The OT thread V1

[cockerpunk]

I consider these kinds of posts a crime worse than Vaguebooking status updates on Facebook.

sorry, wish i could talk more thoroughly about it. i really do, but i can't.

suffice to say ive worked every single day in the month of decemeber up until today, 8 or more hours (sometimes more than 12), to get this cut kicked off. and in 5 more days we'll find out how well it worked or not. we are basically doing something that no one in the world can do, and really, frankly, we can't even do it ourselves, but for a hero effort on everyone's part. the metrology alone we are having to come up with on the fly, because once you are below a certain size nearly every measuring tool is untrustworthy in some way, so we are basically flying blind, with non-deterministic adjustments as well. so we dont know how much we should move, and we don't know how much we have moved, when we move, all to set up something to literally insane tolerances.

all while our nearly biggest customer and the lifeblood of an entire 3M division is riding on the outcome. that they want yesterday of course.

that im in charge of.

and i've been sick the whole time.

stress begets sickness, sickness while working so hard begets stress, begets more sick.

so, vaugebooking is the best i can do for therapy now guys. to be honest, i'd love every second of it, if i wasnt sick. i do love my job and what we are doing, that im leading, is simply amazing.

[Simon]

So what's the point if it's to tolerances that can't even be reliably measured?

Once you get to those levels how you handle the part or the enviroment you are in will change parts. In many ways requiring high tolerances is bad design once you get past a certain level that isn't sustainable due to enviromental factors. Some of the best designs are the best, because they work so well without requiring high tolerance parts.

I guess it depends on the application but at some point doesn't it become irrelevant?

[Fred]

Makes me really glad that I only have to hold contractors in the field in most cases to 1/8th of an inch...

[cockerpunk]

So what's the point if it's to tolerances that can't even be reliably measured?

Once you get to those levels how you handle the part or the enviroment you are in will change parts. In many ways requiring high tolerances is bad design once you get past a certain level that isn't sustainable due to enviromental factors. Some of the best designs are the best, because they work so well without requiring high tolerance parts.

I guess it depends on the application but at some point doesn't it become irrelevant?

you would think, and its a leading reason why lots of initial product ideas with size scale structures like these didn't materialize. but there are applications where the environment is controlled and such things do provide utility. big dollar ones its turns out. big dollar ones my company already plays in.

the fundamental physics are not well understood in this application, so the tolerance is "it works" or "it doesn't work" but such tolerances are not really engineering speak so to say. we need to deal with arc seconds and nanometers, and part of the problem is we don't know, what an arcsecond of angle really does, or doesnt do. what this then turns into is do everything you can possibly think of, in as good of a way as you can possibly think of to do it. and cross your fingers it was good enough. because a lot of the time, it isn't.

[Pump Scout]

The perfect tool for this: the go/no-go gauge for the M-2HB 50 caliber machine gun.



It's either right and works, or it's not and doesn't.

Really, though, I glazed over a bit. My line of work deals in measurements of rods and chains, for God's sake. Flippin' archaic.

[pbjosh]

Psshh, none of you had to deal with the shit, wrong, or missing files left over when KEE bought JT. I think I would rather have no files or drawings than wrong files. Any time I have trusted other people's files I have been bitten.

Psshh!! 1955! No fun until you have to use hand drawn files from the mid-late 1800s. Almost always with completely suspect 'as-built' drawings from past projects. Talk about fun making modern CAD files from that information

The problem is verifying. I normally to walk down and verify an operating plant to make sure. Including full Hazmat, HF Acid (very un cool, and it just drips around the plant) and one area I am climbing under a 60yo reactor that is running a 1400 degree 48" piping that is glowing red. 180 degree rooms full of Urea, leaking platforms (as in, middle of winter we were in the Bruckers ready to abandon ship), verifying parts. Some of it hella cool. 98,000hp pulp grinder? 40,000hp turbine? 250ft up on a Prill Tower? Realizing that the vibrating spherical tank full of liquid propane is about to throw you off, while you verify a stupid gauge on the top? Not so cool. Plus you had to walk up to the top. Ack. Face full of concentrated ammonia? I thought my eyes were bleeding. 5kva/13.5kva lines and switchgear? Don't touch the doors without arcflash suits.

Lucky for me I am in the office just doing CAD 99% of the time. The guys who work that every day... My job is to spec it safe for them.

sorry, wish i could talk more thoroughly about it. i really do, but i can't.

I understand. You have to vent somehow. I am not doing too much that cool, but I have had two projects that I just can't talk about this month, and a 'Need to Know' one I worked on today.

[pbjosh]

Oh, finally say Star Wars. Liked it a lot, the Teens LOVED it. I do have my complaints, but it is spot on consistent with the Star Wars universe, down to the regular flaws.

[ElPanda]

[tyronejk]

That's awesome. I can imagine a guy standing in front of that machine punching in different speeds to pick out notes for hours.

Also, on a machine like that, is the spacing between spindles fixed and assumed to be accurate? or is there typically a way to adjust them?

[ElPanda]

looks fixed to me, and I can almost certify that the distance between them is likely accurate down to millionths