HS Aft Spar

Wed. 31st October 1hr 30min

Clecoed the two HS-903 aft spar halves together with the HS-906 aft spar doubler and match drilled. Clecoed on the eight HS-912 outboard hinge brackets and match drilled.

Calm returned this evening. After the discouraging mess-up yesterday, I had decided to take an evening off but my wife was out at her French class so, as she couldn't benefit from my company, I decided that it would be better to get straight back into the saddle. This piece of work was a good choice as it simply couldn't go wrong. And so it turned out.

The powder coating on the hinge brackets looks great so I don't think I will be painting them. The coating makes getting a 3/32 cleco into the holes somewhat difficult but a 1/8 won't hold them.

Day one, page one

29th October 2007 1hr 30min

Assembled the HS-902 forward spar halves and HS-907 forward spar doubler, match-drilled them. Fabricated the HS-908-L&R attach angles.

After 18 months of planning and over three months of hard work on the shop, I finally got started on building the plane today. What a let down! I had just come from a chastening experience on building the Vans Training Project, so I was aware how important it was to read the plans carefully. Despite this, I really messed up and find myself today ordering a replacement part from Vans.

Step one was to cleco and match-drill the doubler plate to the two forward spar channels. Right away, I couldn't reconcile the instruction to cleco the positioning hole in the attach angles to the ONE matching hole in the appropriate position on the spars. It seemed to me that there were two matching holes, not one. I scratched my head for fully 10 minutes before realising that, while the spar halves are identical left and right, they are not symetrical top and bottom and the one matching hole was on the edge facing away from me. If it takes a similar length of time to figure out other simple instructions like this, I better plan on an eight-year build instead of four-years.

Then I came to make the actual attach angles. Completely failing to read the plans properly, I marked out the first one to the finished width of 1 21/32 inches, and marked an angled line from this point on the edge of the section back to the narrower dimension of 1 1/2inches for the vertical back plate AND CUT IT! Then I realised that the flange is too long and needs to be shortened back to 1 5/32. When I did this, because the part narrows as it goes back along this angled line, the width at the new edge is now shorter by 1/8 inch. Clearly, this will not give me the designed edge distance for the fasteners that go into this horizontal flange. Whether or not it will give me enough edge distance, I was simply too sick to go and measure. Either way, I am not happy with it.

Not content with that, I noticed that the height for the vertical flange of the angle comes really close to the rounded lip of the angle stock. In fact, shortening it back to the designed height would not fully remove the rounded edge profile of the original section. Now it hits me that the original section has flanges of differing depths. It is 2in X 2 1/2 in and I attempted to cut the longer flange of the attach angle from the shorter flange of the angle stock! By now, I am thinking of putting a shot of my best Irish single malt whiskey in my coffee (a once-in-a-decade event) while I try to calm down. It was the nausea that prevented me from wasting good whiskey. I went ahead and finished the part anyway, just for the experience but I am certain it won't pass muster and will have to be redone.

Anyway, picking myself up, I moved on to the second angle. This time I marked it out correctly and cut it on my new band saw with plenty of allowance for smoothing. Maybe a little too much allowance because I am tempted to go back and take a little more metal with the saw. Working by hand with no guide fence this time, I get too close to the line at one point near the ? Disaster! In order to get a straight line, I have to remove all the metal down to the same closest point using a file but when I do, I am on the line and there is still a gouge mark to be removed so I have to take out the line itself, which I marked up in such a way that it needed to be left on the metal. When I remove this gouge, the part sits within the lines on the plan and is about 1/64th too small. Perhaps this is not critical on its own but, combined with the 1/8th error on the other attach angle, there is no getting away from the fact that I have to make a replacement set.

The most upsetting part of this is that my confidence in my workmanship is badly shaken and for the first time, I am beginning to doubt that I can produce the necessary standard. As a result, I have decided that I am going to firstly take a night off and get plenty of sleep (I think tiredness may have been a contributory factor). Secondly, I am going to constantly read and re-read the relevant section of the manual as I go on with the work. I keep a photocopy of the manual with the small preview plans in the office so I can look over them at lunchtime. I will underline each instruction in the manual immediately before I carry it out as well as every part number, material and fastner call-out, note, etc on the relevant plans pages. Finally, I am going to experiment with a clearance of about 3/64th outside the marked dimension when I am making cuts. The original clearance I used on the attach angle was closer to 3/32, which was a bit too much for finishing though, as it turned out, far too little to be removed with a band saw.

So today (30th Oct) I found myself on the Vans website ordering a replacement 5 inch length of angle. It only costs $5.90 and should be easy to ship. To maximise the use of the shipping cost, I also ordered some riveting tape that I wished I had bought at the beginning and I included the smallest quantity of tank sealant available, for use on the trailing edges of the HS and rudder

The Training Project

I included the Vans training project with the order for my empennage and yesterday, having declared the workshop finally ready (not "finished", mind you), I got to work on the project at last.

The project has two elements; a section with two plates and an angle that have to be rivetted together and a short section of a mock aerofoil comprising channel section spar, reinforcing plates, ribs at each end, four stiffners, top and bottom skins, which are curved and pop-riveted at the leading edge like a typical rudder and double-flush rivetted through a wedge extrusion at the trailing edge. The whole thing measures about one foot square when finished.

The instruction say "if the finished project ends up with various dents, holes and other fumbles. . . good! It means you haven't been afraid to try and you are making your "beginners mistakes" here instead of on the actual airplane". Well, I'm glad they said that because otherwise, I would be profoundly depressed today. I can't get over how many mistakes I made. Mainly for my own benefit, here's a list of the main ones:

1. Not reading the plans fully. I put all the stiffners in facing outboard, with the result that the top and bottom flanges were meeting when the section was fully assembled. I drilled two of them out and switched them - unacceptable on a flying aerofoil in my view, as they had been match-drilled to different positions on the skin. However, it demonstrated the exactness of the matched hole system as the stiffners sat perfectly in their new positions. I also missed the fact that the reinforcing plates on the spar were to be fastened with size 4 universal head rivets and did not need to be dimpled. I actually resorted to using flat heads in the dimpler to fix this one by flattening out the dimples. I imagined that far away in Aurora, Oregon, a very distinguished kitplane manufacturer was wondering why he had a sudden urge to cringe! Sorry Mr V. On an actual airplane, that would have meant replacing the spar. I also managed to misread the rivet call-out for the first section of the project and put a universal head into a dimpled hole intended and already dimpled for a countersunk head rivet. At least, I am now quite practiced at drilling out rivets and I do not believe I damaged any of the holes while doing this - a skill that will be used extensively, I am beginning to realise.

2. Deburring holes to be dimpled. The Isham inc. 'manual' makes the case that deburring is over emphasised in the homebuilt community. You can see the effect of deburring very clearly on the underside of a dimple because an edge that should be nice and sharp is clearly 'shaved off' when that edge is turned up and formed into the sharp end of a 'funnel', as in a dimple. This robs effective depth from the finished metal. I have decided to look for a gentler way of removing obvious burrs from holes to be dimpled and leave it at that.

3. Bad rivetting. I encountered every sort of problem with rivetting. I have under-driven rivets, over-driven rivets, slanted heads, off-center heads and hardly a universal factory head without a 'smiley' grinning stupidly back at me. I also seem to have let the mushroom set float off two bucked skin rivets, leaving a slight dent in the skin beside the factory heads. Like I said, I am now very experienced in drilling out these offending examples. I don't even know what I did wrong in some cases. As most riveting is done either in the horizontal or vertical planes, I thought of getting tiny spirit levels to glue to the top, side and 'breech' of my gun as a guide to lining up the gun properly until I can judge it better by eye. One fault (slanted heads) was not really my fault. It happens when back rivetting close to stiffners. The white plastic collar on the back rivet set is quite thick and when the angle used for stiffners is small and the holes are close to the web, it prevents the actual head from sitting squarely on top of the shank. As rivetting progresses, it slides off, leaving a head shaped like a ski jump. Holding the collar firmly with my free hand improved things but I think I will grind off a flat section on one side of this collar when I encounter similar small angles in the elevators.

4. Poor drilling. I managed to get an elongated hole in the trailing edge wedge. This happened while using a drilling guide to achieve the correct angle. Maybe I was concentrating too much on the guide.

I didn't have all the materials available (who has mild steep plate, hardwood blocks and bits of 4 X 2 lumber knocking about the place?) so the project is not yet finished. I can see that the bucking/dimpling bar that I am supposed to make from the steel plate will be necessary on every small rib so I will go to the trouble of making it exactly according to plans. I simply improvised the rest.

All in all, it was quite a chastening experience and far from the simple limbering up exercise that I thought it was going to be. Hopefully, it will turn out to be the best $35 that I will spend on the entire project.

Primer

Most people think that aluminium, from which the RV-9 is built, does not corrode. Unfortunately, this is not quite the case. It is true that pure aluminium reacts with air to form a protective coating of aluminium oxide which is highly resistant to corrosion. In this way, pure aluminium does indeed protect itself. However, the alloys of aluminium used in building aircraft do not have this feature. Some of the parts of the Vans range are coated at the mill with a thin layer of pure alluminium to restore this property. These are refered to as "alclad". The problem is that the coating is very thin and prone to scratching, leaving an entry point for corrosion. You can easily recognise alclad parts because they are covered in a light adhesive film of blue plastic to protect them from such scratches. At some stage however, this film has to come off. Then there is the large number of parts (including all pressed parts such as ribs and bulkheads and the lengths of stock sections used for stringers and longerons) that are not alclad in the first place.

Many US builders debate whether to prime their planes at all. Their warm dry climate may give them a choice but our atlantic climate gives us none at all. I don't know of any builder on this side of the pond who is not priming. I don't have hangerage for the plane sorted out yet and there is a distinct possiblity that it will be stored in the open for at least a period after it is finished. Consequently, it would not be sensible to leave corrosion protection to chance. That disposes of the first battleground of the 'primer wars', which is whether to prime at all or not.

The next question is where to prime. Well the counter question is why wouldn't you prime everything? The answer to that is three-fold; time, cost and weight. These are the main arguments quoted by those who do not favour priming. For me, it is a question of balancing these factors against the need for additional corrosion protection.

One option is to prime everything that is not alclad. Even on an alclad part, the cut edges are not protected. In addition, you are drilling holes through the metal for rivets and sometimes cutting a part such as a stiffner to length, thereby creating more edges. So you would have to prime these areas anyway. Then there is the possibility that condensation will form inside the fuselage and will gather in crevices in the structure. This is where corrosion usually starts. These crevices will normally be at points where two metal parts are joined so it makes sense to prime at all such points. You could, for instance, prime along the narrow strips onto which stiffeners will be riveted. This would ensure that the rivet holes in the skin would be covered also. Even alclad parts riveted to the skin (such as stiffners) have edges and rivet holes where the aluminium alloy is exposed. As they are so small, it would not be worthwhile to prime only holes and edges so in practice the entire part must be done.

At this stage, you have to ask yourself, what areas of the aircraft structure would not be primed under such a regine and the answer is that it would amount to the central areas of the skin that are not close to an edge or close to another part. I started out this blog with the position that I would adopt this strategy but now I feel that leaviing these areas would be more trouble than it was worth.

The rivets themselves have always struck me as being a likely place for corrosion to get a foothold and also the one place you would not want this to happen. The only way of protecting these areas is to 'wet rivet', meaning that every rivet is dunked in primer before being set while still wet. I am interested in giving this a try. If I can develop a workable method, I will use it. Otherwise a compromise may be on the cards.

The next issue is which primer or primer system. There is a variety of products on the market in a variety of formats, such as rattle (aerosol) cans. Some of the primers claim to be self-etching, meaning that they roughen the surface at the microscopic level in order to improve the adhesion of the primer. This is achieved by adding a proportion of acid to the paint, which evaporates away with the rest of the 'carrier' solution in which the paint is delivered. These acids are not easy to incorporate into paint without some other major compromise and I don't believe in them. Etching is a vital step and I think it needs to be a separate step on its own.

One thing that almost all the primers seem to have in common is that they are toxic so, in practice, they are trying to kill you. Among the names you hear are "isocyanate" (cyanide) and "zinc chromate" (chromium). These are not substances you stir into your tea. That is why it is so vital to have really good breathing aparatus and even skin protection.

The automotive world is moving towards non-toxic water-based paints. So far, the only line of products I know of for the aviation world that is water-based are the products produced by a company called Stewart Systems. These have the important added benefits of being very low odor and easy to clean up. In fact, I recently saw a demonstration arranged by Aeropicardie, who are the European Agents for Stewart Systems at a SAAC meeting. Now the SAAC meetings are held in a restaurant. Can you imagine the potential for disaster if we were using toxic smelly chemicals in a restaurant of all places. As it happened, I can vouch for the fact that there was a stronger smell off the excellent cup-cakes that were served with our tea that evening and nobody got sick either.

This feature is particularly important to me, as I am building in such a small space and I don't have space for setting up an elaborate spray booth.

The Stewarts website also mentions another important factor in primer technology, which relates to permeability. The function of a primer is to provide a substrate which is a perfect environment for a finishing coat. One of the important elements of this environment is that is easy to adhere to. For this reason, most primers are actually porous, providing lots of microscopic gaps for the finishing coat to get down into and grip. Inside an aircraft structure, we do not use a finishing coat, for weight, time and cost reasons. So the main reason for using a primer (corrosion prevention) is actually negated by having a porous primer, without the protection of a finishing coat, that does not, in fact, prevent moisture from reaching the metal underneath. Two-pot epoxy primers are an exception here as they are fully impermeable. However they score especially badly on weight, time and cost. The Stewarts product is called a "Primer Sealer". It does completely seal the meetal and is designed to be left without a finishing coat. Together with its non-toxicity and very low odor, that makes it the perfect anti-corrosion system for a homebuilt plane.

So that's it. Decision made. The plane will be fully coated internally with the Stewarts system, incorporating Ekoclean cleaner, Ekoetch etching compound (obviously) and Primer Sealer; all water based, low odor and non-toxic - and friendly to the environment as well.

The Kit Arrives!

10th October 2007 2hrs 30min
Checked the Kit Inventory

I got a call Monday from the shipping company (same one used by Isham inc. - Planetools.com - for the tools) to say they had a consignment for me. Vans ship C.O.D. so I had to pay the shipping charges as well as VAT (sales tax). This was expected and all the calculations checked out. Once again, the shippers cleared the kit through customs without charge and there was no trouble with the duty free status of the aircraft parts. By Wednesday afternoon, the truck driver was ringing from nearby for final directions and suddenly, there it was.


Strangely, I was not particularly excited; just very pleased. It was as if I had been anticipating this milestone for so long that it had begun to be confused with reality. So when it actually happened, it had lost some novelty value.


Anyway, I got the two boxes out to the shop and began the process of checking the inventory. The RV-9 Empennage kit is now a series of sub-kits, each wrapped separately and containing its own tiny inventory list. All the elements in each sub-kit are listed together on the master list so you don't have to go hunting through 4 pages for every item.



Already, in the very first work session, the project has drawn blood. Some of the edges of those skins are sharp and the first thing I knew was that there was blood on the blue plastic. At least the plane is metal so I don't have to worry about the wooden plane builders mantra "don't bleed on the work".


Two of the parts in particular looked hand formed and had some surface scratches, which I will just polish out with the Scotchbrite wheel. Another part (VS808PP - spar doubler) has a very pronounced curve in it and I am wondering how to fix it. Perhaps being rivetted to the VS803PP spar, which is a channel section, will be enough to put manners on it. I will put a question on the bulletin boards and see what everybody thinks.


Now I have to put the kit away until I finish with the trimming of the shop itself and also connect up my new airline manifold (imperial measurements) to the output from the compressor (metric measurements). Fortunately, there are no machine threads involved; just nipples inserted into each end of a section of pressure tubing with jubilee clips to secure them. I do need to think about relocating the regulator out of the soundproof enclosure where I can read the guage and adjust the line presure. Hmmmm!

The Hangar

Though born and raised a country boy, I am an adopted Dubliner and a confirmed townie. As such, I live in a proper townhouse. It is a 1902 edwardian terraced house with minimal garden space front and rear and no off-street parking. There is a laneway to the rear, which is only wide enough to take a sub-compact car (ask me how I know this) and we have only a pedestrian door onto this lane.

While the garden is just about big enough to take the Vans RV-9 kit aircraft I intend to build, the wingtips would almost touch the house and the back garden wall at the same time and the rudder and spinner would be dangerously close to the side walls. As it happens, I quite enjoy a little gardening and I like to look out on some greenery. It is very clear that my wife Dorothy does NOT want to have the entire garden covered with a workshop and in fairness, neither do I. This was actually the biggest obstacle to be overcome in gaining her agreement to the project. I am pretty sure the neighbours would also be relieved at this line of thinking.

Though I have been involved in some fairly ambitious DIY projects concerned with rennovating the house and keeping it maintained, I have no workshop whatsoever. All I have is a boiler house containing a second fridge, laundry machines and storage for my tools. It is about 4 meters long by less than 1.5 meters wide. You can see it on the left in this picture.

I considered renting a workshop but scarcity of appropriate spaces in my area and generally sky-high property prices in the city have driven the rent up beyond reason. I could expect to spend a minimum of €3000 per annum on rent. That's €12000 over the course of the project or over a third of the total of all other building costs. In addition, it would be that much more inefficient in terms of time to travel even a short distance

Several Vans builders have succesfully completed their projects in a one-car garage and I decided that this was the most space I could afford.


I started out with a week off work on 21st July. In the intervening period, I have used a further two weeks leave on the workshop project. I didn' keep track but I would say I now have in excess of 500 hours in this thing.


I worked from a book helpfully called "How to Build a Shed" and loosly followed the construction method for one of the models in this book. However, it had to be adapted to my site and to my own particular requirements. The structure is of 47X75mm (3X2) timber. This is less than the book recommends but the shop only has to last for the duration of the project (about 4 years) and then I will hopefully be able to sell it and restore the garden. It is lined internally with12mm plywood on the walls and 5mm ply on the ceiling. I have left the 'attic' space within the roof trusses open to the shop so that I can easily use it for storage. Also, I have made the infill panel for the end truss on the house/garden end of the shop removeable so that I can get larger and longer items in and out. I have two large windows and one pedestrian door. The overall style I am going for is 'New England'. I thought the light grey colour would be less imposing to look at.



The entire structure is lined with a dense version of rockwool insulation called RW3 in a 50mm thickness. This will not only make the shop useable in the coldest weather but it is also an excellent soundproofing material, which was the main reason I chose it, being so near to neighbours and our own house. As the big windows (which I made from large obsolete perspex signs bearing the old logo of the organisation I run) would allow enormous amounts of sound and heat to escape, I made up polythene covered mats, like gym mats, with the same insulation material, which can be wedged in the window openings while noisy work is going on.



I have taken great care to get the floor perfectly level. If it decides to settle, however, I can wedge the entire building up or down at the appropriate corner.


The finished interior dimensions of the shop are 15'5" by 10' 3". This is not quite enough to fit the engine and propellor spinner to a completed fuselage. I have two strategies planned to cope with this. The first is that 60% of the side wall (up to the inner edge of the second window) is removeable, allowing me to push the tail of the plane out through this opening and work on the engine in the remaining shed. This will still not be an easy operation like opening a door so the idea is to buy a cheap 'party tent' and use it as an extension of the shop while that part of the project is going on - not in winter obviously. The other strategy is more radical and I need to check it out further before I even dare to mention it.


Inside the shop, I have built three standard EEA Chapter 1000 worktables. One of these is used for my three main benchmounted powertools; my bench drill, grinder/sander and bandsaw. Another table was adapted to take the compressor underneath and was clad with more 12mm ply and lined with 2 layers of carpet scraps to make a soundproof enclosure with a removeable front panel. This is remarkeably effective and makes not much more noise than a modern laundry machine. I just need to watch out for heat buildup and possibly build in a vent if it proves to be a problem. The third table has a massive 100mmX75mm aluminium angle along one edge for use as a straight edge and back rivetting plate for the trailing edges of the control surfaces.

All three tables have castors that can be extended by lifting each end of the table in turn, in which case the castors drop down and are locked in position. They can be unlocked by pulling on a string at each end. The weight of the table then carries the legs through to contact the floor as the castors hinge up out of the way. This design, which I saw Norm Abraham use on his New Yankee Workshop cable show, allows the tables to be configured in any way that suits the job at hand

Decisions, Decisions

My first interest in aircraft started with a primary school project on how planes fly. The interest surfaced from time to time over the years, but never in any active way. About 3 years ago I saw a picture of the new Vans RV-10 on the cover of a magazine and thought "Wow, That's beautiful". I bought the magazine and from that moment, I became a Vans fan.

The Mission
What I want to do with this plane is cross country flying. When I was involved in sailing, I never enjoyed racing or just going out for a 'blast'. My interest was in 'seafaring'; planning and executing a voyage from one port to another. I have the same view of flying. The only kind of flying that I look forward to is going places - the further the better. So I need a plane that will eat up the miles and make the most of every weather window.

Two or Four Seats?
I have a five-seater car but the only seat with a shine on it is the driver's seat. I began to realise that the same was very likely to be true of any plane I might own. However, I don't want to be anti-social so I have compromised on a two-seater and, because I think tandem seating is still mildly anti-social, that has to be a side-by-side two-seater. Besides, a four seater costs half as much again to build and operate.

Building space restrictions
Space to build was a huge obstacle. I live in a terraced townhouse with small gardens front and back, no off-street parking and only pedestrian access to the back garden. Unless I took over all of that back garden, I could only afford space for the equivalent of a one-car garage. That is barely enough to build the two-place Vans range and it confirmed my decision to abandon the RV-10.

Too hot to handle
Like about half of the members of the local Society of Amateur Aircraft Constructors (SAAC) who are still building, I am not yet a pilot. I don't see this as a problem. I intend to take an intensive PPL course about one year before I finish the project. It does mean that I will be a very low-hours pilot and this means that I need a plane that is especially easy to fly and suited to the trainer role, while having enough performance to continue to be interesting. That is the essence of the RV-9 concept. I toyed with the idea of a more powerful, popular and aerobatic RV-7 but flight instructors specialising in RV conversion training tell me that the 7 is appreciably more challenging to fly (though not by any means difficult) so I am going to take the more conservative option.

Getting Type-approval
Mine will be the first RV-9 build under Ireland's permit system for experimental aircraft and it looked for a while as if type approval would not be forthcoming. The application process was simplicity itself but the SAAC were having difficulty in making any headway with our Irish Aviation Authority (IAA). This seemed to be just bureaucratic delay rather than any objection to the Vans design. However, just on the point of ordering a 7 afterall, the approval finally came through. I want to pay tribute to Charles O'Shea, Type Approvals Sec of the SAAC for his persistence and his hard work in achieving this approval.

Airplane slut
Vans are thought of as performance planes. So when Rotax-powered LSAs began to emerge like daffodils in spring, I began to wonder whether my mission could be fulfilled even better by one of these new types. The one that caught my eye was a real beauty called the Sportcruiser. I am ashamed to say that I totally and immediately forgot my love for Vans and fell headlong for this sexy Czech 'babe'. The key feature of this kit is its high degree of completeness as sold. It is supplied as a quickbuild with all of the main construction work done. Some panels are even fitted and temporarily riveted with the intention that they be removed to facilitate the inspectors and then permanently fixed back in place. Most of the work to be done by the builder is in the area of cockpit systems, firewall forward installation, panel and paint. While the British PFA never raised a question about this plane satisfying the 51% rule, the SAAC took the view that it could not possibly comply. I think they are wrong about this and that quickbuild kits are the future of homebuilt aircraft and possibly even of personal aviation. As manufacturing gets more automated and efficient, the time taken for the work done in the factory diminishes and therefore the amount of time needed to make up 51% of the total also diminishes. In the end, the SAAC Chairman kindly pointed out to me that such kits were against the ethos of the society which is concerned with construction, not merely assembly. I decided that you can't fight city hall - at least not yet.

As it turns out, the first group of Sportcruiser builders in the UK are having a torid time with their kits. Apparently, factory fitted options that were ordered are not turning up on the kits as delivered. Wrong options are also being supplied and, worst of all, the factory are not being very good about it. Customer service at this point is, by all accounts, appalling. So it seems that the SAAC have done me a favour afterall and saved me from a fate worse than a stall turning onto short finals. On my part, I reflected that the actual build was at least half of the pleasure to be derived from the project so there was little point in shortening it.

So I returned to my first and truest love, was forgiven for my infedelity and embraced in the warmth of knowing that the Vans products are built and flown by real engineers and real pilots.

I have now ordered my empennage kit and I am breathlessly awaiting delivery.