Uber Ultra-light tear

[kennyrayandersen]

OK, so I moved the Uber Ultra thread to the build area now that the general layout has been presented. I'm so far thinking of fiberglass facesheets with a foam core, reinforced locally as required. I liked the A-Frame concept that Andrew laid out because I got a couple of special ideas for that. I'd like to work on a list of weight-saving ideas so if some of you have been thinking light-weight, or have a weight-saving idea, I'd love to hear about it.

So far:
composite sandwich construction
A-frame only -- no other sub frame
torsion axle
cargo netting - no cabinet doors (compare against 1/8 in aircraft-grade door with trim)
1 3/4 ball
10" tires (how much heavier than 8" tires?)
Alloy wheels?
composite shelves and galley divider
camp stove, small propane
If AC, which is lighter 5000 BTU Home Depot AC or Pet Cool
Which heater lighter

Weight costs, so I'm looking for light-weight solutions, even if they are marginally, and maybe significantly more expensive (if cool factor is higher the price can be justified!)

Thanks for the help. I'll post some profiles with a layout soon.

[brian_bp]

...
10" tires (how much heavier than 8" tires?)...

The Carlisle web site allows you to download a copy of their catalog, which has weights for both tires and wheels. They don't have every possible size, but they have a good selection so representative components are likely shown... and it is likely easier to open up the copy which has been downloaded than to hit a web site every time you're curious about a number.

At a glance, it looks like going from 4.80-8 to any 10" tire would greatly increase the weight, because all of the Carlisle 10" tires are very wide... a 12" setup would be lighter! Maybe other brands have narrow 10" tires.

8" - 4.80x8 tire at 5.8 lb, on 8x3.75 steel wheel at 4.7 lb; 21.0 lb total
10" - 20.5x8.0-10 tire at 14.35 lb, on 10x6 steel wheel at 10.6 lb; 49.9 lb total
12" - 4.80-12 tire at 9.7 lb, on 12x4 steel wheel at 10.0 lb total; 39.4 lb total

All tires in this example are Sport Trail model, load range B. In each case they are the lightest and lowest-capacity tire of that diameter, since any of them will have more than enough capacity for the Uber Ultra-Light.

The extra diameter doesn't necessarily cost too much in weight... a taller and skinny tire of the same capacity might not be much heavier than a shorter and wider one. In this case, there are no skinner 10" or 12" tires to match the very low load capacity requirement.

...
Alloy wheels?...

In the Carlisle selection, there are no 8" or 10" alloys, and in the 12x4 size the alloy is heavier than the steel (11 lb versus 10 lb). Although an aluminum alloy wheel can be lighter, often they are not... the same thing occurs in wheels for cars.

[mk10108]

I work in the packaging industry and came across a corrugated sheet (cardboard) infused with a liquid that increased its strength by a factor of 3. Its been a few years but I could try an find out what the liquid is.

Applied to a B-C flute or A-C flute, I think you'll have a low cost light weight structure, however the difficultly would be infuse the liquid throughout the length of the flute

You could build the structure and overlay a layer of fiberglass similiar to a UV blocker for wooden boats.

[kennyrayandersen]

Brian,
Thanks for the link. Oooh! So much for the 10” wheels. Now that I think about it, I do remember another site I was looking at that only had the wide 10 inch ones as well – I was thinking more like the 145/80R10 which at 10 Lb is available through Tire Rack with maybe a Minilite wheel, but even that alloy one weighs10 Lb.; even that’s not any lighter than the 12 in trailer wheels and tires (your observation that the alloy wheels aren’t any lighter seems to be pretty true – there’s always magnesium!$). So, first decision tires and rims – 8 inch – just too big of a weight penalty for the 10s or 12s.

[kennyrayandersen]

MK10108
I think you're tight it would be very difficult to get all of the epoxy, or perhaps a polyester resin down all of the holes. Additionally, it's actually not that light once they're all filled up. I'm pretty sold on the foam core with the glass facesheets -- I'm not seeing anything short of NOMEX and graphite that would be any lighter, and the graphite/NOMEX comb would be prohibitively expense. In fact it's light enough and strong enough that I think I'm using is for all of the shelves as well.

[kennyrayandersen]

Andrew,
When you get around to reading this:

I thought I really liked your A-frame because I didn't much like putting the punch load into the middle of the cabin floor. I wasn't going to use a steel frame but a 2"X2" square pine/fir perimeter frame buried into the floor panel with a 1"X2" cross member at the rear wall and that was about it. The full axle/half axle thing has me a little confused. I was thinking that because I was keeping the cabin mostly frame-less, that a full axle would react the kick loads from the axles without dumping the overturning moment into the floor. I was thinking to reinforce the wall locally at the axle attach with some plywood, also buried into the panel.

The vertical reaction from either the A-frame or single tongue would intersect the forward edge of the cabin, the load being compression and so no problem there (though with the A-frame spread out as it is, the load would be less at two locations (actually 1/2) rather that reacted at a single point), but the balancing load from the tongue then has to come out somewhere in the floor.

If I went with the single, I could bury an additional cross member in the floor where the aft end of the tongue stops. But in the case of the A-frame, it stops at the side next to the axle (the axle load being up and the tongue load being down, so they offset a bit). Initially, I was thinking that this seemed like it would be pretty light. You think the single will be lighter overall? Could you elaborate on what you were thinking?

I was also toying with the idea of making a composite tongue with a combination of wood fiberglass and some carbon rods in the upper and lower flanges (they aren't too expensive like that) -- that might yield a pretty light-weight frame, but I'm not sure about integrating it where the receiver mounts. That might be a little tricky.

[kennyrayandersen]

...

The fiberglass-faced aluminium-honeycomb core panels they used won't be cheap, but the cost of this sort of thing is reducing all the time.

Andrew

That's fascinating stuff, and the end result looks good, but the technique has a worst-of-both-worlds flavour to me. The honeycomb-cored sandwich is fine and relieves the outer body of any structural requirement, but the foam and 'glass over top is the nightmare of massive finishing effort which Andrew has mentioned before - all the work and none of the benefit of other techniques. The foam isn't a structural problem, but it's also not a structural benefit.

To use these panels in a trailer, it would make sense if they could be the final shape themselves, and not need and envelope of foam-and-'glass around them. I don't know how you would get the desired teardrop curve with these inherently flat panels; they seem limited to internal structures or "origami" styling (which I don't mind, but is not classic teardrop).

As for the car project... if stealth aircraft can be built with nothing but flat surfaces and work aerodynamically, perhaps a car can be, too. Skip the whole carved-foam and fiberglass covering stage!

The beauty of using composites is that you CAN put the strength were you need it and in the shape you want, but that has to be done at the time of manufacturing. In the case of the F117A stealth fighter, it performed so poorly [aerodynamically speaking] because of the flat surfaces, it was called the wobbly goblin by those who flew it (I actually know one of the guys who was a test pilot for it). Most all aircraft that use composites (I've worked on the B-2, A-12 (ill-fated), F-22, and F-35) use a mixture of solid laminates as well as some type of core material usually NOMEX honeycomb generally for stability or to increase the load carrying capacity. This can be done for quite complicated and even double curvature panels (with some splicing). The beauty is that after you have built the tooling, the outside becomes the same surface as the mold and the advantage that you pointed out becomes a reality -- the curvature is set at the time that the parts are laid up and the outside finish is excellent. However, when building a one off, or very low production rate, you just can't do it like that because the tooling would eat you alive. The Aussie car was OK for what it was, but the manufacturing process would certainly change for production, and though the outside finishing is difficult, you're still time ahead because you built no tooling. Additionally, after the foam is skinned, it very much becomes 'structural' acting as the shear member between the outer glass facesheets.

Once the flat sheet is set, putting a curve in it is a nightmare. You could use them for the sides and floor, but the cost is just too darn high [so far].

[angib]

I wasn't going to use a steel frame but a 2"X2" square pine/fir perimeter frame buried into the floor panel with a 1"X2" cross member at the rear wall and that was about it.

If you are designing the body as a proper stressed-skin, then much of any framing around the edge of the floor is wasted weight. It is the vertical panels of the body that will carry the vertical loads (as shear panels), so the objective at the tongue/A-frame is to transfer the vertical loads into the front of the roof skin.

Something like a 2"x2" cross-member may be needed at the front just to provide some meat to bolt the tongue/A-frame to, and the same at the sides for the axle/frame bolts - but in both cases you could just put in enlarged local blocks for bolting to.

The one exception to this would be if a full Cub/Modernistic-shape body was built where the front of the roof rolls right under the floor, so there is no near-vertical skin at the front - in that case a front cross-member is likely to be carrying a lot of the tongue/A-frame load.

I don't see a big difference between full axles and half axles - it's just a case of whether you buy the axle cross-member from the axle manufacturer, or you provide it yourself. In practice I think a metal axle cross-member is needed - it's no problem to make the cross-member from composites (or plywood), but by the time you've added the local reinforcement either side to bolt the half-axles to, you might as well have put in a metal cross-member.

Andrew

[Arne]

One thing about a full torsion axle, it is far heavier than I expected. I'm guessing around 70-80 pounds. I might consider using stub axles and welding a piece of light angle iron across the front and back from one stub to the other. I think it should provide the stability not found in just trying to bolt them to wood....

Another way might be to add well a piece of metal a foot long or so to keep the stub from twisting if bolted to wood.

I can almost smell another tear drop build in my life.

--------------------

Instability of design

Stealth aircraft are designed with a focus on minimal radar cross section (RCS) rather than aerodynamic performance. Highly stealth aircraft (the F-117 Nighthawk and B-2 Spirit) are aerodynamically unstable in all three axes and require constant flight corrections from the fly-by-wire system to maintain controlled flight. Most modern non-stealth fighter aircraft (F-16, Su-27, Gripen, Rafale) are unstable on one or two axes only.

[mk10108]

MK10108
I think you're tight it would be very difficult to get all of the epoxy, or perhaps a polyester resin down all of the holes. Additionally, it's actually not that light once they're all filled up. I'm pretty sold on the foam core with the glass facesheets -- I'm not seeing anything short of NOMEX and graphite that would be any lighter, and the graphite/NOMEX comb would be prohibitively expense. In fact it's light enough and strong enough that I think I'm using is for all of the shelves as well.

I need to clarify, the liquid doesn't fill the flutes. The liquid is misted into the flutes and attaches to the fiber board. The air space is retained.

However foam has more engineering information than this product and would give you a know result.

[brian_bp]

... I was thinking more like the 145/80R10 which at 10 Lb is available through Tire Rack with maybe a Minilite wheel, but even that alloy one weighs10 Lb.; even that’s not any lighter than the 12 in trailer wheels and tires...

I was thinking after posting that there might be some automotive tires in a more reasonable width - 145/80-10 is the original Austin Mini size, and even narrower would better suit the trailer. Too bad the Mini tires are not lighter; they're certainly overkill in capacity.

... ...(your observation that the alloy wheels aren’t any lighter seems to be pretty true – there’s always magnesium!$)...

Aluminum alloy wheels certainly can be lighter than the same size of steel wheel, but in some cases the alloy construction is just used to make a better looking wheel, and that seems to be the case with the small trailer wheels. Light wheels should be available in 10" for the Mini (e.g. Minilites), but the bolt pattern and offset may be problematic if using a common trailer hub.

[brian_bp]

To be fair, I didn't say that flat-surfaced stealth aircraft (the F-117 being the best known) were good aerodynamically... just that they work. The trailer doesn't need to fly (indeed, staying on the ground is good), it just needs tolerable drag and decent stability. I suspect that any random shape won't be much worse than many trailers.

A flat-surfaced design can certainly be stable; the F-117 has many challenges, and passive aerodynamic stability was not a priority.

[brian_bp]

Real data for the suspension components would be good. UCF lists the weight of half-axles on their web site, but I didn't see it for full axle assemblies... but Andrew apparently has been been working with some numbers. Can we put some numbers out there?

The smallest Flexiride individual units (FR-425) appear to be 6.48 lb (2.94 kg) per side; I don't know if a full axle assembly direct from UCF even exists in the smallest range. The tube across the trailer in 2,000 lb capacity Flexiride full axle assemblies from UCF is only 2.5" square 0.125" wall tube, and would replace some of the structure required by a half-axle design.

[aggie79]

One thing about a full torsion axle, it is far heavier than I expected. I'm guessing around 70-80 pounds.

I second Arne. I have a derated Dexter #9 (no brakes or backing plate) and it is heavy. Maybe tonight I can grab that thing and stand on the scales and let you know an actual weight.

The #8 Dexter is all a teardrop really needs. but it is more difficult (or impossible) to get the 5-bolt hub in the pattern to match my tow vehicle. I don't intend to use the same wheels as the tow vehicle, but I did want to be able to use the tow vehicle spare tire instead of carrying a spare of different size for the teardrop.

[aggie79]

So, first decision tires and rims – 8 inch – just too big of a weight penalty for the 10s or 12s.

Just a thought - if you go with a wheel/tire/bolt pattern for the teardrop that is different than the tow vehicle, then your weight multiplier for the teardrop wheel and tire should be "3x" rather than "2x" to account for the additional spare tire for the teardrop.