Ryan confirmed that the tube used was of the original Dogbone design with a 4mm ID and a 6.35mm (1/4") OD. This has a wall thickness of only 1.18mm compared to Parkhomov's 2.5mm wall thickness. For the strength of the tube used, the amount of LiAlH4 inserted was just too much.
Fortunately they were all behind a safety shield. This should be a lesson to all replicators. Bob Higgins On Fri, Feb 6, 2015 at 2:02 PM, Blaze Spinnaker <[email protected]> wrote: > Be careful not to get the integrity too strong, otherwise you could have a > real pipe bomb on your hands.. > > On Fri, Feb 6, 2015 at 12:59 PM, Bob Higgins <[email protected]> > wrote: > >> Ryan Hunt reports that the failure mode was NOT the compression fitting >> giving way under pressure - the fitting remained intact. This experiment >> was of the "easier Parkhomov" design, posted previously where the seal was >> made with a compression fitting, in this case with the use of a soft >> aluminum ferrule at the suggestion of Alan Goldwater. Alan's tests >> suggested the compression fitting would hold and it did! Using the >> compression fitting is a real win because it completely avoids the >> problematic sealing of the ends with cement while providing an opportunity >> to instrument the reaction vessel. >> >> When this failure occurred, it appeared to be a raw ceramic body >> failure. This could easily have come from too much pressure coming from a >> too large charge of LiAlH4 for the vacant volume inside the apparatus. >> MFMP will extract that volume information and relate it to the weight of >> LiAlH4 that was added, as being a benchmark for too much LiAlH4. The tube >> used was 1/4" OD, but at the moment, I am not sure if it was a 4mm ID tube >> or a 1/8" ID tube. The Parkhomov tube had an ID of half of its OD. >> >> Bob Higgins >> >> On Fri, Feb 6, 2015 at 11:39 AM, MarkI-ZeroPoint <[email protected]> >> wrote: >> >>> At 2:29/2:30 into the short segment posted by Craig, it looks like the >>> right-side end-plug, or whatever is sticking out that end, blows out. And >>> I use that term specifically since one also sees some hint of a pressure >>> release. Whether that release is at an appropriate level is apparently >>> debatable... >>> -mark iverson >>> >> >
