James Webb Telescope

My argument is the dodgy kind, from incredulity. It is the best I have, but partly that machining a large mirror seems to make this last spec a nonsensical tolerance.
 
Samson said:
My argument is the dodgy kind, from incredulity. It is the best I have, but partly that machining a large mirror seems to make this last spec a nonsensical tolerance.
Click to expand...
This machine is intended to be veeeerrrrryyyyy precise. It's looking for tiny fluctuations in the heat received from absurdly distant objects. It needs nanometre tolerances.
 
Samson said:
My argument is the dodgy kind, from incredulity. It is the best I have, but partly that machining a large mirror seems to make this last spec a nonsensical tolerance.
Click to expand...

You understand that astronomical mirrors are ground & figured to an accuracy measured in fractions of a wavelength of light? The error in the equipment that ground the Hubble mirror caused it to be ground with an aberration 1/50th the thickness of a human hair. It terms of mirror grinding.. that is a massive error.

When you are trying to resolve objects that are a matter of extremely tiny fractions of an arcsecond in angular diameter, using a mirror that is made of multiple discrete segments, those segments have to be aligned very, very, very, very, very, VERY precisely.

For example the nearest star, Alpha Centauri, has an angular diameter of 0.007" ... that is 7/1000ths of an arcsecond (2 millionths of a degree)

The star with the largest angular diameter, Alpha Orionis (Betelgeuse) is 0.044" ... 44/1000ths of an arcsecond (about 12 millionths of a degree)
 
Last edited:
Puppycow said:
Of course, it varies, but one figure given is 25.4 μm as a standardized result.
Click to expand...
Side note: that may look at first blush like it’s got 3 digit of accuracy, but it doesn’t. That’s just 0.001 inches, converted to metric. So it’s got barely 1 digit of accuracy.
 
Ziggurat said:
Side note: that may look at first blush like it’s got 3 digit of accuracy, but it doesn’t. That’s just 0.001 inches, converted to metric. So it’s got barely 1 digit of accuracy.
Click to expand...

No. Yards, Feet and Inches are Imperial units. In science we use SI units.

https://www.nist.gov/pml/weights-and-measures/metric-si/si-units

Things that are expressed in SI units are measured in SI units, they are not measured in Imperial units and then converted.

ETA: However, I know why you said what you did. You are correct that someone took 0.001in and converted it to metric, but I just wanted to make it clear that is not how things are done in science.
 
Last edited:
Just as an aside from the main topic of the thread. I do hope one day that the
United States adopts the metric system as a legal standard, so that we could
convert English units like the inch into meters. Like have the inch be exactly
256 ten thousands of a meter which makes conversions like a sixteenth of an
inch real easy. But the political situation right makes that impossible.

Until then we will have to work with rough approximations like an inch only
being 254 ten thousandths of a meter. Sigh. Carry on with the main topic.
 
smartcooky said:
No. Yards, Feet and Inches are Imperial units. In science we use SI units.
Click to expand...

Most commonly, but there’s no rule requiring that, nor would such a rule be enforceable or even make sense. Science can be done with any set of units, so long as it is done properly. And metric, while convenient, is no guarantee that something is done properly.

Things that are expressed in SI units are measured in SI units, they are not measured in Imperial units and then converted.
Click to expand...

That’s also common practice, but again, no rule, no way to enforce, and not even any good reason you can’t do otherwise. Units are arbitrary, as long as you convert correctly it doesn’t matter. But correct conversion includes error bars, and if you use implicit error bars based on digits rather than explicit ones (as 0.001 inches likely did), they won’t generally remain implicit when you convert. That ultimately is the problem with 25.4 um.
 
Solitaire said:
Just as an aside from the main topic of the thread. I do hope one day that the United States adopts the metric system as a legal standard, so that we could convert English units like the inch into meters. Like have the inch be exactly 256 ten thousands of a meter which makes conversions like a sixteenth of an inch real easy. But the political situation right makes that impossible.

Until then we will have to work with rough approximations like an inch only being 254 ten thousandths of a meter. Sigh. Carry on with the main topic.
Click to expand...

See also: Mars Climate Orbiter, which was lost on Mars orbital insertion due to a mixup where NASA used metric units (newton-seconds) in its specifications but Lockheed Martin used US customary units (foot-seconds) in its calculations.
 
Blue Mountain said:
See also: Mars Climate Orbiter, which was lost on Mars orbital insertion due to a mixup where NASA used metric units (newton-seconds) in its specifications but Lockheed Martin used US customary units (foot-seconds) in its calculations.
Click to expand...

There have even been air accidents such as the Gimli Glider incident where the primary cause was mistaking pounds for kilograms, which resulted in the aircraft carrying only 45% of its required fuel load, and then running out of fuel at 12,500m (41,000 ft)
 
Solitaire said:
Until then we will have to work with rough approximations like an inch only
being 254 ten thousandths of a meter. Sigh. Carry on with the main topic.
Click to expand...

That's not an approximation. The legal definition of an inch is 25.4 mm.

Meanwhile, the measurement that set all this off, 1/10 the diameter of a human hair, is neither metric nor imperial, but one of those measurements based on familiar things, like "as long a three football fields" or "as tall as a ten story building."
 
Ziggurat said:
Side note: that may look at first blush like it’s got 3 digit of accuracy, but it doesn’t. That’s just 0.001 inches, converted to metric. So it’s got barely 1 digit of accuracy.
Click to expand...

Yeah, I see what you mean. And it really doesn't matter anyway because human hairs do not have a uniform width.

But it's very common, when trying to describe very small things, for someone to compare that thing to "the width of a human hair", so it might be a good idea to come to some sort of consensus about what exactly we mean by "the width of a human hair".
 
Puppycow said:
Yeah, I see what you mean. And it really doesn't matter anyway because human hairs do not have a uniform width.

But it's very common, when trying to describe very small things, for someone to compare that thing to "the width of a human hair", so it might be a good idea to come to some sort of consensus about what exactly we mean by "the width of a human hair".
Click to expand...

There is a simple two word expression that can easily substitute for things like "the width of a human hair".... and it uses neither SI units, nor CGS units nor Imperial...


.....******* small!!
 
That sort of precision is amazing. I can't even get the water-filled ice cube tray to the freezer without spilling it.

Blue Mountain said:
I'm not very impressed with the site. It put two popups in my face while I was trying to read it. I have very little tolerance for that sort of rudeness on the web.
Click to expand...

Almost every site has at least one nowadays -- they have to get you to accept or decline their cookie policy. I think it's a new law.
 
Last edited:
smartcooky said:
There is a simple two word expression that can easily substitute for things like "the width of a human hair".... and it uses neither SI units, nor CGS units nor Imperial...


.....******* small!!
Click to expand...
In English engineering: a gnat's cock.
 
You must log in or register to reply here.

Back
Top Bottom