is big enough, sometimes, to fill several suitcases. That paperwork costs several times what the black box itself costs.
So one of the things that we did is, wherever possible, we ordered those components, but we ordered what is called "best commercial". Essentially it is the same parts, tested the same way, on the same equipment, by the same people and put together according to the same plan. The only difference is it doesn't have all that paper following it around.
And you know what that does, It reduces the price from 30-70%.
Some of those black boxes did have to be full milspec. Specifically, our Command Destruct receivers. The entire command destruct system from beginning to end, because the range, in our case the Western Test Range at Vandenberg, will not let us fly unless there is an Air Force officer there with his finger on the pickle, who can push the button and take us out if we at anytime start to head toward Santa Barbara or Los Angeles, for exactly the same reason that the Air Force does that with its own vehicles. The result of that, incidentally, has been that the history of range safety is 100% over 30 years. So, we want,d something in the range safety department that, when we walked onto the range, the Air Force would salute and we would go right through. That is indeed what has happened. That is the story about the electronics.
Structure
The structure that we build is very different than what is built in aerospace, and it is defined this way. Our engineers from Northrup and Lockheed, places like that, laugh about it, because when they design a part for a space vehicle, they sit down a their computers and they design a part, and they try to design it to a safety factor of perhaps 1.5 Then they take that over and they move it onto another computer, and they do what is called a finite element analysis. They stress that part, and they see where it is going to bend and break, and they put in the right extra strength. But they are also trying to take out weight.Then they go down to their contracting department and they say buy 10 of these suckers, and six months later, if they are real lucky, they get 10 of them in. Then they go over to the shop, instrument them up - strain gauges, whatever - twist them, break them, bend them and see where they bend and break and where they can take a little more weight out of them. When they get it down to a safety factor of about 1.33, if it is a manned vehicle, or about 1.1 if it is unmanned, and after they have gone through this cycle maybe 2 or 3 times, which take about 2 or 3 years, then they order what is actually going to fly, and they build it and fly it.
And, again, that is very appropriate if you are building a military missile where the absolute performance is the figure of merit. In our case, what we wanted, remember, was safety and our figure of merit is lowest cost per pound to orbit. So when the engineers sit down to design something for us, the joke is any engineer found designing below a safety factor of 2 will be taken out and shot. And that is the way we work. Most of the stuff we design often to a safety factor of 3. For example, the safety factors that I quoted you are the actual safety factors for NASA pressure vessels, 1..3 for manned, 1.1 for unmanned. Our pressure vessels are designed generally between 2 and 3. The result is that we will build something and will often put it up on a finite element model, try to bend it and break it to see how it works on the computer, then we will go out and build and then we will take it out and try to break it. If it doesn't break below 2 or 3 or wherever it was designed to break, then we build a lot more of them and we fly them.
Another difference is, of course, that our engineers don't have to wait the six months to try and break them, because we don't have to go through all of the hassle associated with a government acquisition system.
One of the things that we decided at the beginning of the company was that we would accept no government R & D money. We will find out soon how smart that turned out to be. But, very simply, I spent 10 years running a defense prime contractor. If we wanted to move something from here to here, I had to write up a justification for it. I had to write up a budget for what it would cost. I had to tell somebody how that was going to affect my projected G & A and overhead rates in 1991, 92, and 93. I had to take the whole package, make sure it was exactly the right form, forward it to the technical contracting officer who would forward it to the administrative contracting officer who would forward it then to the board who looked at it, and, if I was lucky, 45 days later I would get permission to move something from here to here.
Right now, if we want to move a valve, my chief engineer walks around the corner and argues with my VP of Operations and Engineering. They go in the shop and change a plan, initial it in all the right ways. We still do all of those things and we do all of those properly. We keep track of that, and later that afternoon, the valve is moved from there to there. That is also why we have been able to do a development program ve,y rapidly.
Next installment: Propulsion
Read Part 8: After The Fall, Resurrection
Read Part 7: The Proof Of The Pudding: SET-1
Read Part 6: The Rocket
Read Part 5: Questions & Answers
Read Part 4: Propulsion
Read Part 3: Electronics
Read Part 2: In The Beginning
Read Part 1: Introduction
Get exclusive space-related downloads, mission coverage and information
