584,829 active members*
5,150 visitors online*
Register for free
Login
IndustryArena Forum > Mechanical Engineering > Mechanical Calculations/Engineering Design > Help My Part Survive Implosion! A Great Challenge- Calling Hydrostatic Engineers...
Page 1 of 2 12
Results 1 to 20 of 40
  1. #1
    Join Date
    Apr 2007
    Posts
    83

    Lightbulb Help My Part Survive Implosion! A Great Challenge- Calling Hydrostatic Engineers...

    Ok, I have one crazy problem. Let me first explain I'm not a professional engineer. But I have one crazy idea, and I need advice from Hydrostatic or Mechanical engineers for my object to survive in.....are you ready for it....

    The CHALLENGER DEEP. Yes, at the bottom of the Marianas Trench!

    If you think I'm a total nutjob/insane at this point, you're right. But it's (maybe) not as difficult as you think! Granted, I feel like a fool for asking here, but I know no other place for so many professional engineers who know their stuff.

    I'm not an engineer by trade. I'm, let's say, an inventor. I have knowledge on a broad range of subjects, and normally I make very normal stuff. My object is an encapsulated purely mechanical device, spring powered. Not sure how to describe it, but basically, I have moving mechanical parts and gears only, that I want to design to operate at the bottom of the Challenger Deep. Normally this would be a monumental problem to say the least.

    Why I'm not totally mad- the device in question is almost entirely a solid object, with very little internal volume. We're talking less volume than a shot of expresso. Not a submarine, a very small device- therefore it should be possible, as there is less area to overcome imploding due to huge hydrostatic pressure of about 108.6 MPa, at around 36,000 ft. deep in the ocean.

    Here's where it gets technical, and hopefully fun- most of the device is situated between 2 thick and perfectly mated flat industrial sapphire domed crystals. The internal open volume for the device takes the forms of spherical hollows, in the center of the crystals. The main hollow (75% of the total internal volume) is a spherical cavity of about 1.5 CENTIMETERS in diameter.

    Here's where I feel embarrased & lost- I'm not sure how to calculate for implosion due to hydrostatic pressure, or what materials variables come into play to do so. I figure tensile strength of the material is the main point, but maybe that's an amateur mistake? I chose synthetic industrial sapphire for 2 reasons- 1, the device must be transparent. 2, Sapphire is the only material I know of that is clear, chemically inert, and has an extremely high tensile strength. The central cavities to hold the instuments are all spherical for a reason- that is the shape that should resist implosion the best/distribute pressure the most evenly, according to common sense, and studies using glass spheres imploding at depth in the sea to test for hydro-acoustic properties within implosion depth ranges.

    I know of the near incompressibility of fluids under high hydrostatic pressure, and normally, that would have solved it- I could just fill the entire thing with gasoline, or something, and be done with it. After all, gasoline was used in the hull of the Trieste for incompressible bouyancy. The problem is, the main hollow of the device must either be gas-filled or ideally be under vacuum- part of the device in that portion uses mechanical movement based on a natural spring that must keep regular motion- that would never work properly in a liquid. This is why I am thinking it should be a vacuum- oxygen at that depth would liquefy under pressure? Something tells me I don't get that part well. Forgive me, I normally design very normal, simple things. This is something I thought up "off the deep end", far outside my normal range.

    The idea is, the sapphire has a high enough tensile strength that, given a specific total thickness of the 2 mated crystals, with the sphere situated as half in each, in the center, it should survive imploding due to water pressure, and allow the spring motion to work unimpeded in a very small volume of oxygen or vacuum.

    Can anyone shed some light on how I calculate for resisting implosion/material thickness needed here? I have basic engineering calculus under my belt from college and did well (A-), but I'm not a scientist/an experienced engineer. I have experience fabbing with synthetic sapphire and metals at small scale, but I can't find anything remotely helpful to aide me in my design! Any submarine engineers here? Somehow I think they might know what I need to do.

    And yes, go ahead and call me crazy, I know how nuts this is! I am only attempting this build because the small internal volumes needed look manageable to me. And I only posted this insane question because this is a forum for mechanical design- so it's a relevant question, and a great challenge at that! Maybe one of you better than I will see the fun in addressing such a design and offer some advice. Finally, forgive me if I reply to anyone late- I live in Japan currently (from the US), and am going to bed at 3am!

    Thanks.

  2. #2
    Join Date
    Jul 2005
    Posts
    12177
    So what are you inventing? An instrument for measuring the Earth's gravitational field at the bottom of the Challenger Deep?

    You mention 2 thick sapphire hemispheres surrounding a spherical cavity 1.5cm in diameter.

    How thick is 'thick' and how precisely are these hemispheres going to be made?

    I have worked on deep sea (not Challenger Deep deep) pressure housings from 6" to about 30" diameter and the most important thing was to make sure they were true cylinders or spheres, or as true as they could be; either that or simply make the walls really thick.

    My gut feeling is that if your hemispheres are precise to within +/-0.05mm for sphericity and have a wall thickness about the same as the radius of the inner cavity you will have no problems.
    An open mind is a virtue...so long as all the common sense has not leaked out.

  3. #3
    Join Date
    Jan 2009
    Posts
    29
    Hi
    First off, I'm pretty sure you could gas fill without any fear of liquefaction,
    Hydrogen, helium,nitrogen, oxygen need low temp, and are shipped/used in insulated flasks, boiling off constantly. No attempt is made to store them in pressure vessels because they boil regardless.

    Second, thre's a dodgy rule of thumb for tubes resisting INTERNAL pressure.
    This is approximate, AND SHOULD NOT BE EMPLOYED FOR DESIGN CALCULATION!!!

    For steel tube, a t/D ratio of 1/1000 will withstand 100 PSI.
    (I've seen a Swagelok burst test that went near 80% better, but the fitting survived!)

    Obviously, a very thin-wall tube is better internally pressurised (coke can),
    but you'd be looking at way thicker than this, (t/D= 0.2+) and the compression strength should be even better once clear of buckling issues.

    I know the pressure looks high ( partly because the Pascal is such a piss-ant unit!) but there's plenty of stuff out there for the problem to be solvable.
    John.

  4. #4
    Join Date
    Jun 2004
    Posts
    6618
    For testing, you could run over it with a fully loaded steam roller. If it survived that, chances are good that it might survive the deep.
    Try not to make it look edible though. Plenty of things down there willing to scarf up something that size.
    Lee

  5. #5
    Join Date
    Apr 2007
    Posts
    83

    Smile

    :cheers:I knew someone would ask what it is- unfortunately, I can't be any more specific to it's use, only parts. The reason being, if anyone got wind of my design, well, let's just say there's definite commercial application. There are people who would kill for my design if it works!

    I know it's cliche to treat something in a thread like this this way, but given the nature of what came before me, I'm being cautious. All I'll say is this- it is indeed a scientific instrument of measurement of sorts. That's all I'll say!

    I'm not sure yet how true I can get the spherical cavities. I should be able to make a custom tool to the specifications, and then have that vapor coated with diamond, and use that kind of like a drill bit to slowly bore out the hemispheres on each side. This is a rather complex machining exercise for me, something I've never done before like this. But it should be doable. I haven't decided how thick to make the sapphire crystals on each side yet, because I felt that depends on what I need for pressure resistance. The crystals would be slightly domed, per my design. The central spherical cavity is under the center of this- that's where the crystal would be the thinnest. I can make it any thickness there, though, that would be needed.

    So Geof, do you mean each side of the crystal above each half of the hemisphere would have to be roughly the same thickness as the hemisphere on each side? Ouch... the design is trying to keep things as slim as possible, but I knew certain thickness was unavoidable.The final shape just needs to be smooth- not exceptionally bulbous. I can shape the rest accordingly, but am still unsure of what thickness to use.

    JLSD, thanks. Outside the main sphere, I'm thinking of using mini titanium pipes to carry axles leading to/from the central cavity, and these would be very small (like 3 millimeters diameter MAX), set into carved mated channels in each half of crystal. This is where things get hard- I'm hoping the very small (1 or .75 mm) inner axles will be unaffected by pressure evenly distributed on the outer titanium sheaths. I could use vandium, or something else, I just picked titanium for strength to weight ratio.

    To both JLSD & Geof- those axles leading away from the center are what worry me most- even though they would be mated to the crystal halves perfectly (half channel on each side for each pipe), that design would somehow compromise the structural integrity of the whole unit.

    As far as I can, I could build most of the mechanism into smaller spheres around the perimeter of the central one, and connect it to the central core with these axle tubes I mentioned, maybe 3 or 5.

    To Leeway- are you serious? If so, how many megapascals could an average steamroller concentrate on something? I never even thought of that. It sounds crazy simple, but it might work! I'd just have to encase the instrument in something that would evenly distribute the force over the entire face.

    If anyone has any more thoughts, please share. I'll see if I can sketch up a rough picture sometime tonight of what the form will look like. I can show the case, but not what goes inside. Obviously, this discussion is about the case.

    Thanks everyone so far!

  6. #6
    Join Date
    Jul 2005
    Posts
    12177
    [QUOTE=Odin;593034... There are people who would kill for my design if it works!...[/QUOTE]

    You are safe, no-one is going to kill for it because it will not work. You did not mention putting shafts through the hemispheres; that makes it impossible.
    An open mind is a virtue...so long as all the common sense has not leaked out.

  7. #7
    Join Date
    Apr 2007
    Posts
    83
    Really? I don't mean through the hemispheres- I mean between them. Ie: you have 2 felt disks, and a dowel rod. You put the dowel rod between them, and sandwich it half in each. Think a "Y" pressure fitting type shape.

    The axles DO NOT EXTEND outside the crystals- they are within them. Nothing in the design goes outside the crystals- it is meant to be permanently sealed. The axles sit within the crystals. Thus, the full outer perimeter of the crystals is just that- solid crystal.

    I'll draw a picture.
    Attached Thumbnails Attached Thumbnails Part Design (Simple).jpg  

  8. #8
    Join Date
    Jul 2005
    Posts
    12177
    They introduce stress concentrators which are a strict No No when you are dealing with extreme pressures. The most diffiult thing to accommodate in a deep diving submersible is the unavoidable variations in the mechanical propertise of a sphere when you start maikg holes in it. This is why when you see these things there are large reinforcing flanges around portholes and things like that.

    Quite apart from the fact that making sapphire hemispheres is a challenge in itself, making shaft holes is a greater challenge.
    An open mind is a virtue...so long as all the common sense has not leaked out.

  9. #9
    Join Date
    Apr 2007
    Posts
    83
    Geof, I totally understand what you mean, and I'm still worried about that. Take a look at the picture I added to get an idea. If we assume the outer cores of the gear shafts are solid, and perfectly mated between the crystals with no gap, and minimal compression distortion on the crystals due to thickness, I wonder if it can work. Inside those shaft cores would be where smaller shafts turn- they wouldn't necessarily be the same diameter of the hole in the center, maybe a bit smaller to allow them to turn dispite a minimal distortion of their outer shafts under some minor deformation, maybe.

    What do you think? The final design would have the subsidiary spheres (3 or 5) spread out around the perimeter of the central hollow evenly, with each at the same distance from the center. In effect, a shape with hard spokes, and hollow spheres, like one of those ball-clocks from the 1950's (with less balls) sandwiched between two perfectly mating sapphire crystals.

    Are you still sure this can't be done? Maybe lack of the picture made things confusing.

  10. #10
    Join Date
    May 2004
    Posts
    600
    Things are looking up guys. We just got to figure out how to get the other 300 million Americans to use terms like mm, Kpa, etc. Naw, just kidding! Couldn't let the opportunity pass...

  11. #11
    Join Date
    Apr 2007
    Posts
    83
    ?

    I get that maybe I don't get what I'm getting myself into, but if I've learned anything in life, I've learned I have more patience than most people on EARTH when it comes to something I want to do. And anything I've had the tools to build, I've built, and then some.

    I get that people seeing this thread must think I'm a fool, crazy, ignorant, or all three. But being an inventor, I've learned if you try something no one else ever has simply because they thought it was crazy or it would never work, sometimes, you're right- and they're wrong. And I've been right before, where people called me crazy. So make fun if it helps.

    So I'm serious about this, and I'm humble enough to post my idea here, and ask for the help of people who I KNOW are a hell of a lot smarter than me, so I can LEARN. If you never ask for help when you know that you don't know something, I'd say THAT'S the only person who's a real fool/crazy.

    If someone just referenced an online source with the relevant calculations/math involved in this, I can do it. I have access at home to an entire university engineering library- and I've used it before. Right now, I'm stuck overseas for another month, but I'm designing now.


    So make fun of me- or help. After all, none of you would be machinists or engineers now unless someone took the time to tell you what you needed to know, or where to find it. I'm just asking for help like that. And yeah, I'm long winded, I need to work on that.

    Any takers?

  12. #12
    Join Date
    Mar 2008
    Posts
    683
    This needs to be optically clear? Make your mechanical assembly then vacuum cast (encase) the assembly using optically clear epoxy and polish it to size. How you power the device or transfer data is another problem. I imagine you'll also need wires or batteries to power the device.

    One has to wonder though why does the case need to be clear.

    Sounds like a variation of the old vibration triggered AP mines.

  13. #13
    Join Date
    Apr 2007
    Posts
    83
    Again, a normal sugguestion I would normally agree with.

    However- the item in question is completely mechanical- while being a scientific instrument, it is also completely mechanical- therefore it needs to be calibrated & servicable. Hence, being able to open it by being two halves of sapphire.

    The other reason for sapphire, being clear- the insides have to be visible from the outside- it's an instrument that must be able to be seen at depth, thus, it's clear.

    If I was able to tell everyone WHAT it was and did, everything would become apparent instantly. Unfortunately, I can't be any more specific than that!

    Can we assume it's 2 halves of sapphire? I'd use diamond, but that's right out impossible at this point. Industrial sapphire in large sizes is actually not that expensive anymore, and they are still working on CVD diamond tech. Plus, I have great experience/knowledge of industrial sapphire, which is part of why I decided to take up this project. So back to calculating pressure implosion issues?

    As a side note, are there any places that can simulate on small scale a pressurized water environment at the levels of the Marianas Trench, say, in a lab?

  14. #14
    Join Date
    Aug 2008
    Posts
    573
    must be able to be seen at depth
    Ah! It's deep sea advertising


    I have no specific knowledge but my 2p worth is:

    Are you sure your mechanism would not run in a liquefied gas? If, as a result, you can reduce the required strength of the required pressure vessel, you might be able to include a larger power source (to keep the device running/ticking for the same period) for the same money/effort as a crystal sphere.

    I think someone has already mentioned cast polymers; If you can live with a larger sphere and reduced clarity (not a problem if the output of your device is just a
    semaphore) they would be worth looking into (no pun intended).
    Bill

  15. #15
    Join Date
    Apr 2007
    Posts
    777
    I'm baffled about what you are trying to do but it sounds like your goal is at least worthy. Sounds like an underwater Sterling Engine prototype designed to harness black smoker vents in the sea floor to me

    Here's the help I can give you:

    You need to worry about the compressive strength rather than the tensile strength. These are generally very different quantities for non-metals.

    Secondly, I'm not sure I agree with Geoff that it is doomed to fail with multiple cavities but I'd have to say that it is likely to be a lot easier to do with one big cavity. He has more experience than I do. He's definitely right that any sharp transitions inside will provide stress concentration points which could cause the crystals to shatter.

    I expect you could make a very thick sapphire bubble in two halves with a flange in the middle to preload whatever type of gasket you would use for that kind of high pressure application.

    I would suggest that you look at an engineering handbook for the equations for collapse of an externally pressurized sphere. It will probably be somewhere near wherever the index says you will find hoop stress.

    Good luck.

    --Cameron

  16. #16
    Join Date
    Aug 2008
    Posts
    573
    One more thing...

    Given that the visual range (visibility) is not that great at that depth, wouldn't some kind of acoustic transponder be of more use? (although quite how one could make it work at that pressure escapes me )
    Bill

  17. #17
    Join Date
    Apr 2007
    Posts
    83
    wow, some great ideas here, thanks.
    Ah! It's deep sea advertising
    HAHAHA! That cracked me up.

    I love the guesses as to what it is for. I know visibility is total null at that depth, sunlight is basically non-existent. Again, knowing it's a form of scientific instrument, the idea is to have an unmanned submersible be able to shine a light at it and make sense of the readout. Again, I can't explain what it's for- but the final clue I'll give is that it's designed to last a VERY long time at depth. We're talking my decendants. Of course, that's just wishful thinking, but in engineering it to withstand the trench, it should be able to do that by the nature of it's design- and would definitely increase it's value to mankind.

    I've been thinking of adding a small flange around the perimeter with a custom gasket, but I've been thinking no gasket could seal properly at that depth- I've come up with a couple of other possible solutions for that, like a liquid-filled gasket, incompressible, and another design that uses the water pressure as a kind of screw to keep the crystals under locking tension. Just some ideas.

    The central mechanism running in a liquified gas? I honestly don't think that is possible- the central device is a mechanical timing device that relies partly on a spring, and this part of the device has other analog equivalents that have been used for some time in the technical field now, and can only operate properly in normal atmosphere or vacuum- the normal use for these instruments has been on land, and the design using a spring for timing would just not work in a confined liquid, I feel. It's never been attempted to put one in such an environment, I think, but the fact that it's a mechanical construction that thus uses special oils and movement frequencies tells me it could never work in a liquid to proper timing movement. Since you know it's a mechanical timer in that portion, think weighted oscillating spring escapement. That would never move properly in a liquid, would it? Maybe if I overclocked the oscillation speed...?

    Cameron, thank you- compressive strength is a major part of what I am looking for. I don't have engineering library access now, but with your idea of where to look, I can try internet textbook databases. Thanks to google books, a lot of good and perfectly practical older engineering/mathematics texts are free.

    I'll try looking today, but I've gotta get some sleep- I haven't slept at all, and it's 9am here! I'll check back this evening (American morning tommorrow).

  18. #18
    Join Date
    Mar 2008
    Posts
    51
    Hello Odin,
    It's great to see someone on this site talking about sapphire. I grind an polish sapphire at work everyday. I can't tell you what kind of pressures it can handle but I can tell you that the orientation of the material and the quality of the polish you put on the surfaces will make a difference in it's strength. Wev'e have always used c-plane material when making domes. You could try contacting Crystal Systems for the informatin you are looking for. http://www.crystalsystems.com/sapphys.html

    Good Luck
    Ed
    Attached Thumbnails Attached Thumbnails SapphireDome.jpg  

  19. #19
    Join Date
    Apr 2007
    Posts
    777
    Odin,

    emd68 makes a good point that the compressive strength depends on how polished sapphire is. Compressive strength can be found by googling compressive strength. Concrete is one example of a material with high compressive strength but almost no tensile strength. Note here that the compressive strength is an order of magnitude greater than the tensile strength. http://www.roditi.com/SingleCrystal/...roperties.html

    emd68 also makes a good point that the strength of sapphire is anisotropic: different in different directions.

    You may also want to google fracture mechanics and griffith stress. Sapphire is a brittle material and fails by crack propagation. If you have flaws in the sapphire, fracture mechanics will tell you how that impacts the strength.

    Best of luck,
    Cameron

  20. #20
    Join Date
    Jan 2006
    Posts
    17
    The formula for thin wall pressure vessels is stress= (pressure * Radius)/(2 * Thickness) This formula works for either internal pressures where the stress is in tension or external pressures then the stress is in compression.

    Taking the data from above. I decided to use the minimum yeild stress of sapphire to be 275 mpa (worst case senerio compressive yeild is 10 times greater). So 275Mpa=108.6Mpa[(.75cm+T)/2T] After doing the math t = .184 cm.

    Now for the equation we used to be vaild we are assuming a thin walled pressure vessel. A thin walled vessel is Radius > 5 * Thickness. R = .934 5 * T = .92
    Our assumintions are pretty close to the limit of the equation but they still should work. Your pressure vessel would have to be a minimum of 1.868cm in dia with a 1.5cm hollow.

    How ever the method of manufacture is going to play the biggest role on wether the sphere will survive. By having the two halfs that are some how screwed bolted glued together, you are creating stress concentrations that will envitably cause a failure in a brittle material. Also once you make this thing I think it would be wise to coat the sphere in some kind of impact resistance coating to prevent scratches / nicks.

Page 1 of 2 12

Similar Threads

  1. hydrostatic bearings
    By MaxxD in forum Linear and Rotary Motion
    Replies: 4
    Last Post: 12-25-2008, 02:15 PM
  2. Application Engineers TEXAS!!!
    By Jamie Bostick in forum Employment Opportunity
    Replies: 0
    Last Post: 06-09-2008, 04:13 PM
  3. Anyone have a Clarke engineers workbench?
    By daedalus in forum Benchtop Machines
    Replies: 5
    Last Post: 10-27-2007, 11:11 PM
  4. understanding engineers
    By daleman in forum Community Club House
    Replies: 8
    Last Post: 08-29-2006, 01:18 AM
  5. EDM Applications Engineers Needed
    By kimb in forum Employment Opportunity
    Replies: 0
    Last Post: 10-06-2005, 05:28 PM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •