Categories: all aviation Building a Biplane bicycle gadgets misc motorcycle theater

Sun, 19 Jul 2020

Building a Biplane: Getting Unstuck

For the last four days, I've been trying to figure out how to make this part:

[Image showing
plans to make the tie-down ring, part number -282, for a Marquart
Charger]
The -282 tie-down ring

It looks simple enough on the plans, all innocent and perfectly drawn. Yet it's taken me four days so far, 17.5 hours of work, with more to come. Most of these brackets I've been making take maybe 4-8 hours to sort out. But this one's special.

The journey starts with: How to bend that hoop? Quarter-inch 4130 steel is no slouch in the strength department, though with a long enough lever it's easily bent. My first try was just to clamp the thing in a vise, perpendicular to a length of stout 1" tubing I had lying around (for the 1/2" radius bend), then bend it over the tubing by grabbing the long end and giving it a shove. This worked, but the resulting bends had a disagreeable curve to the "straight" parts that I didn't love. My math was also poorly understood, so it took a few tries before I got the right size, and I only had 17" of the stuff to work with. Finally I gave up and decided I could live with the poorly-bent hoops, just so I could make progress.

So, hoops done, check. I had the plates already, thanks to my various Ken Brock bracket kit purchases. I held the hoop up to the plate and tried to imagine how I'd actually hold any of this together. It's pretty fiddly, and clearly a quality jig was in order.

The first problem with the jig is that the space between the plates needs to be 2 1/8". I had 2" square tubing, which I'd used to make jigs for other brackets, but I had to find that extra 1/8" of thickness somewhere. I have some 1/16" thick sheet lying around from a previous project (making the drag strips, which I ultimately scrapped in favor of wires), so I decided to weld two layers of that onto my jig, which should work out to exactly 2 1/8" thick.

Of course, it did not. It was a little bit thin for some reason, and since these brackets tend to shrink a little after they're welded, I really didn't want to start off too small, or I'd never get the bracket onto the wing when the time came. So I did the only reasonable thing I could think of, and ran a weld bead down each of the 1/16" strips to make them a bit thicker.

[Image showing the welding jig with weld beads built up]
The welded-up jig

Then it was a whole lot of filing to get them back down to the right size, after I managed to crash the head of my milling machine when the jig leapt up into the cutter and stopped the works very quickly indeed -- I still need to open that thing up and replace the busted gears. But I got it to the right size, and I'm going to have very odd musculature in my arms when this is all over.

However, my skill with a file is not expert-level, so the strips are only about right, they're not perfectly flat. I decided this was an acceptable compromise because it's all getting welded anyway, and striving for perfection there is an exercise in frustration thanks to the expansion and contraction of the steel as it's heated and cooled.

In any case, I took my imperfect jig to the drill press, and carefully drilled out the hole that would locate the side plates. Not carefully enough, though: the entry hole was perfect, but the exit hole was at least 1/16" off, probably because I'd been resting the jig on those imperfectly welded and filed strips. So I tried a trick I've never done before, and welded up the hole, having to weld from both sides. It worked, to my amazement.

None of this addresses the original problem, though: how to deal with that floppy hoop that has practically no mechanical connection to the plates until it's welded. Finally, I decided that I'd make a crossbar with slots in it that would hold the hoop, and the milling machine came out again (I'm telling this out of order; I made the crossbar before the head crashed). That piece at least was pretty easy to make.

The final problem to solve (so I naively thought) was how to make sure the side plates, which are held down by a single bolt, couldn't rotate. When installed on the plane, they interlock with a big round piece, but I didn't want to have to make that up for my jig. Instead, I ended up making a little shelf out of metal for the legs of the plate to rest on.

[Image showing the finished welding jig]
The finished jig, in the ready-to-weld setup

Finally, late in the evening, having fussed with the spacing on the imperfect hoop for several hours off and on, I turned on the torch, and applied heat to metal, tacking the hoop in place. As soon as I did, I realized that my design was lacking, and there was no way to get the tacked bracket off the jig. The crossbar was locking it in place, and I ended up (the next morning) having to saw the ends off the crossbar, and file rounded ends in it so it could get past the plates while still holding on to the arms of the hoop.

This was ok, because a discussion on the Biplane Forum had convinced me that I needed to try making stainless steel hoops. The idea was that there was no way to chip the paint off them, and they wouldn't rust. There followed a feverish study of stainless steel properties and welding techniques, and a quick order to the metal shop. I was glad to be able to get it on Friday afternoon, so I could have the project to work on over the weekend.

I had called my dad to discuss the stainless steel (he knows more about stainless than I do), and he reminded me that I had a better way to bend the rods than just clamping them next to a tube and pushing hard: I have a press. And, I realized, I built a really moosey tailspring rebending tool which might just work as an improvised press brake. The idea is that you use the press to push on the center of the thing you want to bend, while it's resting between two bars, or a V-block, or something like that. The idea struck me, so I spent most of Friday trying it out. Fortunately, I got three feet of the stainless steel rod, because I made bad bend after bad bend (not the press brake's fault, I just kept getting the math wrong). Eventually, though, I ended up with two perfect hoops, exactly the right size, and with lovely square corners. Shown here before the ends were trimmed, so their perfection isn't quite as obvious:

[Image showing two nicely bent-up tie-down hoops]
Hoops!

I decided, wisely, that before I would commit my hard-to-reproduce side plates to this project, I would try welding some of my incorrectly bent stainless rod to some mild steel, just to see if I could do this dissimilar-metals welding well enough to put on an airplane.

I welded the first hunk of rod to a failed experiment that was a sort of box of square tubing, which was certainly a decent stand-in for the side plate. Once it cooled down I hit it with a hammer a few different ways, finally sticking a piece of MDF (not the strongest material, but the right size, so why not?) through the center of the box and tried pressing the rod away from its plate. With one moderate hit of the hammer, it ripped right off.

Well, that's not good.

I tried again, on a different spot. Same thing. I made a smaller test piece, thinking maybe the big box was sucking up too much heat. Nope, same thing. For some reason, the welding rod I was using to build up the weld was porous and quite weak.

Just to make sure it wasn't my technique for welding a rod to a plate that was at fault, I tried the same thing with a piece of the 4130 that didn't get bent well. Making the same test, the MDF was destroyed, and it barely moved the rod. But it did move it enough that I could flip the piece and whale on it with the big hammer. It bent, then bent further, then bent almost double. The weld held, finally just starting to tear as the rod took its last few hits. That was much more like it. That's a weld I'd trust to hold down an airplane, or jack up an airplane.

[Image showing an amusingly bent piece of 4130 rod welded to a
mild steel structure]
At least the weld didn't break

So, the stainless steel experiment was a bit of a bust, but at least I found out before making a bit expensive mistake, or even worse, putting it on the plane and thinking it was doing well until the one thing that would doubtless be a catastrophic failure.

Now I just have to find more 4130 rod. But once I have it, I have the perfect bending technique all lined up and ready to use. Five days to produce two tie-down rings? Why not.

Posted at 00:43 permanent link category: /charger


Sat, 04 Jul 2020

Building a Biplane - A Quickie

In my video on making aileron pivots, I said I'd spent 16-some hours so far, and expected to spend 20 more before the pivots were all done. I've finished them (or at least gotten to the condition I was thinking of when I made the 20 hour estimate, which means they're ready to be sandblasted, painted, and finish-reamed, so still several hours from being done) as of last night.

I've spent some time revamping the build log (coming soon), and while I was dinking around in the database this morning, I decided to see how many hours I'd actually spent building aileron pivots.

sqlite> select sum(hours) from events where date >= "2020-06-05 19:27:02";
57.7

My original estimate? Less than 40 hours. Actually? Almost 60, over the course of almost exactly a month. It's silly, but I find this very amusing.

Posted at 10:15 permanent link category: /charger


Sat, 20 Jun 2020

Building a Biplane: Some Updates

I've been making good progress on the Charger build lately, and wanted to share some news.

The first thing is that I've made a dashboard showing my build progress. It was amusing to put together, and it's interesting to see in charts and graphs how I've been working on the plane. Having to take 2019 off due to lack of shop space made a pretty healthy dent in my progress.

The other thing is that I made a video about a tiny part of the build process, which may or may not interest you.

This shows one of the many many little steps that go into making the aileron pivots, which are deceptively simple-looking on the plans. It's just these three arms, how bad could it be? As I mention in the video, it's going to take something like 40 hours to go from looking at the plans to having 12 functional pivots in my hand, and that's not even counting getting them sandblasted and painting them, which is a completely separate adventure.

As someone commented about the build when I was describing it to them, "Oh man, even the details have details." Yep, the top-level step of "make 12 aileron pivots" touches on all sorts of questions: how to buy steel that hasn't been made for decades; what's the best compromise among the available sizes of steel; where to find that size; justifying paying $18 a foot (not including shipping or tax) for a weirdo size of steel that is also going to make the plane unnecessarily heavy; where is the steel after the seller put the wrong zip code on the box; how to cut steel; how to cut steel so it's square; how to cut steel so it's got the correct angle to the cut; how to figure out the correct angle to the cut (CAD to the rescue); how to locate two holes in a part so that they have the exact same placement on every piece; how to assemble the pieces once they're cut; how to build a welding jig for assembling the pieces once they're cut; how to make a welding jig that will lock everything into place, and then let it go once it's welded together; how to weld these two pins so they actually match the piece they're supposed to locate... It never ends.

To be clear, this is a big part of the fun: taking a job and breaking it down into its component parts, and then doing all those little jobs as efficiently and well as possible. In the end, you have a finished product that reflects all the hours and effort. It's slow going, but it is rewarding, eventually.

Posted at 23:44 permanent link category: /charger


Tue, 26 May 2020

Building a Biplane: Welding Brackets

I have reached a significant milestone, but it's not exactly about building an airplane. No, what I've done is finally finished the shop.

Of course, the shop will never be finished-finished, but it's now good enough that I can stop thinking about it, and get on with building an airplane. The door is replaced, the drywall is done, the electrical is inspected and passed. Finally.

Which leads me to the next phase of building a biplane: welding brackets. A typical bracket assembly looks like this:


Lower rear interplane strut bracket

The bracket starts life as three separate pieces: two -202 plates, and one -220 filler piece. The challenge is to assemble it and weld it so that all the holes are still lined up and the bracket is straight. Welding tends to make pieces move all over the place, because the welded metal heats up and expands, then contracts when it cools down, just a bit smaller than before it was welded.

This means that the pieces to be welded need to be held in a solid fixture or jig, that keeps them from wandering too far. They'll never be perfect, but perfection is always the goal.

So, my job is to come up with some way to keep these plates in line with each other while heating them to white hot. My first thought, which I'm glad I was talked out of, was to make some MDF pieces (a type of engineered wood that's basically sawdust and glue pressed together), which would basically take the place of the spar and wedge blocks shown above.

MDF is probably rigid enough, but the heat would have set it to smoking badly, and might have caused the structure to weaken enough that the welded pieces wouldn't have been held tight. So that was out.

I was resistant to making them out of metal because I didn't have any metal to speak of. As we live through a global pandemic, I'm trying my best not to venture out, but I decided this was probably worth it. I ordered some square tube stock from Online Metals and picked it up the next day.

So, the jig for this piece needs to do a number of things. It needs to:

I surveyed the jigs that needed to be made, and identified two main spacings I would need: 1.5 inches, and 2 inches. Thus I ordered lengths of those two sizes of square tube, so I wouldn't have to try to build up the right thickness. It turns out I missed the 1.25" brackets that go at the root end of the wings, but I'll work on those later.

The jigs would also need a flat piece across one end, for the spacer to rest against. Somewhere in this process I found the small collection of welding steel I had packed up so carefully when I moved, and decided I'd use the 4130 sheet left over from making the drag strips for this, since it just needed to provide a reference surface, not bear any weight.

What I came up with was this:


Partially completed -202 bracket jig

That's only partially completed, it would eventually have a second hole drilled, so that it would hold the brackets like so:


-202 jig holding all the pieces of the bracket

This plan worked out pretty well, but it took me a couple different jigs before I figured out the best technique for making them, specifically for drilling the holes so they were straight, and would align with the bracket plates. I'm making heavy use of the mill now, simply because it allows me to be so precise. It's almost certainly overkill, since these pieces are getting welded and will necessarily distort themselves all over the place, but you might as well start from a place of precision if possible.

The end result has turned out well, and the brackets look pretty good:


Finished brackets

The biggest problem I actually ran into is that the first set of wing brackets I bought a few years ago turns out to be coated in cadmium as a rust preventative. It worked really well, those pieces are all rust-free. Unfortunately, welding cadmium-coated metal is a good way to severely damage your health, and getting rid of the cad plating involves either muriatic acid, which is quite dangerous, or burning it off, which pollutes and is quite dangerous.

I figured out they were plated when I welded up my first piece, and noticed the horrible black smoke coming off the metal, and the yellow fume that seemed to stain near where I had welded. I clued in enough to put on a respirator after making the tack-welds, but that was not an ideal situation to be in.

Fortunately, one of the Biplane Forum members recently sold me a second set of brackets, duplicating and expanding the first set. This new set turns out to be not plated with cadmium. So, they're a trifle rusty, but clean up nicely, and don't rapidly poison me as I weld them. Win-win, honestly.

I'm on my way to getting my brackets done, another small step accomplished in the impossibly large project of building a biplane from scratch.


Slowly chewing that elephant, one bite at a time

Posted at 00:02 permanent link category: /charger


Tue, 21 Apr 2020

Building a Biplane: Improbably, a Sort of Milestone

When I describe the process of building a biplane, I invariably describe it as one that's going to take "ten to twenty years," because that's just how life is. Most credible estimates suggest that doing something like building a Marquart Charger from scratch is a 3000-5000 hour job. I'm probably going to end up on the 5000 hour end of things (particularly if I try for a round engine, but that's a discussion for a different time).

However, taking the middle of that estimate, at 4000 hours, may be a reasonable best-guess.

I mentioned a couple entries ago (and more than a year ago) that I was maintaining a log of all the work I've been doing. I promised to make it more than it started as, and I've taken the first step along that path: I added a "total hours" row at the bottom of the table. It was a lark, but it reveals a sobering truth: despite my jests about being at the very start of the process (which it truly feels like I am), I'm actually making significant progress.

As I write this, the total hours number stands at 388.8 hours. That is to say, nearly 400 hours, which would be about 10% of the build.

I'm 9.72% of the way done with building a Marquart Charger.

It's not a huge number, sure, but it belies the sense I've had for a long time that I was just farting around, making no real progress. No, in fact, I have a complete set of wing ribs finished, an appreciable portion (I'd say around 90%) of the wing parts built, on hand, or poised on the edge of ordering (the cross-brace wires are waiting on me properly inventorying my need for more spruce bits so I can combine the surprisingly expensive shipping), and am in the final stages of shaping the last little bits and pieces that need to be fabricated.

Once I have the cross-brace wires and fork-ends ordered, all that's left is painting a bunch of steel brackets and profiling the spars before I can start assembling the wings in earnest, at which point I suspect things will seem to move very quickly, compared to the visually unimpressive stack of wing ribs.

There's always more work to do (at least 90% to go), but it's nice to take a step back and appreciate that I really have made some good progress.

Posted at 14:21 permanent link category: /charger


Wed, 15 Apr 2020

Building a Biplane: Back In the Build

My last entry was pretty brief, with the simple aim of showing you my build log. Looking at the log, and at this journal, you would be correct in guessing that there was some kind of an interruption.

In late 2018, it was becoming obvious that I'd have to move soon, which would mean giving up the shop I'd put together for my Charger build. It would mean an unknown delay until I could find another house, ideally with a shop, and resume thinking about this insane plan of building my own full-size biplane.

And yet, here we are. I've found my new house, and I managed to find one with a 2-car garage that was ripe to be converted into a workshop. That process has taken far longer than I wanted it to, but is nearly coming to an end as I wait for the new garage door to be installed -- the whole COVID-19 situation means I'll probably be waiting for a while, though.

When I left off on the project, I felt that my next logical step was to profile the spars, and start assembling a wing. I had the ribs built, and all the compression tubes. I hadn't yet ordered the cross-brace wires, but I've still got that shopping cart loaded up at Aircraft Spruce and ready to go. I had found all the internal wing brackets in a Ken Brock kit on Ebay.

However, there was a significant piece missing. The wedge blocks.

Long story short, the Marquart Charger has a very classy-looking 10° sweep to the wings, but all the ribs and bits that go inside are arranged parallel to the fuselage. This means that you have to make up that 10° difference somehow. This is where the wedge blocks come in.

The use of the wedge blocks is pretty obvious. For instance, in the diagram below, you can see how the wedges, in green, are used to make a 90° surface against the wing spar, shown in blue:

So, that made a lot of sense to me. Wedges, cool.

Then my eye would always trace back to this cyclopean horror:

I had no idea what this thing was trying to tell me. For years, I would see it, and immediately get confused. It looked for all the world like a single piece of wood, that I was somehow supposed to cut up to form all those wedges, and it just boggled me to no end. How do you cut that -238 piece from the corner of the -239 piece? Wouldn't it leave a huge gap? How are you supposed to cut all those angles without the saw taking out its kerf and leaving you with undersized pieces? It just absolutely confused me.

Finally, shop nearly done, but not quite done enough to go around cutting thousands of dollars worth of spar material and assembling wobbly half-completed wing structures which might need to be moved (inviting any number of broken wing ribs), I decided to bite the bullet and ask for help with this wedge horror. Not that I have been shy about asking for help before, but it's always a bit of a struggle to admit that I don't understand something that everyone else seems to get.

So I fired up a graphic editor and cut out the section of plans I'd be posting about, but something about it made me pause. Suddenly, out of the blue, the picture that had for all time looked like a white vase on a black background snapped around, and became two human faces, facing each other with a white space between them.

It wasn't showing a single piece of wood. It was showing all the wedges overlaid on top of each other. My brain felt like it was going to explode.

Instead of drawing out six side profiles and six face profiles, with the attendant measurements for each, Ed Marquart had done the sensible thing and drawn them together to save space and drawing effort. This sudden understanding was, I must say, a huge relief.

Here is an annotated diagram with improved dimension lines, which shows in red how to interpret the dimensions for making a -239 wedge:

Looking over the plans and dimensions, it looks like I've probably got enough off-cut spruce to make most of the larger wedges, though I'm going to order some 1/4" thick stock to make the -238 wedges, since there are so many of them.

Now, I just have to figure out how I'm going to cut all these extremely sharp angles. I've started on a sled for the table saw, similar to this design. I don't have any MDF lying around at the moment, so I'm starting with some not terribly good plywood as a base. I'm pretty sure a sheet of MDF is in my near future, but it's good to experiment with the technique for the moment.

Honestly, I'm just excited to have something I can do during COVID times that doesn't require a helper, like the spars, and can be accomplished with the shop in a relative state of disarray while I wait for a new door. Progress is finally being made again, and it's a gratifying feeling after such a long delay.

Posted at 13:37 permanent link category: /charger


Wed, 26 Dec 2018

Building a Biplane: The Data Edition

Just a short take for this one: I've finally put together a page to display my live buildlog. One of my first tasks for myself before I started building my airplane was to code up a simple system to record my activities in a database. One of the big reasons I wanted to do that was so that I could display that log of activity on a web page. I finally found some free time, and put together that web page last week:

My Marquart Charger Buildlog

It's got data (with more to come, I just haven't had time to code it up yet), and it's got photos. Enjoy!

Posted at 10:58 permanent link category: /charger


Fri, 07 Dec 2018

Building a Biplane: In Which Mistakes are Made

I've been warned repeatedly that I was going to make a bunch of parts, then realize I'd have to completely re-make them due to some small error or flaw I built in without realizing it. So, in a way, I'm better off than I could have been.

Having now completed all 36 drag strips (why 36 when I needed 32? Let's charitably say I was building in my 10% extra and not just mis-remembering), I finally sat down to figure out what was the deal with these drag wires everyone else uses. Good timing, right?

So, I worked out the math. The drag strips are .063" thick and 1/2" wide. The total length of drag strip material is 916", if you add all the plans-specified lengths together. Add another 32" for the 1" doublers that go on the flat end of each strip, another 16" for the 1/2" of length after the flat-end hole, and another 51.2" for the 1.6" length each strip is folded over. This amounts to 1015.2" of strip length. With a cross-section of .0315 square inches, this adds up to about 31.98 cubic inches of 4130. It takes about 3.6 cubic inches of 4130 to make a pound, so the final weight of all this metal is:

8.88 pounds

4130 steel has an annealed yield strength (the strength at which deformations like stretch or bend become a permanent part of the metal) of 52,200 PSI. So one square inch of 4130 will lift 52,200 lbs before it will yield. .0315 square inches of 4130 will lift 1644 lbs if the metal is in the annealed heat-treat condition. This annealed condition is important because it's the weakest state of the metal, and after welding, you can't count on the heat treatment state of the metal around the weld. For normalized 4130, the yield strength is 63,100 PSI.

The turnbuckles, at .06 lbs each, add another 1.92 lbs of weight. We will ignore the weight of the pins, since the alternative wire construction will also use pins (though they'll be shorter, and so will contribute to any weight savings). The turnbuckles are rated for 1600 lbs as their "strength" (I'm not sure if that's working load or yield load or breaking load, but I'll guess it's either yield or working load), so they match nicely with the 1644 lb strength of the drag strips.

Total weight is now: 10.8 pounds for the drag strips. Not bad.

But what about wires?

This is where things got real for me. I checked, and realized that although a 3/16" wire has a smaller cross-section than a drag strip (.0276 in2 instead of .0315 in2), the wires are guaranteed to be normalized. You don't have to weld them, so you don't mess up the heat treatment. This means you get to use the higher 63,100 PSI number, and a 3/16" 4130 wire has a yield strength of 1741 lbs. So... it's stronger and weighs less? What's the downside?


AN665-21R clevis fork rod ends (like you didn't know that already)

There really isn't one. The wires by themselves weigh 8.12 lbs (916" length plus 144" to account for the now-missing turnbuckles, for a total length of 1060"; making 29.26 in3 of steel). You have to add AN665-21 clevis fork rod ends and locknuts, which add up to 1.16 lbs for a total weight of:

9.28 lbs for drag wires.

Hmm. Compared to 10.8, 9.28 sounds pretty good. Saving a pound and a half is nothing to sneeze at -- I've heard tales of people substituting titanium fasteners on a plane for a total savings of less than a pound, at the expense of several thousand dollars. Here, I can do it by returning a thousand dollars worth of turnbuckles and buying $800 worth of rod ends and $200 worth of steel rod. So, I spend the exact same amount of money and save 1.5 lbs? Sign me up!

Now, it would have been really awesome if I'd figured all this out before spending a bunch of hours making drag strips. So, if you're building a Charger, this is my gift to you. Go figure out drag wires rather than wasting a bunch of time on heavier strips. Fortunately, I had a good time with making the strips, and I don't consider it time wasted. I got to meet a local Charger owner to have the strips cut, and learned about fabricating an interesting part. And now I get to play with threading a bunch of 4130 rod and RMA-ing a small army of turnbuckles.

Posted at 21:44 permanent link category: /charger


Mon, 03 Dec 2018

Building a Biplane: The Many Steps to Actually Make a Part

I've been impressed by the seemingly countless steps necessary to actually make a part for this biplane build. I'm currently in the middle of making the drag and anti-drag strips, which will go inside the wings as crossing "wires" to keep the wing structure from racking forward and aft.

At first glance, the drag strip idea makes a lot of sense -- no special tooling required, sheet metal is widely available (and probably more available or cheaper in the late 1960s than 4130 wire of the appropriate size), it's easy to cut on a sufficiently beefy shear, etc. However, on reflection and having accomplished 90% of the construction of the strips, I think I would have been happier buying wires.

Here's what it takes to make a batch of 36 drag strips:

  1. Acquire sheet metal. $80 at Online Metals for a 2x3' sheet of .063" normalized 4130. This was by far the simplest part of the process.

  2. Find someone with a big enough shear to tackle 1/16" 4130. I managed to find someone through the Biplane Forum who offered to do it for free at his shop. We cut my sheet into about 40 1/2" by 36" strips.

  3. Polish the edges of the cut strips. I finally figured out that I could hit multiple strips if I was careful by using the belt sander that has served me so well making ribs. This process alone took over 2 hours between finding the right procedure and actually doing it. I started out with draw filing individual strips clamped in the vise, which doesn't count the 2+ hours I spent making perfectly smooth vise jaw covers. Such a simple operation that took so much time and head-scratching.

  4. Bend the teardrop end of the drag strip (see the above picture). 4130 is no slouch when it comes to the strength department, and I wanted to get these bends perfect the first time. The solution I eventually came up with was to build a bending jig out of beefy 1/4" thick angle iron and 1/4" plate (7.3 hours, not counting driving around hardware stores for a couple hours, looking for 1/4" angle iron). This can be used in the 12-ton el-cheapo hydraulic press I got back in my engine rebuilding days to get the strip most of the way bent, then it's off to the vise with a hunk of 1/4" round to close the end of the loop, but keep the teardrop open.

  5. Weld the edges of the loop's tail. For all my initial fears of welding, this was among the simplest operations in this process. It still took 3 hours to weld all 36 of them.

  6. Cut the slot into the teardrop. This process was the subject of much heartache for me. My first idea was to use my milling machine with a very tiny endmill to cut the slot before the strip was bent. This worked, but was very slow, and ran a substantial risk of breaking the cutter, for it was ever so wee. Then I had the idea to use a slitting saw in the mill once the loop had been bent. Much better, but I discovered that after welding, the slot would expand a bit, and the plans are very particular that the slot should be a particular size. So I thought it better to weld first, then cut the slot to eliminate welding as a possible source of error. I welded up the first 17 strips, and tried to cut the slot in one of them, only to be horrified to find that welding modified the strength of the metal such that the saw (which had worked perfectly before welding) suddenly sounded like it was about to explode. Finally someone on the forum suggested using an abrasive cutoff wheel, which worked well, and was the method I finally used to cut the rest of the strips. I spent nearly 5 hours just figuring out how to cut slots and then slotting all the strips. I'm glad I didn't have to use a hacksaw and file to do the job, as someone suggested when I asked how Ed might have expected the average homebuilder to do it.

  7. Clean up the welding scale. I started out by hitting the strips with sandpaper to clean up the scale, but it didn't do a very good job. I finally switched to a wire brush and had much better success. I also destroyed two partially-used wire brushes in the process. If I have to do any more of that, I'm getting the wire wheel set up on my bench grinder again. I need to do that anyway, as long as I'm welding parts.

That's the process so far. I'm not done yet, but that's as far as I'm taking it until the wings are built, and I can see exactly how long each strip needs to be.

By contrast, the steps necessary to deal with wires would be:

  1. Buy wire. This requires an order from Aircraft Spruce, and costs more than the sheet metal ($128 + shipping vs. $80 for sheet metal).

  2. Cut wire to length. Hacksaw and file to clean up the cut.

  3. Thread ends. Time-consuming, and requires a left-hand threading die for one end, but a well-denfined problem with a well-defined solution.

That's it. If I had it to do over again, the extra $50 would be 100% worth it. I think the turnbuckle situation would be a bit cheaper too, by at least half. If you find yourself building a Marquart Charger, may I humbly suggest that you don't follow the masochistic "Do it exactly per plans" path that I have, at least for this particular situation.

For your edification and edumacation, I also filmed much of the process and turned it into this moderately interesting video:

Posted at 23:06 permanent link category: /charger


Sun, 19 Aug 2018

Building a Biplane: Good to the Last Rib


That's a lotta ribs

I reached a pleasant milestone at the end of July: I finished all the ribs for the Charger. I had run out of materials at the start of May, having underestimated capstrip and 1/16" plywood by about 5%. Aside from kicking myself for the mistake, there wasn't a lot I could do, so I went off and did other things while I waited for my order of new materials to ship. I took advantage of the announcement of the new steel and aluminum tariffs to order my fuselage steel at the same time, although that delayed the order by another month or so. What's a month on a 10-20 year project, right? (Sigh)

In any case, on the 22nd of July, I glued together my final two ribs. This is the first major-ish milestone I've reached in the build. It was great to achieve it, but it was distressing how long it took -- more than a year from first rib to final rib. I had known it would take a while, but I was figuring 6 months. I blew my time estimate by 100%.

The downside to reaching this milestone is that I was now without a singular task to work on. When I was building ribs, it was easy to go out to the garage, and pick up wherever I left off on the ribs, going until I was at an obvious stopping place. Now, I suddenly had a variety of tasks to accomplish: profile the spars; machine bushings for compression tubes; weld bushings to compression tubes; produce drag strip.

Profiling the spars is the scariest task for me. Each spar costs between $100 and $150 to replace, but much more importantly, would take something like 6 months to arrive once ordered. Spar-grade spruce is hard to find. I mean, it's easy to order, but Aircraft Spruce and Specialty, despite their name, don't have a pile of spruce sitting around ready to ship. So, if I mess up a spar, that's 6 months of waiting, and about $300 lost (because they're 10-11 feet long, the spars have to ship by truck, which is a minimum of $150 on top of just buying the wood).

Because of this trepidation, I've been dithering. I got a couple of 2x6s at the hardware store, and crudely resawed them to 1" thick to mimic the spar blanks. I tried a variety of methods of cutting them down until I found a method I like: cut the bevels on the edges with a table saw, but cut it oversize, then finish with a hand-powered bench plane. I got a bigger plane (a Stanley #5), but discovered that Stanley has gone substantially downhill in the quality department, and then went on a week-long research bender getting deeply, deeply nerdy about bench planes. I ended up ordering a Lie-Nielsen #5, and have a bid out on a vintage Stanley #7. For those of my loyal readers (ie, all of you) who don't know planes that well, the Lie-Nielsen is the ridiculous-but-worth-it Cadillac of bench planes, and a vintage Stanley is the they-made-them-well-back-in-the-day winner. A #5 is a good all-around size, and a #7 is a bit of a monster, but good for making sure you don't accidentally introduce some hills-n-valleys on your piece of wood.


That's a lotta tubes

I also found myself dithering on the welding front. Although I now have much of the metal in stock (plus some) that I should need, I've been worried that my welding skills aren't up to scratch. I welded up practice piece after practice piece, but they didn't quite seem good enough. Finally, a couple nights ago, I made some new test pieces, and took a methodical approach to solving the problem, setting up the best jigging system I could think of. I welded my four test pieces, and by the fourth I was actually feeling pretty good. Not great, but good enough that I was willing to try with the expensive aircraft steel.

Fortunately, it looks like I had the technique right, and the resulting welds have met with approval from the experienced folks who've seen them. I now have twelve half-compression tubes. Only 24 to go. (Actually, they went really fast. I should have the compression tubes done in a couple of work sessions if I can maintain that rate.) Not quite worthy of the name "milestone," but it was good to get some real aircraft welding done.

I also made my first foray into the world of waterjet contractors. My original plan was to draw up all the little metal brackets and bits that I would need in CAD, and ship the drawings off to a waterjet shop to have the pieces cut. Then all I would have to do was bend the pieces appropriately (easier said than done, but new pieces would be as far away as the waterjet shop when I messed up), and voila! All done. It turned out that I found a good deal on a set of Ken Brock wing fittings, which took care of 90% of my waterjet work for 1/3 the likely cost, and that was an easy decision when it came up on Ebay. However, it didn't include the drag strips (criss-crossed strips of steel inside the wing, which stiffen it and make it so it won't rack side to side), so I've sent those DXF files out to a bunch of shops to get estimates. I'll be curious to see what they say. Of course, the drag strips would have been cut on a metal shear back in Ed Marquart's day, and I need to explore that option as well, since it may be substantially cheaper. I can cut my own slots and holes if it saves hundreds of dollars.

It's nice to be making progress. I wish I was making progress faster, but I need to let go of any concept of building to a schedule unless I want to upend my life to do it. I enjoy still having relationships and friends and other activities, though, so the airplane building will continue to be a "when I have time" priority.

Posted at 18:42 permanent link category: /charger


Sun, 10 Jun 2018

Building a Biplane: The Waiting Game

When I started building the ribs last year, I ordered a bunch of materials: Spruce capstrip (1/4" x 1/4" by 4' long pieces), plywood in 1/16" (!), 1/8" and 1/4" thicknesses, and so on. The Charger plans include some estimates of of the material required to build the ribs: 600 lineal feet of 1/4" capstrip, 24 square feet of 1/16" plywood, 16 square feet of 1/8", 8 square feet of 1/4". I took these numbers as my basis, and ordered about 120% of each of them, since I figured I'd make some mistakes, and it never hurts to have extra.

As I mentioned about ordering wing spars, the shipping on some of this stuff can take a long time. Well, more fulfillment than shipping. Shipping itself is pretty quick, but it can take a long time for the businesses involved to get all the bits and pieces sorted out for an order. Particularly with orders like these, where it has to go by freight instead of normal package shipping, you can't just split it up and ship some of the stuff separately. A freight shipment is a $150-200 proposition for the things I'm ordering.

So the rib materials arrived after a not-too-bad wait last year, and I got to building. I quickly discovered that some of the capstrip I'd received had gouges in the side, and I was assured that this is simply par for the course. However, I lost about 10% of my order to these gouges. Without being finicky about the math, that meant that I only had 10% extra for mistakes and, as it happened, extra ribs "just in case."

This brings me to a month and a half ago, when I ran out of materials with three and a half ribs left to make. I guess I didn't use the 1/16" plywood as efficiently as the designer or something, as I was out of that around the same time I was out of usable capstrip. Since I can't get either material locally, I was stuck.

So, I loaded up my shopping cart at a national aviation supplier to build the rest of the ribs. Then I thought, "You know, I will just need to launch into building wings once the ribs are done, let's see what else I'm missing." More items into the shopping cart until I had about $750 worth of stuff. It took me a week to be sure I had everything I could reasonably predict into the cart, and I finally hit submit. I got a call a few days later: UPS has updated their shipping prices, and it's now cheapest (at $190) to ship this particular order by truck freight. That was weird, but I double-checked, and indeed, the shipping charges for normal UPS package delivery was nearly $400. Fine, ship by truck.

Unfortunately (fortunately?) I had ordered an item that was backordered -- 11 sheets of 0.8mm Baltic Birch plywood for the leading edges -- and it took most of a month before the order was finally assembled for shipment. As it happens, that was June 1st.

Something else happened on June 1st, of course: Mr. Art-of-the-Deal announced 25% tariffs on imported steel. Well, what is my plane going to be made of? About $2300 (pre-tariff) worth of steel tubing, most of it not produced domestically, whether I wanted domestic or not. Faced with a likely $500+ surcharge to stroke his frail ego if I waited, I decided now was the time to pull the trigger on the steel order, too.

Fortunately, I've had the materials list assembled and ready to order for a few months now, hastened by the first rumblings of a steel tariff. I quickly put together the order, and called to get it added to the existing order if it hadn't shipped yet: might as well make the truck shipment charge really work for me, since UPS definitely won't be taking 20 foot lengths of steel tubing in the brown trucks. I was in luck, and the shipment was scheduled to go out on June 1st, but they were able to put a hold on it before it was actually loaded onto the truck.

Downside: more waiting. Upside: reduced overall shipping cost, and I'll have much or all of the hard-to-ship material in hand for the next 5+ years of the project. There will still be many orders to place as I come on new phases and realize just how much stuff I still need.

Unfortunately, all this means that I've had a half-completed rib sitting in the jig for a month and a half, and not much else to do. I've occupied myself in the meantime with some steel work, trying out a technique to machine bushings for the wing compression struts which was successful, but is likely to be time-consuming if I pursue it.


Bushings. I'll be using the right-hand design.

I had thought up until just now that I could only make prototypes for this bushing out of the material I have on hand: mild steel rod. Then I gave it some serious thought (and this is why I need to write more often), and realized that no, the bushing is under practically no stress in any direction, and can be made from mild steel or aluminum, or whatever works. For some reason I was thinking it had to be made from 4130 chromoly steel, which is very strong, but also expensive and likely to be harder to machine. It's nice to occasionally make little positive discoveries like this.

I still have to wait for the 4130 tubing to arrive so I can weld everything together. But at least I have a project I can work on now, while I'm waiting for things to arrive.

Posted at 11:09 permanent link category: /charger


Sun, 18 Mar 2018

Building a Biplane: The Landing Gear Problem

Marquart Charger forward view

The Marquart Charger is generally very well designed, but one of the problem areas over the years has been the landing gear. Thinking about this problem is years ahead of where I am in the build, but it's been bugging me for a while, and I wanted to get some ideas for a fix before I start building any fuselage pieces.

The Charger landing gear is a cantilever design that uses a box-frame leg, which pivots at the lower outside corner of the fuselage frame, and operates a sort of rubber shock absorber/spring under the passenger's feet. It's very clean-looking, since it only has one "tube" going down (most small biplanes have several tubes, forming a kind of 3D triangle).

Because of this single tube design, the pivot point for that tube has to be very strong. In practice, it hasn't ended up being quite strong enough. On top of this, the rubber shock absorber starts out a little bit too stiff, and only gets stiffer with age. The result of these problems combine to show up as cracked fuselages near the landing gear attachments.

Finding a Solution

The problem really breaks down into two issues:

  1. The rubber shock absorber is too stiff, and doesn't provide an effective energy buffer for the gear
  2. The fore-and-aft attachment points for the gear are pretty close together, meaning that braking and rolling-resistance forces are very high

Fortunately, there is a reasonably simple solution for problem #1, which is to replace the rubber spring with a different material. Univair sells the SK-35 Belleville spring kit for the Ercoupe, which is a drop-in replacement for the rubber donuts (also from an Ercoupe according to the Charger plans), and which by all accounts provides a much smoother ride.

The solution to the second problem, though, is one which required (and still requires) a bit more thought.

Marquart Charger gear attachment diagram
The Marquart Charger gear attachment pivot

To this end, I solicited help on the FATPNW Facebook group, and got a couple offers. I just finished up meeting with one gent, B. (to be named later if he is amenable), who discussed the gear situation with me.

A surprising amount of our time was spent getting the existing structure adequately described for him. It's far too easy for me to forget that I've been thinking hard about this subject for a long time, and how clearly I have it visualized. For the purposes of this discussion, I will refer to the part numbers where they exist, from the diagram above.

The chief problem we ended up on was that the -479 part is a little bit too narrow where it meets the fuselage. In the left-hand section of the diagram, you can see the -479 piece edge-on, and you're looking from the side of the plane in toward the inside of the fuselage. In the right-hand section, you see it looking forward. This setup is about the same for both the forward and aft gear attachment points.

Looking at the red arrow on the diagram you can see how the faded red line is a straight line from the surface of the vertical tube to the beginning of the curve around the bottom of the -479 piece. The plans show that divot inward, and B. said that this is likely to be an area of high stress. Flattening out that curve, and making it more tangential to the vertical tube's surface will help to remove that stress concentration. Doing a similar change on the right-hand side of -479, do smooth the transition between -479 and fuselage was also recommended, although it's not drawn in.

The other change he suggested was to increase the length of the strap (the red faded shape on the left side of the drawing), to increase the welded area that attaches the gear pivot to the fuselage.

Charger gear leg diagram
The Charger gear leg

We discussed and discarded a number of other ideas:

These ideas were either pointless (changing part thicknesses), or too complicated (moving the gear attach points).

Generally, the idea he had was to increase the amount of material bearing the stress of the landing gear pivot. As he said several times, it's pretty intuitive. Adding a little bit more steel, and a little bit more weld area, can pay big dividends. It was a gratifying conversation in a way, since that was the idea I'd had as well, but I didn't know if it would negatively impact the stresses in the fuselage in some non-intuitive way.

I particularly like that his proposed solutions amount to a couple ounces of additional weight at most. They're not big changes, but should make a difference in the longevity of the landing gear (assuming I understood the problems correctly; a possibly big "if").

Posted at 14:32 permanent link category: /charger


Tue, 27 Feb 2018

Building a Biplane: The Saga of the Spars

Imagine for a moment that you're embarking on a decade-long project. You know the rough order in which you're going to take your steps, but not how much time each step will take, beyond very rough "Maybe two years?" type estimates.

That was me last June. I made my first wing rib, and started on the long and repetitive task of making a big stack of ribs. I knew from others' experience that the wing spars would take a long time to actually ship. Most people seemed to place an order, and see the shipment about 6 months later. So, I figured, I would order my spars early, and maybe they'd arrive about when I'd be finishing ribs, in the December-ish time-frame.

December has come and gone, and I'm still slowly grinding out ribs, so that timing was inaccurate. Un/fortunately the timing of finishing ribs was the least of my worries.

I placed the spar order with Wicks Aircraft Supply in June. The price was a bit steep, at almost $800 for eight 11- and 12-foot spars. Still, I'd understood that Wicks was the best place to buy aircraft grade Spruce, and I didn't get into building an airplane thinking I'd do it for free.

They didn't provide a shipping or order completion estimate when I ordered, but I knew it would take months, so I put it out of my mind, and worked on ribs.

I got a phone call around September or October with an update on my order: they were still trying to locate sufficient wood of a high enough quality. They knew I was waiting, and asked me to be patient. I said I was, and thanks for the update. I figured we were more or less on track.

Then I got another call on December 1st: we're unable to find enough wood to fill your order. In fact... we're getting out of the Spruce business. Sorry. Good luck.

Wicks explained that they'd had multiple large shipments of Spruce come in, and they simply couldn't find enough high-quality wood to fill the orders they were getting. They were taking a bath on wood orders, so they decided to cut their losses. They implied heavily that the supply of Spruce was simply inadequate to the demand, and that I was probably out of luck getting spars anywhere.

I was, to put it mildly, discouraged.

The next thing I did, though, was go check out the Aircraft Spruce wood selection (I found it ironic too, that Wicks was supposed to be a higher quality supplier of aircraft Spruce than Aircraft Spruce the business was). They listed everything I wanted as "in stock," which seemed like a potentially good sign. Previous experience with them had shown that they were generally on the ball, and if they listed something as in stock, that's because it was.

I also checked a few other options. I called up Steen Aero Lab to see how much they'd want to for a set of laminated spars. A week later, I got the answer: $2300 or so. That seemed like a choice of last resort to me. I contacted an aluminum extrusion company to see if they'd be willing to make a custom extrusion, and how much it would cost, but never heard back.

It seemed like Aircraft Spruce was the best choice, so I called them up and asked them what their expected leadtime was on a set of spars. They went into consultation with their wood department and mailed back a few days later: two to three weeks. Awesome, thought I, and placed the order with them on December 6th.

I'm sure you can imagine my surprise when, 3 weeks later, I had not heard from them. Around the 30th, I called and asked for an update. They eventually got back to me: my shipment would most definitely go out on Janury 30th. Ok, sure. I would have been perfectly happy if they'd told me it would take 3-6 months to ship the order, but they said 2-3 weeks, so now I was unhappy. This is pretty basic business logic: don't make up numbers you can't meet.

It was with a sense of resignation that I watched the 30th come and go without contact from ACS. A week or two into February, I put in a "customer contact" form asking for more information and expressing my displeasure at being misled on multiple occasions. Don't lie to me, I explained, just tell me you don't know, if you don't know.

Several days later, I got a terse response saying that my order would be shipping the next day. Of course, that would put the shipment arriving at my house in the middle of a week I'd be away in Hawaii for my parents' 50th wedding anniversary, so I had the weird role reversal of calling ACS and asking them to delay my order by a week. With visions of one of the spar horror stories in my head, there was no way I was going to allow delivery without me being there to personally inspect the package before signing: one of the Biplane Forum members had his spars delivered in similar circumstances, and discovered that a forklift operator had put the forks clean through the shipping container, destroying the spars inside. He turned the shipment back and had to wait another 6 months for replacements. I don't expect that to happen to me, but for a shipment of this price and fragility, I want to minimize all the risks I can.

And that puts me at now. The spars are supposed to arrive today, in the afternoon. Look for an update below on the actual condition they arrive in. Hopefully they'll be in perfect condition and give me a kick in the pants to get on with my slow-paced rib building so I can move on to building some wings already.

And the expected update:

The spars did, in fact, arrive with damage to the box. The UPS delivery driver and I cut away the outer box and determined that the abrasion damage was only to the outer layers, and didn't go past the second layer of three of sturdy cardboard around the wood. Without realizing I had an choice other than "Send them back" and "Accept the shipment without reservation," I signed on the form, since it looked like the damage was light, and the wood had been spared.

I turns out (news to me), you do in fact have another option. If you suspect that a shipment has been damaged, but it's not so bad you want to send it back, you can note your concerns on the form, something like "Possible hidden damage." Then you sign with that on the form, and UPS is still on the hook even though you've accepted the package. Future damage claims will be substantially easier. Of course, I learned this after I signed for the damaged package.

In any case, a day later, I was able to take the package apart and inspect the contents. The damage did not indeed go further than the cardboard, and the wood appeared to be intact and without problems. I am now the slightly nervous owner of eight lengths of nearly perfect vertical grain Sitka Spruce, which will be carefully stored against the day that I actually finish making ribs, and can move on to the next phase of building a wing.

Posted at 09:15 permanent link category: /charger


Wed, 10 Jan 2018

Building a Biplane: The Glue/Temperature Problem

Back in October of 2017 (it seems so long ago, and yet so recent), the temperature started dropping, and I kept on building ribs at my slightly glacial pace. Soon, it was 40° F in the garage. This poses a bit of a problem. BTW, Deep Nerd alert -- if you think the idea of a technical discussion of epoxy mixing sounds boring, you can probably skip this one.

System Three's T-88 Structural Epoxy lists the minimum application temperature of their glue at 35° F. I can't find the reference right now, but I have read from System Three that the epoxy will cure at any temperature, but every 18° F rise will halve the curing time. So it'll cure at 40°, but it might take a while.


Test joints; the one on the right has failed badly

Along with some of the ribs, I would make up these little test joints, to be destroyed later to see if the glue holds up. I've made perhaps a dozen of these as I've made the ribs, and never had a problem with any of them. Until December.

I tested the joint I'd made along with ribs #22 and 23, the little rib nose pieces that go in front of the gas tanks, in the upper wings. The test joint separated neatly at the glue line with a mushy-jam kind of feeling as they pulled apart. What it should do is tear out the wood somewhere. This was bad news.

My first suspicion was that I was mixing incorrectly. A variety of people on the Biplane Forum had told me that T-88 is tolerant of sloppy mixing -- that is, mixing ratios that aren't exactly 1:1 by volume (or 1:0.83 by weight). I was mixing up these little tiny batches, though, and I had quickly decided to use a scale to measure my mixture when I discovered that the bottles were uneven after a few batches of measuring by eye.

So, I got more or less the cheapest little scale I could find. It measured down to 0.1g, which seemed like plenty of accuracy to me: a tenth of a gram is pretty small. I happily mixed away with my little drug-dealer's scale, until that failed joint happened.

Then, I stopped what I was doing, and started testing my technique. First, I deliberately mis-mixed a batch at a ratio of 1:0.9 by weight, which I was sure I'd done multiple times on previous ribs ("It's very tolerant of ratio errors," they said). I tested these joints after varying degrees of being left out in the cold, and brought into the relatively warm house, and they failed in the glue line regardless of temperature. So I definitely have some suspect ribs in the stack, though I hadn't been taking notes on my glue weights until after I discovered this problem, so knowing which ribs are affected is impossible.

However, compounding my confusion was the fact that I'd tested a handful of joints previously, and they'd been good, and probably (or maybe not?) at least one of those joints was made with a poor mixture. So now I was unsure. The house isn't kept ridiculously warm, perhaps 66-68° F, and that only for the parts of the day we're actually home and awake. Perhaps the temperature just wasn't getting high enough.

One thing I was reasonably sure of was that the low-resolution scale was causing problems. On my test joints where I mixed at 1:0.9 instead of the recommended 1:0.83, part of the reason my mixture was so off was the scale. I had poured out 1.0g of resin, which meant I should have been aiming for about 0.8g of hardener. Ideally, I'd pour out exactly 0.83g of hardener, but the best I could hope for was 0.8, with only decigram resolution. As I was carefully squeezing hardener out of the bottle, the scale jumped from 0.7g to 0.9g. How much over 0.9g? Who knows! I shrugged, and made my test pieces, figuring this was a valid test.

So, the next order of business was a more precise scale. I located another inexpensive scale, but this one read down to milligrams (0.001g). I didn't need that last digit, but better too much precision than too little. This scale, at least, wouldn't jump from 0.7g to 0.9g without ever showing 0.8g.

Of course, the larger the batch of glue you mix, the less important it is to have a super-precise scale. Unfortunately, for now, I need to mix a roughly 3.0g and 2.52g batch, and it doesn't take much variance on the hardener to take the ratio far from its ideal place. The new scale gives me much greater confidence that I've got the mix right. My tests with the 1:0.9 batch are too worrying to allow that kind of mixing error to continue.

Discussion on the forum started to convince me that what I really had was a temperature problem. That my summer ribs were all good because they'd cured in a 60-70° F average temperature, and the cold Northwest winter was causing my problems.


The Easy Bake™ Rib Oven

I pondered my possible options, and decided I would make the grown-up version of that venerable child's favorite: the Easy Bake™ Oven. I cut up a sheet of plywood I had sitting around from some previous project, and scrounged around in my project supplies to come up with a pair of light bulb bases, an electrical box, a short electrical cord, a discarded computer fan, and some Romex to connect it all together.

In two days of casual bodging, I put together my Easy Bake™, and after scientifically determining that it wouldn't catch itself on fire by leaving it on overnight with nothing inside, I stuck a temperature probe inside the box, and measured 82° F. Perfect.

My next test was to make up some test joints and put them in the Easy Bake™ to see if that made a difference. I mixed up a very close-to-perfect batch of T-88, and assembled three test joints: one with a thick layer of glue and no pressure before stapling, one with heavy glue and normal pressure, and one with a normal thickness of glue and normal pressure. I had read over the T-88 FAQ again, and developed a vague fear that I might be suffering from glue starvation, where there's not enough epoxy in the joint to form a good bond. A starved joint sounded like it might fail in a way similar to the failures I was seeing. All three test pieces went into the oven for a days' cure.

The heavy-glue, no-pressure joint failed on one side but not the other, which suggests that the glue at least isn't the issue (and one day may not have been enough to fully cure it, as System Three says it takes 72 hours at 77° F for a full cure). The heavy-glue, normal-pressure joint broke exactly where it should, in the wood. Likewise, the normal-glue, normal-pressure joint failed in the wood, confirming that at least my normal technique wasn't causing problems.

So, with any luck, I now have a system which will result in full-strength ribs going forward. Doesn't help me with past ribs that may or may not be strong enough, but I can deal with those later. At least now I can bake my ribs to be sure they're getting the cure temperature they need, with a mixture of glue that's as close as humanly possible to perfect.

Posted at 22:34 permanent link category: /charger


Fri, 01 Dec 2017

A Minor Setback

I have been slowly but surely building up my stack of ribs in the Charger project. I'm up to rib #33 (out of 44), with each taking about 2 hours, plus about 6 hours to produce all the parts to support every 15 or so ribs. So, slow but steady progress.

The wings are built up of ribs, some metal bits and pieces, and spars. These are the long pieces of solid wood that stretch out from the fuselage, and provide the foundation of the wings. The plans call for aircraft-grade Sitka Spruce, which basically means wood with straight, tightly-packed grain and a minimum of flaws and defects. The spars are a bit under 12' and a bit under 11' long, depending on whether you're talking about the top or bottom wing.

In June of this year, I placed an order for my spars with Wicks Aircraft Supply. I knew they would take months to actually ship, so I figured that by the time they shipped, I'd probably have all my ribs done, and everything would move along hunkey-dorey.

We have encountered a small snag in that plan.

Wicks called today and said that, effectively, they can't get high enough quality wood any more, and they're quitting the Spruce spar business. My order is canceled, so sorry, have a nice day. They were very kind about it, and I understand, but this still leaves me in a bit of a lurch. I still need spars.

Fortunately, I have some alternatives:

  1. Laminated Spruce: just because Wicks can't find any 12' lengths of aircraft-quality Spruce doesn't mean there's no aircraft-quality Spruce to be had. It just means that finding it in 12' lengths is no longer possible. Shorter, smaller pieces of Spruce carefully laminated with high quality glue can make up a spar, and there are distinct advantages. Laminated wood is stronger (see plywood vs. a plank of the same thickness), fails more gracefully, and is not prone to warping. Because it's made up of a lot of small pieces of wood, you get to preview the wood for defects that would be hidden in larger pieces. I can do the lamination myself, which sounds both enjoyable and inexpensive. Steen Aero Lab sells laminated spars, and I have a request in to them for a quote.

  2. Douglas Fir: Fir is about 20% stronger than Spruce, and is about 20% heavier (rough numbers from memory, don't hold me to them). I can't change the shape of my spars, that ship has already sailed unless I want to remake all 33 of the ribs and re-calculate a bunch of other stuff. So, I'd add 20% to the weight of the spars, which would amount to probably 5-10 lbs at a guess. Douglas Fir is much easier to find in plain old lumberyards, so I'd be looking at cheaper lumber, too.

  3. Aircraft Spruce still sells spars: the cost is about the same as Wicks, and at least on the phone when I asked, there was no suggestion that they would have trouble finding the wood I was asking for. I have a request for a quote in to ACS to see how long their leadtime would be, and what shipping would cost.

So I've got some options, but it's a bit disappointing that my well-planned order fell apart like this, putting me further behind on my construction plans than I already was. We'll see what Steen and Aircraft Spruce have to say, and that may determine the way forward.

Posted at 15:21 permanent link category: /charger


Sat, 29 Jul 2017

Building a Biplane: Boring Video Edition

If you're following along this journey because you're maybe going to do it yourself, or you really want to know the nitty-gritty of how an airplane is built, have I got a treat for you! If you're just kind of interested but not that interested, probably skip this one.

Posted at 23:34 permanent link category: /charger


Tue, 18 Jul 2017

Building a Biplane: Makin' Ribs

With the completion of rib #5 (of 44) last night, I've entered the phase of building where things get kind of boring. It's still enjoyable and gratifying work, but it's hard to describe to anyone else in a way that makes it sound interesting. So, when you get bored half-way through this description, I won't be offended if you wander off to look at cat pictures or something.

I've built up my kit of parts, and am progressing through the ribs. I'm building the -293 (the part number, pronounced "dash two nine three") ribs now, and will be discarding the first two for various technical faults. There are sixteen -293 ribs, and the Skybolt builder I visited in Oregon advised me to have at least one spare built up and on hand, so I have 14 to go.

Each rib takes about an hour and a half to build. I had hoped they'd go quicker with the kit ready, but I guess that's what they take. I don't want to rush, since rushing just leads to mistakes. It's a ten-year project anyway, so a difference of half an hour per rib is absolutely inconsequential.

The procedure I'm following now is as follows. First, I put the nose piece and all the truss pieces into the jig. Some of the pieces need to be trimmed slightly on the sander, but some just drop into place. Then, lay out the gussets for each joint. Mark the gussets' positions with pencil so I know how far out to put glue. Wipe down all the sanded plywood surfaces with MEK to remove dust. At this point the compressor is already at pressure, and the stapler has a full load of staples. The two gussets that overlay the nose piece need to be trimmed beforehand, but the remainder get trimmed with the router once they're stapled down.

Now, mix the glue. I was trying the "squirt out N equal lines of each" method, but kept coming up short on the hardener, as marked by uneven levels in the bottles. T-88 epoxy should be mixed 1:1 by volume, or 100:83 by weight. Since my volume attempts were noticeably off, I switch to weight, and have been happy with the result. The $9 drug-dealer scale on Amazon is perfect for this task.

Once the glue is mixed, you have 30 minutes before it starts to thicken and get hard to work with. That's just enough time to carefully coat each joint in glue, coat each gusset in glue, position the gussets, and staple them down, working one joint at a time. I'm careful to align the gussets such that they don't impinge on the space for the spar (the rectangular openings near the ends of the rib). Any excess glue I see gets wiped up with a disposable rag.

Once all the gussets for a side are glued and stapled, the rib comes out and goes over to the router table, where I trim down any gussets that overlap the outer edge of the rib. Have to wipe the glue off the joint first, so as not to gum up the router blade too badly. The rib is then flipped over, and the same MEK wipe/mark/glue/staple/cleanup/router routine happens again on the flip side. The single rib in the photo above has just come off the router table after its flip-side gussets were attached.

So it's not the most challenging of work to do, but it is definitely satisfying to see the stack of ribs grow one by one. I need to figure out a better storage solution than just stacking them on a shelf, but that's what I've got for now. Before you know it, I'll be swimming in ribs, and trying hard not to damage them.

Posted at 14:05 permanent link category: /charger


Tue, 04 Jul 2017

Building a Biplane: Building Your Own Kit


A completed rib for the Charger

The Marquart Charger is typical of most amateur-built biplanes in that it doesn't have a kit. You get a bunch of sheets of paper with ink marks on them, and your wits. If you're really lucky (I'm not), there's some variety of build manual to give you a hint how to do things.

I am very fortunate that I have the Biplane Forum (where I'm known as IanJ) as a backup, and between reading existing threads and information, and asking questions, I get about as much support as if I had a designer I could call. Unfortunately Ed Marquart, who designed the Charger, passed away in 2007, so calling him is not an option (though I understand he was a very helpful guy).

Because there's no kit for the Charger, I get to make everything myself. However, that doesn't mean that I completely ignore the kit world. For instance, I can take some of their ideas, and make them my own.


Eight little rib noses, ready to go

This weekend's idea that I'm taking for my own is to produce my own mini-"kit" of parts to build ribs. It's hardly revoluionary, but it will give me a little taste of what a kit-builder's life is like, as I take nearly-finished parts from their labelled containers, and put them together with a minimum of fuss to crank out completed ribs. It also satisfies my slightly OCD heart to be able to bag and tag parts and keep them perfectly organized to facilitate the task.

So, I've cut out 15 of each truss-piece, and 15 top- and bottom-pieces, and will soon have 15 nose pieces cut out, and enough 1/16" plywood gussets cut to make the remaining 15 ribs of the first type (there are around 60 ribs in total, so this is nowhere near finished, but it's a start). Then I can knock out a rib an hour, or maybe even faster. It's debatable if this actually saves me much time in the grand scheme of things, but it's a little bit more efficient to mass-produce some parts, then use them later.

As I've finally been able to tell people, "I'm building an airplane," and now, finally finally finally, it's actually true.

Posted at 22:40 permanent link category: /charger


Tue, 27 Jun 2017

Building a Biplane: Finally, A Real Airplane Part

As the clock rolled past 10 pm, and bedtime was looming, I mixed up the second batch of epoxy, and set to work. I had been working for the last hour and a half making the final preparation for finishing off the first Real Airplane Part of my build: a wing rib.

It all started, for the purposes of this story, on March 19th, 2017, when I placed the order for the 1/4" capstrip and aircraft-grade plywood that would make up the majority of the materials needed to build the wing ribs. Wing ribs are the airfoil shaped pieces that are parallel to the airflow over a wing, and give it the correctly shaped cross-section. There are around 60 in the Marquart Charger, in a variety of forms.

I ordered the capstrip knowing it would take a while to arrive, but I didn't imagine it would take as long as it did: almost exactly three months. Still, I had a plan to use up some of that time productively, as previously documented. I would build up the templates and jigs that would enable me to build ribs once I had all the materials on hand.


In-progress rib jig

After about two months of waiting with gradually decreasing patience, I went out to a local hardwood retailer (with huge eyes, and massive self-restraint) and bought a few sticks of wonderfully tight-grained Douglas Fir. This was cut down into 1/4" sticks which would fill in until the proper Spruce arrived. I used this and the previously created nose template to make up a gloriously fake rib jig (seen above in progress).

Finally, the day arrived, and my order from Wicks arrived: a selection of mahogany plywood (1/16", 1/8" and 1/4" thick -- imagine for a moment 1/16" plywood, that's less than 2mm thick), my 650 feet of 1/4" capstrip, and some random spruce blocks that will doubtless come in handy in the future.

Still, I couldn't leap in like I might have wanted to: my schedule was packed to the gills with other projects -- I hadn't known when the wood would ship, and wasn't going to put my life on hold waiting. So, I had some theatrical engagements take up my evenings, and a weekend of lovely weather saw us painting the outside of the garage rather than me working on ribs inside it.

The process also ended up taking longer than I had expected, with more setbacks than I had anticipated: cutting plywood for the gussets with a razor knife, the tip of the knife broke off without me realizing it, and I spent some very unproductive minutes scraping unsharpened steel against the final ply; mixing the glue demonstrated that my glue-dispensing skills are rusty (it needs to be mixed 1:1 to work right); cutting template pieces out of cardboard proved useful, but took more than an hour; and the staple gun started misfiring until I realized I'd never oiled it (it's brand new). I had cut the sticks for the rib last week, so I spent two evenings this week just getting the gussets and staple gun sorted out.


Completed wing rib

Still, last night I was able to hold up a completed wing rib for the first time ever. It's only the first of about a zillion pieces, and more head-scratching than I care to think about. I may not even use it in the plane as I answer my own questions about glue mixture and staple placement. But it's done, and I've taken the first real step on the path to building a biplane.

How do you eat an elephant? Well, one bite at a time.

Posted at 16:13 permanent link category: /charger


Mon, 01 May 2017

Building a Biplane: The First (Almost) Real Part

I have just finished the first physical part that is very nearly a real airplane part, in my nascent project to build a Marquart Charger biplane.

As I mentioned last time, I'm going to start on the ribs. In the intervening time, I've gotten the floor painted, built my (first) worktable, and started moving stuff in to the garage. I also acquired my welding equipment, but that's a topic for a different post.

The worktable, conveniently, is built with a 3/4" MDF top surface. It's 33" by 8', so there was a convenient strip of MDF left over from that project. I've now measured and marked the airfoil coordinates onto a piece of that MDF for the rib jig, but lacking 1/4" wood, had to set it aside for a moment. I am not going to print out the rib from the plans at full scale, since I've already found a couple of minor drafting errors, and will work off the printed measurements instead. To some extent, the exact dimensions of the internal structure of the rib isn't very important, and long as it's consistent between ribs. I will naturally be striving to be as close to plans as possible, but there's likely to come a time when I'll have to choose between how it's drawn, and how it's measured.

In any case, lacking the capstrip, but having the MDF, I decided to proceed with the rib nose template. This ended up being (like probably every single thing I do on this project) easier said than done.

First, I grabbed a sheet of graph paper, and measured out the coordinates of the airfoil, as far as the nose would go, and drew it out with the help of a somewhat precariously grasped steel ruler to approximate the curves between points. I also discovered that my graph paper grid was neither quite exactly straight nor quite exactly 1/4" per square. So, marginal. Still, it looked nice on paper.

One problem quickly presented itself, though: what was up with these notches? The nosepiece is notched with 1/4" by 3/8" cutouts, for stringers that run along the leading edge, supporting the sheet of thin aluminum that makes up the leading edge of the wing. But the stringers called out in the later plans were 3/8" by 3/8", not 1/4" by 3/8". Hmm. I asked the question on the Biplane Forum, and the general consensus was that since I was planning on plywood leading edges anyway (generally stiffer and more resilient than aluminum, though probably a touch heavier than the very thin Al called for in the plans), I could just omit the stringers. The plywood would be plenty strong enough to support itself without stringers. Also, bonus: one less order of expensive spruce to buy.

So, that bit of mental gymnastics out of the way, I decided to draw out the nosepiece in CAD, so I could be 100% sure I had the airfoil shape right. One of the comments in the thread pointed me to airfoiltools.com, which gave me a plot of 80 points of the NACA 2412 airfoil that the Charger uses. I saved that list of points to a file (here if you want it), and wrote a quick Python script, which turned into a quick FreeCAD macro (here, but you'll need to modify the file location to suit your installation) that draws out the airfoil for you. If there's enough interest, I may update that macro to handle other airfoils, but at the moment it just does the one.

Anyway, I now had a way to quickly and easily draw a correct NACA 2412 airfoil in FreeCAD, but making the nosepiece shape was more complicated. I ended up spending about 5 hours farting around trying to make FreeCAD do what I wanted, before I figured it out (email me if you want details, it's too far into the weeds for this post). Of course, now that I've figured out how to do it, I can make the nosepiece in about 10 minutes.

All that work resulted in this PDF file of the drawing, which, if you print it out at 100% scale (it slightly overprints on Letter size paper, but the nosepiece is all there), gives you a 100% size, accurate drawing of the nosepiece.

Finally, finally, I was ready to make the MDF template. I carefully cut out the template from the printed sheet, and glued it down to the MDF. The jigsaw took away most of the excess, except on the inside edge, where I couldn't quite turn quickly enough, and had to leave a big chunk. I busted out the files, and started the long, tedious work of filing down the last little bit of MDF to exactly match the printout.

Two hours later, I was the comparatively proud owner of one rib nose template, lovingly handcrafted out of artisanal fiberboard.

This will be the master template, from which I'll use the router to make a copy in MDF, which will then be used to cut out actual rib nosepieces. The master will get locked away somewhere safe, so that if I mess up the copy template, I can make another one that will hopefully be exactly the same. It's a good theory, anyway.

It's nice, finally, to be making airplane parts (or very nearly airplane parts). The build feels like it's almost started. Only 11 years after I first started thinking seriously about it. As occurs to me every time I ponder it, if I had only started then, I'd be done by now!

Posted at 22:32 permanent link category: /charger


Categories: all aviation Building a Biplane bicycle gadgets misc motorcycle theater