Thursday, November 29, 2007

another power outage

about 200 hours down the drain

I haven't been thinking of any other things to do while we wait, but now I guess my hand is forced.


MTF

Wednesday, November 21, 2007

green light on the centrifuge


I ran a short simulation to see if everything stayed together. For the most part it did. There were about a dozen molecules that few apart; I assume from being in a too energetic position. I played around with moving them around, but it was just a lot easier to delete them, so we are playing with 12-14 less specimens. The above arrangement is what I am going with for the long sort sim. It is going to take a while, but I'll try to come up with some stuff to entertain you while we are waiting.

MTF

Tuesday, November 20, 2007

the nitrogen and the acetylene


Think this is asking too much? I better run a quick test.

after one revolution of the carbon nanotube centrifuge



Red is Helium and green is Krypton. Next up N2 and Acetylene with a few more rotations.

MTF

Saturday, November 17, 2007

I think we are ready for some RPMs


I was having trouble building a stable funnel with such a small diameter but moving to a wider tube solves that problem. It also apparently makes the check valve unnecessary in the first place.

There must be some electrostatic interaction between the atoms in the tube and tube walls that increasing this distance lowers significantly-as it would. In fact if you look real close in the simulation, the He atoms never touch the wall and are repelled by it.

I think for now I will leave the funnel on for looks but remove the C60, fill the tube with some different things and spin it.


MTF

Friday, November 16, 2007

adding a back flow preventer -or- atoms enter but they can't leave


First, I want to make sure everyone knows my last post meant to be humorous, and I am sure our core group does, but for any interlopers-there is no crying in molecular dynamics.

Now, back to the task at hand. This makes the pressure of the escaping gas work, work to seal its doom! It could push the C60 into the funnel, sealing the tube, or blow up into dozens of pieces. Either way it should be fun to watch.


MTF

sort of a disapointing start to the weekend


And I went and left the power supply to Big Red at work
:(
.
.
. <--blog tears

Thursday, November 15, 2007

first try cheating then use a more realistic tube


Capping the tube works, but complicates build anything approaching a practical sorter. Also with using helium it still leaks a little, but through the walls.

Next I figured using a tube 32 Angstroms long would make for a pretty short tube. I tripled the length . The contents stay in. It is my hope they stay in long enough for things to get up to speed.

Other things I tried before these two:
  • deleting half the atoms
  • plugging it with a fullerene- it made a cannon
  • decreasing the rotation
I hope to spend some time on the diamondiod systems this weekend on a high-end gaming rig.

Longish sim on the longish tube overnight.

MTF

Tuesday, November 13, 2007

diary of a man whose passions pull him in many directions

First I had to over come this:

Thanks D.

Now this:


I had been putting off working on the diamondiod parts because the performance wall on my system is just a little too high for it to be enjoyable. However I may be able to score some time on a really tricked out system, so I'll give it another shot.

Friday, November 9, 2007

Wednesday, November 7, 2007

Big idea for a Big project

It seems a lot of people are interested in sorters of one type or another and with good reason. I was thinking it might be interesting/entertaining to build one. The type I have in mind right now is based on the centripetal force, basically a nanoscale centrifuge. I am sure anyone that has taken a general chemistry class is familiar with type you drop test tubes in and let rip. I think I could recreate this with a diamond hub with capped carbon nanotubes radiating outward like test tubes. I could even try to use all my diamond horological parts to start and stop the rotation, maybe stop the rotation in a precise way so the tubes are in a known position.

It would be neat to pack these tubes with various things[

chunks of DNA
molecules trapped in C60s
amino acids
whatever

]run a big old simulation on it, then on the last frame of the simulation cut the tube open to see the position of the contents.

Anyone have any thoughts on this?

follow up on the cam simulation

Damian is sitting next to me right now and has viewed the cam simulation from last night. He believes the flying off of the cnts and hydrogens are an expected consequence of electrostatic repulsion, so it is most likely not a bug at all.

Tuesday, November 6, 2007

more on the double escape movement


This one could get a little rough. I added first tries of the pinion and spiral gears. I was a little surprised the spiral gears cam out spiral. That's a trade secret I just developed tonight* I still need to reconnect the forks, but I was expecting the first rack design to be way too narrow for a robust pinion anyway. However I really need to spend some time on school stuff for the next couple of days, so give me a few.

MTF






*Just kidding. If anyone wants a set I'll send them to you.

cam simulation inconclusive


I raced home today because I knew the simulation would be finished. I have to admit one of the post titles I thought of while stuck in traffic was "pay no attention to the cam exploding on your left". That title was inspired by the numerous excessive energy warnings reported during the simulation, apparently from the shattering force of the cnts flying through the slab.

I am not sure if they should do that unless there are some parameters in the dynamics' code that basically says "if you are sitting on top of a monster diamondiod cam, run like hell!". I'll need to ask around about that one too. It could be a bug, but don't worry. That's part of what this is about.

I wanted to make sure I had some frames of the cam whizzing around, but it isn't exactly blurring.

Should I call this a success? The cam holds together and appears pretty rigid. I could design a more complicated test, like with forces pushing back on the cam. Any thoughts?

I haven't spent too much time building the double escape movement more, but I always get a rush of inspiration when stuff works, even like above.

A note for The Guy-DA says Boron Nitride performs a lot different at these scales, and we probably shouldn't bother with it to make something more rigid.

MTF

Friday, November 2, 2007

restarting the cam simulation

Turns out that it is technically possible to salvage a truncated simulation, but it is a little beyond me, like it would probably take me three days to figure out how to do it. The good news is some where down the line these feature will be included in NE1. I will lobby for it every few months.

True Story: I was driving home when I found myself behind a large flatbed truck with a bumper sticker you see a lot around Michigan: "Construction Workers: Give'm a brake". In reference to my last post, homonyms have never been my strong suit when typing fast, and I hope my statement of needing a "break" instead of brake does not turn out to be some kind of self fulfilling prophecy. I think I should add apologizing for my english skills every few months to my list of things to do.

MTF

Thursday, November 1, 2007

Rats

So the simulation of the cam system was somewhere between 94 and 100% done when there was an apparent power outage. NE1 did manage to write a 283Mb simulation file, but at this time I don't know of a way to salvage it; I'll ask around. It is most likely three days down the drain.

Now if I was of the paranoid persuasion, I might think that someone broke in after I left in the morning and flipped the circuit breakers. I am not of the paranoid persuasion, but I went ahead and established a perimeter with multiple booby-traps on all the strategic targets, just in case.

You never seem to notice the small reliability problems of the power grid until you depend on it for glory.

MTF