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.

Wednesday, August 27, 2014

Brake/Rudder pedals - part 4

Bracket testing (6.0 hrs)

No soon had we finished machining the first short bracket, than Wade and I started beating the crap out of it with a rubber mallet! 

Though silly, the abuse might well represent the one inflicted by boots stepping on it over the years, and highlighted the fact that the top rails and webbing needed to be made thicker. Actually, they should have been thicker already, but I had made a drawing mistake while reducing its dimensions in CAD from 12” (30 cm) to 9.5” (24 cm), and didn’t catch it until we had already machined the bracket.

Doh!

Because this had been Wade’s first foray into CNC, I decided I let him keep the bracket as a souvenir of his hard work. In turn, he promised to send me picture of the mangled bracket next to the unit he had purchased. 


Short bracket after being struck with a rubber mallet multiple times, next to Wade's (any similarity between brackets is mere happenstance). Photo courtesy of Wade.

Meanwhile, I decided to do some load testing on the 12” bracket I still had in the EZ shop, and depending on the outcome, make additional changes.





Overload testing of the 12" bracket




Satisfied and encouraged by the latest testing results, I moved on to machining the first one incorporating a few improvements (mostly increased webbing thickness, and edge distances)

This beauty will end up in the nose of my plane shortly.


The chamfering came out just perfect

3 bolts will hold it down, and its position will be adjustable to plus or minus 1" (2.5 cm)

It will attach to nuts, secured to an aviation-grade plywood, itself floxed and glassed to the nose floor.


A common misconception about CNC is that you can just press the start button, then walk away from the machine, but that's not even remotely true, at least not at my level. 

Not including the weeks spent designing it, and testing its geometry on the computer, this bracket took over 6 hours to produce (2 hours per side x 3).

What made the process so slow were the 6 repositioning moves necessary due to its size (compared to my mill) and shape, and the countless tool changes. Actual milling time added up to "only" 3 hours! 






Time lapse bracket machining




Master cylinder attach point

Sadly, once installed it might never be seen again.


Here is the 9.5" bracket next to the 12" one (aka "the original").


"Shorty" right below "the original"


Note the thicker webbing. No part of it is now any thinner than 0.200" (5 mm).


Sturdier and even lighter new design



Looks and strength, what else can a bracket hope for?



Tuesday, August 19, 2014

Brake/Rudder pedals - part 3

Shorter bracket (10.0 hrs)

My friend and fellow Long EZ builder Wade, came over to visit me this past weekend, and caught me right in  the middle of machining the shorter version of the brake pedal bracket.


Left side of bracket machined out of a 1 ¼" (3.2 cm) U channel aluminum extrusion


Since he had expressed an interest in CNC machining in the past, and the mill was not operational the last time he stopped by, I decided to seize the opportunity to bring him up to speed with a little "CNC boot camp".

I showed him the basics of CAD, CAM, Mach3, and G-code then gave him some hands on time on the CNC mill.


Wade changing the cutting bit in the spindle


He’s such a quick learner, and did a fantastic job, so much so in-fact that he ended up machining the whole center section of the bracket on his own.


First half of the center section completed, Wade is about to machine the far half.


I think he really enjoyed creating with metal in such a precise fashion, and the next day the little seed I planted showed signs of germination when he started inquiring about CNC machines, prices, etc. 


New 9.5" bracket vs the older 12" one


I mostly gave him the same advice I gave Phil from the west coast, don’t do as I did. Buy a more capable machine from the get-go, then add a pre-made CNC kit. 

I am coaching Phil right now in setting up his own hobby CNC shop. 

Phil, a former Boeing 737 Captain that now manufactures high fidelity flight simulators, took my advice and returned his new Harbor Freight mill, purchased a much better quality Little Machine Shop one, and will end up modifying it with a CNCfusion kit later on. That’s the way I’d do it, short of getting a Vertical Machining Center already made.


Phil's handy work, a fully functional Boeing 737 simulator (with visuals)... amazing!  "I want to fly it!"


The new bracket came out really well, and the smaller size of the track will allow for more flexibility in mounting it in the confined spaces of the Long EZ's nose. 


The shorter bracket makes it an easier fit in the nose


The geometry of the pedal assembly also worked great in all the positions for which I had drilled holes in the flange. Which position I will choose will have to be determined by sitting in the fuselage, and playing with it a bit.


Pedal fully aft in the middle hole position

Pedal fully aft in the bottom hole position

Pedal fully aft in the top hole position


Adjustability was a major driving factor in my design, and I think I will have plenty of it in the form of track holes selection (5 positions to choose from), flange hole selection (5 options there), and rotation of the piston rod extension (fine adjustment with at least 5 or more usable options).

That’s at least 125 combinations to choose from! Even a picky guy like me ought to find more than one suitable position.


Tuesday, August 12, 2014

Brake/Rudder pedals - part 2

Prototyping the pedals (27.5 hrs)

The original brake pedals are pretty ingenious, they are light, sturdy, and as small as the designer dared making them. Better yet, I own a set of never-before-used ones already. The left one was a gift from my friend Walter, and it’s an original Brock specimen (1980’s vintage), the right one I threw together on a “slow day” using 4130 steel from my leftover pile.


Old and new together


I can only imagine improving a couple of things. First, I’d like the pedals to be a tad bigger so that I can steel feel them through winter shoes, and second, I’d like to do away with the cross-tubes, this is just a nuisance to have to get over every time I’d like to stretch my legs in flight, and a newly designed pedal set could easily eliminate them.





Initial brake assembly idea




These issues alone wouldn’t justify “reinventing the wheel”, but the availability of CNC tipped the balance in favor of more R&D (research and development), and so it was that I started down another rabbit hole.

My original intention was to use a 3/4” (1.9 cm) U channel. This forced me to carve a lot of relief around the master cylinder base, or it wouldn’t be able to swing properly.


A lot of aluminum was trimmed to remove the interference

Close up of the critical joint

Later, when I tried to purchase a 3/4” U channel, I found that nobody carried it, so had to settle on a 1-1/4” (3.2 cm) wide extrusion. This negated the requirement for the relief, since the body of the master cylinder now easily fits within the channel.

However, I didn't want to design a new bracket since I actually liked its curvy profile, and I rationalized that it would be lighter to boost.  So, since I was going to have to machine it anyway, I decided to make it even lighter (read: fancier) by incorporating a few lightening holes.


Why not!



The biggest problem I faced was the length of the bracket. It was just too big for my little mill, so I decided to split the work in two zones, the “brake pedal” half, and the “master cylinder” half.

A small reference hole in the middle of the bracket would tie the two zones together. Because this hole would be the only mean to zero out the mill’s coordinates after repositioning the U channel in the vise, it would be one of the first features to get machined.

This arrangement worked fairly well, given that precision was only needed for looks, rather than fit.


First half done. Getting ready to machine the second half.

finished bracket 
Spacer had to be machined on the lathe because of the wider U channel

Inch reference marks






This is how it was done




With all the cool stuff starting to come out of the garage, I thought it would be appropriate to give it a name. 

I decided to call it the “EZ shop”.




What do you think?

Anyway... I decided that machining aluminum brake pedals would be too difficult for me at this stage, and that I would start out with welded steel ones, perhaps swapping them in the future.

I started out with 2 different 4130 steel tubes, the big one would become the pedal, while the smaller and thicker tube would become the hinge point, and get welded to the former.


4130 chromoly steel tubing ready to become a brake pedal


I started working on the hinge by precision cutting the small tube on the lathe, and fitting it to the bracket.


Pedal hinge


Next, I drilled a hole in the big tube the same size as the small one.


Creating the mating hole


The two got welded together, then the assembly went back on the mill for some more drilling on the opposite end. This time I enlarged the hole until the tube was severed, leaving the perfect notch to fit the part that my foot will actually stomp on.


Tubes mated and welded together (left), and top of pedal getting prepared (right)

A drill with the same diameter as the tube cut it in two

Tube welded to pedal hinge


I crosshatched a pattern with the lathe on one end, then welded it in the shape of a capital T.


Top tube crosshatched and welded 

A view of the whole assembly


The non-crosshatched side will connect to the rudder cable somehow, I haven’t figured it out yet.

The more immediate problem to solve was how to connect the pedal to the master cylinder in a way that would activate the rudder first, then the brakes.

One option I liked was the way my friend Mike did it, so I came up with a similar design of my own.




Virtual testing of the blade type pedal prototype




The advantage of this design is that the connection between the pedal and the master cylinder is retained even in the aft part of the pedal’s travel envelope, in other words the pedal does not disconnect and fall backward under the rudder’s rearward pull.

I just wasn’t too fond of the size of the “blade”, and decided to go a different route.

I’d use a smaller blade, and concentric tubes to activate rudders and brakes.


Smaller, yet much trickier design.





CNC machining the small pedal tab




The only problem with my new design, as you might have guessed, was pedal retention in the aft region of travel. The tube could just slide off of the rod with nothing stopping it from happening.

The way I solved this new puzzle was to machine a slot on the slider tube and insert a set screw into the rod that stuck out about 0.070” (1.8 mm), and rode in the slot.

Positioning of the screw was critical since I didn’t want it to bear any load during brake applications.

Brake application itself does not require additional items like washers or nuts for the siding tube to press against, but occurs when the steel flange contacts the inner steel rod. 

Fine adjustment for length is maintained at 1 turn intervals, by removing the set screw (or the bolt), screwing or unscrewing the steel rod, and reinserting the set screw after the desired position is achieved.

Coarse adjustment is through the 5 hole pattern drilled into the steel flange.

At this stage, rudder throw before brake activation is 2" (5 cm). Should more be required once the coarse adjustment has been decided, the slider's slot can be lengthened to provide more aft travel of the pedal. I will have to complete my wings and rudders first though, before I can confirm that.




Exciting (not!) video shot while machining the slot into the steel slider tube





The slot length and position are essentially the brains behind the pedal retention






Demonstrating the rudder pedal retention mechanism




The prototype bracket served its purpose well. I now know that a shorter one is more advantageous, both in terms of cockpit space, as well as pedal mechanics.

I will redesign it down from 12" (30.5 cm) to just 9.5" (24.1 cm), and machine two new specimens soon.