<html><head></head><body><p dir="ltr">I will preface these comments with a standard disclaimer that what works and what is explicitly permitted by the rules may not be entirely congruent.</p>
<p dir="ltr">That said, when you cut any opening in your pressure hull, you ordinarily must reinforce the opening with additional material, in order to carry the additional shell stress which can no longer be carried by the material that would have been in the opening. Certainly this is the case for window openings, because an acrylic window is structurally insufficient to carry the anticipated hoop stresses. Where the opening is to be filled with material equal or greater in strength than the hull shell material, it is less clear cut. We had a discussion some time ago about essentially replacing the cutout shell with an equivalent load path through a substantial hatch, and there is no reason why this shouldn't be permissible, provided the "shell" which is integral to your hatch provides an uninterrupted load path. In the case of a spherical pressure hull, this means that the hatch "shell" would need to be of equal or lesser inner radius than the hull, of equal or greater outer
radius than the hull, be appropriately centered on the adjoining shell at the interface, and that the interface itself be exactly normal to the spherical surface, so the angle passes through the centre of curvature and hoop stresses do not induce bending. Any elastomeric seal would have to be positioned outside of the theoretical "shell", so that you retain full bearing area in the metal to metal interface. You would also have to ensure that the hatch itself is appropriately reinforced if you have penetrations for windows or dogging mechanisms in it. What you would not be able to get away with is a hatch which sat substantially proud of the neutral axis within the hull shell (i.e. a "lid" on top), as that would provided no structural support, and thus would require the same reinforcement around the hole, integral to the hull shell wall, that would be required for a window or other non structural penetration.</p>
<p dir="ltr">Apart from that, I imagine that a hatch seat similar to the recommended methods for penetrating "Nemo" windows, as outlined in PVHO-1 and Stachiw's book, would prove adequate. You will note that in no instance do these methods permit screwing or bolting to the acrylic hull shell, and I am of the same mind with regard to a metal shell. Whatever you come up with, you can bolt two halves of an insert to each other across a hole, but must avoid drilling into your base hull shell, or indeed removing material from it for other purposes if not adding material to compensate, or if you already have excess shell thickness, or are prepared to derate the hull. Instead of machining sealing features on the hull, I would look at using an adhesive sealant to seal to your designed insert (which must already be structurally seated against the hull opening), and restrict the o-ring features to the insert and hatch exclusively.</p>
<p dir="ltr">As for sealing two hemispheres together with an o-ring seal, there's no reason that you can't do that, but you need to pay attention to centering / concentricity, and of course your design o-ring gaps / extrusion considerations, etc. I would probably be inclined to simply face the hemispheres, and employ a double sided seat ring which would incorporate the sealing features, or attach a ring to each hemisphere with an adhesive sealant, and have the rings seal to each other, in which case you could also incorporate assembly / centering features into the rings so that they consistently seat in the correct position. While not strictly required (a seal ring between two hemispheres is functionally a short cylindrical section), I might give consideration to facing the hemispheres slightly short to account for the ring length, such that in the fully assembled and seated condition the hemispheres remain concentric and there is no jog in the load path through the sealing
ring(s).</p>
<p dir="ltr">Hope that helps.</p>
<p dir="ltr">Sean</p>
<br><br><div class="gmail_quote">On November 10, 2016 6:51:51 PM MST, hank pronk via Personal_Submersibles <personal_submersibles@psubs.org> wrote:<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<div style="color:#000; background-color:#fff; font-family:HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif;font-size:12px"><div id="yui_3_16_0_ym19_1_1478827787399_7512" dir="ltr">Sean, I know we talked a bit about this idea before but I need a refresher. Can I machine a hole in a heavy sphere and insert a hatch land\reinforcing ring assembly that bolts to the sphere. It looks like Karl Stanley may have done that with his front dome. This idea if feasible makes building a deep diver more realistic. I am thinking of trying this on my current sphere just to see if I can do it. Better to screw up a cheap sphere first. ;-)</div><div id="yui_3_16_0_ym19_1_1478827787399_7512" dir="ltr"><br /></div><div id="yui_3_16_0_ym19_1_1478827787399_7512" dir="ltr">Next question, when I look at the Deep Rover sub, it has two hemispherical domes coming together on a two sided land. I am sure the land also gives the domes side
support. Why would this design not work with two heavy steel hemispherical domes. I am talking 3 to 4 inches thick. The steel domes can easily be machined with a flat land and o-ring grooves. The idea is to eliminate a very costly weld and it would be pretty darn nice to build with the top half of the hull removed. </div><div id="yui_3_16_0_ym19_1_1478827787399_7512" dir="ltr">Hank</div></div><p style="margin-top: 2.5em; margin-bottom: 1em; border-bottom: 1px solid #000"></p><pre class="k9mail"><hr /><br />Personal_Submersibles mailing list<br />Personal_Submersibles@psubs.org<br /><a href="http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles">http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles</a><br /></pre></blockquote></div></body></html>