Hacker News Reader

Comment by cube00 | Hacker News Reader

> The internal illumination for all those switch panels was accomplished via 1,600 tiny incandescent bulbs that were soldered onto printed circuit boards on the backs of the panels.

That's a strange design decision given it'd be known that those incandescent bulbs would need regular replacement.

It's a shame they went with a static display rather then add a "ride" aspect. While some may argue this is a museum not a theme park, the more engaging it is the more people will visit and learn.

View on HN

> It's a shame they went with a static display rather then add a "ride" aspect.

This simulator is over 50 years old; anyone actively using it, especially unsupervised, would cause damage.

I mean I'm all for someone building a newer and more resilient simulator, or reproducing it in VR, but at this point this is a genuine museum piece and the focus should be on conservation, not entertainment.

>This simulator is over 50 years old; anyone actively using it, especially unsupervised, would cause damage.

It's expensive and takes up a lot of space. Letting people use it, even if they can't truly "fly" it, is probably the best shot it has at having any staying power in its new home. Sure that'll probably require some motor, lightbulb and upholstery replacements over time but it should be fine as long as they stay on top of it.

When the museum comes under hard times a decade or more in the future, after the novelty has worn off, being a cost positive attraction vs a cost neutral use of space that could be used for something that can attract people is likely the difference between the simulator staying or going.

> This simulator is over 50 years old; anyone actively using it, especially unsupervised, would cause damage.

The Delta museum near the Atlanta airport has an old, full-motion, level D 737 flight simulator anyone can pay to "fly". A operator joins you in the simulator and sets up and operates the simulator for you. $425 gets you 45 minutes and it seems to stay relatively busy.

I would imagine the operator of a Space Shuttle simulator could charge quite a bit more for the opportunity to fly one of the most famous aircraft ever built.

It would be cheaper to pay someone like CAE to build a modern hardware space-shuttle simulator if you wanted rides. Then it would be easy to replace parts, since they actively maintain hundreds of aircraft simulators.

It wouldn't really have the reliability to be a ride.

Theme park rides are designed from scratch for high-uptime usage. Like real aircraft, something like this probably required several hours of maintenance for every hour of "flight-time" back in the day.

It was often said the the shuttle was more rebuildable than reusable.

Just doing some quick back-of-the-napkin math, 1,600 miniature bayonet sockets potentially add up to hundreds of pounds of mass. That may have been a factor in the decision to solder the bulbs directly onto the boards. Especially when replacement probably required disassembling panels, so the soldering isn't really time worth saving. Just a guess though, and the significance varies quite a bit depending on where you set the slider for "mass of one socket." Still, it's the same order of magnitude that they shaved off when they decided not to paint the external tank, so it's plausible.

The simulator was designed in the days when white LEDs did not yet exist, and likely the time on display is well in excess of what it normally would have gone through when in use as a training device.

I don't think GP's comment was about the use of incadescent bulbs, I think it was about soldering them onto the board.

In vibration sensitive applications that makes perfect sense.

I mean, yeah, I wasn't in the room when the Space Shuttle was built. I don't know the "surprising level of detail" involved. I guess it's a round-about way of saying, "I'd like to know what those details were to understand why having to desolder burnt-out bulbs is better than using a spring-loaded, locking fitting."

While it is always possible that a whole army of experts would miss something so obvious as the existence of spring loaded locking fittings (aka banjonet bulbs), the safe assumption is that they thought of it and rejected it for valid reasons. Vibration is a very nasty problem.

i dont know that burnt out bulbs are a problem: you can trade reliability for efficiency: don’t burn the bulb at its brightest and it’s unlikely it will burn out

The shuttle experiences severe vibration during launch. You don't want your indicator lights shaking loose during launch, so therefore they should be soldered.

LEDs were available at the time the shuttles were built, but were they considered too new to take a bet on?

Also, I'm bummed they replaced them because LEDs have a very different feel.

The only thing you would be using LEDs for in the late 70s would be indicator lights and 7-segment displays. The only colors available are red, green and yellow, and they wen't bright enough to illuminate dials, instrument panels or cockpits. Even if they did use LEDs where they made sense, they would still need hundreds of incandescent bulbs.

> LEDs were available at the time the shuttles were built

Assembly of Endeavour, the final STS to be built, was completed in 1990. Blue LEDs appeared in 1993 and white ones several years alter. During the shuttle era, LEDS were invariably red, green and yellow. I'd be surprised if those colours were considered suitable for general illumination of flight instrumentation.

> Blue LEDs appeared in 1993

Then it is interesting that my father's 1989 Wolfsburg Edition Jetta has a blue LED in the dash

"The first blue-violet LED using magnesium-doped gallium nitride was made at Stanford University in 1972 by Herb Maruska and Wally Rhines, doctoral students in materials science and engineering." [1]

"In August 1989, Cree introduced the first commercially available blue LED based on the indirect bandgap semiconductor, silicon carbide (SiC)." [2]

What you must be thinking of: "Two years later, in 1993, high-brightness blue LEDs were demonstrated by Shuji Nakamura of Nichia Corporation using a gallium nitride growth process." [3]

[1,2,3] https://en.m.wikipedia.org/wiki/Light-emitting_diode#blue_LE...

Blue LEDs were not in commercial products in the 1980s, even if the technology existed. Perhaps due to cost, I don’t really know, but I was there and remember it well. In the 90s, companies went nuts with blue LEDs using them as power indicators in everything.

You are simply mistaken. Volkswagen used blue LED in their 1989 model year dash, models presumably available in 1988. [1]

[1] https://www.youtube.com/watch?v=snld2cT83qg&t=1m56s

I think you might be mistaken. Those are lenses that are shaped like LEDs but there are bulbs behind them.

https://www.youtube.com/watch?v=Oqpyc8huRLE&t=20s

god bless youtube, what a specific piece of history to have a reference for

We've reached peak YouTube. But that does answer the poster's claim. The lens is perfectly shaped to emulate an LED.

Clearly I was referring to commercial off-the-shelf products rated for aerospace use, not experimental lab prototypes.

Regarding your father's car, I'd suggest the indicator in question is either an after-market addition, or an original incandescent mini-bulb behind a blue plastic filter. The latter was a common way to implement coloured dashboard lights back then.

> Clearly I was referring to commercial off-the-shelf products rated for aerospace use, not experimental lab prototypes.

Except for "clearly," then sure. I should have known because when most people talk about LED they mean "hardened for aerospace LED." Right? Come on.

> Regarding your father's car

Nope. Volkswagen used what anyone would recognize as an ordinary blue LED in their 1989 model year dash, along with ordinary red, green & yellow LEDs.

In my 1984 (A1) Rabbit GTI the indicator that looks like a blue LED is a piece of blue plastic backlit with a white lamp. The A2 cars (including the 1989 Jetta) used the same setup.

No. Those are blue lenses only. They did not use blue LEDS in 1989:

https://www.youtube.com/watch?v=Oqpyc8huRLE&t=20s

Incandescent bulbs can be very reliable when run below the rated voltage. The color temperature will be a bit lower but they last a long time.

They are also incredibly tolerant to current/voltage spikes because they are purely resistive, which makes them ideal for situations when power could be 'interesting' as they will still function unless they are run well beyond normal limits. So when it's more important that it just works, it's not a bad design.

Also soldering them would be least of the cost concerns for the shuttle program, which had much bigger and nastier cost centers.

I think the concern with soldering goes beyond just cost. Vibrations and micro-gravity do have a lot of concerns when it comes to shorting and other reliability issues.

I thought it was a strange restoration decisions to replace them, this was built before led's were around.

Reply on news.ycombinator.com