New kit lines

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

Joined
Jan 25, 2003
Messages
430
Location
Shirley airport MA
I would like to see the cost break down of those ribs machined from a solid block of one inch thick aluminum. From looking at it it seems to me it would take something in the order of a thousand pounds of aluminum for a set of ribs for one plane plus all the machining cost ???. That would probably cost as much for the ribs alone as all material for entire airframe of comparable plane, I wouldn’t be surprised to see a government agency to build a space shuttle this way but a kit plane ?

George
 

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
The one inch aluminum plate is fairly pricey (about $2,300) but I can get all the airplane's components from one sheet, including all the fusleage frames, all the ribs, as well as several other miscellaneous bits of hardware. I'm not sure of the cost of using a router to get the parts but using a waterjet cutter approximately doubles the raw material cost. As such, all the framework components can be had for less than $5,000, give or take a bit.

All that's left of the primary structure then are the skin components, the longitudinal tubes, the secondary frames, the spar webs and the carrythrough fittings. I don't as of yet have an estimated cost of producing all the net shapes of each airframe kit but the current ballpark guesstimate is somewhere around $8,500 to $9,500. Add in all the other items such as the landing gear, the control systems, the fuel system, etc. and so far it looks like the material cost of each airframe will run in the neighborhood of $15,000. However don't hold me to that as not all the parts have been finalized nor have I gotten any firm numbers for some of the other parts yet to be designed.

Yes, this is more expensive than if I used more coventional components and items like stamped ribs, but then this is not meant to be a conventional airplane. So far I think the price increase is well balanced by the substantial difference in the ease of assembly and of course, the airframe's strength and durability. I wouldn't be doing this if I wanted to build just another RV.
 

Midniteoyl

Well-Known Member
Joined
Sep 3, 2003
Messages
2,406
Location
Indiana
If you can get all of the airframe from one sheet, I would think that it would actually be cheaper in the long run compared to using sandwiched plate/honeycomb just in the setup time alone...

Agree with the 'just another RV..'

Nice plane Bill
 

Othman

Well-Known Member
Joined
Aug 23, 2004
Messages
355
Location
Ottawa, ON, Canada
Pretty impressive design... looks like a Boeng engineer just stepped out of the 7E7 design office and started working on a kit plane.

All you're missing now are chemically milled skins.

I would be careful about publishing "your cost" to build the airplane. All the people here are gonna freak out when you quote them "their cost" to buy it (after the markup).
 

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
Finally got around to conducting the new bonding tests. The test consits of 18 samples that cover two material types and three different surface preparations. The two different materials being bonded are a bare sheet of 6061-T6 and a Clad sheet of 2024-T3.

The surface preparations include a simple acetone wipe, a mild scuff (220 grit sand paper) with acetone wipe, and an aggressive scuff using a three inch diameter 3M Scotchbrite pad on an angled die grider/polisher, also with an acetone wipe.

I am only interested in the peel characteristics since the supplied infomation indicates that the shear strength on bare untreated aluminum is substantially more than the structure requires (about 4,100 psi.). The tests I am conducting are not exactly to ASTM specifications but as I am more interested in qualatative information rather than quantatative, the results will be more than sufficient.

The first set of pulls were conducted on the two types of materials, with the three surface preps each. The adhesive has now cured for approximately a week at room temperature. It is still slightly soft (you can indent it with a firmly pressed finger nail) but seems to exhibit excellent properties. The eventual full cure is expected to be quite hard.

There was one problem with the coupons in that the clamping fixture was not able to exert the level of pressure that I wanted to achieve and as such each sample had a bit more adhesive between the coupons than I would have liked. In several samples the lack of clamping force also resulted in poor contact and a few minor voids. None of this however has seemed to affect the results to any appreciable degree.

All the test coupons were pulled to failure - in each case the failure was in the adhesive itself. In other words, the adhesive did not pull away from the metal - the adhesive itself failed in what would essentially be a tensile failure. Peel did not occur.

The picture below shows two things. First, it shows the rather crude setup I'm using for this - however it works.

Second, it shows the strength of the bond. The vise this is clamped in is a 6" milling vise, which weighs just short of 100 pounds. The pull has lifted most of the weight off the ground. The coupon is 1.65 inches wide.

After the picture was taken, I pulled harder, and lifted the entire vise off the ground - the bond did not fail.
 

Attachments

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
This next picture is one of the failed coupons. Although not really a great photo, it clearly shows where at the coupon corners the bare metal ends and the adhesive begins, and the fact that only the adhesive has failed, not the bond to the metal. Also, the curvature of the failed coupons shows a gradual failure, rather than a catastrophic one, where the bond retained much of its strength even with the stress concentration created by the propegating sharp corner of the failed adhesive.

The failure type seen here was observed in all the coupons, regardless of the material or surface preparation.

The next set of tests will be done in about a week or two, in order to see how the material behaves when it is more fully cured and thus a bit more brittle.

The third test will be in about a month, after the remaining coupons had a chance to be exposed to outdoor conditions.
 

Attachments

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
01/31 update - a couple of steps forward, but also one or two back.

Today I conducted the next set of pull tests on the bonded aluminum coupons. The resin has now cured to a much harder state - enough so that a pressed finger nail will not leave any marks. If one presses hard, an Xacto blade will penetrate the surface, but barely.

The first pulls were conducted on the samples with no surface prep, just an acetone wipe. The pulls lifted the entire 100 pound vise off the ground. As a result, the coupons were then cut on a band saw to a width of .75". The pulls again lifted the entire vise off the ground, even with the narrower width. Only by lifting the vise off the ground and then giving the line a good jerk did the coupon finally fail. (A secondary effect was that when the coupon failed, the vise nearly flatened my foot - but that's not really a part of the study).

As in the first test, the failure was in the resin, not in the resin/aluminum interface.

The same results were evidenced in the coupons that had the scuffed surfaces.

There will be a third set of tests in another week or two, which will look at the coupons that have been outside for the past couple of weeks. Prior to testing this batch, all the samples will be placed in a freezer, as suggested earlier by Eric (CNCRouterman).

Now the setbacks - it turns out that the bonding agent is quite caustic, or I am quite sensitive to its properties, or both. Shortly after I assembled these coupons I noticed that a bit of my forehead broke out in what looked like a mild form of Exzema. I didn't think anything of it until about a week ago when I assembled another sample structure. Although I used a set of fans to ventilate the shop the first time, I did not do so the second time around.

This time my skin had a worse reaction, simulating almost an equivalent of a chemical or acid burn - not pleasant. This means that in a production setting, this material will have to be used in a very well ventilated space, along with personal chemical protection for anyone in the immediate area. At this point I don't think this is unmanagable, but it is a hitch I did not foresee.

The second setback is the issue of cost. For the last two weeks I've been trying to reduce the cost of the design. A few changes in the fabrication methods helped (CNC routing is much more cost effective than water-jet cutting) but the raw material costs are still a bit too high for my tastes (not to mention the eventual customers').

Although the structural concept is still the same, I am currently re-evaluating changes in the primary internal framework. With a few minor changes, it looks like the interior assembly might be able to be constructed from prebent square tubing. I'm still waiting for feedback from a company that utilizes CNC bending machines and specializes in forming square tubing extrusions but our initial discussions have indicated that the bends could be done without serious effects to the tubing cross sectional geometry. I should know more next week.

And the last stumbling block - funding. I was anticipating at least four contracts this Spring but due to their own funding issues, at least three of the customers decided to delay their developments for the foreseeable future. As a result, I'll continue with the design process but the fabrication of the prototypes may have to be delayed for a couple of months - we'll see.
 

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
02/17/05 update - Today I conducted the final set of coupon pulls. This last set of test pieces were left outside for nearly a month, exposed to rain, sun and for the past week or so, freezing temperatures. The results were pretty much unremarkable in that they were identical to that of all the previous tests. In all except one case, the bond was sufficient to lift the 100 pound vise off the floor - failure was achieved only after repeatedly heaving on the line once the vise was unsupported - substantially more load than such a structure would ever see on an airplane.

The one early failure was determined to be caused by poor contact forces at the time of the initial assembly.

As a result, I am confident that the material and process of bonding the aluminum will be more than satisfactory for a general aviation product.

Now that that's done, I've also been looking at several different approaches for reducing the part cost of the airframe. The primary drivers thus far were the cost of the 1" thick aluminum plate and the machining that was involved in cutting the parts. So far I've come up with two alternate approaches.

The first replaces the previously pictured frames with pre-bent square aluminum tubes. With today's CNC tube benders, one supplier is indicating that maintaining .020" tolerances should not be a problem. These frames would then be joined to the longerons with a series of specially flanged fittings, which would be used to position and bond the entire assembly together.

There is still the issue of being able to bend the tubes without serious wall distortion but those tests will be done within the next week (material was sent to the fabricator earlier today) so I should know soon.

The second candidate approach is to make the frames using investment castings. The patterns would be machined from a special tooling wax using a CNC router, then sent to a facility that can do the casting itself. The wax is easy to machine and is capable of holding amazingly good tolerances.

For those interested, both of the alternate methods dramatically decrease the part costs. The original firewall framework, for instance, was going to cost somewhere around $450 for the materials and machining, possibly even a bit more.

Using the bent tube approach, the framework cost drops to less than $130, including the material, the bending, and the additional tube lengths that will comprise the diagonal braces.

The investment casting approach will be a bit more as it will cost about $60 for the coating and casting process and maybe something over $100 for the machining of the wax.

In volume, obviously these costs will decrease, thus showing good promise for keeping the kit prices at the level estimated earlier.

I have however slightly redesigned the airplane (mainly to account for the bending requirements of the square tubes) so once I finish the lofting work I'll post the new pictures.
 

sonex293

Well-Known Member
Joined
Aug 27, 2004
Messages
46
Location
NC
Thanks for the update. Looking forward to the new photos.

--Michael
Sonex #293
 

CNCRouterman

Well-Known Member
Joined
Nov 26, 2003
Messages
52
Location
Minnesota, USA
Stereo Lithography

Bill,
I checked into Stereo Lithography. The cost of admission to be able to lith` a 20" x 20" x 23" work area is around $300,000 or so. The published tolerances are .005" per inch per inch (.005/in^2). Pretty sloppy even for a wood worker. Now, the actual achievable accuracies are much better, but that is what the manufacturer is willing to stand behind. The rate of throughput for the S.L. equip is about 7 hours for a coffee cup sized item. Assuming you leased it, and wanted a 5 year payback, you would have to charge a pretty penny per hour to crack a $6000 per month nut. Then of course, the usual assortment of associated expenses are on top of that, like electric, consumables, operator expense, insurance, etc...

Bottom line, for robust geometry items, a CNC Router like mine can rout things for much less money and in much less time than you could do with a laser cure/ laser sintering machine, not to mention that most CNC routers with have work volumes substantially larger in at least two if not all three dimensions.

One huge advantage for stereo lith and laser sintering equipment is that the material induced limitations for thin wall machining do not exist. Also, internal or hidden features that would be impossible to machine in one block can be easily done with these fancy rapid prototyping innovations. Lucky for me, I do not expect any serious competition from this technology.

The sales rep indicated that the typical buyer of these stereo lith machines is a company that needs it to support another process (such as investment casting) as opposed to a stand-alone profit center. He also said that improvements in the process are being made to facilitate single step - laser to application use in some plastics such as nylon. I don't expect to be in the market for one of these puppies for quite some time!

I found two shops in my area that do investment casting, both of which also have one form or another of RPL equipment.


--edit: I corrected my horrendous spelling..
 
Last edited:

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
Good morning Eric;

Sounds like the info you got on STL is about what I've seen. The only reason I mentioned it is because the local investment casting shop mentioned it as one possible approach - however he too indicated the possibility of some pretty impressively high costs.

Once I get feedback and the samples from the tube bender (probably by the end of next week or so), I'll have a better idea as to which process will deliver the better results. If the investment casting idea comes out ahead, I'll go ahead and redesign the frames and send you the files so you can take a look at what will be involved in machining the patterns and also what your local investment casting outfits will want to do the parts.
 

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
One of the drawbacks of trying to do something different from the norm is that despite the most dilligent of efforts, the best of conceptual approaches may not work due to more practical considerations. Such is the case in our latest approach to the framing of the fusleage.

To review, the basic approach that I'm taking towards the configuration of our airplane is to develop a structural concept that is not only strong and durable, but one that is also easy to fabricate and assemble, thus allowing the design to be built quicker and cheaper.

So far it has become clear that the first idea, cutting the fuse framework out of aluminum plate, is out of the question due to the cost of the aluminum, as well as those costs associated with the cutting of each structural framework. Although router cutting turned out to be the most cost effective, the parts still generated way too many chips for the concept to work effectively in a production setting.

The follow-on idea was to fabricate the cross sections for the cockpit out of extruded square tubes. With today's CNC driven tube benders, several potential suppliers assured me that they would be able to maintain about a .020" accuracy and repeatability for each of the cross sectional members. Given the fact that extruded aluminum tubing is not all that expesive, and that the bending costs were quoted at $37 per cross section, this seemed like an ideal solution. The only unknown was the supplier's ability to bend the tube without significant wall distortion.

Yesterday we discovered that this requirement was virtually impossible to meet. Although the fabricator was not able to generate the small corner radius to the magnitude I specified, they did come close. Close enough as a matter of fact that I was willing to reloft the airframe to take the larger bend radius into consideration. The problem though was that the bending process caused serious distortion of the tube walls. Since the distortion is virtually impossible to eliminate, and at times is responsible for cracks in the faces, it looks lke this idea is going by the wayside also.

Oh well, back to the drawing board. I still have three other ideas so there's still a chance one of them will deliver the properties and low cost fabrication that I'm after. Back to the testing.
 

Attachments

Last edited:

Midniteoyl

Well-Known Member
Joined
Sep 3, 2003
Messages
2,406
Location
Indiana
Would a couple relief cuts on the inside radius help with that? Course, the tig welding required afterwards would prolly defeat the low-cost... just a thought.
 

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
The only idea I've been able to come up with is something similar to what is used for forming tubing for trumpets and trombones - the tube is filled with a special liquid, which is then frozen solid. After that the tube is put on the bending tool and formed to the necessary shapes. The frozen interor keeps the tube from forming any distortions.

For this case I thought to fill the tube with sand and water, then freeze it solid. That mix should provide sufficient resistance to the distortion, however how much, I don't know. The problem though is the cost - I don't know how much it would add but off the top of my head I'm guessing it might again make the parts a bitt too prohibitive.

But as of today, I'm off onto some of my other ideas, so we'll see. (Boy, composites are a lot easier.)
 

Midniteoyl

Well-Known Member
Joined
Sep 3, 2003
Messages
2,406
Location
Indiana
(Boy, composites are a lot easier.)
Ok, how about composite? At least for this? A way of fastening the skin would be needed... Like a aluminum 90 degree angle or 'T' molded/sandwiched along the edge of each composite frame to allow for riveting of the skin...?
 

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
Well, the original idea behind this approach was to come up with a design that was not only strong, durable and easy to assemble, but also one that would allow the company to bring in reasonably priced components from outside vendors. In that way the company is not saddled with the costs and floor requirements of producing laminated components.

Yes, I know there are companies out there that are setup to produce laminated parts for their customers, but having seen the success of that (or lack thereof) at companies like Glasair, I really did not want to take that road. Metal parts are easier since once you provide the specific drawings to a fabrictor, all he has to do is maintain the tolerances. After all, "tin-bashing" is a well known and established process.

One would think that that could be done for composites also however the reality seems to be that there are several more issues to deal with which make the process a bit more difficult (and costly).

The other issue of that is that if I went to the compsites, I would want to substantially refine the loft of the body. But in doing that, I would now of course have to make the tools. I have done that on many occasions before, and I really did not want to do that now. The only two choices of that route are to either spend the next six months building a plug and molds, or spend about $10,000 and have them machined. Not my favorite choices.

The other drawback of doing the latter is that once you're done with the tools, you can only produce that aircraft. The idea behind what I'm trying to do with this airplane is to develop a structural approach that would allow the company to produce a line of airplanes that had the major parts in common.

Yes, I know - picky, picky. But then if one is starting from scratch and spending his own money, one can be picky to meet the desired goals.

So now I'm going through the process of evaluating the variables. Is it more important to meet the original goals or can I relax a few ideals and take a more compromised approach. I sure wish I had more funding for this to play around more with several of my ideas.

Well, I still have three other ideas to try before I change the goals - we'll see.
 

whirlybomber

Well-Known Member
Joined
May 4, 2004
Messages
67
Location
sydney, australia
orion,

perhaps this wont suit your low cost approach, but harley-davidson have a method of forming complex al. tube shapes very accurately (well i saw them doing it, don't know if anyone else does).

Basically they get the shape roughly correct with a tube (bent, cut and welded), then put the pipe in a steel machined mould. They close the mold around the pipe and whack HIGH pressure fluid through it, to swell the tube out to the shape of the steel mold.

AFAIK they get +/- .5mm tolerances on it.

Only problem is going to be the cost of tooling.

Cheers,
Brad
 
2
Group Builder
Top