SpaceX and Learning Curves

I was having a recent discussion with a friend on a space related discussion board, when he made a rather flawed argument that I think deserves further discussion. Recently, I’ve been pointing out a lot of the problems with the approach that NASA wants to take for its return to the Moon. This has made some von Braunian sorts start to bash private companies like SpaceX in a rather uninformed and knee-jerk fashion.

He brought up some statements made by Elon Musk back in June of 2002 about what SpaceX was trying to achieve, how much he’d do it for, and when it would be ready. Since then, the schedule has slipped substantially (Q3 2005 instead of Q4 2003), and the budget has gone over by quite a bit. His analysis was that Musk was naive, and is now learning how tough rockets really are. The implication being that now that SpaceX knows “how tough space is”, that they will now budget a lot more time and money in the future for development projects, and maybe will stop making as big of claims for what it will achieve.

The problem with this as I see it is that my friend is ignoring the effect of the learning curve. When any company starts doing something, there is a learning curve. The company starts out with some important lessons that they haven’t learned yet. This is true for any industry, not just space.

For example, my first engineering job, after I graduated back in ’99 had me doing sheet metal design for the first six months I was there. I had never done sheetmetal design before in my life. It took me quite a while to come up with the first iteration of the sheet metal design for our first prototype. It had many problems with it, and I learned a lot about CAD design, working with suppliers, tolerances, doing interference checks, designing sheet metal parts, etc. It was a lot harder and more involved then I had thought. Does that mean that next time I should budget a lot more time to do sheet metal design? Not neccessarily. Now that I understood the basic rules, the next iteration went a lot faster. By the time I left the company to go serve a mission, we had gone through nearly a dozen iterations, almost none of which took as long as that first one.

Back to the example of SpaceX, I think there is good reason to believe that their next engine development program will go faster. Yes, engine design is tougher than they may have thought in June 2002, but now they know a lot more about it. They can avoid previous mistakes, capitalize on previous experience, leverage off of previous work, and previously installed infrastructure. I wouldn’t be surprised if SpaceX could build a 200,000lbf Merlin-2 type engine in half the time that it took them to develop the Merlin-1. Likewise, I think they will be able to come a lot closer to their time and cost goals for Falcon V than they did with Falcon I.

Space isn’t inherently that much harder than almost any other field of modern engineering design. It’s tough, but so is automotive or airplane design. There is a learning curve. Companies on their first project, or trying to do cutting edge work in a new field will often have cost and schedule slips. One should be very careful though in how one extrapolates from the experience they had with thier first forays into predictions of how things will go in the future.

The following two tabs change content below.
Jonathan Goff

Jonathan Goff

President/CEO at Altius Space Machines
Jonathan Goff is a space technologist, inventor, and serial space entrepreneur who created the Selenian Boondocks blog. Jon was a co-founder of Masten Space Systems, and is the founder and CEO of Altius Space Machines, a space robotics startup in Broomfield, CO. His family includes his wife, Tiffany, and five boys: Jarom (deceased), Jonathan, James, Peter, and Andrew. Jon has a BS in Manufacturing Engineering (1999) and an MS in Mechanical Engineering (2007) from Brigham Young University, and served an LDS proselytizing mission in Olongapo, Philippines from 2000-2002.
Jonathan Goff

About Jonathan Goff

Jonathan Goff is a space technologist, inventor, and serial space entrepreneur who created the Selenian Boondocks blog. Jon was a co-founder of Masten Space Systems, and is the founder and CEO of Altius Space Machines, a space robotics startup in Broomfield, CO. His family includes his wife, Tiffany, and five boys: Jarom (deceased), Jonathan, James, Peter, and Andrew. Jon has a BS in Manufacturing Engineering (1999) and an MS in Mechanical Engineering (2007) from Brigham Young University, and served an LDS proselytizing mission in Olongapo, Philippines from 2000-2002.
This entry was posted in Uncategorized. Bookmark the permalink.

3 Responses to SpaceX and Learning Curves

  1. Michael Mealling says:

    Also don’t discount the effects of having to wait for pad access in order to launch. How much of that schedule slip was due to actual design issues with the vehicle and how much were due to waiting on pad-hogs to get out of the way, forcing a costly and time consuming move to Kwajalein.

  2. Mr. X says:

    They’ve been ready to go from Vandenberg for quite some time (I’d guess around three months or more,) but they’ve had to wait for the pad-queen Titan IV to launch first.

    The “move” to Kwaj isn’t a move in the sense that the original rocket is relocating–it isn’t. Kwaj had always been a SpaceX launch site, and Falconsat had always been manifest on the Falcon-I. It’s just that the delays have caused a shakeup in the launch schedule (same for Delta IV, with the next launch now scheduled from Vandenberg.)

    The switch is all for the better. Falconsat is a lightweight student project that will not upset too many people if the booster fails. Better to launch Falconsat first to make sure that the booster will work properly for the heavier and more expensive TacSat-1.

  3. Jon Goff says:

    Michael,

    Well, if you assume about 120 people like Elon’s currently saying, and take a swag at the salary costs for a four month delay…something like $6-10M, not counting facilities and other types of overhead, not counting the cost of getting that other pad up and running, etc. So, something like 5-10% of the total Falcon I development costs.

Leave a Reply

Your email address will not be published. Required fields are marked *