Old Man Emu thinks about building his own

[old man emu]

This plane building disease is catching. In the 80's the bloke I work for built himself a Pitts S-1, which he still owns and will fly when we get around to completing the 100 hourly. When he gets some time to himself, he's working on a Gee Bee Sportster.

 

Two of the young AMEs I work with have decided to get started on their builds. One is going to be a Pitts of some type (probably an S1) and the other is a replica of a plane christened "Miss Los Angeles"

 

That leaves the apprentice and the storeman (me) not building anything. And here I am, selling nuts and bolts to everyone else who's building their own.

 

I was ratting through the Boss' library and found the original 1930's book that had the plans for the Pietenpol Aircamper. He also has the following year's edition and in it I found the plans for the Powell Racer. This is a little single seat biplane that contested and won some competitions at the 1925 New York Air Races.

 

It's a pretty straight forward aircraft of rag and tube construction. It's 14' 1/2" long with a span of 15'9". The original had a Bristol Cherub 2 cylinder engine of 32HP swinging a 48" prop. Top speed 80 Kts (95mph); Cruising speed 70 Kts (80mph), and Landing speed of 28 Kts (32mph).

 

The aerofoil is the RAF15 which is used in de Haviland biplanes including the Tiger Moth. The wing is wood and rag. No flaps.

 

I'm going to start with a "Proof of Concept" on the fuselage. The concept to be proven is whether or not I can get my big fat body into it. My GP strongly supports this project if it means I shed a third of my body beautiful. I'll be laying out the fuselage jig and am going to use electrical conduit in place of the chromemoly tubing.

 

I'll start blogging my efforts on Old Man Emu's Blog | Thoughts and Ideas scratched up by Old Man Emu starting this weekend.

 

Old Man Emu

 

 

 

 

 

 

 

[Bubbleboy]

Dead sexy!...and thank you for the rib stitching tutorial. I will be using it when covering time comes!

 

Scotty

 

 

[old man emu]

I'm stonkered.

 

Today I sat in a Jabiru J160 to see how I fitted, because the distance from the centre console to the side of the door is a tad narrower than the Powell's cockpit. I was also testing if I had to bend my legs when I had my feet on the rudder pedals. Everything fitted OK, so I'm beginning to believe that the plane will fit me.

 

I was in the Jab with the owner. We did some quick calculations and decided that with both of us on board, we could carry enough fuel for 1 hour's flight with 45 min reserves.

 

Now this Jab has a 80Hp donk and I know it would fly with both of us on board and at its MTOW. The owner said the Jab had a high Wing Loading value. It's 13.8 lb.sq ft.

 

When I got home I did some number crunching and calculated that the MTOW of the Powell is 520lbs, based on a wing loading of 6.2 lbs/sq ft. With an Empty weight of 310 lbs, that only leaves 210 lb load. If I drop my weight to 80 Kg, that only leaves enough load for 20 litres of fuel, or about 5 US gallons.

 

If all other factors are held the same I obviously need to increase the Wing Loading,

 

How do I do that?

 

Old Man Emu

 

 

[old man emu]

The jury has returned its verdict. The Powell PH Racer is out. It's too small and can't carry any realistic load. A bloke would have to be the sixe of a Melbourne Cup jockey to fly it any distance. The aerofoil would have to be changed. There's not low price (cheap) engine that's suitable, yada, yada and so forth and fifth.

 

So. What now?

 

From the same source comes the Gere Biplane. A single seater with a 19ft span and 16ft length. It was originally designed for Ford A or T engines, so it will take a VW with ease.

 

If you Google "Gere Biplane" you find the one that has been built in New South Wales. Looks like it originally flew with an horizontally opposed engine of some sort, but that has been replaced by another type you'll soon identify.

 

It was designed with square ended wings, but I don't think they look nice. I think it will be worth the effort to make a rounded end.

 

Here's some elevations. What won me was the design of the instrument panel.

 

Old Man Emu

 

Gere Elevations.pdf

 

Gere Elevations.pdf

 

Gere Elevations.pdf

[Mazda]

Enjoy the build!

 

It's about time Dave got back into "Fly Off Quick."

 

 

[old man emu]

I've got lots of thoughts about building my own plane, and I'd like to air them amongst like-minded people. I could start writing a blog, but I suspect most people on this forum just jump on, check out the newest posts, make a comment or two and then jump off to check out other websites, and so, a blog wouldn't be read as much as a Thread entry.

 

So here goes. Look forward to your responses.

 

Why do we still make ribs from little sticks?

 

How many aircraft wing plans show the ribs built up with sticks of spruce?

 

The plan for the rib for the Gere Biplane calls up 1/8 x 1/2" spruce for the caps, struts and braces. The caption to the diagram says, "The Gere rib ... follows modern construction practice." So, if you build it that way, you end up with a rib that internally has a typical truss appearance. Published in the same book, in the same year is the rib diagram for the Powell Racer which shows the rib for that plane cut from 1/16" plywood. It has lightening holes and spaces cut into it.

 

Now, which rib construction shows "modern construction practice"?

 

Both plans identify which airfoil is used in each airplane, and there is data on chord length and spar location. Both have tables which show the rib boundary offset from the chord line at set distances from a stated datum point (in both cases, the leading edge). From these figures one can produce a set of (X,Y) co-ordinates that can be used to plot the curvature of the ribs to produce a template for assembly. In fact, I can import the set of co-ordinates into a CAD program and it will draw the template at the press of a button. I can then export the template as a *dxf file to a CAM program which will create the control code so the rib can be cut by a CNC machine.

 

I can tell the CAM program to cut however many ribs I need, and if I tell it the length and width of the plywood sheet I'm using, it will nest the cuts so that the least amount of material is used to produce the ribs. Press and I'll have all my ribs cut exactly the same in about an hour or so.

 

Why should I use 1930's "modern construction practice" in 2010, when it would be easier and less time and material consuming to use 21st Century technology?

 

Old Man Emu

 

 

[Yenn]

Good on you OME

 

Reading from the start I was concerned as it didn't look to me as if what you were goung to attempt would work, but I delayed comment and found you had beaten me to it. I wasn't impressed with your idea of using conduit, at least if it is anything like the conduit I know.

 

The little itty, bitty stick ribs were made that way because they were light and strong. My Corby starlet full wing weighs less than a part completed half of an RV4s wing. Ply has been used for ribs, but it still needs a cap to provide a surface for the skin. The supercats used foam instead of ply and they were seriously light, plus strong.

 

Hava a look at the Corby, it is timber and ply with a fabric covering, light, easily powered by a Jab or similar of up to 85hp, but best of all it is a delight to fly. Unless you are over 6' or 90kg approx.

 

 

[old man emu]

I wasn't going to use electrical conduit for the fuselage of a real plane. The idea was to use it as a cheap substitute for chromemoly during the mock-up stage. The conduit only costs about $2.50 for four metres, so it's a cheap as.

 

Creating a mock-up allows me to see if the cockpit is going to be big enough for me. I can also mock-up control components and such to see how they go together. I can cut a lot of the brackets etc from plywood using my CNC machine to check fit them. It's also a lot easier to cut and birdmouth the crossmembers by using the conduit. I can label each bit and when the time comes to construct the tubing frame, I'll have a template for every piece.

 

Regarding "sticks -v- plywood ribs".

 

Making ribs from sticks is very time consuming, and each rib, no matter how much attention to detail is made, will be unique. The strength of each rib is dependent on the stength of the glued joints between the sticks and the gussets.

 

Using CNC technology, each rib will be exactly the same, and there is no reliance on surface-to-surface glue joints. As I wrote, CNC cutting will produce a wing load of ribs in a couple of hours, tops.

 

I'm just looking to use some 21st Century technology to save time.

 

Old Man Emu

 

 

[djpacro]

Unless you use very thick ply, my guess is that you'll find that the ply ribs will still need spruce capstrips and perhaps vertical stiffeners (especially to aid in attachment to the spars). The ribs carry the local airload to the spars so the ribs need to carry vertical shear and bending loads - thin ply alone is not very good at that. Pitts production airplanes changed from spruce trusses to ply some years ago. Drawings now on CAD and the ply is cut by water from memory. Worth the effort if you're making a bunch of airplanes.

 

The S2C Wing Construction - Pitts Factory Tour

 

Complete all the ribs in one hour plus time to set it all up. Plus time spent discussing it here (OK, disregard that if you like).

 

How long will it take to simply make all the ribs per the existing drawings and how much time would you really save on one aeroplane?

 

How many hours do you expect to put into building the whole aeroplane?

 

My guess is that the answer to the first question is a small number and the answer to the second question is a big number. Anyway, whichever way you go it will keep you off the streets and out of trouble.

 

 

[old man emu]

Well, compare the two types: plywood RAF15 and stick USA Airfoil No 27.

 

Don't forget, we are building a puddle jumper here, not a competition standard fully aerobatic jobby that's going to pull big G's. We don't have to build it as if it was a Mack truck. The Pitts uses 1/4x1/4 sticks in its ribs. The RAF15 rib is 1/16 ply with 1/4x1/4 cap strip. The USA No 27 calls up 1/8x1/2 material.

 

Isn't it ironic that in cross-sectional area, Pitts = 1/4x1/4 = 1/16; USA No 27 = 1/8x1/2 = 1/16, and the plywood has the same cross-sectional area, plus more material overall. I think the RAF15, and a CNC cut rib would be a lot stronger than a built up rib.

 

Old Man Emu

 

 

 

 

[old man emu]

After putting up the Christmas lights, and helping trim the tree,I finally got to do some building tasks today.

 

I started to prepare the fuselage jig by drawing upper and lower fuselage tubing layouts on a sheet of MDF. Next step is to make a mock-up of the fuselage using conduit. If teh Good Lord's willing, and the creeks don't rise, I may have the mock-up done by the end of next weekend.

 

I'm having a fellow CNC some ribs from plywood to see how that comes out. On the first run it took 4 minutes to cut one rib. That's a lot quicker than making one from sticks.

 

OME

 

 

[old man emu]

Well, the end of the weekend came and went. I managed to get some conduit placed on the MDF. I did a fuselage side truss, but didn't know how to fit the diagonal struts.

 

I picked up a copy of Construction of Tubular Steel Fuselages by Dave Russo and published by Aircraft Technical Book Company. Every aspect of making a steel tube fuselage is covered in it. Looks like it's head in a book time!

 

Here's a picture of what I've done so far.

 

OME

 

 

 

[djpacro]

Thanks OME, I don't understand your comment about the diagonals?

 

Keep us informed of progress.

 

PS - the park bench needs some attention.

 

 

[Deskpilot]

PS - the park bench needs some attention.

Never mind the bench, look at the state of the runway. Soon be overtaken by wild grasses etc.

 

 

[old man emu]

The diagonals I mean are the ones which go from the bottom longeron to the top longeron in each rectangular bay. Please ignore the white strips. These are strips of duct tape I used to cover some lines I drew which were incorrect. You should only take notice of the grey tubing.

 

And what do you nmean by "wild grasses". I mowed that lawn the day before it took the picture. Do you wnat me to stop working on the plane and spend a day digging out paspalum?

 

OME

 

 

[mechfx]

Test run of a profile I drew from scratch

 

 

 

[old man emu]

As discussed. This is a good first attempt, but we need to allow for the spars. That will probably just mean we have to reposition the lightening holes.

 

OME

 

 

[Deskpilot]

And what do you nmean by "wild grasses". I mowed that lawn the day before it took the picture. Do you wnat me to stop working on the plane and spend a day digging out paspalum?

OME

Yes, or you'll regret it later.

 

 

[old man emu]

4 Jan 2011

 

I went to MechFX today and had a session with Art. We (he mainly) drew up the rib using a CAD program. We also made some files for a false nose rib and the aileron rib. It didn't take long, really. With all the starts, stops, go backs and do agains, it only took us about two hours to get the drawing finished and saved.

 

Then we set up the CNC machine to cut some test pieces. We figure that it took about 12 minutes to cut a full rib, including internal cut outs. Bet no one can construct a rib from sticks and have it ready to fit to the spar in that time!

 

Here is a picture of the cut rib; false nose rib and aileron rib. There are some minor alterations to be made to the cuts, but that will only take five minutes.

 

Thanks heaps to Art. You've got me moving on, and I hope you profited from the experience.

 

OME

 

 

 

[old man emu]

Well! I have finally got a CAD file for my wing rib that I am happy with. I replotted the various station points using the data from the plan on a piece of MDF. Then I hit in a nail at each point. Finally, I got a length of brazing rod and bent it around the nails until it formed a nice smooth curve. I drew the shape by tracing along the rod with a pencil.

 

To get the co-ordinates necessary to draw the curve with a CAD package, I drew in an offfset base line and measured from it. The first few times, I measured in 1/8ths and parts thereof. I found that the CAD software did not like the measurements in up to 5 decimal places which you get when you have a measurement of something like 29/32. I decided to try using 1/10ths and parts thereof. This worked much better as this produces units with a maximum of 3 decimal places.

 

When I plotted the data from the 1/10ths measurements, I still had a curve with some small bumps in it. I found that if I reduced the total number of points used to draw the curve, the CAD software was happier and produced a neater curve.

 

So now I've sent the file to my CNC friend who is going to put in a line to represent the edge thickness. Then I can get it back and draw in the spar locations and internal bracing.

 

This task has made me realise why it takes so long to build a plane from scratch. You spend three of four times the time just thinking how things are going to work. The actual construction when you have all the parts at hand and techniques worked out should be less time consuming.

 

Back to the CAD, I've got bulkhead drawing to complete, then I have to design the engine mount.

 

OME

 

 

[old man emu]

Actually, the data I obtained from this attempt sucked. I found that with CAD, the more data you give, the bumpier the curves become. So I dumped all my earlier work and started at Step 1.

 

The data I had from the plan gave the distance offset of the lower surfaceof the aerofoil from a longitudinal datum line, so I marked these points on my drawing board. The data for the upper surface was in the form of measurements of the thickness of the aerofoil at each station along the datum line, so I marked these points. The plan called for cap strips 1/8" thick to make the upper and lower surfaces, so I got a bit of cap strip and used it to draw a line which was offset 1/8" outside the template at each station. Then I hammered in a fine nail at each intersection of the line perpendicular to the datum line at each station.

 

I got some fishing line and wrapped it around the shape, using the nails to keep the fishing line tight. That gave me my aerofoil shape, and allowed me to create a smooth curve by minor adjustments to the offset distance of the nails.

 

I marked in the positions of the fore and aft edges of the spars. Then I used masking tape over the fishing line to hold it in place and to make it easier to see. I drew a new datum line1" below the original datum line so that the CAD drawing would be above the X-axis, and easier to see. Then I measured the distance from the front of the rib to the important points on the rib, such as where there were changes in curvature. Although all the measurements on the plan had been given in parts of 1/8", I measured in 1/10" as this made the data more acceptable to the CAD program.

 

Finally, I entered the data into the CAD program, and. Hey Presto!, it drew the aerofoil shape. From Then I was able to draw in the spaces for the spars, and now All I have to do is draw in the lightening spaces and cut the drawing up into the sections that will be compression ribs, false nose ribs and aileron ribs.

 

Once that is all done, I just have to create teh machine code for my CNC machine to be able to cut the shapes.

 

Old Man Emu

 

 

[old man emu]

Yeah, well that was a wasted effort. I wasn't happy with the result I got from measurements off the drawing I did, so I went back to basics. One of the CAD programs I use will read a text file which contains a list of X,Y coordinates. and then draw a shape from them. So I used the data from the plans to write the file, but this time I converted the Imperial measurements to Metric before I added each one to the file.

 

I don't know why, but the curves produced by the metric measurements came out virtually spot on, with no unwanted bumps. I think that this could have been helped by the fact that I had draw the shape so often that I knew where to look for errors in thepublished data.

 

After I had my initial drawing, I saved it as a *dxf file, and opened another, simpler CAD program calles eMachineshop [ www.emachineshop.com ]. This program has few of the advanced features of the other program, but its procedures fro drawing lines and making curves are much easier for a novice CAD user. And it's free

 

After several attempts, I have finally come up with a drawing that I can use to great the G-Code for the CNC machine. One thing that I did notice was that the depth of the spars was greater in my drawing than on the original plan. I gave this some thought, and figured that if the spars were the correct distance apart, then I could locate the wing attachment lugs on the fuselage to suit the depth of the spar.

 

Since I have the drawing of the whole rib, I can do some cut and paste work to produce the drawings for the false nose ribs, the ailerons, wing tip, copmpression rib pieces and the shorter ribs in where the ailerons are.

 

OME

 

 

 

[sain]

Hi Emu, as a mater of interest are you planning on using the emachineshop service to cut the ribs out for you, or are you wanting to use your local guys for that?

 

 

[old man emu]

Given the size and weight (but mostly the weight) of a full set of ribs, it's cheaper to cut them here. Also, the drawing can be cut up into sections so I can have separate drawing files for the false nose ribs, compression ribs, end ribs, short ribs where the ailerons are, and the aileron ribs.

 

I am thinking of getting eMachineshop to do all the metal fittings since they can cut and bend. The postal charge for all the fittings would probably be about $US35. I'l have to see how much the manufacturing cost will be.

 

OME

 

 

[Yenn]

In the CAD program I use I can draw lines between all the points, then turn them into a "Polyline" and there are several ways of smoothing them. The program is Autocad, version 6 I think.