New Ultralight and LSA Trainer design PAIR 2

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,751
Location
Rotterdam, Netherlands
But for the aft fuselage as an example here, it all breaks down to a few worst cases of this much bending in this axis, that much bending in another, "x" amount of drag load, and so much torsion.
Not really and that's the problem. Different load case combine for higher local loads.
I don't believe you. My own experience says otherwise. HITC's own experience says otherwise. HotWing's own experience says otherwise.
What experience?
HitC actually confirms my point. Make it heavier and physically smaller and it's less complex to design. But we can't do that for a FAR103-like design, since size (stall limit) is by definition big and weight (weight limit) by definition very low.
I suspect you're doing a Boeing-level structural analysis on something that absolutely doesn't need it. No, the product won't be as optimized. It doesn't need to be. I'm talking about balance and you're talking about absolutes.
Yes, it does, unless you want to fly with an airplane that fails below limit load.
Simple fact. NO single design I'm aware of has passed actual load tests without failures and I can't think of one that didn't fail at least once before limit load. All designs now flying and tested either did the engineering properly, or after multiple failures concluded that they had to do it, since continuing breaking wings/tails was too expensive. Walking around all those broken constructions does give one a great understanding of actual design (vs analysis, which is what the books teach you) ;)

Going through dozens of photos from different load tests is sobering and I've seen hundreds of thousands of euro's of trashed wings/fuselages etc because they didn't do their engineering. I'm not trying to argue Marc and this is not about belief.

This is about sharing some hard-learned lessons. That's what HBA is all about in the end.


Another then. You can replace LSA with FAR103. Exactly the same story:
Mainly because of experience, but a bit of analysis is behind it too. So far I've been asked to either design or consult to probably half a dozen different LSA developers, all of whom interestingly enough, had very similar attitudes regarding their airplanes and the development programs. I can sum them up into a few categories:

1) The first has several variations - basically this is where they show me an artistic sketch and a bit of what they call design work and then ask me to go from there and design the airplane. When I proceed to do so however, they respond with an attitude where they think that they already designed the airplane but what they really need from me is to essentially rubber-stamp their ideas. That's without any analysis or engineering though. When I refuse to do so they come back with an insulted attitude - one actually told me that I should feel privileged to be asked to work on their program and so should provide the verification they asked for.

2) The second is sort of a corollary to the first - it's where their "team" has come up with a design and they don't need me to actually design the plane but they do want me to look over a couple of the details. Usually I can't do this simply because I need to verify what I'm working on. Eventually they agree but when my BS detector is verified and I have to inform them that they have an unworkable design (and I do document my reasoning), they again come back very huffy, usually telling that I don't know what I'm talking about. One even threatened to sue me. When I provide them with my initial report (one that they actually never pay for), they often try arguing the points however usually with a lot of arm waving (and statements about having faith) but no actual verifiable proof.

3) The third is more basic and very common - an unrealistic budget. They think that because this is a light airplane that the design work should take no time (or money) at all. Interestingly enough, when I explain that the work to design an LSA and something high performance like a Glasair III is nearly identical they indicate that they understand, but they still think their project should be much cheaper. Virtually all of these had very unrealistic budgets to begin with and none have actually gone anywhere.

4) The last is more technical - I've had several discussions with other designers regarding this category and most of us feel that these are very marginal airplanes. If they were being used and marketed as originally proposed (fat ultralights) then I think there would be no problem. But instead these are now being marketed as trainers or cross country airplanes, equivalent to Part 23 production units. And here is where we have a problem - they simply are not. The structural design of many of these new airframes is short of what we'd have with Part 23 standards, and more so when you consider that most of them will be operated over weight. That's of course not to say that all are marginal (after all, some of the older production planes fall within the LSA limits and they did certify to Part 23) and some are honest enough to publish the lower load factors to which they were developed, but in our opinion there just is not sufficient structural mass margin in a two place LSA configuration to make us feel comfortable.

In short, the reason we stay away is due to many of the somewhat unpleasant personalities we've run into, and because we feel the category is too restricted and mostly a big step backwards for the light airplane industry.
 

Rienk

Well-Known Member
Joined
Oct 11, 2008
Messages
1,366
Location
Santa Maria, CA (SMX)
I appreciate the discussions going on here, as it helps me to weigh the options for the title of this thread, designing a pair of aircraft - a 103 legal Ultralight, and a similar two-seat trainer (E-AB or S-LSA).

I don't remember which thread it was in, but one of my considerations was to not actually try to go legal UL, but instead to go for a motorglider. Since flying a UL should entail adequate training, why not just ensure there is enough by going the MG route? I don't remember the certification requirements (separate from performance parameters) but I recall that it may actually be similar - or simpler - than going the LSA route.

In spite of the title, let's remember the primary goal of this thread - developing a pair of aircraft that are inexpensive to fly and acquire, with the goal of exponentially increasing participation in sport aviation.

Though I appreciate all of you who have an interest in developing low cost Plans or Kit built planes, such as a MPG, etc., I am convinced that the only way to increase the number of pilots is to offer RTF aircraft - which means production manufacturing. That is why, based on a lot of what I have gleaned from this forum, I envisioned using UL as the means to (hopefully) mass produce a single-place "entry level" plane. And of course, a similar two-seat trainer would need to be developed, along with a training infrastructure, so that people aren't in the predicament that Topaz is in. There is no way that sport aviation will see a renesaunce unless two-such aircraft are developed together.

So, if anyone wants to chime in on the slight diversion of MG vs. UL, I would appreciate revisiting that comparison - with the understanding that we are looking to low cost production manufacturing, not low cost kit/plans building.


PS, I would very much like to collaborate with others on such a project; if anyone is interested, please PM me!
(and by the way, I do have access to engineers who can help with Catia)

BTW, I just picked up a CNC plasma cutter yesterday, with a tube cutting attachment. I picked it up from a bicycle company who is moving manufacturing to China; unfortunately, it apparently doesn't do well for small diameter, thin wall tubing - so I will have to wait for a CNC laser to be able to do some of the fancy stuff for
R&T airplanes... will have to do that by hand for now
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,751
Location
Rotterdam, Netherlands
This may be a source of our miscommunication and is a relative of the SPDS syndrome. I'm not interested in "reasonable weight" for my tail boom or designing one from scratch. It's a simple off the shelf tube with one sleeve. As a mater of fact right now it's plain old 1015 steel or "muffler molly". It's double the weight of an equivalent off the shelf 2024 tube but I can get it locally for a fraction of the cost, weld it with MIG and not worry about any HAZ, which lets me make all kinds of simple light fittings and attachments. Lots of planes, some of them certified, use tubular spars even though we (and presumably the designers) all know that they aren't an efficient use of materials for this application.

I could save maybe as much as 10Kg with a designed from scratch tail boom but even with the 254# limit I don't think I'm going to need to.
Then I'd be most interested in how the weight break-down looks like. From my experience, making one that's below the weight limit, meets min stall speeds and can withstand ultimate loads is challenging to say the least.
The sailplane limit is a bit easier (no huge wing/engine), but it's still hard enough to meet the 155 lbs limit. You cannot hang a motor on one unless you go back to the full-sized wing again (135-ish sqft).
In the mean time aren't we kind of taking this thread of on a tangent that doesn't really add to Rinek's topic?
I think it's relevant to his topic. There's a lot of wishful thinking in this topic and the general though of simple/cheap seems very unrealistic to me, because something FAR103 compliant is by definition tough to do well and expensive. But what do I know... I only do this for a living :roll:
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,751
Location
Rotterdam, Netherlands
Though I appreciate all of you who have an interest in developing low cost Plans or Kit built planes, such as a MPG, etc., I am convinced that the only way to increase the number of pilots is to offer RTF aircraft - which means production manufacturing. That is why, based on a lot of what I have gleaned from this forum, I envisioned using UL as the means to (hopefully) mass produce a single-place "entry level" plane. And of course, a similar two-seat trainer would need to be developed, along with a training infrastructure, so that people aren't in the predicament that Topaz is in. There is no way that sport aviation will see a renesaunce unless two-such aircraft are developed together.
That single-seat aircraft isn't going to be anything cheaper as the two-seater, except for the powerplant. If you can limit the 2-seater to something like the Aerovee, there is no market for the single-seater.

I would personally go for a composite, sleek pusher.

But for production of a basic training aircraft, it's almost impossible to beat a tractor with a welded steel tube fuselage, with a single composite interior (hardpoints integral to the part) and a composite monopiece wing with flaperons. Both in terms of engineering and actual production the required time is minimal with low initial cost.
 

Topaz

Super Moderator
Staff member
Log Member
Joined
Jul 29, 2005
Messages
14,111
Location
Orange County, California
I'm out of this part of the discussion. This is pointless. Rienk, sorry to have disrupted your thread.
 
Last edited:

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,751
Location
Rotterdam, Netherlands
I'm out of this part of the discussion. This is pointless. Jarno, we have a similar educational background and you haven't completed an flown any more of your own designs than I have.
Incorrect, on both accounts.
but minimizing HITC's actual experience of designing and completing multiple aircraft - using which experience he disagrees with you - is beginning to sound like hubris. I respect the heck out of your skills and knowledge, but you're going too far here.
?
I'm not disagreeing with most of what HitC said. Make it smaller and heavier and reduce build time and engineering complexity/time.

I'm simply sharing my experience after almost a decade of designing, building and testing. I know you get pretty much the same response from every engineer that's working in the small aircraft industry (I have and learned a lot from that), except that they're wise enough not to engage the debate anymore ;)
 

Topaz

Super Moderator
Staff member
Log Member
Joined
Jul 29, 2005
Messages
14,111
Location
Orange County, California
Incorrect, on both accounts.
Please post a picture of the full-sized aircraft, of your own design, that you have completed and flown. Not the one your employer brought you in to work on, not something you modifed. Your own design, from clean sheet of paper.

Unless something's changed radically that I haven't heard about, your first self-designed airplane is still on the drawing board.
 

Hot Wings

Grumpy Cynic
HBA Supporter
Log Member
Joined
Nov 14, 2009
Messages
7,085
Location
Rocky Mountains
The sailplane limit is a bit easier (no huge wing/engine), but it's still hard enough to meet the 155 lbs limit. You cannot hang a motor on one unless you go back to the full-sized wing again (135-ish sqft).
If you hang a typical Rotax on the plane then it does get harder to meet the 254 weight limit. The only reason I brought up this part 103 concept is that I think/believe that we need to seriously consider this type of plane and the solo instruction model as a way to get to Rienk's objectives.

His idea of a cheap to build 2 place trainer is a good one, but not a viable first step without a lot of cash to invest. It will cost a lot more to build in moderate numbers. It will have to be at least LSA certified to be able to be used in a commercial setting, other than in a flying club here in the US, because of our regulations. Existing powered single seat ultralights simply aren't suitable for solo instruction. I proved this to myself long ago the hard way. I still have part of one wheel as a wall ornament.

This lead to my trying to find a way to build within the part 103 regulations. We already have things like the Sandlin airchairs that could probably be adapted for solo instruction. While they may be simple planes to build they have extremely high parts count, are time consuming to build, and not suitable for production in moderate numbers. So I started to look at different ways to build an equivalent plane that could be produced economically in small numbers, or built in minimum time individually. Making them simpler always led to heavier - at least for me. The 155 pound limit is tough to meet if you want stall speeds low enough for a solo training aircraft..

Adding power, even if it wasn't really adequate lets us use the full 254 pounds. Working back from the classic primary glider that weigh in the 200 to 220 pound range it looked like if a bit more weight could be saved then there would be enough room to add some power. Electric has a better power to weight ratio than a Rotax. Range, or duration is still far less, but we don't need much range for a training aircraft, especially one that can be expected to use ridge lift or thermals to increase duration.

A 6kw motor, controller, wiring and batteries for 5 minutes weighs 10 to 12 Kg. That leaves ~210 pounds for the structure and still be part 103 legal. This was the starting point for my version of a Modern Primary Glider.
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,751
Location
Rotterdam, Netherlands
I can't do that as you know.
His idea of a cheap to build 2 place trainer is a good one, but not a viable first step without a lot of cash to invest. It will cost a lot more to build in moderate numbers. It will have to be at least LSA certified to be able to be used in a commercial setting, other than in a flying club here in the US, because of our regulations. Existing powered single seat ultralights simply aren't suitable for solo instruction. I proved this to myself long ago the hard way. I still have part of one wheel as a wall ornament.
I'm not so sure developing a 2-seat trainer (LSA-approved) has to be expensive. Sticking to simple structures and focussing on reducing building time (essentially parts count), you're essentially building a first plane and certifying that. Must be possible for what, 100-200K if done right?
This lead to my trying to find a way to build within the part 103 regulations. We already have things like the Sandlin airchairs that could probably be adapted for solo instruction. While they may be simple planes to build they have extremely high parts count, are time consuming to build, and not suitable for production in moderate numbers. So I started to look at different ways to build an equivalent plane that could be produced economically in small numbers, or built in minimum time individually. Making them simpler always led to heavier - at least for me. The 155 pound limit is tough to meet if you want stall speeds low enough for a solo training aircraft..

Adding power, even if it wasn't really adequate lets us use the full 254 pounds. Working back from the classic primary glider that weigh in the 200 to 220 pound range it looked like if a bit more weight could be saved then there would be enough room to add some power. Electric has a better power to weight ratio than a Rotax. Range, or duration is still far less, but we don't need much range for a training aircraft, especially one that can be expected to use ridge lift or thermals to increase duration.

A 6kw motor, controller, wiring and batteries for 5 minutes weighs 10 to 12 Kg. That leaves ~210 pounds for the structure and still be part 103 legal. This was the starting point for my version of a Modern Primary Glider.
Ah, clear. Yep, exchanging the Rotax for electric gives you just enough extra weight margin to greatly simplify the structure. Still tough to keep the wing structure simple. We've been thinking of making full molds (upper/lower skin), but then only have ribs aft of the LE. This way you only have 2 molds and the rear-front spar are integral. Glue some pultrusions in, layup over the (still split) shear web and LE and you have a wing that requires very low building time (hours).
Build the plugs up from hotwired foam (D-section, cover with Mylar, then use some alu U-profiles in the plug, make the trailing edge/rear spar via hotwired foam and mylar too and make the mold. Zero sanding, zero CNC.
 

Rienk

Well-Known Member
Joined
Oct 11, 2008
Messages
1,366
Location
Santa Maria, CA (SMX)
I'm not so sure developing a 2-seat trainer (LSA-approved) has to be expensive. Sticking to simple structures and focussing on reducing building time (essentially parts count), you're essentially building a first plane and certifying that. Must be possible for what, 100-200K if done right?
I agree that a two-seat trainer doesn't need to be expensive to develop - especially since I would intend to start it as an E-AB to see if it proves out, and then while the ASTM certification process takes place.
But it is a typical "chicken or the egg" scenario. People may opt for flight training in the two-seater by itself (and may actually choose to go the Sport Pilot route, since the two-seater may make better financial sense.
However, the goal to having an inexpensive single-seat airplane has pushed us to the UL route, for two critical reasons. First is that I believe the plane must be sold RTF (ready to fly), and having no certification costs to amortize will help keep costs down. Second is that, no matter what the price, we will need to be able to offer financing - and that is not possible with a kit (and will take some financial backing to even offer RTF). Same with the two-place trainers; not many people who will be enthused enough to become a dealer/training center will have $120k cash for the necessary start up (3 UL's and 1 trainer).
 

Jay Kempf

Curmudgeon in Training (CIT)
Lifetime Supporter
Joined
Apr 13, 2009
Messages
3,944
Location
Warren, VT USA
I agree that a two-seat trainer doesn't need to be expensive to develop - especially since I would intend to start it as an E-AB to see if it proves out, and then while the ASTM certification process takes place.
But it is a typical "chicken or the egg" scenario. People may opt for flight training in the two-seater by itself (and may actually choose to go the Sport Pilot route, since the two-seater may make better financial sense.
However, the goal to having an inexpensive single-seat airplane has pushed us to the UL route, for two critical reasons. First is that I believe the plane must be sold RTF (ready to fly), and having no certification costs to amortize will help keep costs down. Second is that, no matter what the price, we will need to be able to offer financing - and that is not possible with a kit (and will take some financial backing to even offer RTF). Same with the two-place trainers; not many people who will be enthused enough to become a dealer/training center will have $120k cash for the necessary start up (3 UL's and 1 trainer).
Starlite and Pulsar. Buy the molds and there ya go.
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,751
Location
Rotterdam, Netherlands
How expensive are those certification costs? I am clueless, but looking at the equivalents here, I doubt they're significant (for LSA).

As said before, I don't think you can build a legal, safe FAR103 plane for considerably less than a full-blown trainer. I'd personally just ditch the whole UL part and focus on an affordable LSA-trainer.
 

Rienk

Well-Known Member
Joined
Oct 11, 2008
Messages
1,366
Location
Santa Maria, CA (SMX)
Yep, exchanging the Rotax for electric gives you just enough extra weight margin to greatly simplify the structure. Still tough to keep the wing structure simple. We've been thinking of making full molds (upper/lower skin), but then only have ribs aft of the LE. This way you only have 2 molds and the rear-front spar are integral. Glue some pultrusions in, layup over the (still split) shear web and LE and you have a wing that requires very low building time (hours).
Build the plugs up from hotwired foam (D-section, cover with Mylar, then use some alu U-profiles in the plug, make the trailing edge/rear spar via hotwired foam and mylar too and make the mold. Zero sanding, zero CNC.
I've also thought about using a light weight PPG engine for the UL. For me, the problem with those is their noise, high revs, and low TBO.
I don't know if it's possible, but I'd really like to use a fourstroke - and an industrial engine at that.

As to how to build the plane, I've contemplated the hotwire foam cores (though we would use CNC :) ).

Here's an idea... I hope I can explain it verbally.
What if 1-2" solid foam ribs were cut, and spaced every 8-16". And then hollow wing section (airfoil shape, about 1/2" thick caps, maybe with some internal X-bracing as well) were cut out to fit in between the solid ribs, so that the entire wing had a foam surface, to which fiberglass sheets were laminated (VIP to a glass table, so the outside surface would be glass smooth). Tons of contact area, and very strong... the biggest problem would be weight?

Another thought. In my other thread, I contemplated a way to get a higher useful speed out of a UL, but the formula required a number of struts to be used, even if they were not structurally needed. The wings we're now discussing would be cantelevered, and to meet the stall speed formula requirement, would need to be at least 117 sq.ft.

I like the idea of a (maybe laminar) wing that is much smaller in area, using high lift devices... it could keep the wing weight down?

I also considered using one of our vendors to make the entire airframe (and wing spars) out of Carbon Fiber tube, which could make the structure very strong and stiff - but the cost would obviously be a factor.

I'm trying to envision how you are proposing building this wing. I can build molds for very low cost, so I'm open to such ideas. The Solo has only two molds for the wing skins (top and bottom - including wing tips, clever molds), a mold for each spar, and then some ribs. About as simple as it can get (though designing the molds to make the system simple was a fair amount of work).

Thanks!
 

Rienk

Well-Known Member
Joined
Oct 11, 2008
Messages
1,366
Location
Santa Maria, CA (SMX)
Starlite and Pulsar. Buy the molds and there ya go.
I wish they were available - do you know anything about them?
Rich Trickle was a friend of mine (the designer of the KIS) which was taken over by a South American company that also has the Pulsar, but the last I heard (years ago) the project disappeared. Rich was living down there, heading up the program for a while, but then moved to another project (another country) before he died.

Still, the Pulsar is too small for two "normal" size people, which is one of the reasons I gave up on my KR2 project 25 years ago.
As mentioned before, 6'6" and 250 lbs for the UL, and also for the trainer.

(still, the Pulsar is a cute little plane - like the Lightning?)
 

Rienk

Well-Known Member
Joined
Oct 11, 2008
Messages
1,366
Location
Santa Maria, CA (SMX)
How expensive are those certification costs? I am clueless, but looking at the equivalents here, I doubt they're significant (for LSA).

As said before, I don't think you can build a legal, safe FAR103 plane for considerably less than a full-blown trainer. I'd personally just ditch the whole UL part and focus on an affordable LSA-trainer.
That may be the result of this whole exercise.
I do have an ulterior motive - my kids want to be able to fly their own UL at age 14 (but by the time I get to this, they'll probably be 17 anyway).

To meet the "magic" price point of $20k, the profit margin would be slim - much more healthy for a two-seat trainer, which I believe could sell well in the $30-40k range. But then it would have to be certified, and that means S-LSA or Motorglider.

So ignoring the UL for the moment, what is the least expensive way to manufacture a trainer?
If it's going to be all composite, then I may as well stick with the Solo and Duet ($30-40k, and $50-60k, respectively).

I am enamored with the idea of building a sexy Rag-and-Tube airplane, and I really believe that a good way to invigorate GA is to get something safe and fun for the $20k and $40k price point.
How do we get there?


Edit: Certification costs may not be too much for the airframe, but an adequate "low cost" power plant might also need to be certified, which could significantly increase the program cost.

The reality is that such a program would only be profitable in significant numbers. Thus, anyone who undertakes this would have to do so more for the "love of the game" than for any potential financial gain. In other words, for the fun of it! Any other takers?
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,751
Location
Rotterdam, Netherlands
Some random remarks about an earlier post.

Laminar wings with high lift devices. The Dyn'Aero MCR01 is hard to beat. Clever, simple, VERY good performance (lift).
Solid foam ribs are heavy. Well, they're not, but once you glue them in, total weight is heavy.
The design (two molds, upper and lower skins) I talked about has a lot of potential. It takes some brain-twisting, but can potentially be very light, because you have an order of magnitude less glue joints and hard points (which is where most of the weigh is)
To meet the "magic" price point of $20k, the profit margin would be slim - much more healthy for a two-seat trainer, which I believe could sell well in the $30-40k range. But then it would have to be certified, and that means S-LSA or Motorglider.
20K might be marginally achieveable for an electric UL (much simpler/heavier structure). An LSA isn't possible for 30K, even assuming a 10K (vs 30K for a Rotax) engine cost. 40K... well, even that might be impossible once we figure in the costs of running a company.
So ignoring the UL for the moment, what is the least expensive way to manufacture a trainer?
If it's going to be all composite, then I may as well stick with the Solo and Duet ($30-40k, and $50-60k, respectively).
Steel tube fuselage (covered) and a very advanced (molds) composite wing. A single integral composite cockpit with ALL the hard points integrated. Stock gear, a taildragger is far simpler, but might not be feasible commercially. A single mold for all six halves of the tail (left stab, fin, right stab). I don't like Junkers flaperons (draggy), but they make a lot of sense for trainers.

The cockpit with dozens of hard points is usually a big deal (lots of work). Spend some time developing a mold that does all in one step and you save at least a hundred hours of build time.

The wing is another area where you need massive time and effort for a traditional plane. A composite warren-truss wing has enormous potential (currently working on that). Half the weight and 3 parts for an entire wing...
Edit: Certification costs may not be too much for the airframe, but an adequate "low cost" power plant might also need to be certified, which could significantly increase the program cost.
Without a "low-cost" engine you're lost. A Rotax 912S sets you back around 30K US$.
 

Head in the clouds

Well-Known Member
Joined
Mar 11, 2012
Messages
1,983
Location
Gold Coast, East Coast of Australia
Friends ... I didn't mean to hornet nest.gif

I think I've not properly explained what I meant.

I can quite see what autoreply means when he says that all smallish planes require the same amount of design work and quite probably he's right that the smaller they get the more demanding they might be. However in the real world I've never seen it work that way and I've seen a hell of a lot of successful ultralights get designed, built and quite a number of them put into production.

Our (Australia's) first two seaters did get a fairly comprehensive structural analysis as part of their certification package when they went through that process but it's worth keeping in mind that all of them had been flying for years before the regulations changed sufficiently for them to have to prove that they met JAR, BCAR or FAR standards. What had actually happened at the earlier stage was that the more critical parts of the structure (spars and struts) were either calculated or sized from similar existing craft and then the prototype was built and physically tested. The certification process required physical testing whether or not a full structural analysis was supplied with the certification application package.

In the case of the planes that I have built, although I have not needed to have any of them certified I have adopted the same approach. Much as I love and am capable of the math, and my engineering computing experience goes back to the glory days when any self-respecting computer owner also had a large room filled with lines and lines of delightful young girls in mini-skirts punching cards full of holes to feed the machine code or Fortran into the humming behemoth that had advanced past the age of needing preheated valves and actually used the miracle of the semi-conducting transistor ... I digress, it's the memories of those fabulous lassies in the punch-room I think.

Anyway, when I said the design time taken for a two seater was four or six times that for a single seater, I was referring to the time spent to model all the components and sort out their connections and all that, not so much the calculations of the structure. In reality we are limited to just a few sizes of material for each part of the structure so there's no point in calculating the exact loads on each part of each part. Take a cantilever wing like the Macro I posted pics of. The spar caps are 6061T6 structural angle, commercial/marine grade. The two smallest sizes were 1.5"x1.5"x0.1875" or 2"x2"x0.25". I would never use the offerings in aircraft grade even though they might optimise the structure just a little more and save a little weight but the weight saving would be just a few ounces over the whole airframe and no-one would ever buy the plane because it would be too expensive (I know autoreply's response is that the cost of materials isn't relevant, and that may be true when the materials are available in your country but not when you have to import them to our side of the planet). So you do a quick calc (5mins) and decide on 2" or 1.5" spar caps and get the bandsaw and planer started to taper them down (Yes, electric planer works great on aly if you use TC blades).

So the member sizing of small single and to a degree, two seaters, is mainly determined by what materials are available and what they need to cope with i.e. will the airloads on the tailcone be greater than the tailwheel loads during taxi, bad landings, trailering? My first Macro tailcone was 0.16" alclad and it passed the sandbagging tests with ease but I ended up having to add several laminations near the rudder post because of localised damage caused by the tailwheel mounting from taxiing over rough ground. The later Macros had 0.025" talicones and I was able to have half the amount/weight of internal angles forming the stiffening/truss so the weight came out the same and it reduced the amount of work to build it.

As far as the build time is concerned it's just my experience, and that of all the folks I've seen build single or two seaters - my estimate is that the build time is at least a square function of the number of seats i.e. if a single seater takes one person working alone a month, a two seater would take 2x2=4 months, a four seater would take 4x4=16 months, an 8 seater would take 8x8=64 months etc. As I see it if someone suggests that a two seater takes the same time as a single then does that mean an eight seater also takes the same time? (and cost?). And that estimate has proved about right, I could easily build four of my Macros in the time it takes to build one two seater of the same design.

Once the build is completed it only takes a day to complete the structural testing whereas it might have taken many months (or lots of dollars to an engineer) to complete all the calculations, and frankly even if a fully qualified and experienced aeronautical engineer provided all the calcs I'd still spend a day sandbagging it, it gives you so much more confidence when you hit heavy turbulence ... When I sandbagged the two seater it was a very easy affair, we removed the canopy and sat the sides (armrest area) of the fuselage on the garden wall in the front yard, nose down 15* so that the weight would impart an anti-drag load similar to that at the stall, fitted the wings and laid sandbags to approximate the lift distribution, a bit over 3/4 tonne on each wing IIRC. We had some pretty nice metronomy equipment from a toolmaker friend so we plotted the deflections and were able to make a good estimate of where the yield point would be, it was around 6-7G as the 5min calc had suggested. Later we found that we'd forgotten to put the second pair of bolts in the wing stubs and hadn't put any nuts on the first pair so we had to have the spar ends and carry-through x-rayed/dye tested. There wasn't any distortion or cracking/tearing which was surprising, the material exceeded the calcs in that respect.

After the wing bagging we loaded the HS, and to do that we had to hold the nose down so we drove a couple of star pickets 6ft/2m into the lawn at an angle and chained the nose down. Then we put a bit under half a tonne of bags on the HS, when we unloaded it we unloaded all of one side first. It put a lot of twist in the tailcone and you could see wrinkles in the metal but it all went back to normal when we unloaded it.

OK, it might not be the most scientific of tests but we flew it hard and with confidence after that and never had a structural failure on any of my designs so it worked for me.

I'm still wading my way through my re-discovered old photos, they got submerged in a flood following a cyclone so many of them are destroyed and the rest are stuck together. I have been soaking the negatives and slides in distilled water and slowly saving some of them and bought a small scanner that converts them to digital. I haven't got to any of my own sandbagging ones but here are ones I took when I was a witness to the load testing of the Jabiru, it was around 1986/7. Below that I have posted a video of the current Jabiru hydraulic test rig which is much more engineered. You may be astounded to see what the structure of that plane will withstand - it's worth having a look! It's worth noting too that although some people have crashed their Jabs into rocks and trees the airframe seems to take it really well and protect the occupants, it's a tough little plane!

Since I started to compose this I see the conversation is 'moving on' but I might as well post it now and I hope folks see that it is relevant to the cost of production thing, you can get bogged down in design and calculation and never get around to building anything, or you can do some basic calcs and otherwise use estimates from similar structures for member sizing, build it and test it ... you'll get a perfectly satisfactory result quicker and cheaper using the latter method.

jabtest1.jpg jabtest2.jpg jabtest3.jpg

 
Last edited by a moderator:

Rienk

Well-Known Member
Joined
Oct 11, 2008
Messages
1,366
Location
Santa Maria, CA (SMX)
Laminar wings with high lift devices. The Dyn'Aero MCR01 is hard to beat. Clever, simple, VERY good performance (lift).
Solid foam ribs are heavy. Well, they're not, but once you glue them in, total weight is heavy.
The design (two molds, upper and lower skins) I talked about has a lot of potential. It takes some brain-twisting, but can potentially be very light, because you have an order of magnitude less glue joints and hard points (which is where most of the weigh is).
I agree about the Dyn'Aero aircraft (is the MCR01 the original Colomban design?). That is essentially what I'm hoping to do with the Solo and Duet.
On the current Solo, not only do we only use a top and bottom mold for the wing skins, but the flared tips are built in. Additionally, the front seam (and around the wing tip) will be covered by a vinyl sticker, and the back edges are on the bottom side of the trailing edges, so not visible unless you crawl underneath the plane. This allows us to build the finish into the parts, and not require much fit and finish at all.
Though the Solo will have a limited market, it is also a POC for the Duet. Using a low cost industrial engine for the Solo lets us keep development costs down, and if all works as anticipated, we can invest in the larger and more costly (mostly engine and panel) two-seat Duet.

As far as this tentative trainer is concerned, I am still looking for a way to develop a lower cost production plane. If we are successful with offering the Duet at $50-60k, their isn't much point of doing another trainer unless it can retail for under $40k (and still make a good profit margin). To do this, it needs an basic, inexpensive powerplant (VW derivative, LSA certified?), and thus needs to be fairly light. If it is a little draggy, that is fine.

BUT - this project primarily seems to make sense if it is a trainer for a similar UL. Otherwise, all we're trying to do is offer a modernized/certified KitFox (larger cabin).
That doesn't really excite me...
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,751
Location
Rotterdam, Netherlands
I can quite see what autoreply means when he says that all smallish planes require the same amount of design work and quite probably he's right that the smaller they get the more demanding they might be. However in the real world I've never seen it work that way and I've seen a hell of a lot of successful ultralights get designed, built and quite a number of them put into production.
Australian ultralight I presume?

That's a totally different story. The extra weight and less ridiculous stall speeds allow a much simpler structure. We're really talking an order of magnitude less engineering and building time. Same for the EU MLA (35 kts stall, 1000 lbs MTOW, 660 lbs for a single-seater).
In the case of the planes that I have built, although I have not needed to have any of them certified I have adopted the same approach.
Sure. Any "engineering" without testing is just nice talk. But what pops up is that if you're forced to the extremities of FAR103, your parts get so small (thin mostly) that you get the most unexpected failure modes. Which makes it far more difficult to engineer properly.
IMG_3295.jpg

That's what 2 tonnes of compressive load does to an ultralight (FAR103-like) optimized (read:skinny) wing...
Anyway, when I said the design time taken for a two seater was four or six times that for a single seater, I was referring to the time spent to model all the components and sort out their connections and all that, not so much the calculations of the structure. In reality we are limited to just a few sizes of material for each part of the structure so there's no point in calculating the exact loads on each part of each part. Take a cantilever wing like the Macro I posted pics of. The spar caps are 6061T6 structural angle, commercial/marine grade. The two smallest sizes were 1.5"x1.5"x0.1875" or 2"x2"x0.25". I would never use the offerings in aircraft grade even though they might optimise the structure just a little more and save a little weight but the weight saving would be just a few ounces over the whole airframe and no-one would ever buy the plane because it would be too expensive (I know autoreply's response is that the cost of materials isn't relevant, and that may be true when the materials are available in your country but not when you have to import them to our side of the planet). So you do a quick calc (5mins) and decide on 2" or 1.5" spar caps and get the bandsaw and planer started to taper them down (Yes, electric planer works great on aly if you use TC blades).
Yep, the downside of living on the wrong side of the moon :gig:

Joking aside, any aircraft, save FAR103 ultralight (and LTF-L/SSDR) will end up with a given skin gauge for the "moron factor" and daily loads, whether you're building in wood, alu or composites. Using proven practises, it means we can "dumb-down" engineering enormously.
As far as the build time is concerned it's just my experience, and that of all the folks I've seen build single or two seaters - my estimate is that the build time is at least a square function of the number of seats i.e. if a single seater takes one person working alone a month, a two seater would take 2x2=4 months, a four seater would take 4x4=16 months, an 8 seater would take 8x8=64 months etc. As I see it if someone suggests that a two seater takes the same time as a single then does that mean an eight seater also takes the same time? (and cost?). And that estimate has proved about right, I could easily build four of my Macros in the time it takes to build one two seater of the same design.
I don't get that. Why would parts count or assembly time go up exponentially? In my experience it doesn't. I've seen many builds and I don't see that trend anywhere. Still the same parts count (for similar planes) and the same amount of actions of the builder?
Compare the build time of a hummelbird or an Onex to a Sonex. RV7 to RV10. Berkut to Velocity. Sonerai to Bearhawk.
Same for cost, save the obvious engines?
 
Last edited:

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,751
Location
Rotterdam, Netherlands
I agree about the Dyn'Aero aircraft (is the MCR01 the original Colomban design?). That is essentially what I'm hoping to do with the Solo and Duet.
The MCR100 (Michel Columban 100) is the original one. Alu sheet-covered, unlike the production version.
Additionally, the front seam (and around the wing tip) will be covered by a vinyl sticker, and the back edges are on the bottom side of the trailing edges, so not visible unless you crawl underneath the plane.
So, how do you react shear? You need a pretty beefy joint to react it away.
 
Top