Bede 4 Bonded Wing

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

DaveK

Well-Known Member
Joined
Apr 21, 2007
Messages
382
Location
Northern California
A fellow in the local EAA chapter has a BD-4 that he has replaced the original wing with a new one from Bede that is uses bonded aluminum. The thing is it looks like the wing is bonded using something like Proseal (grey and sort of elastic). The wing skin is made up of several sheets of aluminum that wrap from rear spar up around and back again to the rear spar, so there are several spanwise joints. There are a few rivets by the rear spar, but otherwise it is clean except for the gaps at the butt joints between sheets (you can see the gray Proseal along the joints). Now the owner reports he is very happy with the new wing. It went together quickly and he flew it to Oshkosh and back this last summer. But the Proseal kind of worries me. I know from my background that bonding aluminum has traditionally been difficult at best (for a homebuilder) and I couldn't find much info regarding actual strength properties of polysulfides (Proseal). So anybody have any good info on this type of adhesive, used in this way? I was thinking of setting up a small test using a sheet of aluminum over a closed box that I could then cyclically vary the pressure inside to simulate loading and unloading a wing skin to see if the Proseal would actually last that many cycles. Don't know the guy well, but I would still hate to see him kill himself.
 

orion

R.I.P.
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
In my view, the bottom line here is that Proseal is a sealant, not an adhesive. It is formulated for the sealing of various surfaces and a such, it does exhibit relatively good adhesion properties that allow it to attach to a variety of substrates. It is also ductile, which allows it to stay in place regardless of service environment. However its use generally does require that the structure it's sealing is connected together in some other physical manner.

There are several variation of Proseal, each of which exhibits slightly different characteristics in service and so it is difficult to make any conclusive statements regarding the specific properties of your friend's structure.

About two years ago we conducted a series of aluminum adhesive tests using several different bonding agents. Shortly before the tests, I had to use Proseal on my fuel tanks - the results on my airplane seemed to be quite good so I decided to include the material in the tests. The bonding samples included four different aluminum alloys (6061-T6 bare, 6061-T6 Clad, 2024-T3 bare, 2024-T3 anodized clear), with three different surface prep techniques (degreased, very light fine scuff, coarse and aggressive scuff). The coupons were bonded in a continuous sheet, which was cut apart after cure to form the actual coupons for the tests. The purpose of the tests was primarily to examine peel characteristics.

Out of the half dozen or so bonding agents, the Proseal was most disappointing - the coupons separated and fell apart during the cutting of the sheets. The separated material exhibited clean and shiny surfaces, showing that there was little or no physical adhesion to any of the tested samples.

This was further verified later when I had to do some work on the tanks in my airplane. Initially I thought I'd have to grind off the Proseal but it turned out that I was able to peel it off the parts and fuel lines with relative ease. Yes, it was tough and required a bit of work but it separated from the aluminum and fiberglass quite easily.

As such, based on that experience my assessment would be that a Proseal assembled wing structure would be an accident waiting to happen.
 
Last edited:

DaveK

Well-Known Member
Joined
Apr 21, 2007
Messages
382
Location
Northern California
That is exactly what I was afraid of.

Bede is marketing this as the next great way to build a wing and the guy I know is just the first customer. I'll probably get hate mail for this but it seems like everywhere Bede goes disaster of some sort follows. I can't believe people keep buying his stuff.

Hopefully a small debond will happen and will be caught before the whole wing unzips:shock: I don't think the single row of rivets on the back will really do that much.
 

orion

R.I.P.
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
If history is any judge, Jim Bede's products have proven numerous times that he is a rather inadequate designer, especially as his abilities relate to structures. Even the FAA and the NTSB have both remarked, very bluntly at times, on the inadequate structural considerations used in the design of several of his past products.

I would strongly urge the gentleman to go ahead and rivet his wing as if it had no bonding agent in it.
 

DaveK

Well-Known Member
Joined
Apr 21, 2007
Messages
382
Location
Northern California
Well the problem is the ribs are aluminum honeycomb with thin facesheets. There isn't a flange to rivet to except along the rear spar. As it was described the 1-inch thick honeycomb ribs are routed out, slide over the tube spar and bonded using the Proseal, then the edge of each rib is slathered with Proseal and the skins wrapped around them and riveted to the rear spar. I assume some type of strapping was used to keep it tight while the goop cures. As I mentioned, he has been flying this wing for a year, so I don't think there is much he can do but trash the wing and start over again. Which is something that will likely be hard to convince him to do.
 

orion

R.I.P.
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
Yep, you're right, I don't see a way to fix this unless he'd be up to quite a bit of rework, and fabrication of the secondary components.

As I said, the different formulations of Proseal behave in slightly different manners so this one may have better adhesion than the others but even so, from what you're describing, there isn't even a structural connection between the ribs and the spar - just the adhesive, which tends to "tear" relatively easily. As such, he is depending only on the goop to hold all the torsional stresses that are generated by the airfoil and the deflection of the controls.

Assuming at least a reasonable bond, this type of structure may hold for normal day to day operations. The problem is that there is no way to know what it would do during more severe maneuvers, either pilot induced or forced on the airplane by the environment. As such, my own opinion is that this is very severe risk for him and anyone flying along (or living underneath his flight path).

It would be nice to get a hold of some of the bonded structure, maybe sample coupons, and simply test them in a variety of ways to see how the Proseal actually bonds and how it holds up in shear.
 

Midniteoyl

Well-Known Member
Joined
Sep 3, 2003
Messages
2,406
Location
Indiana
The thing is it looks like the wing is bonded using something like Proseal (grey and sort of elastic).
While I too dont really care for Bede's methods and history, are we even sure this is Proseal?

And while on the subject of Bede:
Bill, Bede is the new 'engineer' for Exosphere - the company that bought Express - and is designing a new LSA. I fear the Express is officially 'dead'...
 

DaveK

Well-Known Member
Joined
Apr 21, 2007
Messages
382
Location
Northern California
Without going and talking to the guy again I found this piece from another EAA Chapter newsletter describing another "bonded" BD-4 wing. The guy I know also described the adhesive as "Proseal like" or "the same as Proseal" but I can't remember exactly.

In his place, Joe Lineau showed his
new BD-4 wing and described the differences
between his new BD-4 and his old one.
Although the old wing has a fiberglass skin
and ribs, the new wing is all metal. Both use
the same construction method, that is, twofoot-
long wing segments stacked on a tubular
aluminum spar. Only a few rivets hold the
0.032-in-thick skin to the ribs. Proseal does
most of the fastening, and the lack of rivets
leaves a very smooth skin.
The new BD-4 will be larger than the old one.
It will have 28 bays in the wing instead of 24,
and the fuselage will be 2 ft longer as well.
 

Jeremy

Well-Known Member
Joined
Feb 23, 2003
Messages
75
Location
Salisbury, England
I've checked the materials list for the BD4 (it's here: http://www.bedecorp.com/images/pricelists/bd4_bill_materials.doc) and it does look very much as if the wing is bonded with "fuel sealant" (unspecified). The BoM calls out 15 pints of the stuff for the wings, which is way more than would be needed to seal tanks.

Having stripped Proseal from the inside of a leaking tank, and discovered that the stuff really doesn't have great peel strength when bonded to alloy, I share the concern expressed here.

What I find odd is that there are other, well proven, bonding systems around that should have been used. I have used Hysol for years with great success, provided that the joints are very clean, dry and grease free. I've also experimented with some polyurethane adhesives (but not tried for real) and found that they give results that look quite promising. My current project is using foam/composite ribs bonded to an alloy tube spar with Hysol.
 

orion

R.I.P.
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
In the test's I've done the Hysol 9430 was the best (I primarily tested for peel). The failure mode though was a relatively clean break so I'd say you still have to be concerned a bit with exposed edges that might initiate a peel condition.

The biggest promise of aluminum bonding seems to be in the Methycrylates. So far I've only tested the products from Extreme Adhesives but Plexus and Loctite also make similar products. The Methylcrylate seems to actually react chemically with the aluminum and in the test that I ran about two years ago it was the bonding agent that failed, not the bond itself.
 
T

Tinbuzzard

Just to add another data point, when I built my BD-5 ten years ago I was dubious about the prosealed joints. The trailing edge of all of the control surfaces consisted of the skin being pressed together between two clamped bars and bonded with only a single rivet near the ends to prevent peeling. The very thin trailing edges were quite flexible and I worried about handling damage and temperture cycling. I added 3/32 rivets every few inches along the TE. To date, the trailing edge seams all look fine.

The rest of the airframe also has all joints bonded and fastened with pull rivets. I drove solid rivets wherever possible and substituted 3/32 instead of 1/8 along some of the joints between thinner sheets. (using an appropriatly greater number) These seams also seem OK to date. EXCEPT for the ones near the mainspar to fuselage connection and the wing roots! I had a landing accident where I removed the maingear involuntarily (since fixed) and the shock of the inpact broke the adhesion of the proseal in several places including many that showed no distortion of the structure or loose rivets. I have also found that the surfaces must be perfectly clean before bonding to get good adhesion as well as scuffed. Just letting a cleaned surface sit exposed to the air for more than half hour or so before bonding resulted in poorer bonding. I made a few test coupons and witness samples while I was building to monitor my technique. Bottom line, don't trust the proseal as a structural material by itself.
 

bmcj

Well-Known Member
HBA Supporter
Joined
Apr 10, 2007
Messages
14,007
Location
Fresno, California
The Methylcrylate seems to actually react chemically with the aluminum and in the test that I ran about two years ago it was the bonding agent that failed, not the bond itself.
Hi Orion,

Can you please clarify your statement about failure of the bonding agent vs the bond? I'm not quote sure what you mean here.

Thanks,
Bruce
 

orion

R.I.P.
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
When I peeled the two strips of aluminum apart, both had an even coating of the Methylcrylate adhesive remaining - in other words, when I peeled the coupon apart, it peeled due to the tensile failure of the adhesive at the point of peel, not due to the bond failure of the adhesive to the aluminum. The amount of force it took to peel the parts apart was substantial - I had to fix one coupon in a vise while I rolled the other around long nose pliers, sort of like opening the old peel type of sardine tins.
 

MalcolmW

Well-Known Member
Joined
Jan 21, 2007
Messages
118
Hello, Dave K.,

for additional info on bonding aluminum, the discussion on 'Bonded Aluminum Construction,' (January 23, 2007) goes into some detail as to the nitty-gritty details of adhesive bonding.

That being said, Orion and I do not agree completely about adhesives, particularly as to whether epoxy or acrylic adhesives are the best choice for aluminum. Both work quite well in different applications. Both provide superior strength and durability as long as the substrate (the surfaces being bonded together) are meticulously clean (wear white cotton gloves while handling after cleaning), and adhesives are mixed exactly to specification.

The truck body industry uses acrylics extensively. The military and commercial aviation industries use expoxies.

I suspect that a metal Bede wing (properly cleaned) bonded with either an acrylic or an epoxy would provide long service. Do not be too concerned about honeycomb aluminum panels within the wing structure - commercially prepared, they will last a long, long time.

All the best. Fly safe.
 

orion

R.I.P.
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
The issue here is not the bonded honeycomb panel itself - it's just that it's essentially butt-joined to the aluminum tube spar and bonded with a compound that is a sealant, not a bonding agent.

As far as our agreeing on adhesives, we may actually be closer than you think. It's just that I do not have experience with Polyurethanes and as such, cannot make a qualitative analysis of their use.

I however have personally used and tested numerous epoxies and Methylcrylates, both within large aerospace firms and on my own and only based on that experience I've come up with a list of my favorites.

Speaking of tests, I'm now conducting a few more informal tests on the working and cure properties of several other epoxy agents for the bonding of graphite laminates. Hysol's EA9460 is showing great promise, although the others look fairly good also.
 

DaveK

Well-Known Member
Joined
Apr 21, 2007
Messages
382
Location
Northern California
I suspect that a metal Bede wing (properly cleaned) bonded with either an acrylic or an epoxy would provide long service. Do not be too concerned about honeycomb aluminum panels within the wing structure - commercially prepared, they will last a long, long time.
No, I'm not concerned about the honeycomb panels, they are likely the most well fastened pieces of this entire wing. What I was/am worried about is that from my experiance, and this has been echoed by others including you and Orion, that bonding is a precise process that needs utmost care or it will turn out badly and Bede is pushing this thing as a easy, "don't need to care about it" type process. To make matters worse he's promoting the use of a sealant as an adhesive.

So what I was really going for was backup from some experts that I could point to that would maybe get to see this guy to see the light and understand that he is taking a huge risk with that wing. I was also toying with the idea of setting up some form of test to show that the sealant would debond after repeated cycles to maybe prove the point.

On a side note I noticed a BD-4 carcass with engine siting out at Rio Vista Airport (between Sacramento and Oakland) with the old fiberglass with metal spar wing in tatters. Doesn't seem like the original wing turned out that well either. The fuselage looked dirty but otherwise intact.
Dave
 

MalcolmW

Well-Known Member
Joined
Jan 21, 2007
Messages
118
Orion, Dave K.,

I have experience bonding aluminun - not recent - but extensive.

Bonding aluminum requires (essential) good surface preparation, no matter what adhesive you use.

Cleanliness is next to Godliness when it comes to adhesive bonding.

First, solvent wash (MEK, acetone, etc.) to remove grease and printing from the aluminum surface. Then aluminum must be surface etched. The best etching is a chromate conversion using a chromic acid solution (or equivalent). This is best done at elevated temperatures (160 F) for 3 - 30 minutes. If elevated temperatures are not possible (this requires ventilation and masks), use a room temperature solution for at least 10 minutes.

In either case, follow with an immediate rinse using distilled water. Use white cotton gloves to prevent 'fingerprint' contamination once aluminum is washed. If distilled water is not available, clear rainwater (low in ions) works reasonably well. Air dry (do not use blower). Protect from dust and or contamination.

Bond within twenty-four hours, preferably with a holding jig.

Honeycomb panels (commercially produced): Prior to bonding, degrease and 'edge-dip' in cold chromate conversion solution (2 - 6 minutes), rinse with distilled water, shake off excess water, and air dry. Bond within twenty-four hours.

Epoxies produce the highest strength (both in peel and shear). Military and commercial aviation industry uses a variety of expoxy or modified expoxies for structural construction.

Acrylics are a close second in strength, however, they should not be used in a location which has constant exposure to 100 LL gasoline (wing fuel tanks). The aluminum truck body industry uses acrylic adhesives extensively, with good results. I do not know of military or commercial aviation use of acrylic adhesives - there may be some.

Use rivets to hold bonded areas in position or as a clamping mechanism. Solvent wash rivets prior to use.

Adhesive bonding requires meticulous preparation and follow-through. The benefits include freedom from joint fatigue, lower weight and higher strength.

As for polysulfide sealants / adhesives, they too, require meticulous surface preparation for maximum performance. The space program has used polysulfide sealants for many years, and in the right task, perform well.

I have previously discussed surface preparation, however, it is a topic that cannot be over-emphasized for those first attempting to use adhesives.

Do make trial runs. Do prepare test coupons and test them. Once you have mastered adhesive bonding, the test results will convince you that the military and commercial aviation construction practices really do offer significant benefits.

Fiberglass, without UV protection, slowly oxidizes. Wings on the BD may be shot, but the fuselage... who knows?

Polyurethanes: There are four classes of polyurethanes:
Polyether polyurethanes - good biological resistance.
Polyester polyurethanes - maximum physical strength properties.
Aromatic polyurethanes - least expensive, however unstable to UV radiation.
Aliphatic polyurethanes - light (UV) stable.
Remember these factors when thinking about using a polyurethane adhesives.

Good luck with your project, and fly safe.
 
Last edited:

orion

R.I.P.
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
Thanks again Malcolm for the great reply. I'm pretty sure that the prep that you describe is the very reason we have not seen much aluminum bonding used in our industry. I too would tend to favor epoxies for aluminum, primarily due to their resistance to fuel, but in my work I have pretty much come to realize that I couldn't really depend on the builder to go through the prep process so as to guarantee the best bond performance. It's one of the reasons I like the Acrylics since they seem to be less affected by surface quality. And now we also have a simple wipe on etching/primer product from Hysol specifically formulated for use with their line of Methylcrylates (it's not recommended for epoxies - darn!). Now all we need is one for the epoxy application.
 

DaveK

Well-Known Member
Joined
Apr 21, 2007
Messages
382
Location
Northern California
Thank you Malcolm and Orion for your informative responses. Nice to be on a board where you actually learn something.

Regarding preparation, as it was described to me, Bede only calls for a scotchbrite scuff and an acetone wipe. Not nearly as thorough as you have described.
Dave K.
 
Top