Chronicling my Long EZ construction (and a few other things).
Disclaimer
This blog is for entertainment purposes only, and is not meant to teach you how to build anything. The author is not responsible for any accident, injury, or loss that occurs as a result of reading this blog. Read this blog at your own risk.
Just in case you might have missed it, I had a terrible construction mishap last year, and the time has come to fix it.
These images are still painful to look at
About 6" (15 cm) of damage on the left longeron
Same amount and type of damage on the other side
The foam and outer skin took a beating as well
Deciding how to fix the longerons was one of those things I didn’t want to do in a vacuum, on in a rush, so I polled as many Long EZ “experts” as I could find on this repair. I even had the head of CSA email Burt Rutan and Mike Melville with sadly to say, no reply. Not that I realistically expected one (though it would have been awesome).
Eventually, I offered as much money as needed to a well known engineer with extensive composite canard construction and modification experience. He graciously would have none of my money, offered expert advice, and cheerfully tried to put my many worries to rest.
“I am very grateful for your input, Sir (you know who you are)!”
After all of this, I went on to build the Center-Section Spar instead, in order to create emotional and time distance from this problem, so that I could be more rational and deliberate when the time came, but even then, I secretly harbored serious reservations about the advice I had received.
Like the doubting Thomas that I obviously am, I ended up trying to test those theories, albeit in a very unscientific manner, and I thought you might enjoy seeing what happened.
Let’s take a look at what this incredible epoxy our airplanes are built with allows us to accomplish…
The first "loading to fracture" experiment on longeron stock vs the cut-then-repaired longeron stock ended with the fracture of the pristine longeron, and the survival of the repaired one.
Sacrificial stock for testing as well as repairs (this stuff ain't cheep!)
Cutting at a 15:1 ratio
The battle of repaired vs new begins...
Flox is the only bonding agent
Spending the night up
Ready to slug it out!
I wish I had taken pictures of me climbing on these pieces. I stood with both feet in the middle of the repaired one, then lifted a foot, and... nothing. Then I did the same with the new stock, and when I lifted the foot... it just gave away.
As I said, unscientific and unrepresentative, but fun.
The results of this experiment were hard to believe
I suppose Aviation Circular 43-13 really knows what it's talking about
According to AC43-13 this is the type of repair to be carried out on broken wooden wing spars, however it was explained to me that these longerons are nowhere nearly as stressed, nor similarly loaded, and was advised to use much less than a 15:1 scarf, and ditch the side reinforcements.
In this 2nd totally unscientific, and once again hardly representative loading experiment, I have cut the spruce longeron stock (the broken one from the previous experiment) at a 6:1 angle (closely matching my future repairs), and used just flox to glue the pieces back together.
If you have never tried doing this yourself, it is the most unbelievable thing you can imagine, I will let the following pictures speak for themselves...
Just a straight cut and some flox
Wanna bet?!
Let's just say... a lot!
Doubling the pressure (same force applied to half the area) on the poor joint
As you can imagine, I am obviously no longer worried about repairing my longerons this way.
“A man’s got to know his limitations” is a great line taken from Dirty Harry, and so true in many ways.
from Dirty Harry (1971)
In my case, I was afraid that enlisting Gina in cutting the foam cores would have sadly stretched the limits of our nearly 30 years old vows beyond the “death do us part” clause.
I am sure most of you are familiar with the name, but should you not be, let it be known that Steve James, owner of Eureka CNC, built his own CNC hot wire cutting machine, and can reproduce a number of canard airplane parts with great precision.
He is also a delightful person to deal with, and operates from Alamogordo, New Mexico, home of Holloman AFB, the same place where I did my T38B flying back in the day (1989).
Good times!
You might say that I am taking the easy way out, and missing out on a fundamental skill set of building a Long EZ, and I would partially agree, but there were other considerations that factored into my decision.
For one thing, I’ve been married most of my life, I like it, and I wanted it to remain that way, so I didn’t have any desire to unduly burden my relation with Gina. Also, while profile accuracy might not be all that important on the fuselage section, it is very much so with the wings, canards, elevators, and winglets.
I was after the most accurate foam cores available, regardless of who made them or where they came from, and in my humble opinion Steve is king. Price was a concern as always, but screwing up a few cores would have put me back in the same price range (though perhaps no longer married)as having Steve do them perfectly the first time, so I basically bought everything he had for sale.
Gulp!
It might have taken a month or so to get the cores from Eureka CNC, and coincidentally they were delivered in my absence by my neighbor whoworks for FedEx, and has already helped me in the build a few times.
My good friend and neighbor delivering the goods
When I finally got to see them, I was astonished by the mountain of stuff I had purchased, and even more so by the quality of Steve’s workmanship. These parts are truly beautiful.
Where the hell am I going to store all of this stuff?!
These Roncz canard and elevator sections are incredible
Wings and winglets parts
Dunno
More wing stuff
Still wing stuff, I think.
I'd have to guess winglet
??
How much wing stuff is there?
...and this is how I knew not to get Gina involved.
Wow again! Steve, you da man!
There was some very minor damage to the corner of one box, but nothing a little micro couldn’t easily fix.
Not bad at all
Gina wasn’t very happy to have me take over a bedroom with the new stuff, but as graciously as a New Yorker could muster when backed against a wall, she allowed the Trojan horse in.
My little DeWalt portable bandsaw has been doing an admirable job for years. With the addition of a steel table and a foot pedal, I converted it to a vertical bandsaw and have cut plenty of aluminum and steel with it.
Poor man's vertical metal cutting bandsaw
Back then however, I could not have imagined I’d eventually come to enjoy machining as much as I do, or that I’d be doing so much of it. Heck, I had never even seen a mill or a lathe before! Needless to say, the little saw has seen more action than I had planned for it on the onset, and I have been looking for a better way to cut larger stock for some time.
In my research, I saw most people going for the Grizzly G0622 light duty style of saws. Many retailers sell them as their own, after painting them with their color scheme, and at less than $300 they are hard to pass up.
Harbor Freight's version. Looks a lot better in photographs than in person, but it's inexpensive.
I was all primed to buy one many-a-time, until I actually saw it in the flesh, and left with my money still in my wallet. It just looked truly cheap, and felt very flimsy.
The one I actually really liked was the next step up in the Grizzly lineup, the G9742, but at twice the price I just could not bring myself to buy it.
Not as much bang for the buck, but a sweet looking saw, with many upgrades.
Then, the other day, I stepped into a Northern Tool store to buy something completely unrelated, and as I walked by the clearance section, I saw (no pun intended) their last one of this kind still in the box for $423.
"Where do I know you from?"
Well, I didn’t even know where I’d put it, but I just couldn’t pass on this deal, and after some forklift magic, the 250 lbs box was in my Honda Element, if only barely.
Make no mistake about it, this thing is heavy, and the box is big (for my truck), but four adults were able to take it out of the car and sit it on the shop floor without too much trouble.
You'll be better appreciated here, than in the clearance section of Northern Tool.
Putting it together was pretty simple, and took about an hour, but don’t count on using the provided instructions, since most of the time you cannot even tell what they are referring to.
She's pretty big!
Packaging was good
Nothing bent, broken, or missing.
I'm going to need some help picking her up
The square piece and the blackbracket make up a table that turns the horizontal saw into a vertical one.
The base had to be reassembled a few times, no thanks to the useless instruction pamphlet.
Putting the saw on the stand is just too much for one person, but is very doable with the help of an assistant (thank you Gina).
The retractable handle (not present in the Grizzly version) makes moving it in the shop a breeze.
No tendency to flip over (thank goodness) in the vertical position
Here's a pretend 45º cut
So, how does it cut?
Slicing 4130 steel with the original blade
I was actually pleasantly surprised. Right out the box the cuts were perfectly square in all directions, and I was actually able to slice very thin rings out of a 4130 steel tube.
I'm not sure if this really tests anything, but it was fun.
0.026" (0.66 mm) steel slice
0.023" (0.58 mm) steel slice
I did not use any coolant for these cuts, but I did set the saw to its slowest speed of 80 fpm (24 m/min). After the cuts, the blade was still cold while the rings were pretty warm.
The hydraulic down-feed of the blade is a beautiful thing, and operates smoothly, with plenty of adjustment. So far I'd recommend it, especially at the clearance price. We'll see how it holds up in time.
Last month I stopped by my friend Wade’s home to check out the incredible progress he’s making on his Long EZ project. We talked about airplanes and other stuff for so long that we ended up going to bed around 4 am.
CP#70 was one of the topics of the night. In it, on page 5, the issue of how to prevent heat damage to the gear legs from the disk brakes is discussed, and fixes are offered. The problem is addressed by venting the top of the wheel pant (when in the "grazing" position), wrapping the bottom of the leg in heat resistant fiberfrax, and by adding heat shields between the disks and the gear legs.
One of the things I literally took away from our get together, was Wade's paper template for the disk brake heat shields.
Wade's heat shield template
I have to admit it looked very clever, covering a greater portion of the disk surface than others I have seen in the past.
Excellent disk coverage
Not wanting to let a good idea go to waste, and with Wade's blessing, I set out to reproduce his design on the computer, and machine a set on the mini-mill. Obviously one could have accomplished this in many other ways, the best of all being a CNC plasma table, but you’ve got to work with what you have, so milling it would be.
Fist though, I’d have to extract some approximate dimensions from Wade’s drawing, then massage them a bit in order to create a symmetrical design. For this purpose, I dusted off my 37 years old drafting table, the same one my parents bought me in high school, and started analyzing Wade’s masterpiece.
This drafting table is ancient, but still works great.
Extracting some approximate dimensions
I used Fusion360 CAD on my MacBook to reproduce the heat shield, then print a 1:1 drawing of it for testing purposes.
1:1 drawing of the new and improved heat shield
This thing's got better coverage than Verizon!
I wish I could say I nailed it the first time, but it actually took 4 revisions to come up with something I’d be willing to try on the mill. I will share the pdf with Wade's permission in the "Downloads" section of this blog.
3 of the 4 iterations
One of the latest update was the slotted mounting holes. I did this to compensate for minor possible misalignment of the holes in the gear legs, since these were drilled by hand.
Test fitting #4
#4 gets the nod of approval
I ordered two 0.125” (2.3mm) thick 2024 aluminum sheets, but it looked like I would be able to squeeze both heat shields on one plate.
This is going to be a tight fit
My biggest problem was that these parts were twice the size of my mini-mill’s machinable area, a big problem indeed. I would have to think outside the box for this to work.
I decided to machine each shield with two separate operations, one would take care of the top half, the other would do the bottom contour. Each part would have to be repositioned between operations, so I created a fixture that would accommodate both ops, and make the milling accurate and repeatable.
Brake shields fixture to be
Bolts for alignment and hold down use
Nearly final configuration of the fixture
How well did it turn out?
Much better than doing it by hand of course, but not quite perfect, I’d say 95%.
I think a better way of doing it would have been to use dowel pins, but there was just not enough room on the plate, so I used the skinniest bolts I had, and ended up with ever so slight misalignment problems. Nothing that a quick pass on the belt sander couldn't fix, but not as exact as I had expected.
This is how the top half would get machined
Plate moved to cut the bottom half
Heat shield bottom completed
Two fully functional generation #4 heat shields
Mounting the heat shields on the gear legs uncovered a problem I had not anticipated, a fiberglass ridge had formed when I originally mounted the axles over wet fiberglass, and now the shields could not lay flat. On top of that, I needed additional clearance for the fiberfrax insulation that would eventually be wrapped around the bottom of the gear legs.
Two steps forward one step back
The step is around 1/16" (1.6 mm) thick
Wise old Wade anticipated this issues, and fabricated the shields first, than used them over the wet fiberglass, guaranteeing a flush fit.
“Crap, I need axles spacers!” Back to the computerized drawing board…
Axles spacer design
Actually making them turned out to be pretty easy using the leftover aluminum. I'll put the drawing in the "Download" section as well.
You've got to love it when everything fits this tightly on the first try
Fit turned out great and all clearance issues gone.
Of course, Icouldn’t just mount bare aluminum to the only place on the plane that would encounter every single puddle I will ever taxi through. With my Alodine too old to yield acceptable results, I decided to dust off my barely used 1 year old Anodizing setup, and finally get my money’s worth out that rig, but we'll talk about that on the next installment of the heat shields manufacturing story. UPDATE: (4-29-16) I decided to make the shields over again, and record the machining on camera. This time I used dowel pins to make the whole process more repeatable, and they actually came out great. I have had some issues with the anodizing though, with one of the shields not taking the color as well as it should have. I believe it had to do with some corrosion build up on the "fishing pole" I use to suspend the parts in the acid, an electrical issue, since all parts went through the anodizing together. I'll try to figure that out later, but since these parts are otherwise perfect, I will use them on the plane.
