[emd68]

Odin,
I should add that I think sapphire is an excellent choice for what you are trying to do but just make sure that all surfaces get a good polish. The profile that you are trying to make on the two inside surfaces will be very difficult to cut and polish but not impossible. Can I ask what your experience is with sapphire?

Take care
Ed

[daemonfly]

Interesting read. I'm just a general hobbyist, but what popped into my had is if you were trying to keep points of stress concentrations at absolute minimum, couldn't you completely separate the outer sapphire sphere from the device itself? Build the device within it's own spherical framework and have that "floating" entirely within a single spherical hollow of the sapphire case.

[Odin]

ok, some really good feedback lately. Ringleboy, thanks for that math. I'm still taking in if the variables you used apply to what I'm doing, I haven't had any time to sit down and review it all yet- the timezone difference is making my life hell.

Emd68, good to meet someone who works with the stuff daily! Actually, I've been to your company's site several times. You seem to make good stuff!
My background with sapphire isn't professional- I cut & polish industrial sapphire by hand as a lapidary when I get time, and I spent most of college pouring over texts and reports on synthetic sapphire production & furnace design, as a personal interest. I've been wanting to put together my own crystal furnace sometime for Verneuil ruby, as they are rather simple, but ideally I'd like to work on making a Czochralski furnace when I get a little more capital. Yes, I know, I'm crazy. This is why I'm not an engineer- I'm convinced anything can be done if you know how it's made. So I'm familiar with very exacting polish on industrial sapphire, but on a small scale. I would kill to get a tour of a place like your business, but most companies don't take non-super rich inventors seriously.

Also- everyone, I know it would be simpler to build this device spherically and deposit into a large sapphire sphere. Problem solved, but-

I can't use that form! The device is about the proportions of a large p-o-c-k-e-t-w-a-t-c-h, and from studying that industry I'm quite familiar with normal and extreme design uses of synthetic sapphire, just not much on what it's capable of withstanding at depth. There's a big hint to what it is, just don't mention that word at all, ok? Don't mention anything of the sort in these posts, I'm paranoid, yes, but there really are people who would take this idea before I get a chance to. Whether you think I'm crazy or not, fine, just treat things with a grain of salt for me, ok?

So the reason this can't be in a sapphire sphere, etc, is that it's a set shape. I have solutions for the "crown" not coming through the crystal, I have my own design. I need to make sure it still somehow resembles that shape when I'm done, though- basically, a large p-o-c-k-e-t-w-a-t-c-h sized sapphire "burger".

I'm attaching another poor sketch to give you an idea for the crystal shape- keep in mind I can't, due to form, set this in a sapphire sphere. In the picture, I'm shooting for a #2 type of crystal on each half, but could do #1 too, just thicker. The black/shaded areas are a side view of the hollows. Yes, I can't draw well, I know.

Especially Emd68, I would love any input you care to provide about whether or not this could work, and how. You know the material better than anyone on here, I'm sure, with credentials like that. I've even considered ordering the blanks from you guys, and then finish them with the channels and hollows myself, as anything else would be far outside my current budget.

Then again, my budget may be much higher- this isn't a project I assume I will finish right away- this is going to take a few years of planning & design, if I really take this seriously, and I do want to. The insides I can handle, the case is what I needed advice with!

[emd68]

Odin I don't work for Crystal Systems. They are a sapphire grower, We buy material from them to fab optics for others and sometimes we fab optics for them. The picture of the sapphire dome in my earlier post was one we made for them back in the late 90's and then they sold it to Raytheon I believe. Lets see if I have this right you would like to make #2 so you would have about a 1.5"- 2" diameter sapphire optic with say a 10" convex radius on the outside and you would cut a .295" radius in the center with two half tubes running out to two other smaller radiuses on the inside. The outside radius and the center radius on the inside are easy, the half tubes and the other two radiuses is were things get difficult. Also how do you plan on holding the two halfs together?

Ed

[emd68]

Odin also do you have a max thickness? I'm thinking you need to be about 1/2" or better for each half.

Ed

[Odin]

Ah, Ed, sorry for the mistake there. I thought you worked with them. Still, I admire your job- I've thought myself at some points to grind my own lenses from some sapphire I cut for better loupe lenses- my current ones get scratched up pretty easily!

Again, forgive me for not being more specific- that's an inventor's curse- you see in your own mind exactly what you want, but sometimes it's hard to/you forget to specify everything exactly for others.

I work in metric, now, but if I did inches, each crystal, again, due to the shape/size I'm trying to do to conform to said previous object, would be around 7 cm (2.75") in diameter. As far as thickness, I'm not sure yet, that's why I started this thread.

[Odin]

Ok- new size definitions.

Based on realizing what I need to fit inside, I have to change somethings.

Total diameter for each crystal half I'll stick with the above 7cm (about 2.75").

Internal central chamber I want to do 2 CENTIMETERS, instead of the original 1.5. And subsidiary sphere chambers all 1.5 centimeters in diameter, or if possible, 2 centimeters max.

Total allowable thickness for both crystal halves together- 4 cm.

[Odin]

Actually guys, the more I look at my own idea, the more this looks impossible by my own size restrictions indeed.

I'll have to go back to the drawing board with this one, maybe I could try a 7cm diameter crystal "squished" totally hollow sphere. That would make everything very simple, case wise.

Let's try that- maybe 2 circular hollow crystal halves, still half-domed shape of that #2 shape I posted earlier.

If I went with that, I wonder if using some math, I could create a chart of possible total thicknesses for both crystals together versus total allowable internal hollow volume at what could be withstood at that depth, assuming a perfect perimeter seal. I could go as thick as 2 centimeters for each half, creating a 4 centimeter thick crystal "case", with a yet unknown internal volume.

If only there were a computer simulator program I could run for a shape in a certain material like sapphire, to see how different shapes would handle a 108 MPa load.

Honestly, that would be the easiest thing to do- if I had a program that I could "design" my case in, and test it for failure based on material parameters, I could find the best design balance for my need, without trying to specify "I need this size/shape" to work.

Anything like that out there?

[ckelloug]

Odin,

There is software out there to do simulations but it is generally outrageously expensive. It usually also requires substantial domain knowledge to interpret the results. One such package is NASTRAN and the various offshoots from it.

Can you build your invention with an outer spherical containment that holds against the pressure and then some kind of transparent sapphire chamber internally that holds your parts where you need them but isn't directly physically coupled to the shell?

[emd68]

Wow Odin,
Your mind sounds like it is going at 100 mph. Well I can tell you that 2 sapphire blanks at 7cm diameter and 2cm thick are not going to be cheap. My guess $400-$600 each for the raw material. The single inner radius would be the simplest to fabricate if you can go that way.

Take care

[Odin]

Emd, yeah, I kinda figured around that range. A few thousand dollars is allowable assuming I get to be the first to claim my feat a success if it works!

And the 100mph comment- that's how my mind always works- when I get an idea, it stays in overdrive mode until I can come up with something that satisfies my mind. For me, honestly, I've already come up with maybe 20 different variations of all this, and hundreds of permeutations of the total design- my problem is concretely settling on the best of them, and/or explaining the variation choices to others. I'm the kind that can think up multiple solutions to a problem, and then my problem is picking/being able to make one! Couple this with being a terrible artist, and someone trying to learn parametric CAD to help that, and I'm constantly frustrated by my own lack of ability to visualize my own designs on paper/screen because I suck at drawing. Basically, I create if I have tools. I always make models right away, because I can't draw them.

Ckelloug, looked into NASTRAN- it's programmed in FORTRAN?? Jesus, that's some old code! I'm integrated into the Linux software community- if it's that old, there may be something out there in freeware that can compare, maybe... you'd be surprised what's actually available as freeware for Linux.

That said, as long as the final outer shape resembles a large p-o-c-k-e-t-w-a-t-c-h (again, forgive my paranoid obfuscation), I can have the outer "shell" in that size/shape, and the internal hollow, I can fill with a metal "movement", or even a custom carved sapphire "movement" plate to hold all the gears/winding barrels/mechanical escapement.

The most important thing is the final shape & size, and that being able to withstand the pressure at depth- otherwise, I'd just make a giant sphere and put the stuff inside.

The challenge here isn't sending this thing to the bottom of the sea to do what it does- anyone could do that now for a few thousand. That's not what I'm after- I want to be the first to send one of these to the bottom, on a design that is close to the shape and feel of what traditionally exists, THAT'S the challenge. Think the ultimate expression of a diving w-a-t-c-h, except I've given myself leeway to not do it in wrist-form factor, going with the tradtional size. If you don't understand where I'm coming from now, you don't get it.

[PEU]

You can use saphire or whatever translucent material, but unless you take care of micro organisms that live down there your glass (the entire body actually) will be covered with goo over a year or so, check these photos:

This dive flashlight was submerged for a year and a half at 38 meters in the sea:





Microorganisms did a better job than saltwater at damaging a milspec hard annodized type 3 class 1 surface.


Pablo

[Odin]

O.o.....

I will have to coat the metal case perimeter I plan on using in ceramic, or something, yikes! I know that glass does degrade in saltwater, so I was aware my design would have to protect against this corrosion.

Obviously, bare polished metal won't cut it. I have to put a non-porous coating of something on the surface- or it will be covered in this stuff. Especially when I remember I want this thing to survive a nuclear apocalypse at the bottom of the sea for a few thousand years or so. Sapphire, yes- as inert and non-porus as I can imagine. Metals, however, are another issue.

I have to find a superior coating, as non-porous as possible, otherwise barnicles/stuff like you show here will be able to attach. Surely by now there's something out there....

How about PURE GOLD (thick plating)? That always seems to survive shipwrecks quite intact. If it worked for the Spaniards, it could work here...

[emd68]

Odin I did a little drawing so I can visualize this better. You said and outside diameter of 7cm and a max thickness of 2cm so based on that the largest radius you could put on the inside would be about 14mm and still maintain a wall thickness of 6mm at the thinnest part of the lens. I don't think you want any part of this to be less then 6mm in order to hold up to the pressure. But it sounds like you need a lot more room then that, one object at 2cm and two others at 1.5cm each. Hopefully you can see what I mean from the attached drwaing.

Ed

[Odin]

Emd, thanks for the sketch- you must have drafting tools. I have a compass, drafting ruler, and a few other things around here somewhere- I'll make a couple DETAILED sketches for everyone today (early morning US time) of what I proposed before, and what I propose now- we'll see what works best.

I think you misinterpreted my previous comment, before I decided to change the design- those 2 other 1.5 cm items would each have been separate hollows, next to the main 2 cm hollow. It wouldn't have been all in one big hollow- it would have been spread out. Still, it would have been pushing it.

So as the resident sapphire expert, your sugguestion is around 6mm? That's all I'd need to withstand the challenger deep? I would have thought it a bit more.

Be patient, and I'll draw up some scale stuff today. It will help me as well. EDIT- Actually, since I'm in the middle of it now, I'm thinking I can go thicker than 4cm total (2cm each half)- with a better idea of the outer shell's shape/case seal.

[emd68]

Odin I don't know if 6mm is enough, I was mostly trying to show that if you are going to go with one central sphere that based on the max thickness of 2cm per side you would have very limited space on the inside to work with. Some detailed drawings would be great. I was thinking today about your original design and remembered that from time to time we get parts ultrasonically drilled. Check out the web site Sonic-Mill. We have not done work with this company but it will show you what I am talking about. If they could drill it for you then maybe you could find a way to polish it. http://www.sonicmill.com/ultrasonic.html

Take care

[Odin]

Everyone- sorry, it's taking me a bit longer than I planned. I've decided to find a decent Freeware CAD program to do my drawing- I have run out of specialty lead for my drafting pencils, and nowhere here in the sticks carries it. Give me a few days, to find a freeware CAD program, and I'll draw up some simple stuff. I need to do that anyway- we'll start up again in a few days after I've finished. Sorry! You'll know when we can start up again as the thread will be bumped up as new. I've been very busy lately, and getting dates with beautiful Japanese women... ha. No time to draw by hand anyway, if you know what I mean

EMD- WHOA. I mean, holy. Jesus. SONIC MILLING???!?! I thought I'd seen every machining technique known to man, and now I see THIS. You just.blew.my.MIND!

With that, this entire project is SO much easier, in many ways. I wonder how expensive it is, and how much those machines cost? I was thinking someday of buying a wire EDM machine, as they're only a few thousand on ebay nowadays. How much are these sonic things?

Actually, this opens an entire other WORLD of possibilities now- I've thought up a lot of projects in the past that used custom milled blocks of Sapphire for a large range of things, in applications only sapphire could work, and I abandoned them because I never knew of any way to machine it on the scale I wanted. This Sonic Milling would let me realize my wildest dreams- I could shape anything, ANYTHING.

You've made me very, VERY happy!!! My dream projects often used materials that could only be shaped by that tech!

[emd68]

Well Odin I'm glad I made your day, I do not know the cost of the Sonic Milling machine, you will have to give them a call. If what you are going to be doing is low volume then it would be easier to just let them do the work. Looking forward to seeing some drawings.

[Odin]

A couple things have come up.

One, I've been looking for days for something I could use to draw with, and gone through 3 free CAD programs- 2 didn't work with my build of linux properly, and one didn't allow for construction of spherical hollows. So I'm still looking for something free that works in linux, even with wine.

Two, and mainly, I don't even know that there's a point to posting my original design- as I've realized it has a fatal flaw. The original design would put the mechanical escapement in the center, while the gears regulating this have to be directly in the middle of the spring barrel and the escapement. My design would screw up that order based on position- there might be a way around this, if my design could incorporate another gear ring in the center sphere to bypass the escapement once, go to the other subsidary sphere this way, and finally go to the central mechanical escapement at the end via a final transmission link, but that's ridiculously complex. Either I make the biggest sphere off-center (ie: switching with one of the perimeter spheres), or I make a long channeled axle pipe to route around the central sphere first. In any case, my original design might never work at all anyway due to this, unless I can incorporate the ring transmission, and I'd rather not (however, it would definitely look cool..)

So, it looks like this thread is dead, at least for some time. I will post again, at some point, when I have a working CAD solution (I may have to save up 500$ or so and buy one, like variCAD....). Should I just start a new thread then, or post again to this one? It will be some time, it seems, before I get things in shape, unless I have a breakthrough with that transmission design.

[emd68]

Odin best of luck with your search for a CAD Program and your project. I will keep an eye out for this thread or another to start sometime in the furture. If I come across any other info that may be of use to you I will pass it on in a private message.

Take care
Ed....................