### Author Topic: How NASA Solved a \$100 Million Problem for Five Bucks  (Read 17035 times)

#### agman25

• Full Member
• Posts: 452
• Liked: 0
• Likes Given: 1
##### How NASA Solved a \$100 Million Problem for Five Bucks
« on: 02/01/2012 04:38 PM »

#### robertross

• Senior Member
• Posts: 17804
• Westphal, Nova Scotia
• Liked: 462
• Likes Given: 4278
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #1 on: 02/02/2012 12:47 AM »
http://gizmodo.com/5880850/how-nasa-solved-a-100-million-problem-for-five-bucks

A bit more than \$5...the accelerometers are a pretty penny.

From the article:
"NASA could do better. So they grabbed a few accelerometers and attached them to the chair. With the vibration and the strobing now perfectly in sync, the display became crystal clear. And the final cost was a fraction of a fraction of a fraction of what they'd anticipated. Victory."

And obviously a lab result. Put into effect as a flight qual unit: \$100M
Remembering those who made the ultimate sacrifice for our rights & freedoms, and for those injured, visible or otherwise, in that fight.

#### Robotbeat

• Senior Member
• Posts: 28469
• Minnesota
• Liked: 8338
• Likes Given: 5481
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #2 on: 02/02/2012 04:48 PM »
http://gizmodo.com/5880850/how-nasa-solved-a-100-million-problem-for-five-bucks

A bit more than \$5...the accelerometers are a pretty penny.

From the article:
"NASA could do better. So they grabbed a few accelerometers and attached them to the chair. With the vibration and the strobing now perfectly in sync, the display became crystal clear. And the final cost was a fraction of a fraction of a fraction of what they'd anticipated. Victory."

And obviously a lab result. Put into effect as a flight qual unit: \$100M
Accelerometers actually ARE cheap. If you have an iPhone or an Android, you've got 'em (most likely). Here's a \$3 triple-axis accelerometer that would work (\$10 for the break-out board version):
http://www.sparkfun.com/categories/80

(Made in America, too, I believe.)
« Last Edit: 02/02/2012 04:53 PM by Robotbeat »
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

#### deltaV

• Full Member
• Posts: 1538
• Change in velocity
• Liked: 166
• Likes Given: 480
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #3 on: 02/02/2012 04:56 PM »
Just don't try to build an inertial navigation system with that \$3 accelerometer.

#### synchrotron

• Member
• Full Member
• Posts: 300
• Liked: 5
• Likes Given: 13
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #4 on: 02/02/2012 05:16 PM »
Just don't try to build an inertial navigation system with that \$3 accelerometer.

Of course not.  but these only need stability in the 10's or 100's of milliseconds to antialias the visuals.

It'll be more than \$5 when/if the fault-tolerant system is rolled out.

#### Robotbeat

• Senior Member
• Posts: 28469
• Minnesota
• Liked: 8338
• Likes Given: 5481
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #5 on: 02/02/2012 07:00 PM »
Just don't try to build an inertial navigation system with that \$3 accelerometer.
Why not? What sort of drift rate is acceptable?

It totally depends on the application. In some applications, a \$3 accelerometer is more than adequate.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

• Full Member
• Posts: 954
• Liked: 100
• Likes Given: 47
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #6 on: 02/02/2012 07:21 PM »
There was quite a bit more to the issue than being able to read panel displays, was there not? I think I remember fears that the TO could have been severe enough to incapacitate the crew. So, Gizmodo's story about strobing panel displays wasn't quite the solution to the problem.

Don't depend on those gadget sites for real science. Especially not Giz.
« Last Edit: 02/02/2012 07:21 PM by DMeader »

#### Patchouli

• Senior Member
• Posts: 4400
• Liked: 182
• Likes Given: 359
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #7 on: 02/03/2012 02:19 AM »
Just don't try to build an inertial navigation system with that \$3 accelerometer.
Why not? What sort of drift rate is acceptable?

It totally depends on the application. In some applications, a \$3 accelerometer is more than adequate.

For this application it just needs to be able to sense the frequency of the vibration so the cheap ones are good enough.
« Last Edit: 02/03/2012 02:19 AM by Patchouli »

#### Proponent

• Senior Member
• Posts: 5529
• Liked: 1066
• Likes Given: 664
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #8 on: 02/03/2012 04:25 AM »
There was quite a bit more to the issue than being able to read panel displays, was there not? I think I remember fears that the TO could have been severe enough to incapacitate the crew.

The Gizmodo article itself mentions that when the author rode the TO simulator, the acceleration was limited to 0.5 G, and that the people conducting the test wouldn't let him try it at the 0.7-G level that was anticipated for the actual Ares I.  That tells you right there that there's a considerably more serious risk to the crew than just illegible displays.

EDIT: "he" -> "the author" in first sentence.
« Last Edit: 02/06/2012 02:57 AM by Proponent »

#### robertross

• Senior Member
• Posts: 17804
• Westphal, Nova Scotia
• Liked: 462
• Likes Given: 4278
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #9 on: 02/03/2012 03:52 PM »
Just don't try to build an inertial navigation system with that \$3 accelerometer.
Why not? What sort of drift rate is acceptable?

It totally depends on the application. In some applications, a \$3 accelerometer is more than adequate.

For this application it just needs to be able to sense the frequency of the vibration so the cheap ones are good enough.

Bet you life on that?

I wouldn't.
Remembering those who made the ultimate sacrifice for our rights & freedoms, and for those injured, visible or otherwise, in that fight.

#### Robotbeat

• Senior Member
• Posts: 28469
• Minnesota
• Liked: 8338
• Likes Given: 5481
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #10 on: 02/03/2012 05:27 PM »
Just don't try to build an inertial navigation system with that \$3 accelerometer.
Why not? What sort of drift rate is acceptable?

It totally depends on the application. In some applications, a \$3 accelerometer is more than adequate.

For this application it just needs to be able to sense the frequency of the vibration so the cheap ones are good enough.

Bet you life on that?

I wouldn't.
That is non-sense. There are \$3 pieces in an airplane that if a few of the right ones failed at once, you'd die.

Just because you spend a lot more on something doesn't mean it's necessarily better or safer.

Good on NASA for finding a clever and inexpensive solution to an expensive problem. Creative thinking like that to solve a safety problem in an inexpensive manner should be PRAISED.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

#### deltaV

• Full Member
• Posts: 1538
• Change in velocity
• Liked: 166
• Likes Given: 480
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #11 on: 02/03/2012 06:26 PM »
Bet you life on that?

I wouldn't.

An astronaut no longer being able to read his/her displays is not a particularly life-threatening situation. There are multiple backups, namely several flight computers to fly the rocket, the other pilot to watch the flight computers, a few hundred people on the ground to read their non-shaking displays and advise the other pilot, and the pilot who can't read his/her displays to make sure the other pilot doesn't touch anything.

(Edit: also the anti-vibration system could be designed to always flash at between say 10.5 and 12.5 Hz, so even if the accelerometers conspired to mess things up the displays would still be sort of readable.)

The airbag deployment system in some NASA employee's car activating improperly, on the other hand, is a very serious matter since an air bag deploying at the wrong time or failing to deploy when it should can easily cause serious injury or death. This greater consequence of failure suggests that airbag deployment systems should be engineered for greater reliability than the anti-vibration Orion display system. Inexpensive accelerometers (although maybe not \$3 ones) are sufficient for airbag deployment, so why not for Orion's displays?
« Last Edit: 02/03/2012 06:36 PM by deltaV »

#### jongoff

• Recovering Rocket Plumber/Space Entrepreneur
• Senior Member
• Posts: 6190
• Lafayette/Broomfield, CO
• Liked: 2294
• Likes Given: 775
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #12 on: 02/06/2012 04:20 PM »
It's amazing how cheap a solution can be when you don't factor in the labor cost of finding it, the cost of implementing it, and only count the cost of the cheapest first-draft proof of concept prototype hardware.

Still a clever workaround, just think the article is rather deceptive.

~Jon

#### Jim

• Night Gator
• Senior Member
• Posts: 32422
• Cape Canaveral Spaceport
• Liked: 11163
• Likes Given: 331
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #13 on: 02/06/2012 04:31 PM »
It's amazing how cheap a solution can be when you don't factor in the labor cost of finding it, the cost of implementing it, and only count the cost of the cheapest first-draft proof of concept prototype hardware.

Still a clever workaround, just think the article is rather deceptive.

~Jon

Also, when it is just a bandaid and the real problems still exists.

• Senior Member
• Posts: 3246
• Boca Chica, Texas
• Liked: 5383
• Likes Given: 340
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #14 on: 02/06/2012 04:36 PM »
They could have just strapped an iPhone to it. A vibration app called "Vibration" is remarkable accurate and records Gs and frequency.

#### JMS

• Full Member
• Posts: 327
• Liked: 7
• Likes Given: 14
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #15 on: 02/06/2012 04:48 PM »
That is non-sense. There are \$3 pieces in an airplane that if a few of the right ones failed at once, you'd die.

There are very few parts on even a general aviation aircraft that costs only \$3. Hell, even Dzus fasteners cost more than that...

#### Robotbeat

• Senior Member
• Posts: 28469
• Minnesota
• Liked: 8338
• Likes Given: 5481
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #16 on: 02/06/2012 05:24 PM »
That is non-sense. There are \$3 pieces in an airplane that if a few of the right ones failed at once, you'd die.

There are very few parts on even a general aviation aircraft that costs only \$3. Hell, even Dzus fasteners cost more than that...
There are 1-cent surface-mount resistors that if several happened to fail in the different avionics strings at once, you'd be out of luck for fly-by-wire avionics systems. (Of course, the possibility of that happening is very small, especially for simultaneous failures.)

No reason you wouldn't use a \$3 part for a solution like this. If you felt bad about it, I could sell it to you for \$300 if you'd like.

Of course, Jon Goff (and Jim) is right. The individual parts aren't usually the bulk of the expense.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

#### Patchouli

• Senior Member
• Posts: 4400
• Liked: 182
• Likes Given: 359
##### Re: How NASA Solved a \$100 Million Problem for Five Bucks
« Reply #17 on: 02/07/2012 06:08 PM »
That is non-sense. There are \$3 pieces in an airplane that if a few of the right ones failed at once, you'd die.

There are very few parts on even a general aviation aircraft that costs only \$3. Hell, even Dzus fasteners cost more than that...

Most of the fasteners used in a typical aircraft are rivets that only cost a few cents each.
But if enough of them fail you'll have a very bad day.
« Last Edit: 02/07/2012 06:13 PM by Patchouli »

Tags: