Slashdot Mirror
NASA Sticking To Imperial Units For Shuttle Replacement
JerryQ sends in a story at New Scientist about the criticism NASA is taking for deciding to use Imperial units in the development of the Constellation program, their project to replace the space shuttle. "The sticking point is that Ares is a shuttle-derived design — it uses solid rocket boosters whose dimensions and technology are based on those currently strapped to either side of the shuttle's giant liquid fuel tank. And the shuttle's 30-year-old specifications, design drawings and software are rooted in pounds and feet rather than newtons and meters. ... NASA recently calculated that converting the relevant drawings, software and documentation to the 'International System' of units (SI) would cost a total of $370 million — almost half the cost of a 2009 shuttle launch, which costs a total of $759 million. 'We found the cost of converting to SI would exceed what we can afford,' says [NASA spokesman Grey Hautaluoma]."
You know, a lot of Europeans probably think that U.S. reluctance to embrace the metric system is just another example of our arrogance. But a lot of Americans (like me) are genuinely interested in adopting this system. We even passed a law in 1975 trying to mandate it.
The real problem is that it is surprisingly hard to embrace a new system of measurement when you've spent your entire life thinking in different terms. Try as I might, I still can't picture a kilometer without converting it to a mile first, and still can't picture a centimeter without converting it to inches. The meter is a lot easier because it's pretty analogous to the yard. I think maybe your brain gets locked into a certain measurement pattern pretty early in life and it's very difficult to get out of it, even though many of us would happily embrace it. I'm still trying to think more in metric, but it requires a surprising amount mental effort to do so.
It's not that Americans are really all that arrogant or stubborn about the imperial system. We've actually been trying to embrace the metric system for some time.
SJW: Someone who has run out of real oppression, and has to fake it.
Which is the difference between scientists and engineers.... Sometimes the right decision is to listen to the engineers and not the scientists.
The scientists have it easy. They work in theories and numbers. The engineers have to produce usable physical objects. They have to do so in an environment that had significantly established manufacturing infrastructure before the SI standard existed. The countries that have converted to SI are the countries that were late to the industrial revolution party. It is expensive and difficult to overcome a massive established base of equipment. And it's a self perpetuating problem, because you can't just replace individual tools and machines as they wear out. An individual replacement has to be compatible with the rest of your infrastructure.
Sigh all you like. Short of a massive cash investment (Many Trillions of Dollars), or all manufacturing leaving the US and UK for good, Imperial units will stay and be indifferent to the sighs of the "rest" of the world.
(Incidentally, this would have been a *great* thing to spend stimulus money on instead of government employee salaries and other stupid programs.)
Maybe they should be re-thinking their plan to use 30-year-old technology on their flagship 21st Century project. Really: what does it say about the technical competence of NASA that they admit to being unable to use SI units, even though they would like to?
You make it sound so simple...when, in fact, this is quite literally rocket science here.
One of the common stories here is people needing to rewrite an entire project because of a new language fad. The old project worked. Rewriting it first means you have to replicate the old project and then deal with new bugs while the old project had all the bugs mostly ironed out.
Why do we insist NASA to reinvent the wheel when we're so against it in our own profession?
say 5:1 at each level for management span of control. Figure 100 enginers, 20 supervisors, 4 middle managers, 1 project lead, 10 admin people, and another team about 20% that big for QA. Each of these people costs twice as much as their salary, plus the cost of a building for a year to house 200 people. at 100 sq/ft per person and $20/sq ft/mo if they're in a decent area for office space, there's an extra chunk of money. it has to be contracted out, so the contractor has to make a profit on it, and carry a shitton of liability insurance. There's also the NASA guys to QA this. When changing the units, every rounding error has to be validated. Poof, you just blew through $300M or so in a few years. Congrats.
Since you clearly know all the details about NASA's technical documentation, I'm curious to know what format the drawings are in. You must be implying a knowledge that they're all in a convenient digital vector form, because any rasters or hard-copies would be quite a chore to change. What is the total volume of documentation we're talking about here? How many copies are kept at how many locations, and in what forms? Also, perhaps you can elaborate on the tolerances of the measurements and how that would relate to rounding errors when multiplying a measurement by a factor like 0.3048006096012 (meters per foot). It's a relief to hear, as you are clearly implying, that NASA doesn't use any off-the-shelf fasteners. If they did, and if all of their specs were to be in metric, they'd probaby have to convert to metric parts, which could require some subtle redesign even if the above-mentioned rounding issues don't. The fact is, I don't know which of the above are factors -- and I'm willing to bet you don't either, which makes me wonder at the hubris in your post.
the issue isn't just one of redoing the drawings along with the various checks and cross checks to make certain the units were converted properly. I'm sure they could that, but the resulting set of new drawings would be extremely prone to encouraging mistakes. As a minor example. Let's assume that on one piece they currently have a dimension of 12 inches +/- 0.01 inches. So they convert this dimension to metric giving a new value of 30.48 cm +/- 0.025 cm. Excuse me?!?!? That's a rather odd and strange dimensional target to hand off to the machinest. And you'll be getting these rather strange dimensions for everything on the original design. Frankly using the metric measurements would make that rocket utterly hell to construct. So the "proper" solution would be to use the original design and then stretch/shrink various dimensions in order to make the dimensions "rounder" and easier to manufacture. But upon doing that, they have effectively come up with a new design that has to be recertified.
I prefer imperial units for lots of everyday tasks like cooking. Imperial units are much closer to a binary-based system, which is very convenient for human beings. Two cups in a pint. Two pints in a quart. An ounce of water weighs about an ounce. A pint of water weighs about a pound. Human beings are very good at halving or doubling things by eyeball, but we're lousy at dividing into tenths.
But if you're building a fucking spaceship, use SI units for Christ's sake.
Yes...stored in electronic form. That's right. Wait, remind me again, what file format did Autocad use in the 1960s? 1970s? 1980s? Was it DWG back then?
Plus we're not building another shuttle. We're going back to the days of Apollo, with a capsule^Wspacecraft on top of a rocket. Apparently, though, they found that they can utilize the SRB design for part of the new project. (The big white rockets that get reused after launches.) The SRBs date from the start of the Shuttle era which...erm, yeah. 1970s.
So here we have a rocket booster already designed that works like a champion. The blueprints are all done. They work. They're reusable. They've been fieldtested over 100 times.
And you want to redesign them essentially from scratch? As many coders here want to say to their bosses when upper eschelon wants to recode an application in the new flavor of the month language: if it is not broken, do not fix it.
Is that why Top Gear reviews all the cars in terms of miles per hour and horsepowers?
It doesn't hurt to be nice.
after several years it becomes difficult if not impossible to find the original file...We also use a fairly vigorous quality control system
How do you keep quality on your products, but not even keep your original documentation files? What happens if there is a change?
our drawings are created in autocad... unit conversions are not a trivial operation
The engineering team where I work uses Solidworks, and there are macros to do the conversions. Of course, those macros only work on the original files, not the printed documents... :-) So that brings us back to having lost the files...
It's not like they're building anything new or buying raw materials; they just need someone to re-draw plans with new measurements in a different system.
Frankly, and without trying to be insulting, you're so ignorant of what the issue is that it's laughable that you even have an opinion on it.
This isn't a matter of trivia, where we are worried if plans are marked in inches or mm. Change to metric, now every bolt must be metric pitch thread, every nut must be changed to accomodate. Every calculation of mass and structural integrity has to be reexamined and recalculated for new components. You don't just magically say "ok, our 3/8" bolts are now to be called 9.525mm bolts" and call it a day.
$370 million to do undergrad-level (at most) grunt work? Isn't that what unpaid interns are for?
We're just talking about units of measure. If it is easier to use imperial units because previous design and drawings were done in imperial, then that's the smart choice. I would be upset if NASA was wasting taxpayer money just so that the design could be done in metric. I actually applaud NASA for making a smart, cost/benefit engineering decision.
The only real valid arguments I've ever heard for using metric are that (a) it's easy to learn the conversions, and (b) everyone else uses it for all science and egineering.
(b) is the reason that NASA should just use metric... And anyone else in the world doing any kind of science or engineering.
But for everyday life, imperial or American units turn out to have a lot of utility that most people aren't aware of, because most of us of the younger generation have just relied on calculators doing decimal calculations for us most of the time. If instead you picture fractions in your head, imerial or American units are quite handy. They also often match real world objects a bit closer. If you're dividing meters into centimeters, you can really only talk about tenths, hundreds, etc. If you're dividing yards into feet and inches, or pounds into ounces, etc. you have thirds, 16ths, 12ths, and all kinds of other useful fractions to use to think about the divisions. Find a carpenter who is good at this to see what I mean. Same with volume and weight; if you do a lot of cooking and modifying quantities in recipes you can get good at those conversions.
The fact of what it's based off of is irrelevant, everything comes down to the metric system being consistent in staying in base 10, and the imperial system is not.
Next time you should replace 4999 of those hours with a simple BASH script.
The original files were not available. What shell do you use that compiles to paper?
Write your representatives! Repeal the 2nd Law of Thermodynamics!
both SI and Imperial units are pegged to arbitrary things. In the case of Imperial units it was some king's foot. In the case of SI it is the distance light travels in some amount of time. whatever.
That's beside the pont. Yes the SI units are pegged to arbitrary things but they are not arbitrarily pegged to *eachother*.
Converting from centimeters to kilometers requires dividing by 100,000. I can do that in my head.
Converting from miles to inches requires dividing by 63,360. I can't do that in my head.
Exactly. And that's a flaw that's been fixed.
How many flaws will be introduced if they have to be redesigned from scratch?
The problem with imperial is not what it is based upon (actually, these days the US units are all defined by reference to the SI units anyway - since 1959 an inch is defined as 2.54 cm - see http://en.wikipedia.org/wiki/Inch ) the problem with the imperial system is the arbitrariness and inconsistency of the relationships between the units. The SI system has a consistent relationship between all of the units, and a consistent naming system and a consistent abbreviation system. In the imperial system, the relationships between units are not only arbitrary, but they are also inconsistent, there are multiple uses of the same word (ounces for example) used to describe different measurements (weight as well as volume) or dry vs liquid volumes.
Except unit conversions are not the norm. When a grocery store manager orders potatoes, it doesn't really matter if he orders 200 10 pound bags (which is really tough to convert to 2,000 pounds) or if he orders 200 5 kilogram bags (which is really tough to convert to 1,000 kilograms).
Sure, sometimes someone has to get a calculator to figure out how many inches are in 200 feet (but hopefully not most people) before they figure out how many 1.65 inch pieces they can cut that 200 feet into, but the other guy is going to need a calculator (or some scratch paper, whatever) to figure out how many 4.191 cm pieces they can get from 60 meters anyway.
Nerd rage is the funniest rage.
From The Snopes Article:
"True, but for trivial and unremarkable reasons."
But wasn't that kind of the point of the story?
And a liter of water weight exactly a kg. There are exactly 1,000 meters in a kilometer. And there's exactly a year in a light year.
Wait ... one of those is wrong.
let people know that this was meant to be funny, not true. Perhaps it was just too subtle for you.
Perhaps you just fail at "funny". Since your comment was clearly not funny*, the only plausible explanation was "sincerely believed as true, no matter how wrong.
*NB: factually and incontrovertibly not funny. Don't bother trying to salve your ego by blaming your readers' sense of humor. Their sense of humor is not defective; your attempt at humor was.
Welcome to the Panopticon. Used to be a prison, now it's your home.
I'm sorry, it really sounded like you just called the metric system a "new language fad". Seriously?
Look, it's not about saving money now, it's about the future. They've already lost a multi-million dollar project because they haven't gone metric yet, and a similar mistake in the future is not unlikely.
Much of the production for space parts occurs outside of the US, where they use what? Metric. What kind of overhead do you think they charge to supply imperial equipment as well as the metric that they supply to the rest of the world? How much time is wasted converting between yards, cubic feet, gallons, and pounds instead of using a system that mostly just involves moving the decimal point around? What about the fact that all of their scientific data is measured in metric, which is how they share it with the rest of the world? Do you really not see a reason for them to move to the new measurement system that they will undoubtedly move to at some point in the future anyway?
Besides, as the OP points out, the cost estimate for this is insane.
There's a VERY good reason to re-use the SRB's, they are a well tested design with the flaws worked out and the real operating parameters known.
We would be remiss if we did not note that the engineering kinks of the SRBs have been ironed out, by that they killed seven people in the process.
Don't blame me, I voted for Baltar.
I don't know how many times I've tried to read a science forum only to have these "should be in metric" unit Nazi losers fill the thing up with pointless posts. Please, if you have nothing to contribute to a science article...just don't post. We also don't care about the billionth stupid post of "but but but, the earth is only 6,000 years old.
It never ceases to amaze me the resistance to going to metric here in the states for measurements but no one bats an eyelash at the fact our money is basically metric (base 10).
It is in fact soo damn easy that we can instinctively give somone a $5 and a penny for something that costs $4.01 so we can get back a dollar rather then 3 quarters 2 dimes and 4 pennies....
Boo metric it's too damn easy to use! Forget cutting a board 1.46 meters in half. it's too damn hard to cut it .73 meters! Better yet that 3 5/8th inch board needs to be cut in half so we need umm... err... need some scratch paper here....
-=[ Who Is John Galt? ]=-
And once the drawings are re-drawn, you have to verify the individual drawings. Then you have to verify the interfaces to make sure that vendor 'A' didn't round his tolerances in a direction that means his part will no longer properly mate with a part from vendor 'B'. Then you have to withdraw the old drawings from service and replace them with the new in an orderly fashion. Somewhere along the way you also have to not only update the references between drawings, but also the hundreds of thousands of pages of documentation, specifications, etc... that reference these drawings.
The individual steps are bone simple - but there are a lot of individual steps and they interact in various complicated ways.
An additional problem is that all this has to be done while those drawings, specifications, etc... etc... are in daily use at facilities scattered across the country, which means you have a fairly difficult problem not only in making these changes - but in ensuring everybody is 'on the same page'...
Yeah, that's very helpful.
Patents Drive Free Software as Hurricanes Drive Construction Industry
If they paid their engineers $150,000/year, they could hire almost 2500 engineers for a year-long project.
Or, pay 10 engineers to make sure that the adapter between the (imperial) boosters and (metric) Ares is properly sized and be done with it. If you're pulling a boat behind a truck, you don't care if the truck engine's bolts are metric and the boat's are imperial because they don't have anything to do with each other. As long as the hitch pieces are compatible, you're golden.
Dewey, what part of this looks like authorities should be involved?
You scan the entire document in, OCR for text, and try to perform an automatic conversion. Then people go through where they see a graphic of the original scanned text and the OCR'd text. For each one the person selects it is correct, or adds the correction. I've known people that do this for a living, and this was cutting edge methodology 15 years ago. Now it is pretty standard.
The alternative method is to detect the text, but not OCR it. Force the user to enter the text themselves. In either case, you show the graphic to multiple users and check for consistent answers between them. The best part about this is that reading text can be done by minimum wage workers, and can be done extremely fast after a while.
The resulting electronic file has an image containing all lines from the original drawing, with text replacing scanned in text. This can be printed out and should look identical to the original, but with converted units.
(The conversion is not necessary to see along the OCR'd text as proper conversions are trivial to perform automatically. All you really need to see is that the original OCR occurred properly.)
So your owner there... he wants 200 10 pound bags, good... good...easy enough...
How many ounces is that? Tons?
It's a stupid system which is held onto simply because it's what we're used to. (and yes, it's easy to google the answer, but I could give you the kilogram converstions faster then you could type it)
Beware of he who would deny you access to information, for in his heart, he dreams himself your master.
Since I was born and live in the U.S., I tend to think in Imperial units, too. However, the fallacy with your argument is that it is only intuitive to you and I because that is what we are accustomed to using. I submit that, had you and I been brought up using SI units, 5'3" would sound just as foreign as 1.6m currently does to us.
In other words, it only takes a generation for your argument to no longer be valid.
MCSE? No, sir...I don't do Windows. Yes, I am an idealist. What's your point?
Please advise us on how you're going to train every machinist, and QC agent, just to name two job categories on how to measure 30.22mm with calipers that are intended to measure in Imperial? If they can't do this, do it reliably, and do it accurately then you're going to have some funny fitting parts on those Ares.
That is just one very simple example in two very limited job categories where changing from SI to Metric would introduce horrible, and potentially disastrous, difficulties.
That *is* the lesson learned, and being followed.
There are other issues. A machine setup that can make a .5 inch bore to ten-thousandths of an inch precision, cannot necessarily be changed to make a 1.27cm bore with the same precision. Many of the machine tools used in aerospace are calibrated in SAE units, and the machines cannot be replaced economically, if at all -- lathes, milling machines, grinders etc., still in service since the 1960s or even 1940s, refit for CNC, still turning out high-precision work, and some of these cannot really be replaced.
Yes, you can calibrate a CNC milling machine to work in metric units, but precision in one system is not equivalent to precision in another system, especially when you're talking about a screw-calibrated device that is extremely accurate at whole units in its reference scale.
I'm only considering linear measurement here. When you're talking about rocket engines, I can only wonder how many physical equations have to change if you change the reference units of all the components.
In an ideal world, changing between measurement scales creates an equivalence set, but in the real world this is not such a foregone conclusion.
-fb Everything not expressly forbidden is now mandatory.
Yes, it's an artificial example, but you're more likely to divide things in halves (or even thirds) than into fifths simply because it's easier. If I had to do fifths, I'd probably start with sixths (1/3 * 1/2) and estimate 1/5 to be slightly bigger.
My point was not that any dummy off the street could get 1 fl oz from 1 gal. My point was that any dummy off the street could successfully divide the liquid in half repeatedly if that's what you told him to do.
I'd forgotten just how darn tiny 1 mL is. Give me 10 mL and call it fair enough.
Again, even thirds is a lot easier than fifths. Imagine cutting a cake: cut almost in the middle, but not; cut halves from the small side, and thirds from the larger one. The halves and thirds are going to be easy, but good luck getting the first cut right. Luckily nobody will care, since somebody probably wanted the bigger piece anyway.
1760 is the conversion from yards to miles. I can't say I've ever had to do that, so it really doesn't concern me. Yards actually aren't used that much, and distances long enough to be expressed in miles aren't usually thought of in feet, much less yards.
Decimal is easier? On paper, maybe. On a computer, absolutely not. Binary can't express decimals very well anyway. I see no reason that a base 10 system would be better, on a computer, than a base 12 system.
Alexander Peter Kristopeit bought his basement from his mommy for one dollar.
Well, maybe because 200 grams of carrots would be two carrots and most people need higher amounts.
"It's such a fine line between stupid and clever" -- David St. Hubbins, Spinal Tap
Does it matter? I assume the TARDIS does unit translation, so you hear whatever units you prefer.