[PSUBS-MAILIST] casting parts

Tue Apr 19 07:51:42 EDT 2016

Not sure what you mean by overbuild? I only ever design things to be as strong as they need to be. A casting is not a forging - you must accept that the strength gains you get by working as with forgings and plate materials don't exist in castings - all processes have tradeoffs.

I would specify an appropriate material, such as ASTM A352 LCC, which is analogous to the A516 grade 70 plate specification with regard to low temperature toughness. Also could use A216, but I note that the ABS steel vessel rules point out that all 216 castings should use a 0.8 quality factor unless NDT is carried out beyond that required by the material specification, which would necessitate the part being 25% larger for equivalent load carrying capacity. For a sub I would NDT the hell out of it anyway, so it comes down to a confidence thing. This is something I would work with the foundry on in order to minimize voids, and then a NDT professional in order to detect any that do exist.

It's always better to use a stringent material specification (i.e. A352), a reputable foundry who can discuss these issues with you, and do material testing (both exhaustive NDT on the completed part and e.g. notch toughness tests using samples from the same heat) so you can be assured of exactly what you have, versus not being confident and overbuilding to compensate.

In any case, cast parts don't necessarily need to be larger than machined equivalents. In fact, often the freedom of form afforded by casting allows you to eliminate geometric stress concentrations that would be unavoidable with plate fabrication and machining operations exclusively, such that even if you have less load carrying capacity in the material, you also have less load to carry because you can delocalize it.

I think the bigger consideration for the amateur is the tradeoff between the advantages of casting any particular part, versus the cost of doing the NDT.

Sean

On April 18, 2016 7:58:12 PM MDT, hank pronk via Personal_Submersibles <personal_submersibles at psubs.org> wrote:
>Sean,I have to admit,I find the process of mould making quite romantic.
> I think that is one of the reasons I keep looking at concrete.  With
>concrete you have total freedom of form.  I have it all worked out how
>I would make a perfect sphere.  
>I thought you had to over build when you cast parts, does this not mess
>with your buoyancy.  That can be costly- your sub is likely big enough
>to handle the weight. Hank 
>
>On Monday, April 18, 2016 7:18 PM, Sean T. Stevenson via
>Personal_Submersibles <personal_submersibles at psubs.org> wrote:
> 
>
>Actually, I would think that casting would be a useful process for a
>hands-on guy like you, Hank. You could probably do your own patterns
>from plywood and Bondo and then deal with a foundry without having to
>pay for the pattern making.  Just do some research first - you need to
>accommodate shrinkage, and the draft requirement.  Also, you do need to
>100% NDT the completed castings. Unlike billets / plates, voids can be
>quite common in cast parts, and these need to be detected and corrected
>for our application.Sean
>
>
>On April 18, 2016 7:05:01 PM MDT, "Sean T. Stevenson via
>Personal_Submersibles" <personal_submersibles at psubs.org> wrote:
>There are a few reasons:1) Economy of process. Creating a casting
>entails creating a master pattern, usually by a pattern maker at the
>foundry or associated with them. Once this is done, additional parts
>can be obtained for the cost of an additional heat / pour.  For
>multiple similar parts (six identical hatches) this is cheaper than
>machining them all from billets.2) Economy of waste. Machining from
>billets obviously is less material efficient, as much of the billet
>ends up as chips on the machine room floor. With a casting, extraneous
>material exists in the form of risers, flow channels and flash that
>must be trimmed, but in many cases these wastes can be cleaned and
>remelted in the furnace for reuse.3) Suitability to size.  Some parts
>which would be awkward to handle or difficult to machine without access
>to suitably large machines can be more readily cast at lesser cost.4)
>Efficacy of form. The casting process allows you to readily incorporate
>large sweeping or compound curvatures and other features to reduce
>local stress concentrations or improve aesthetics, without
>necessitating multiple machining operations or tooling change outs to
>do so. Often, recreating a cast part with machining operations
>exclusively is prohibitively expensive. The design process is different
>for each - with machined parts, you need to think about machine
>envelope, repositioning, cutter clearances and interferences, tooling,
>avoiding impossible operations, and so forth. With castings, there is
>greater freedom of form, although you have to avoid thin-walled parts,
>apply draft to parallel edges to facilitate mold release, and consider
>material flow into the mold and how the part will cool to avoid
>warping. Often, combining processes gives you the best of both worlds:
>casting a blank to get the overall desired shape where surface profile
>is generally notcritical, and then subsequently machining specific
>faces or features to establish critical dimensions.This is the approach
>I am taking - casting the transition rings from the hull shell
>thickness to the hatch seat thickness to get that smooth large
>curvature (and a sexier looking part, IMO), where the contact interface
>is machined in situ after stress relieving the hull, and then also
>casting the hatch blanks and machining their critical features (contact
>interface, O-ring grooves) into the blank.Sean
>
>On April 18, 2016 4:11:09 PM MDT, hank pronk via Personal_Submersibles
><personal_submersibles at psubs.org> wrote:
>Hi Sean,I am curious why you wouldn't machine your hatch from a heavy
>disk, it would be one piece and any shape you want.Hank
>Personal_Submersibles mailing list
>Personal_Submersibles at psubs.org
>http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
>
>
>Personal_Submersibles mailing list
>Personal_Submersibles at psubs.! org
>http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
>
>
>_______________________________________________
>Personal_Submersibles mailing list
>Personal_Submersibles at psubs.org
>http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
>
>
>
>
>------------------------------------------------------------------------
>
>_______________________________________________
>Personal_Submersibles mailing list
>Personal_Submersibles at psubs.org
>http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.whoweb.com/pipermail/personal_submersibles/attachments/20160419/a88a0d9d/attachment.html>