Print

[Bill Mangus]

Let's think about exactly what information/photos we want to get from the Electra being restored by Wichita Air Service.
My thoughts:

• We need a straight-on photo of the interior starboard wall from Sta. 293 5/8 (the bulkhead at the front of the lavatory) to Sta. 320, the circumferential structure to which the aft edge of both the window and the patch were riveted to.  That will, of course, encompass the narrow vertical stiffener at Sta. 307 and the heavy former that goes across the top of the cabin.  The photo needs to include tape measure or yard stick held against the wall so that we can measure the interval between stringers.

• We need detail photos of all of the stringers, bulkheads, and stiffeners with a tape measure or ruler laid across to show the width of the structure.

• We need an exterior photo of that same area.

• We need the opinion of the people who have been doing the restoration about the consequences of cutting a hole for the window and how you would go about replacing the window with a patch.

Anything else?

Any pictures taken as you describe will be useful.  Blowing them up and hanging them on a board to fit check 2-2-V-1 would be tricky it that is the intent.  Here's another idea to supplement any pictures.  (Be kind, everyone - I'm making this up as I go along).

Obtain a roll of butcher paper or some other kind of heavy-weight roll stock, maybe like that paper on the table in the doctor's office.  Use masking tape and lay down enough paper on the outside skin of the aircraft to cover the area of interest.  You'll have to overlap several rows of the paper, so make index marks to match-up the sections later.  Once the paper is in place, make a rubbing or tracing if you will of the rivet patterns.  I'm thinking of a blackboard/chalkboard eraser and some graphite dust.  Put a little of the graphite dust on the eraser and with the right amount of pressure rub the paper with the eraser over the rivet lines. You should get a nice impression of the raised rivet heads.  You could try do the same thing on the inside with the ribs and stringers but getting the paper stretched properly and accurately might be a problem and would require 2 or 3 people.  You'd for sure need accurate measurements for the width and height of these pieces.  Once the paper is hung on a wall back home, you could draw the ribs and stringers using the measurements, etc.  Almost as good as a mock-up for show-and-tell, fit checks, or whatever. 

[Dave Thaker]

If the patch was about 25 3/8" long [the distance between stations 293.625 and 320, less one inch,] and 2-2-V-1 is about 24 3/8" long, the scaling in the overlay photo needs to be adjusted. 

I don't think you understand what's going on here. We can't "adjust" the the scaling of the overlay to make everything fit the way we'd prefer.  Jeff Glickman scaled the photo of the aircraft-with-patch and the photo of the pressed-down artifact as best he could to the same scale.  Then he overlaid the photos.  The artifact is smaller than the full patch, as it would have to be because none of the edges on the artifact is an original finished edge. Positioning the artifact vertically within the patch was easy because of the alignment of the rivet lines but Jeff didn't know how to position the artifact fore-and-aft until I pointed out that we can anchor the lower right-hand edge because we can see that it failed from metal fatigue against a rigid underlying structure.  Once we positioned the artifact that way it became obvious that the mysterious impression of an un-riveted vertical structure aligned perfectly with Station 307.

In other words, the overlay is correct.  If anything needs to be adjusted it's the measurements for the exact size of the patch and artifact.

I am also not understanding this, perhaps if I explain what I think is correct someone can explain where I am going wrong.

I thought the dimensions of the Miami Patch were indicated by the taped area on the side of the Electra as shown in the photo taken at the NE Aviation Museum:

http://tighar.org/smf/index.php?topic=1490.420

What are the dimensions of the Miami Patch, as determined by Jeff Glickman?

Also, I am still wondering about the irregularly spaced 5/8 inch rivet holes on 2-2-v-1.  Wouldn't this line of holes correspond to the line of staggered rivets just below the navigator's window?  If so, how do those irregularly spaced holes conform with the (presumably) regularly spaced holes on AE's Electra? Why don't we see staggered rivet holes on the 'tab'?

[Ric Gillespie]

I thought the dimensions of the Miami Patch were indicated by the taped area on the side of the Electra as shown in the photo taken at the NE Aviation Museum

The taped area on the side of the New England Air Museum Electra was a rough approximation I did on June 15.  A little background:
During the 2-2-V-1 Commission's visit to the Air Force Museum on March 28 we learned that, in a repair, the pitch of the rivets can't change.  That squashed our theory that the artifact fits a repaired area on the belly.  The rivet pitch where it seemed to fit is 1.5 inches.  The rivet pitch on the artifact is 1 inch.  We were back to square one.
On May 30, several members of the 2-2-V-1 Commission and I visited the New England Air Museum to try to find somewhere the artifact would fit.  No luck.  At that point I was ready to concede that whatever airplane the artifact came from, it wasn't NR16020.

Then Jeff Neville suggested it might be the Miami Patch.  It sounded like a desperate Hail Mary but I had to agree that the patch was the one part of the airplane that was not built or repaired by Lockheed so it did not have to conform to engineering drawings or repair orders. On June 15, I was back at the New England Air Museum to do two Father's Day talks in front of their Electra. During the break between talks I borrowed a roll of Scotch tape and taped off the patch as best I could eyeball it.  Then I drafted an innocent bystander and had him hold the artifact up against the airplane.  The artifact seemed to fit but this was just an initial what-the-heck experiment and should not be taken as rocket science.

What are the dimensions of the Miami Patch, as determined by Jeff Glickman?

I don't think exact measurement are possible.  There's not enough resolution in the photo to see hard, defined edges.  But it doesn't matter.  We know how big the artifact is and we know how big an Electra is. 

Also, I am still wondering about the irregularly spaced 5/8 inch rivet holes on 2-2-v-1.  Wouldn't this line of holes correspond to the line of staggered rivets just below the navigator's window?  If so, how do those irregularly spaced holes conform with the (presumably) regularly spaced holes on AE's Electra? Why don't we see staggered rivet holes on the 'tab'?

Ahh but we do.  The wave-like pattern indicates the presence of another row of rivets.   The attached photo shows the same kind of lateral tearing on wreckage from a Lockheed Electra that flew into Mt. Richmond in New Zealand. In that case the second row of rivets was not staggered.
We don't know why the pitch on the artifact is irregular. We also don't know that it wasn't irregular on the patch.  Maybe a better understanding of underlying structures will suggest a reason.

[John Ousterhout]

I'd want to record the thicknesses of stringers and sheet metal, and rivet diameters and type(s).

[Steve Lyle Gunderson]

Can a mold of the interior surfaces be made? that would give us an accurate representation of the stiffeners, rivets and something we could measure later. Not sure how that could be done but I am sure someone has the knowledge & skill.

[Greg Daspit]

Measure the curves with something like a curve template or a similar tool. It may help in alignment or overlays of 2-2-V-1 and photographs

[Greg Daspit]

Adhesive measuring tape may be a way to get a tight fit to everything.
The yard stick can be used to get the overall dimensions and frame spacing and the tape to get the recessed parts and in between dimensions.
The thickness of a rigid ruler can throw off what you are measuring when viewed at an angle so if you want a more accurate dimension of the structure you need to have the camera or your eye 90 degrees to each piece that you want a dimension of. Even a steel tape can be off when viewed at an angle due to the curved profile of the tape.
Also sticky tape allows you to keep your hands free to photograph vertical parts in a tight space.

[Jim Thwaites]

At this point it might be wise to contact FARO, who makes 3D measuring equipment. They have a portable measuring arm and a portable laser scanner, among other things. They are coordinate measuring machines, both contact and non contact,
which would measure the airframe with more accuracy than you would need. Perhaps someone in TIGHAR with a silver tongue could talk FARO into loaning one for a bit of publicity. I have no connection with the company, but have heard favorable things about their company through many years of owning a machine shop and foundry.

Jim

[Jay Burkett]

Jim,

I agree.  Scan that section of the fuselage AND scan the "patch".  A 3D model can be generated from both scans.  From the model of the "patch" a flat pattern can be generated that would conform to the curvature of the fuesalage in that area.  The "patch" is bent up, but, it does not look like there was a whole lot of material yeilding that would affect the results.  Any flat pattern that results would be fairly accurate.  Scanning and CAD modeling would harm neither the sample airacraft nor the artfiact.

A lower tech quick solution might be to use the "patch" and make silicone mold like a halloween mask.  from that make a silicone "positive".  Offer the silicone "model" up to a sample fueslage.  This may be cheaper and quicker.

[Monty Fowler]

Any flat pattern that results would be fairly accurate. 

These are both great suggestions, and I think Ric is already looking into a mold of some kind based on an earlier suggestion. Why I highlighted your sentence was to call attention to the high level of proof that TIGHAR is held to by its detractors. "Fairly accurate," in our case, will be relentlessly shredded into a huge pile of "maybes" and "could have beens" by The Earhart Conspiracy Theory Industrial Complex, in order to completely discount whatever we come up with.

That said, I'm willing to pony up some bucks to get this done, when the need arises.

LTM,
Monty Fowler, TIGHAR No. 2189 ECSP

[Bill Mangus]

 The Earhart Conspiracy Theory Industrial Complex = TECTIC

Good one, Monty! (and I agree).

[Jay Burkett]

Monty,

The reason that I am advocating that a silicone “negative’, then a silicone “positive” be made is that the “patch” is not substantially stretched or yielded to any extent.  The edges and fastener locations are torn, but, a resulting flat pattern, whether from 3D scans or from a “splash” would essentially give the same results.

The silicone contact mold and positive would be (a) quick, (b) cheap, (c) would not damage the “patch” and (d) would not damage any of the surviving candidate aircraft it was offered up to.

Please don’t get me wrong:  I am a LONG time CAD user (circa 1984) and the biggest advocate of 3D CAD that you will ever find!  The electronic approach has the following drawbacks: (a) it must be performed by skilled technicians (the scanning, modeling, flattening and creating the flat pattern).  The scanning cost (equipment, technicians travel , …) is expensive.  The software is the same.  The patch would have to be scanned.  The resulting point cloud would have to be turned into a 3D model.  That model would have to be “flattened” manually using the old hand lofting techniques within the 3D CAD application being used or by using flat-patterning software.  The surviving Electras (one, or more) would have to be scanned.  A 3D reference model from the original design loft drawings would have to be made.  Note:  The “reference” model would be different than any of the surviving Electras and even AE’s Electra if it was sitting in front of us in pristine condition!  This is because of production and manufacturing tolerances.

Since the “patch” was field fabricated we will probably NEVER have any drawings to compare it to!  Until the aircraft is found we will never know for sure if it is THE “patch” we think it is!  What the rubber model will allow us to determine, with a reasonable amount of certainty, whether or not this could be THE “patch”.

I am only advocating the silicone mold method because it can yield essentially the same results far faster, and cheaper, than using the electronic methods.

Remember:  baby steps (Cheap baby steps! --- I’m all about conserving resources for the long haul!).  Use the cheap and quick method to see if we are barking up the wrong tree. 

If using 3D scanning and CAD is the only way to satify some people I would suggest a short-cut 3D scanning process that would involve a 3D scan of the “patch” and having an elastomeric (rubber) model 3D-printed.  The result would be essentially the same as the direct splash and positive model method.  I recently read an article where jogging shoes were being 3D printed at some sort of exotic boutique.  So, the equipment, processes and material is out there.  I’m just not sure how to go about sourcing it. 

[Jeff Carter]

Anybody have the wherewithal and resources to do a fuselage mock up? Just a small section of course.  Not sure what use it could be, but might prove informative in working out scenarios as to how the window hole was patched.

I believe to accomplish that, you would have to have a bill of materials and a set of working drawings.  Neither of which I believe anyone has.

That's a good point for accuracy's sake, and accuracy is of course vital.

That said, as to a reasonable representation of the original structure, it is dimensions that count the most.  That includes an accurate portrayal of the outer skin - air passage - to include an accurate placement of all rivet lines, window outline and placement of associated window framing rivets.  That means an accurate 'skin'; the underlying structure could be simplified by the accurate placement of plywood ring formers (the vertical pieces), and simple "L" sections for stiffeners.  I think it would be good to leave the window aperture uncut, but clearly drawn, such that the artifact (or a facsimile of the flat pattern of same) could be applied.

We see this in accurate mock-ups in this industry all the time, and it works well - but it has to be high-quality / held to accurate dimensions.  That takes us back to the drawings to get it right, plus, since we have no drawings that we know of for the window installation itself, we have to derive accurate dimensions from the photo evidence.  I'm sure that can be done (already has to some degree by Glickman's work, apparently): it is a matter of scaling off the evident landmarks on the visible skins - rivet lines vs. known stations, etc.

Contours could be a challenge, since the Electra was graced with compound curves - but I don't think the longitudenal axis is critical here - a 'barrel' would do for a reasonable illustration as I think the vertical contour is the more critical.

A good, long weekend project for somebody.  Getting accurate dimensions set to paper would be the biggest part of it and the drawings would surely go a long way toward making it possible.  Short of that, I guess a real Electra could serve as a source for reverse engineering the thing.

As discussed here http://tighar.org/smf/index.php/topic,1490.msg32359.html#msg32359 earlier in the thread, the Electra skin diagrams can be used to calculate the size of the original window mount and the resulting patch.  This method gives the size along the circumference of the surface curvature.

It is then possible to overlay the flattened 2-2-V-1 over the original window mount as shown.
http://i.imgur.com/NbTwQwx.jpg

[JNev]

Jeff Carter, that is an impressive overlay and is very much along the lines of what I have been believing we need to see.

For the moment I am taking your scaling as 'correct' - you seem to have taken care to get that right.  Scaling is vital and does need to be substantiated clearly, of course.  That's not to question you - it's just to underscore exactly what we are all concerned with: accuracy of model and stack-up of 2-2-V-1 against same.

It would be ever so much easier had we a more clear picture of "the patch" itself - but that remains elusive; hats off to Glickman for working with what we have, but it seems a tough path.  What we do have are a) clear pictures of the aperture complete with frame riveting, etc., b) a means thereby to get accurate scale where the window and framing is concerned, and c) 2-2-V-1 which can easily be modeled as an overlay.

Let the chips fall where they will.  If your scaling is accurate, we have an odd fitment question in the now-remaining 'upper row' of fasteners in 2-2-V-1, even as the 'other stars' tend to align in various ways (like the bottom apparent 'double row', picking up as we might see it, the double row at the bottom of the Electra's window frame, etc.). 

If accurate by your picture, is that placement of the upper row a disqualifier?  Or was there some intervening internal member added at that point for rigidity / strength that the new patch was obligated to pick up with this odd row?

I don't know.  Just ideas - and pondering over your excellent draft model here.  This needs further pursuit IMHO.

[Gary Vance]

I was re-reading all the posts in this thread and came across a story in the UK daily mail from July 1 2014.  What interested me was the photo of the Electra being built, clearly shows the window area, with the stringers installed.  With the new spectral imaging results and the discovery of an indentation for a replaced support stringer for the patch, can someone compare/overlay the two and see if they match? 

The article link is here:   http://www.dailymail.co.uk/news/article-2677486/Did-Amelia-Earhart-perish-castaway-desert-island-Long-forgotten-picture-hints-tragic-demise-aviator-disappeared-nearly-80-years-ago-today.html#ixzz36IncUczh

I've attached the picture in question. 

Gary