DooMaw - building a STOL

[Litespeed]

As always Beautiful Work.

 

A continuous inspiration to us all.

 

What colour for the airframe?

 

 

[Head in the clouds]

To bring the log right up to date ...

 

I spent last weekend making parts. I think I'm down to fitting the last few bits to the fuselage frame then I can get the epoxy coat on. The timing has worked out well as far as the season and temperatures are concerned. I was rather optimistic four months ago when I bought the epoxy hoping to get it applied before the winter but at least now in our gorgeous Spring weather the temperature and humidity will be ideal. The coating is Jotun Jotacote 605 high-build high solids 2 pack epoxy which I have used before on steel and aluminium boats and it provides, in my mind, unrivalled adhesion and corrosion protection, and is extremely tough i.e. abrasion resistant and flexible, so it doesn't tend to wear off in high traffic areas or crack either. Like all epoxies it's prone to surface powdering/chalking if exposed to UV over long periods but the airframe is fabric and sheet-metal covered so that won't be an issue. Any areas that might be subject to constant exposure should/will be overcoated with a polyurethane coating.

 

As far as the colour is concerned ... rather boring I'm afraid, I chose white. It's a little cheaper than the colours but that wasn't the reason. White shows up any defects better than colours. In event that you did get a weld crack or some epoxy coating damage, or corrosion under the coating for any reason, the associated rust will show up very quickly whereas on black, red or brown particularly, it takes a very close inspection to reveal any imperfections.

 

Though I've learned a little about the coatings themselves, I'm far from being an expert on applying them, so if there are any folks out there knowledgeable about spray-application I'd be delighted to hear about tips and tricks. For the record I've bought a couple of new spray guns, really small gravity-fed HVLP ones with 0.5, 0.8 and 1 mm nozzles and 60cc and 100cc cups. I don't have a spray booth or any reasonable method of making one so I chose the tiny guns to be able to work closely on each tube and avoid as much overspray as possible.

 

Back to the weekend of making parts -

 

As described in my previous post, I had to work out the geometry for the aileron and flap connections, so I also made the templates for the bellcranks, control horns and the flap detente plate, marked them out on CRMO sheetmetal and started cutting them out and drilling them.

 

Earlier in the week I'd been up to Brisbane for a meeting as part of my day job and for some unexplained reason I decided to take a diversion on the way home and return via my current favourite shop. I didn't have anything specific to buy, I just fancied a browse and I always get some inspiration when I have a look around that place. The shop is called Miniature Bearings Australia, and is an absolute Aladdin's Cave for modeller's, homebuilders and machine engineers. They stock everything from RC car and plane parts to bearings and linkages of every kind, small gearboxes, those bushings that you can't find anywhere, phosphor bronze rod (up to 3" diameter!) for making your own, balsa, carbon fibre, foam sheets, adhesives, wheels, ground silver steel shafting, keysteel of every size, you name it. (Click the link and just keep scrolling and scrolling down, you'll be amazed at their range of products that you just can't find anywhere else). They have a very efficient online shop too, for those unfortunate enough not to live nearby, and their online prices are less than their in-store prices, which more than makes up for the cost of the postage.

 

Anyway, I digress yet again ... probably the only part of the DooMaw project I hadn't either worked out in CAD or had a good picture of it in my head, was the instant connect/disconnect method for the aileron and flap controls when the wings are folded or unfolded. It had to be foolproof i.e. must engage correctly and properly indexed without fail, and most importantly must not be 'sloppy'. I hate 'lost motion' (slack) in the control system, so the coupling method needed to be very precise without introducing tightness or stiffness, which I hate even more. Additionally, the coupling must permit a small amount of off-axis/angle drive, in this case it's the amount of the dihedral angle of the wings. So the coupling method was still playing on my subconscious but I was confident that the answer would present eventually.

 

As it turned out the visit to Miniature Bearings was the catalyst. As soon as I saw the Dogbone Couplings used to transmit shaft drive in RC buggies and helicopters I realised I could adopt that method for the aileron and flap torque-tubes. A couple of pics shamelessly lifted from the net, thank you hot-racing.com and the other contributor -

 

 

I worked out that I would need the ball part to be 7/8" (0.875"/22.23mm) diameter, to engage nicely with the 0.884" internal diameter of the CRMO end of the outboard torque-tubes i.e. with 0.009" (9 thou) clearance. The last time I did any ball-turning on my lathe was quite a while ago and I recalled that the set-up took me a day or two and to turn such small balls integral with the shafts would be fiddly and I'd need a fourth axis slider that I don't presently have. I quickly also gave up the idea of cutting or grinding them freehand to a profile template - if there'd been just one to do I would have, but four would take too long.

 

Instead I decided to fabricate the dogbones from several parts and then spent a whole day searching the local bearing suppliers for self-aligning bearings and/or rod-ends that might have had pre-drilled balls of the right diameter. The local suppliers were hopeless - no stock, just catalogues - so I drove an hour each way to a supplier that I know has a good display stand of them. Vernier in hand I set about measuring their balls (so to speak) and was about to give up when I looked on the other stand and, amazingly, discovered that their Metric rod-ends had Imperial-sized balls and the Imperial rod-ends had Metric-sized balls. I decided that quite clearly ball size has nothing to do with regional origins ... Anyway I found that the 12mm rod-ends had 7/8" balls and that was all that mattered, I could make the cross-pins any size that suited the hole through the ball.

 

$80 might seem a lot for four small balls but compared to the effort to make and drill them I saw it as a bargain. I cut one rod-end apart and got to work to make the rest of the parts. By the end of the weekend I have them all ready to anneal the balls and weld everything together. See the photo captions for more info -

 

Aileron bellcrank and control horns being cut out - paper template and spray paint method to make the first, then scribe around it to make the rest

 

 

Flap detente plate - a pin on the sideways-sprung flap lever will engage in any of the six medium sized holes arranged in an arc on the lower part of the plate, to select the six flap positions. The flap torque tube goes through a bronze bushing which will be installed in the almost-cut-out large hole at the top, which is the centre of the arc of the detente holes below. The smallest holes are to accept 1/8 rivets which will secure a PTFE/Teflon plate on the surface so that the detente pin doesn't make unpleasant scraping while changing flap positions. The top large hole wasn't completely cut through so that I can use the remaining 1/4" hole to align the plate with the outer holes in the same way as shown in the previous post for the aileron torque-tube. After the plate is installed I can complete the cut-out with the hole-saw in a hand-drill. The three larger holes serve two purposes, lightening of course, and also allow me to spray paint through them onto the cabin-top side plate which is very close outboard of it, I'll need similar holes in that plate to paint the back of this one. Second pic shows the bronze bushing to fit the top hole once it's fully cut out.

 

 

The rod-ends, one has been cut to release the ball. Second and third pics shows the parts ready for welding to fabricate the dogbone.

 

 

Bushings and torque-tube end fittings which will later be fitted to the inner ends of the aileron torque-tubes.

 

 

Another 17hrs making a total of 1263hrs so far.

 

 

 

 

 

 

[Head in the clouds]

September 17th 2016 -

 

Yesterday was a good day. As I mentioned earlier the auto-connect/disconnect system for the ailerons and flaps when the wings are folded and unfolded was the last matter still playing on my mind, until last week.

 

In the previous post I described the 'Dogbone Method' and the machining of the parts for it last weekend.

 

The last remaining minor concerns I had were the effect welding would have on the metallurgy of the steel that ball bearings are made from, and whether I would be able to make good enough welds on such small components, while also keeping the parts perfectly aligned so that the assembled items ended up within close tolerances. The crossbars needed to be quite perfectly square to the shaft.

 

To satisfy myself about the issue of welding 4130 CRMO tube to the ball-bearing I decided to make a test piece first. Previously I mentioned that I would anneal the drilled balls prior to welding them but having given it further thought there would be little point as they would reach cherry red during the welding, so it would be the rate of cooling that would affect the hardness and temper of the completed part. The concern I had was that the balls and the CRMO are both high carbon steels and if they cooled too quickly, and with the (probably) fairly high chromium content in the steel the balls are made from, the welded joint could exhibit brittleness because 4130 doesn't like to be welded to high chromium content steels, especially stainless steel. The degree of brittleness I'd end up with was a complete unknown, though I had already determined that these balls are not stainless.

 

So - among my 'things' I found a ball-bearing, not as large as the ones to be used for the dogbones, but large enough for the purpose, and I welded it to a piece of 3/8" CRMO tubing. I was careful to allow it to cool in still air. The steel the ball was made from was quite different to weld from welding CRMO, it welded more like mild steel, but didn't show any noticeable reluctance to meld in the puddle.

 

Once it had cooled I put on a face shield in case it shattered under the crude test methods I had in mind, and used the part like a club to bash a large steel block I use as an anvil. I was actually very surprised that it didn't do anything untoward at all, I did quite expect the weld to break, or the 3/8 tubing to break at the HAZ (heat affected zone on the outer edge of the weld). I welded on an extension to the length of the tubing and gave it some more big swings, as you would with a golf club and the stupid little white ball ... and it still didn't break. Next I put the ball in the vise and whacked the tube with a hammer, and then the other way around with the tube in the vise and whacked the ball. In that last test I could have broken the tube if I hit it hard enough of course, but I wasn't trying to test the tube, just the weld.

 

Examining the ball I could see that the impacts between the ball and steel block had been hard enough to cause quite large indentations in the ball. So the force was quite high which in my mind gave the test reasonable validity, certainly enough for me to be confident that the ball-to-4130 weld is many times stronger than it will ever experience in service as a coupling. After all, it only has to be stronger than the weakest component in the system, which is the 7/8" x 0.049" wall 6061T6 aly tubing that the torque-tubes are made from.

 

Having done that I decided to make another test, just as a matter of interest. I reheated the welded ball and tube to cherry red with a blowtorch, and dropped it in a bucket of water. That would have shock-cooled it and left the metals in their hardest possible state, and without tempering them afterwards they would also be at their most brittle, theoretically ready to shatter with little more than a firm clout of a hammer. Then I adopted the face shield again and swung it against the 'anvil' again, put it in the vise and belted it and so on, and it still held up. That rather makes you wonder about all the brouhaha about making sure there isn't even a zephyr of breeze in the shop while welding the airframe, perhaps that's meant for folk who are welding up an airframe in the Alaskan winter, rather than those of us who've sweltered through our tropical summers making one.

 

Anyway, thus emboldened I set to work to weld the cross-pins into the balls. The first one took a fair bit of concentration but they got easier after that, especially once I made a little set-up to hold them still on the bench.

 

The second stage of the process required a jig of some kind to hold the balls and the stem so that the stem would be exactly at right angles to the cross-pins. Even though it's only a Taiwanese cheapy, my pillar drill has a table that is adjustable in all axes and the vertical motion of the lockable quill was the obvious way to clamp it all together. In the lathe I made a small mandrel from some aly rod to fit one end into the chuck and the other end slip into the stem. Using a small engineer's square I adjusted the table to make sure it was perfectly at right angles to the quill, then sat the cross-pins on a pair of parallel blocks and used the quill to clamp the stem down onto the ball. I then put four small tack welds around the contact points and did the same for the other three balls.

 

Then it was just a case of going back to the bench and completing the welding.

 

Finally I had to attach the flange to the other end of the stem and since I had machined a close-tolerance spigot on that end it was just a case of using the pillar drill and the centre-drill indent in the end of the mandrel to clamp the ball-and-stem firmly down onto the flange and tack it in four places around the perimeter, similar to the other end - then back to the bench to weld those out too.

 

Pictures -

 

Test piece - the second close-up pic shows the indents in the ball after the testing

 

 

Welding the cross-pins in

 

 

Setting the table square to the quill using an engineer's square

 

 

The set-ups to get the cross-pins and stem square to each other

 

 

Completed Dogbones ready to weld into the ends of the torque-tubes

 

 

Another 8hrs in the log, 1271hrs so far

 

 

 

 

 

[Kyle Communications]

AWESOME as we have come to expect from you

 

 

[biggles]

AWESOME as we have come to expect from you

Yep , he's a living legend ! ..... Bob

 

 

[Head in the clouds]

September 18th 2016.

 

Thanks for the kind words fellas

 

A notable milestone was reached yesterday when I completed all the welding on the main part of the fuselage frame. That means I can start the epoxy coating as soon as we have a fine day with low humidity. It will be a pleasant relief to know that the steelwork is sealed away from moisture and I won't have to be so vigilant about rust protection.

 

More to the point though, is that once the frame is painted I can begin the final assembly of all the other components I have built along the way. The floors can go in, then the adjustable pedal assemblies, the control column, instrument panel, control cables with their associated pulleys, bellcranks and walking beams. It will really begin to feel like I'm over the hump and coasting downhill by then.

 

Then, eventually, after I build a new carport for the 4WD so that DooMaw can take over the garage/workshop, I will be able to sit it on it's real gear legs instead of the temporary dolly wheels/legs it's on now. It has to be on these wheels for now because the real landing gear is too tall to allow the fuselage to be stored under the house when I'm not working on it.

 

The next main structure I'm now finalising is the engine mount. Those of you who have built your own plane from scratch or from kits might be able to contribute here - how much side-thrust and down-thrust do your engines have? Each design is different of course, the amount of side thrust is determined by the power of the engine, the length of the fuselage (that's over-simplifying it but is close enough for our purposes), and most particularly the ratio of fin/rudder area above the fuselage longitudinal datum compared with the amount of area below it (again, in simple terms).

 

Most of our planes are fairly conventional in their design, so we can make reasonable comparisons between them. At 100hp and with the fuselage length I have and with the approximately double the fin/rudder area above the datum, it would appear that I need between 2-3 degrees of right thrust. I moved the engine mount around in CAD and when incorporating 3 degrees it looks an awful lot. It's one of those things when building the engine mount though, that is quite important to get right or at least very close, because it's not easy to fix or adjust later. You can, of course, just add packing washers where the mount bolts to the firewall or at the mounting rubber locations but that moves the front of the engine sideways and then the propeller hub would be off-centre. Consequently the mount must be initially set-up so that the engine is rotated at the hub, not at the back of the engine where the rubbers are.

 

Since the 3 degrees looks too much I reduced it a little, to 2.5 degrees and I can certainly adjust half a degree at the mounting rubbers later if I need to, without unduly upsetting the hub position in relation to the centreline of the cowling. But I would be very interested to hear from those of you who know at what angle your engines are set.

 

I've also incorporated 1 degree of down-thrust to compensate for some of the nose's tendency to rise due to increased airspeed as the throttle is increased. I favour that tendency anyway in a plane of this kind, because when you increase throttle it's usually with an intention of climbing rather than just going faster - in a Spitfire it would be a different matter altogether ... (love your new avatar Marty )

 

So - yesterday's progress - first I added a couple of tabs to which I can later attach P clips and/or saddles to hold the throttle, choke, carby heat and cabin heat cables so that they don't flop around in the foot-well. Next I pressed a bend into the bottom of the flap controller plate I'd made previously, so that it sat flat against the cabin-top side rails that it was to be welded to. I drilled the centre holes in the cabin-top side plates in the location of the flap torque-tube and used them to thread a long 1/4" steel rod through, to position the flap controller plate in the correct alignment, clamped the controller plate and tacked it in position.

 

After I'd tacked it in four places I found that it had moved! It was only about half a millimetre but it was enough to make the 1/4" rod tight in the three holes it passed through. The locations of the tacks were quite inaccessible even for a small grinder so the only option for removing the plate was to break it away from the tacks and since the tacks were quite strong it would mean substantial damage to the plate and probably having to re-make it.

 

Since it was such a small amount out of position, I decided that with careful use of the welding order I would be able to make the natural shrinkage of the welds pull it back into alignment, so I changed the usual welding order so that I would complete all the welding on one side first, rather than doing some on one side then some on the other side, which you would normally do to prevent it pulling out of line. The process worked well and pulled the holes back into alignment.

 

I also used a small holesaw to cut three access holes in the sides of the cabin-top plates so that I could get paint to the inner face of the flap controller plate, those holes will serve as inspection holes later and need only be covered with 100mph tape, since they will also be covered by the inner rib of the wing-root.

 

A few pics -

 

A couple of tabs for 'P' or saddle clamps to hold the cables

 

 

Flap controller plate

 

 

fuselage welding finished and ready for epoxy coating

 

 

 

 

 

 

[Head in the clouds]

Calculations ...

 

It's about time for me to organise another couple of orders from the US. I need a lot of fasteners, a couple of instruments, covering materials, some bearings and the airfoil shaped tubing for the struts from AS&S (Aircraft Spruce and Specialities), and from FBI (Freebird Innovations) I need the toe-brake cylinders and some other brake parts. Mostly it's just a matter of making lists and trying not to miss anything out because the freight costs are quite high, so you don't want to be having to make another order for a couple of small parts you didn't think of in time.

 

The struts are the main consideration just now and there are many sizes (and prices) available. Originally I was going to use aluminium extrusions but the folding wing mechanism on DooMaw is quite complex and if I use aly sections for the struts then the end fittings have to be bolted in, and with the folding mechanism it would make the strut ends quite bulky. By using 4130N chromoly I can weld the ends in and make them far more compact.

 

If one had a completely free hand in the design of an aircraft, one would calculate the desired characteristics of the strut and then have the extrusion made to suit. That might work if you were planning on building a few hundred or thousand similar aircraft but it certainly isn't feasible for a one-off design, so there's no point calculating the ideal strut characteristic, instead I need to choose one of the available sections and calculate how it would perform in this particular application. If it proves to be way too strong i.e. over-engineered then I'd calculate the next size (or two) down, or v.v.

 

The other major factor to consider is the huge price differential between almost identical sections. For example (chromoly comes in imperial measurements) 1.99"x 0.87"x0.049" streamline 4130N chromoly tubing (that's about 2in/50mmx 7/8in/23mmx 49thou/1.2mm - the last size is the wall thickness) is about US$15/AU$21 per foot/300mm. Whereas 2.023"x0.857"x0.049", which is virtually identical in size and strength properties, is US$21/ft or AU$30 per foot/300mm. At 300kts the extremely slightly narrower latter one might make a tiny difference in drag but at 60kts or so there wouldn't be any difference at all between the two, except the narrower wallet during the build ...

 

The sizes discussed above are in the mid-range of the available sizes and I had a reasonable idea that they would be in the ballpark of my requirement, and the US$15/ft one is also the most popular and hence best value for money, so my calculation just needed to make sure it was sufficient for the need. At first look it appears really quite small, we're all used to seeing the struts on our sport two seaters being about twice the size of these i.e. more like 4-5"/100-125mm front-to-back and perhaps close to 1.5"/38mm thick. There are two considerations there - most struts we see are aly extrusions and the 4130 is way stronger - particularly in its resistance to buckling under negative G - and most struts on our sport planes are probably quite a bit oversize so they 'look right' to keep the customers happy. My justification for suggesting that is no more than a simple comparison between the struts on a C172 with a MTOW of around 1000kg and the struts on an LSA with a MTOW of 600kg, the LSA struts are sometimes larger than the Cessna's!

 

So - first the positive G calculation - I'll do it all in imperial because I'm more used to doing it that way but anyone who might want to use metric units for their own project would just substitute the equivalent metric values into the same formula.

 

Where planes are concerned I don't try and calculate everything exactly, not only is it rather tedious but there are many variables, so instead of being totally anal and trying to get to the lowest possible value for each and every part, I deliberately err on the side of safety and round everything up rather than down, so -

 

First we need to know the load in the strut during level (un-accelerated) flight. DooMaw's struts are attached at the 60% outboard point of the wing i.e. there is 60% of the wing between the fuselage and the strut point, then the remaining 40% out to the wingtip is cantilevered out from the strut point. A very simple method to determine the proportions of the load being carried by the strut, and by the wing/fuselage attachment point, is by considering what would happen with a free-wing. If the strut was attached at the 50% span point then the wing would fly equally balanced on both sides, so it would fly level. As we move the attach point outboard, the inboard portion of the wing produces more lift than the outboard portion so the inboard portion would start to fly higher. By connecting that inboard end to the fuselage it allows the fuselage to take some of the lift load and keep the wing level.

 

So - if we have 40% of the wing outboard of the strut we can reasonably say that the strut is carrying that 40% of the total wing lift load plus another 40% of it inboard of the strut, which leaves 20% of it being carried by the fuselage, so the strut is carrying 80% of the wing's total lifting load. Then - the strut is at an angle of 60° and just like the G loading at 60° bank angle, the load in a strut is doubled if it has to support the wing when attached at that angle - and that also imparts a compression loading in the wingspar, but we'll deal with the wing components separately later.

 

Just in case we eventually get an increase to the weight restriction I'll calculate this lot for more than the current limitations, and perhaps a bit more than the suggested 750kg, who knows, sometime in the distant future someone might own this plane and convert it to VH category and want to put an 0-360 in it, so let's work on say, 800kg, and see how that pans out.

 

Say the all-up weight was 800kg, then one wing would be carrying 400kg. Well not quite actually, because the wing itself would weigh at least 20kg, so the wing is actually carrying its own 20kg plus 380kg, and the strut is carrying 80% of that, which is 304kg, but we have to double that due to the 60° strut angle, so at 1G the strut has a loading of 608kg which is 1338lbs.

 

We know from a manufacturer-supplied table that the 4130 streamline tubing size I described above is equivalent to a 1.5" round tube, so we can calculate the circumference using the 2πr (or πd) formula i.e. 3.142x1.5=4.713". Then we multiply the circumference by the thickness of the wall to arrive at the cross-sectional area of the material in the tube i.e 4.713x0.049=0.2309in². From Materials Properties data for 4130N steel we can find that chromoly has a tensile yield strength of 460MPa and being an old-schooler I convert that to psi (1MPa = 145psi) so the yield strength is 460x145=66,700psi. In our strut we have 0.2309in² of cross-sectional area so the strut will have a yield strength of 0.2309x66,700lbs = 15401lbs.

 

From the earlier calculation we know that at 1G the strut is loaded at 1338lbs and it won't begin to yield (stretch) until 15401lbs so to determine how many Gs it is capable of we divide its yield strength by the load at 1G i.e. 15401/1338=11.51G. So even if we very conservatively said the strut was carrying all the lift load, and the fuselage not carrying any of it, the strut would still have a positive G capability of 80% of 11.51G = +9.2G which is a safety margin of 50% above the +6G service this aircraft is designed to.

 

And - remembering that's at 800kg MTOW.

 

Next post I'll detail the negative G condition which is a bit more complicated.

 

 

[Head in the clouds]

Strut buckling calculation -

 

A strut in compression i.e. under negative G loading would normally fail as a result of structural buckling rather than compression failure of the metal itself.

 

There are various formulae used for calculation column buckling, I've used the Euler Column Buckling Formula -

 

F=nπ²EI/L²

 

 

 

Where -

 

 

 

F = allowable load

 

n = a factor accounting for the column end conditions

 

E = the modulus of elasticity of the column material

 

L = the unsupported length of the column

 

I = the moment of inertia

 

 

Among the above the unknown which we want to find out is the allowable load 'F'.

 

Regarding 'n', the column end conditions, in DooMaw's case for a buckling calculation each strut needs to be considered to be two struts (or columns) because DooMaw will have jury struts supporting the centre of the strut so each half of the strut needs to be calculated separately. Without jury struts the long slender strut would buckle very easily under negative G loading.

 

There are various end conditions to consider and each of them affect the way, and how easily, the column will buckle so a factor 'n' is applied to the rest of the formula to account for the way the ends of the column are supported. For examples, if both ends of the column are pivoted (pin jointed), as most struts are, then the factor is 1. If one end of the column is held rigid (embedded in concrete perhaps) and the other end is pivoted, you can imagine that the stiffness at one end would increase the column's resistance to buckling, so the factor is 2. If both ends are held rigid (like a steel column embedded in concrete at both ends, between floors of a high-rise building perhaps) then the stiffness provided at both ends increases the resistance to buckling even more, so the factor is 4. Another condition might be where the column is held rigidly at one end but not supported at the other end at all (like a flagpole embedded in concrete and supporting a heavy weight sitting on the top), you can imagine that a column in that condition might buckle very easily, so the factor is 0.25 ... and so on.

 

In DooMaw's case, each end of the strut is pivoted but the centre of the strut is held rigid by the jury struts, so we can consider each end separately as columns that are half the length of the complete strut and where one end is held rigid and the other is pivoted, so the factor we use is 2.

 

E - The more elastic the material of the column, the more easily it will buckle. You can imagine that a rubber column would buckle more easily than a steel one of the same dimensions. The modulus of elasticity for any material is found in its Materials Properties data. In the case of 4130N chromoly it is 205GPa/29,700ksi (i.e. 205,000MPa/29,700,000psi). As a matter for comparison the modulus of elasticity for 6061T6 aluminium, of which many LSA struts are made, is 68.9GPa/10,000ksi which is only about a third of that of chromoly i.e. aly is three times more elastic than chromoly, so you can immediately see why aluminium struts need to be so much larger than chromoly struts.

 

L - DooMaw's struts are 8ft (96") long but they are supported in the middle by the jury struts, so the unsupported length for the formula is 48".

 

I - The moment of inertia is determined by another formula and is affected by the diameter of the column and the thickness of the wall (assuming the column is tubular). If the column is round we only need to calculate its buckling in one direction, if it is ovaloid it would normally buckle in the direction of the narrower (minor) axis but we would need to take into account the end conditions i.e. it might be pivoted in the direction of the major axis but rigid in the direction of the minor axis. In DooMaw's case it is pivoted in the direction of the minor axis which is the worst case, so we calculate it that way.

 

As you can see from the image posted below, the streamline-shaped strut I am intending to use has similar value buckling characteristics to that of a 1" diameter round tube of the same wall thickness, so we use the dimensions of a 1" round tube for the moment of inertia formula -

 

I = π(do^4 - di^4)/64

 

 

 

Where -

 

 

 

do is the outside diameter of the column

 

di is the internal diameter of the column

 

 

 

Note - do^4 means the outside diameter to the fourth power i.e. do x do x do x do, similar for di^4 (my keyboard allows second and third power do² do³ but I can't find the alt code for fourth power ... anyone?)

 

 

So the moment of inertia of DooMaw's strut is 3.142(1 - 0.66195)/64 = 0.016596

 

We can now apply all the variables to the Euler column buckling formula -

 

F = 2 x 3.142² x 29,700,000 x 0.016596/48² and that would give us the allowable load in lbs

 

= 9732003/2304

 

=4224lbs

 

From the previous post we recall that with MTOW 800kg at1G the strut is loaded with 1338lbs, so to determine the G at which the strut will buckle we divide the allowable load by the load at 1G = 4224/1338 = 3.15G.

 

My C172 in the utility category was rated at +4G -2G, so with the applicable safety factor of 1.5 it would have been capable of around +6G and -3G at yield. From the above two posts we can see that the chosen strut size for DooMaw will exceed the C172 G capability by about 50% in positive G and is around the same as the C172 in negative G. We don't hear of Cessnas losing their wings so I'm quite satisfied that the chosen strut size will do the job admirably.

 

 

 

[Head in the clouds]

September 24th - October 2nd 2016.

 

Before starting the epoxy coating of the airframe I wanted to drill the holes in the elevator ribs which will attach the elevator trim tab hinges. I'd intended to have a trim tab attached to the elevator trailing edge. Having drilled the holes I realised that it was going to be very difficult to have a neat arrangement for the push-pull trim-tab control cable due to the need to curve it along the elevator hinge-line then immediately curve it back again so that it then followed the horizontal stabiliser folding hinge-line (for when the stabs are folded up).

 

So - I decided I'd be better off with an inset trim-tab and run the push-pull cable along the inboard edge of it next to the rudder, then the cable could flex with the operation of the elevator, and would already be in the line of the hinge for the stab folding up.

 

As you can see, I made the trim tab quite large, and that's for two reasons. Firstly, to ensure it has plenty of authority to control the elevators at the low speeds DooMaw should be capable of, and secondly, so that only a small deflection angle is required because the more you deflect the tab down, for example, to trim the nose up, the more the tab detracts from the elevator's own power because the tab is providing a reflex to the airfoil shape, so the less deflection the tab has, the less it reduces the elevator's own effectiveness.

 

I cut the tab out of the aft portion of the port elevator, made up bushings and rear/front spars for the elevator and the tab and welded it all back together again, including using a piece of copper strip as a backing while welding up the holes I'd drilled earlier.

 

As part of the fail-safe design of DooMaw all control surfaces will be 100% mass-balanced, so the trim tab has its own mass balance as you can see from the 1/4" tube projecting forward under the elevator. It will have a torpedo shaped piece of lead of approx 100g cast onto it.

 

Then the elevators themselves will have mass balances attached to the aerodynamic balance forward of their hinge line.

 

These balance weights are not only for flutter resistance. The trim-tab would probably be well within its flutter excitation speed at normal cruise should it have a control disconnect so the tab's balance weight is crucial but it's probable that the flutter speed of the elevators would be above the Vne, considering they have aero-balance tabs. So - the reason for the balance weights on the elevators is to ensure that in event of an elevator control disconnect, for whatever reason, the elevators can be controlled throughout the speed range by the trim tab. It's all very well being able to control the elevators via trim at cruise speed but as you slow down for a landing the weight of the elevators makes them hang down if they aren't mass balanced i.e. induces 'down elevator', so as you slow down the nose keeps dropping and that requires more trim which in turn is less effective at lower speed and so on. Consequently, if your elevators aren't mass balanced and you need to make a landing using trim only then you're forced to make a very fast landing. The mass balances will probably be about 2lbs/1kg each but well worth the weight penalty for the added safety, in my mind.

 

So today's project is starting on installing those mass balance weights, time to make molds and start melting lead.

 

Some pics to help tell the story -

 

 

Another 19hrs in that, 1297hrs in total so far.

 

 

 

 

 

[Head in the clouds]

October 3rd 2016.

 

Yesterday I made the mass balance for the trim tab.

 

I balanced the control surface on its hinge line and added test weights to find out how much weight was required, then added a nominal 10% to account for the weight of the fabric covering. Once I knew the required weight I worked out what volume it would be and made up a torpedo shaped template so that I could picture the size and shape that the weight would be.

 

It isn't critical that the weight be exact, within 10% or so is quite close enough, so I then turned a free-hand shape out of acetal on the lathe and cleaned it up with a single-cut file to remove small imperfections and give it a good finish. I drilled and tapped the narrow end so that I could use a 3/16 machine screw to hold it in place while making the mold.

 

Next I turned up a tapered piece of PVC, this would be used to form a gallery into the mold to pour the lead into the cavity.

 

I mounted the acetal and PVC parts to a piece of 2mm aly sheet and then went in search of a suitable container to make the mold in. Those small disposable plastic cups that are used to hold water from a water chiller would have been ideal but I didn't have any so I found a small 'tupperware' which was the right size.

 

I forgot to mention, I was making the mold from silicone, as it's so quick and convenient for one-off molding of lead. Provided you don't overheat the lead, silicone easily manages the heat of molten lead. Lead melts at 330°C and most documentation has silicone melting at around 300°C but it copes with higher temperatures for short periods, so if the lead is 'just melted' and then poured quite quickly so that it doesn't freeze up during the pour it works fine. One caution though, the relatively high temperature does cause the silicone to give off some vapour which affects the surface finish of the lead, so if you want fine moldings it's not the best way to go.

 

If you wanted a better surface finish you'd need to go down the Plaster of Paris road which is rather more arduous because you can't pull the mold off the positive to leave an empty cavity so you have to make an indexed split mold or use the lost wax process, and make sure you bake the plaster mold in a low temperature oven for a few hours to dry it completely before pouring the lead or the heat from the lead will boil any moisture in the plaster and the steam will eject the lead violently and/or explode the mold, either way probably spraying you with liquid lead and causing horrific burns ... beware!

 

I mixed up a quantity of platinum cure 2 part silicone, filled the plastic container and immersed the mounted plastic parts in the silicone. The silicone takes about half an hour to cure sufficiently to de-mold the plastic parts. Before mixing the silicone I had sprayed all the parts including the container with Ease Release mold releasing agent, so to get the mold out of the container I just had to put the nozzle of an airline between the side of the silicone and the container and blow it out. Once the silicone was no longer constrained by the container it was simple to just pull the plastic parts out of the elastic mold.

 

I then created a set-up with the mass-balance support tube dangling into the hole in the top of the silicone mold, heated some lead in an old steel spoon using a gas ring below and a small blowtorch above, and poured the molten lead into the gallery created by the tapered PVC piece.

 

Once the lead had cooled sufficiently I broke the lead sprue off at its narrowest point and pulled the silicone mold off the mass-balance. The last photo shows the control surface now balancing level. After that photo was taken I lightly dressed the surface of the lead with a fine rasp and it's ready for painting -

 

 

Six hours of fiddling about, a total of 1303 hours so far.

 

 

 

 

 

 

 

 

 

 

[Head in the clouds]

It's been a long break since the last update, an international family reunion got in the way ...

 

I did manage to fit in a bit on the plane between other things though.

 

It was time to get the seating sorted out because it affects the positioning of other things like pedals, trim controls, length and shape of the flap handle and so on.

 

The seat base is made from PVC coated polyester tarpaulin material, much like truck tarpaulins. It's supported off the two cabin cross-members and the strut carry-through, and is configured much like a spring-style folding chair. I like that method because it's relatively light and by adjusting the lacing it allows lots of adjustment in height and backrest angle, and it also serves well for my crashworthiness design requirements because the base can 'give' substantially and progressively into the space below which is filled with impact absorbing foam.

 

Additionally, the thickness of the seat cushion and backrest cushion could be varied for taller and shorter people, though that shouldn't be necessary except for very tall or very short people because the rudder pedals also adjust through a range of 250mm.

 

My biggest issue was finding someone who could sew the thick materials for me because I got rid of my sailmaking machine quite a few years ago. That aspect proved to be too hard because there was inevitably going to need to be many trips to and fro for measuring and trimming, to get the fit just right. Consequently I bit the bullet and bought another industrial machine, this time with a walking foot, so even eight layers of heavy tarpaulin is now no hurdle. In fact I found it can sew through 6mm marine ply and polyethylene (kitchen cutting board material) with no hesitation. So - having also bought a beach cat for a bit of fun during the reunion, I'll make a new mainsail for that too - that's what the sewing the polyethylene board is about, to make the new headboard on the sail ... but I digress.

 

I already had a roll of the PVC tarpaulin material so that was just a matter of measuring, cutting out and stitching. The whole base is laced into place to tension it, the straight hems have a 6mm stainless steel rod inserted into them to lace around, and the seat base and back have D rings sewn to them using a light webbing strap in V configuration to spread the load across the fabric. 3.5mm starter cord makes a strong and durable lacing cord.

 

For the seat cushions and upholstery I found a shaped seat back foam billet with headrest and lumbar support at Clark rubber for $50 each and cut them to width and shaped the sides of the headrests with an electric carving knife. The seat bases foam is premium high density grade which should last for many years without softening and breaking down, but quality foam price has certainly jumped at $160 per square metre for 75mm thickness. I needed about $100 worth for the two seat bases.

 

I found some fabric I liked at Spotlight, the colour should complement the intended airframe colour scheme and not show marks too much, yes being brown I'm sure I have an idea what some of the comments might be ... it's a heavy duty faux swede so that also helps to keep it looking reasonable even if it gets a bit roughed up, it's a bush plane after all. 4m of the fabric set me back $60 because it was on sale at 50% discount and 4m of 50mm velcro to hold it securely to the seat base was also $60. The starter cord was another $20 and the light webbing was $30, the D rings were $36, so even DIY seats don't come cheap if you want them comfortable, at a total of over $450 if you include the tarp material.

 

I'm very pleased with the result though, they're exceptionally comfortable, the main reason being the shaped seat back which provides proper support, and of course the headrests are essential to prevent neck injury in event of the unthinkable.

 

The pictures tell the rest of the story -

 

 

Not including the shopping for materials that's another 34hrs for the log, a total of 1337hrs so far.

 

 

 

 

 

 

 

[Head in the clouds]

Before you know it a couple of months has gone by ... a friend pointed out yesterday that he hadn't seen any updates in the DooMaw log for quite a while.

 

I have been busy though and taking a few photos along the way.

 

I've had a couple of bandsaws for decades, one is the small tilt-up-and-down table and drop-saw type that are plagued with band-shedding problems unless you re-engineer them every few years, so I got onto that and it should perform OK again for the next while. I also have a large upright bandsaw that's sheer magic for cutting larger stuff or ripping up sheet material. The problem I've had with it is that work has had me moving around fairly regularly and I've only been able to use it when I've had 3 phase power, which hasn't been all that often. I've been intending to convert it to single phase for ages, it's only 1hp, so not a big issue ... or so I thought.

 

I bought a nice shiny new single phase Chinese motor for about the price of a round of drinks, and set to work to remove the old motor. It was deeply buried in the bowels of the machine of course, which is in a shipping container in the sun in the back yard and in the middle of a series of heat-waves. None of the bolts was either accessible nor had been undone since it was made in the 1950s, so I lost a few kilos in sweat but prevailed in the end. I was then able to refurbish the mounting plate and belt tensioning system, and did the same for the back-gear arrangement while I was about it.

 

Naturally the two motors' shafts were different sizes so I had to machine the drive pulley and I had misgivings about that because unless you get the bore right to within a couple of tenths of a thou a single phase motor will rattle the pulley and mount apart in a few seconds with the accompanying noise of a machine gun. My lathe is very small and the four stage pulley quite large so it (I) was struggling and you guessed it, I went oversize by a couple of thou. Then I had to devise a tool to broach a new keyway, I haven't done a lot of broaching so I didn't have much success with the first two tools ... I rewired the machine while I was about it, and then came the time to test run it. Sure enough - machine gun - why, oh why do they make parallel shafts? Tapered shafts work so much better!

 

I finally solved the problem by providing the pulley with seven grub screws to lock it up and control the harmonics, got some new blades made up, greased the bearings and adjusted the blade guides and it's like a new one.

 

I forgot to mention that all of this was in preparation for cutting up the material I had ordered to make the wing spar caps. The cap material is 6061T6 aluminium and its raw form comes as a rectangular hollow section (RHS) but I want an angle section which will also be tapered toward the wingtips, so I need to cut the RHS lengthwise into four, to create the angle pieces. I could buy the angle from an aviation supplier at about $60/m and I'd need about 18m so it would cost a bit over $1000 and I'd still need the bandsaw to taper the ends. By doing it this way the piece of RHS cost $90 and I'll have about 1/3 of the 6.5m length left over ...

 

I'd no sooner had that job under control than I finally came across a small milling machine which I have been seeking for several years since I got rid of the previous one which was too large to move around regularly, it weighed about 900kg. Actually this one is a mill-drill but the largest of those that I've come across, not the rather flimsy thing that they usually are. It weighs 350kg which is just about manageable to move if you separate the table, head and base.

 

The reason for a milling machine at this stage is to make the rather complex bits at each end of the struts which allow the wings to fold virtually in an instant, and lay flat against the fuselage sides rather than just swinging back the way most of them do.

 

I could have had the parts made by CNC machine for about $2500 but where's the fun and satisfaction in that? Except getting flying sooner of course. And this way I get an effectively 'free' milling machine out of it.

 

The machine didn't have a stand so I built that first, and equipped the rear of the stand with wheels so that I could move the whole thing around the workshop if necessary by slipping in a couple of long steel handles, wheelbarrow fashion. In the event it worked but it is still heavy enough that the handles need to be 2m long to be able to lift the front legs off the ground.

 

There was going to be quite a number of days ahead spent machining and it didn't have a powered table so the next thing was to build a system to motorise it. I had a couple of old Baldor units lying around, one was 19rpm and the other 500rpm i.e one was too slow and one too fast, even though they are both infinitely adjustable with their electronic speed controllers. In any case it would be beneficial to use a couple of pulleys in the drive system, providing a simple way to disconnect the drive system when it's not needed, by slipping the belt off.

 

Finally I was ready to do some work on real aircraft bits ... none of the time spent doing the above strictly counts as DooMaw build time, so no hours added to the build log this post.

 

A pic of the milling table drive unit -

 

 

 

[apm]

The problem I've had with it is that work has had me moving around fairly regularly and I've only been able to use it when I've had 3 phase power, which hasn't been all that often. I've been intending to convert it to single phase for ages, it's only 1hp, so not a big issue ... or so I thought.

In my opinion a VFD is the best way to run small 3phase motors from single phase, as a side benefit they provide precise control of the machine.

 

Andrew

 

 

[Oscar]

Naturally the two motors' shafts were different sizes so I had to machine the drive pulley and I had misgivings about that because unless you get the bore right to within a couple of tenths of a thou a single phase motor will rattle the pulley and mount apart in a few seconds with the accompanying noise of a machine gun. My lathe is very small and the four stage pulley quite large so it (I) was struggling and you guessed it, I went oversize by a couple of thou. Then I had to devise a tool to broach a new keyway, I haven't done a lot of broaching so I didn't have much success with the first two tools ... I rewired the machine while I was about it, and then came the time to test run it. Sure enough - machine gun - why, oh why do they make parallel shafts? Tapered shafts work so much better!

I finally solved the problem by providing the pulley with seven grub screws to lock it up and control the harmonics, got some new blades made up, greased the bearings and adjusted the blade guides and it's like a new one.

 

Finally I was ready to do some work on real aircraft bits ... none of the time spent doing the above strictly counts as DooMaw build time, so no hours added to the build log this post.

People who blithely say 'A poor workman blames his tools" have NEVER spent half-a-day carefully machining something to the very best of one's ability - and then finding it has unacceptable tolerances due to the limitations of the machine, and a further two days or more having to make a work-around for that. I have the same problem with the only FAQ chucks on my own (fairly similar) lathe, and I reckon in the last two years I have half-filled a garbage bin with the swarf just from making adaptors for various jobs to take out the eccentricity!.

 

Welcome back to the wonderful world of building, HITC. It's been a bitch of a summer so far down here, and I believe worse up there, for working comfortably and sometimes a bit of a break is welcome - but the 'itch' starts again, doesn't it!. The old, walk into the workshop and get the 'right, now where was I?' feeling....

 

 

[Marty_d]

The cap material is 6061T6 aluminium and its raw form comes as a rectangular hollow section (RHS) but I want an angle section which will also be tapered toward the wingtips, so I need to cut the RHS lengthwise into four, to create the angle pieces.

Not only the plane but the machines to make the plane... nice work!

Question though... could you not have run the RHS (carefully) through a bench-mounted circular saw? Yes you'd lose some material to the blade thickness and some filing would be required afterward, but I'd have thought it quicker than re-engine-ing a bandsaw.

 

 

[Head in the clouds]

In my opinion a VFD is the best way to run small 3phase motors from single phase, as a side benefit they provide precise control of the machine.Andrew

Hi Andrew, thanks for the comment. I'm embarrassed to say that I'd never heard of a VFD in this application. Mr Google has since told me a fair bit about them, and you're right, it would have been a good option. I'm probably still a fair bit ahead cost-wise though, the new motor only cost $80 and the cheapest VFD seems to be closer to $200. The bandsaw has a very wide range of speeds available, so there wouldn't have been any great benefit from that, but well worth knowing about for the future, thanks!

 

Not only the plane but the machines to make the plane... nice work!Question though... could you not have run the RHS (carefully) through a bench-mounted circular saw? Yes you'd lose some material to the blade thickness and some filing would be required afterward, but I'd have thought it quicker than re-engine-ing a bandsaw.

Yes, I could have Marty, if I had a bench saw ... it was just that I've had the bandsaw for so long, and it's such a good thing that I was pleased to be forced into getting it operational again. You're quite right about the benefits of a bench saw though, I do have a compound mitre saw with a special carbide blade for cutting aly and it does a fantastic job. I also use an electric planer (hand-held type) with the blade set to take a cut of only about 0.2mm, to clean up the edges of the aly after cutting on the bandsaw. Word of warning if you try it though, never go near the ends, start and stop a good 50mm or so before the ends or the blade can/will bite the end viciously and can throw the planer back at you, as well as pretty much destroy the job. Hasn't happened to me cutting aly but I know someone to whom it has ...

 

 

[Kyle Communications]

My CNC machine uses a VFD it has a 3 phase motor but is driven from single phase via the VFD. I can run the spindle from 50 rpm to 6000 and its a 2 kw drive

 

I control it from the CNC program of course but you can select that mode and select a manual mode to either select the speed via a arrow setup on the keypad or a std pot type control

 

 

[Head in the clouds]

A few months back, when I was going through the strut calculations, forumite Oscar mentioned to me that he knew of a pair of aly struts that might be available and might be a suitable size for DooMaw. I had a bit of difficulty getting information from the owner of them as he was very busy, and I gave up because I was quite set on having chromoly ones in any case.

 

In November I was finalising the design of the strut ends and it became apparent that aly struts would suit better because aly struts end up with inserts bolted into the ends of them whereas chromoly struts have a cut-out with a welded perimeter which becomes a pair of vertical straps to attach to, at least those are the most 'conventional' methods, and which probably are the best strength-to-weight solutions. Since DooMaw's wing-folding mechanism at the strut ends requires something solid (as in a solid block), rather than a couple of plates to bolt to, then the aly struts started to look a lot more appealing than previously.

 

Consequently I renewed my pursuit of them and following a few weeks of persistence I was very well rewarded when a brand new pair of extrusions arrived for the princely sum of just $350 including the freight. Thanks heaps for the tip-off Oscar, it saved me heaps in time, hassle and money compared to having to get them in from USA - and Trump's now withdrawn USA from the TPP so I guess he doesn't want any more of our money, does he?

 

Then it was time to concentrate on machining the strut end parts because once I get them out of the way most of the rest of the build is fairly straightforward - there's still lots of it to do, but most of the complex work which takes up so much time with little to show for it, will be done.

 

I modelled the parts in CAD first and produced the drawings, though I knew there would need to be a little refinement along the way as machining difficulties revealed themselves.

 

To feel that I'd made a start I began with a couple of parts that will be bolted into the upper ends of the struts, and the machining just involved an accurate drilling setout. It's just made from some 16mm 5083 plate. Two are made at the same time in one piece and then sawed down the middle and milled square and to size -

 

 

A simple job like that was one thing but it's been years since I did much complex machining requiring careful attention to machining order. In which case I thought it would be prudent to get my head back in shape with a fairly easy part next, and I'm glad I did because, although apparently simple enough, it still threw a few challenges at me. I should say 'pair of parts' because it's much quicker to make two at a time, one on each end of a piece mounted through the vise, than to do one and then another the same, because you can use the same setup for each machining process, just move the table back and forth between each end.

 

One of the things that always tests me with these kinds of parts is how to get hold of them during the latter stages, to complete the final machining once I've cut off the main boss that clamped in the vise at first. Another thing with any clevis type of part is that you're effectively making a tuning fork so it want's to sing like Dame Nellie and so it is very prone to chatter and running out dimensionally. These parts do show more than a little evidence of that but I'm not concerned from a structural viewpoint because the parts are designed to withstand the dynamic loads of supporting the wings while trailering on rough roads, which are much higher than the flight loads would ever be, and they can't actually deform during trailering because the travel will be limited by fuselage/wing support cradles in the trailer. So the only matter to really take account of is the bearing surface areas of the swivel and hinge pins which I've made very substantial, and the holes are carefully reamed to exact size, rather than just drilled - the accurate fit of the hinge pin massively reduces the rate of wear in service.

 

A few pics show the first effort, the ends haven't been radiused yet, I'll do that when I've finished all the linear machining and can remove the vise from the mill and mount a small rotary table -

 

 

In the last picture you can see a tapped thread, I made that so that I could screw in a mandrel to mount the part in the lathe and machine the boss end circular instead of square, the thread will be drilled/reamed out in due course.

 

Having broken the ice I was keen to get on with the most complex of the parts because once they're done it'll feel like more of a downhill run. These present quite a challenge because they're very long tuning forks with lots of machining along the length of the tangs, and also because later I have to mount the parts up in the lathe to machine them circular, similar to the last parts, but this time they're much longer and achieving concentricity is going to be much harder.

 

I had worked out a method to stop the tangs harmonising though, and that would be by leaving a full web all down one side and also a web across the end. The end web would also help with the concentricity thing in the lathe because I could drill though it centrally to provide an end support for the mandrel, and additionally run a live centre in the boss end.

 

The long web down the side would add a significant complication though, because I would have to mill a very deep trench from one side rather than half the depth from each side but that would just be a matter of taking my time and being careful.

 

The end web posed a hazard, because the machining travel would have to be terminated accurately at both ends, rather than just one. It's so much easier if you only have to concentrate at one end and can relax as the cut runs out the other end - that's the beauty of CNC machines of course, taking the long periods of concentration out of it - because with manual machining, one stuff up late in the job can cost you days of doing it all over again from the beginning.

 

The pictures below show where I've reached so far, the parts are still in the machine. I've cut the deep (35mm) trenches on both of them and turned the workpiece twice to cut the shallower trenches on the outsides of each of the tangs. Next I will do the setout for the drilling, then drill and ream for the bolts - at that stage I'll introduce the first pieces I made, between the tangs, and drill/ream through them at the same time. Then I'll turn it in the vise again and machine away the side webs and create the tapers down each side. After that it can come out of the vise to be cut apart for longitudinal drilling/tapping to fit the mandrel for the lathe.

 

 

I haven't wanted to rush it and then need to do it again, so it's been slow and steady work, 43 hours in that so far, making a total of 1380hrs to date.

 

 

 

 

 

 

 

 

 

 

 

 

[Oscar]

HITC: very happy that that 'lead' turned out well for you!

 

And, very impressed with your machining - bloody great work!.

 

Here are two piccies of the strut end fittings (for the same strut material) that we had CAMit machine up for us about two years ago for our Jab. and I reckon anybody can see that your work is right up there for quality with stuff produced on mega-buck cnc machines.. They are to the design of an aero-engineer who is somewhat notorious for not 'underdesigning' things - they are in 4140 and should last for about another millenium... but certainly not featherweights.. Since he will be the test pilot when it's back in the air, and we'll be doing the full suite including full spin testing, I guess I can't argue with what he feels comfortable to fly with..

 

 

 

 

[Head in the clouds]

Still machining the same pair of parts.

 

It gets more nerve-wracking towards the end of jobs like these because there's an ever-increasing investment in time spent on the job and it would be infuriating to have to go back and start again. It constantly brings to mind the words of my toolmaker 'extraordinaire' friend and mentor, the inimitable Barry Hughes, who, when he was working on something similar and was asked how it was going, would usually respond "it must be OK, I haven't stuffed it up yet".

 

That investment in time and reluctance to have to do it again is also a good reminder of the flying creed which extols the virtue of maintaining concentration right to the end, or in other words, 'not going to sleep on the home stretch'.

 

So my Australia Day was spent with another 9 hours on the milling machine making swarf.

 

The pictures show -

 

My regular companion, QA inspector and workshop foreman, he spends a lot of time standing on that stool on hot days because it gets him up into the breeze from the fan - and he can keep an eye on the work progress of course.

 

The first side of one end with the caps tapered to reduce weight, the heavy web on the bottom is still there to stabilise it while the machining on the sides takes place.

 

After all four sides have been tapered the workpiece has been turned over and the main webs machined off, the trench down the centre now goes right through.

 

The same thing from a different angle.

 

The first pieces that were made, the drilled pieces of 16mm plate, are inserted into the trench (which will be a clevis later, once the end webs have been removed), positioned with a spacer at the inboard end and then clamped, drilled and reamed for a series of 3/16" bolts.

 

 

 

 

 

 

 

[Guest SrPilot]

HitC, I've been following your project and I must say I'm impressed although I am glad that I bought a kit when I went to build a plane. At my age, I would never have finished had I built it the Australian way despite the fact that you chaps have a three-step process.

 

1. Build or rebuild your machines.

 

2. Manufacture your parts.

 

3. Assemble the plane.

 

I would never have flown in my lifetime but I would have left the wife with a nice machine shop and maybe some parts had I gotten that far along. Thankfully I already had the tools (or most of them anyway) and with the kit I had to fabricate many fewer parts.

 

I'm impressed. You fellows are operating at a different level!

 

 

[rgmwa]

I'm impressed. You fellows are operating at a different level!

Don't be fooled SrPilot. HITC's got it easy. Now, I had to go out and dig the ore and do my own smelting before I could make a start. He just went out and bought stuff, so apart from the odd bit of fabrication and a few simple welds, he's just about done.

rgmwa

 

 

[Oscar]

Don't be fooled SrPilot. HITC's got it easy. Now, I had to go out and dig the ore and do my own smelting before I could make a start. He just went out and bought stuff, so apart from the odd bit of fabrication, he's just about done.rgmwa

LUXURY!. I had to make the smelter... and Dad would beat us with a belt full of razor-blades it we didn't clean out forge wi' tongue before the clinker set hard..

 

You tell that to the kids of today, and they'll never believe you...

 

 

[Head in the clouds]

Another weekend a bit like the preceding ones - machining, machining, with a lot of sweat to go with it, we're into the third consecutive heat-wave now.

 

However - sheer bliss, I completed the trickiest pair of parts, the long clevises, without stuffing them up and having to start again. I'm a lot less concerned about messing the rest of them up because if I do, by comparison they're a minor task to repeat.

 

So, as the pictures show, I completed all the milling processes then cut the two parts away from the central billet which I'd used to hold them in the vise. Then I created a setup to hold each of the parts in turn in the vise and drill centrally through the end web. I made up a long centre-drill by boring the end of a piece of ground stainless shafting in the lathe and silver-soldering a centre drill into it. I then used that extended drill through the hole in the end web to start the hole in the butt end of the part. By carefully scribing the position of the centre on the inside of the butt end, and finely adjusting the part in the vise, when the centre-drill met the scribed position I could be sure that the part was exactly vertical.

 

Then I used a progression of long-series drill bits to increase the bottom hole up to 13/32" and the top hole to 12mm at which time I could tap a 12mm thread in the butt end and use a 200mm long cut-off 12mm bolt as a mandrel to mount the part in the lathe.

 

Beforehand I had used a centre drill in the threaded end of the mandrel so that I could employ a live centre mounted in the tailstock to support the end of the workpiece, and then it was just a few minutes of machining to turn the butt end circular so that it matches the first pieces I made.

 

After the turning was completed I no longer needed the webs on the end of the parts so I cut them off in the bandsaw. Those square-cut ends will be radiused later, when I have finished with the parallel vise on the mill and can remove it to mount the rotary table.

 

Finally I drilled out the threads in all four parts with which I had used the threaded mandrels, to 1/2", ready for the sleeve which goes through them, and about which the wings rotate as they're folded.

 

Moving onto the next parts, I didn't need the billet to be the full thickness and was able to make use of the newly refurbished upright bandsaw to cut away the excess as shown in the second last photo. It took just three minutes a side to cut that, whereas to create the setups and mill that amount of material away would have been more like an hour, so it's really nice to have the big saw working again.

 

I got as far as milling a full depth slot each end before the sunset, these will be the clevises that attach directly to the cleats on the wing-spar.

 

 

Another 17hrs in that lot, and I forgot to add to the total last post, so it's 26 more hours, making a total of 1406hrs to date.

 

I'm impressed. You fellows are operating at a different level!

Many thanks for the kind words srPilot but it's nothing remarkable really. It just comes from having a tight budget and the scavenging resourcefulness that results from being born and bred in the bush where there's plenty of opportunity but often there's not a lot of equipment unless you make or fix it yourself.

 

rgmwa and Oscar are so right, it's FLAMIN' LUXURY to have old broken tools to fix up and work with, you tell that to the kids of today ...

 

Just in case you're not a Monty Python fan, this is what they're on about -

 

 

 

 

 

 

 

[Guest SrPilot]

Many thanks for the kind words srPilot but it's nothing remarkable really. It just comes from having a tight budget and the scavenging resourcefulness that results from being born and bred in the bush where there's plenty of opportunity but often there's not a lot of equipment unless you make or fix it yourself. . . . t's FLAMIN' LUXURY to have old broken tools to fix up and work with, you tell that to the kids of today ... Just in case you're not a Monty Python fan, this is what they're on about -https://www.youtube.com/watch?v=FatHLHG2uGY

Thanks HitC. I get it. Came from the country myself - long ago though. And I too am a Monty Python fan although I usually revisit the witch's scene -- which I have used in the classroom when discussing trial procedure.