The XPB Stage 1 underway.

[Nobody]

The impact of the firewall into the soil is a very important factor in the crash worthiness of an aircraft. In many accidents into soft soil the firewall digs in resulting rapid deceleration of the occupants. This can be the difference between walking away with some bruises and death.

 

You may have seen video of some of the full scale testing nasa did on old GA aircraft using a pendulum type apparatus into the ground. The learning from that testing has been compiled into the Crashworthiness design guide at the link below. Have a read of section 6.8

 

http://www.niar.wichita.edu/agate/Documents/Crashworthiness/WP3.4-034043-036.pdf

 

 

[gandalph]

Co-incidentally I was talking to some Chinese today and they reckon with a mix of some melted down coke cans, old urine specimen jars and melamine that some lightweight imitation oak could be had.

I thought all their melamine went into their baby food milk substitute formula.

 

 

[kasper]

The impact of the firewall into the soil is a very important factor in the crash worthiness of an aircraft. In many accidents into soft soil the firewall digs in resulting rapid deceleration of the occupants. This can be the difference between walking away with some bruises and death.You may have seen video of some of the full scale testing nasa did on old GA aircraft using a pendulum type apparatus into the ground. The learning from that testing has been compiled into the Crashworthiness design guide at the link below. Have a read of section 6.8

 

http://www.niar.wichita.edu/agate/Documents/Crashworthiness/WP3.4-034043-036.pdf

Agreed ... in theory ... but in practice I have seen an ultralight under full power at near takeoff speed plow into an earthen bank collapsing/crushing the entire 'soft' fibreglass cowl, tearing off the tailwheel undercart main legs come to a halt quickly enough for the parasol wing to snap the box spar into matchsticks on an absolutely vertical firewall with a V in the centre of the fuselage to 'assist' with digging in ... and it ended with a pilot sitting in the middle of the mess with absolutely nothing more than bruised ego and a very, very broken airframe.

And with a metal airframe - even one with very strong edges as per the Bex design - the force of impact great enough to collapse the engine frame/tear off the engine will almost certainly collapse/deform the structural members on the lower fuselage in a rearward direction creating a slope you want to design in ...

 

I am not convinced that designing it in is actually adding much in terms of safety and it makes the build that much more challenging. I'll stick to my vertical firewalls and if I am looking for a new plane in the future that I do not build myself its not really an aspect of design I will be concerned with.

 

And yes this is anecdotal and one event only BUT the Wichita document you reference is looking at 30deg nose down arrivals and design elements to get the forward fuselage and undersurface of the fuselage to retain sufficient strength to allow the airframe to rotate to the level attitude and allow the horizontal energy to dissipate over a longer time interval without plowing an earthern bank in front of the firewall ... if I am hitting 30 deg nose down at speed I am not flying into the crash at minimum airspeed because I am clearly not stalled ... sorry but the design to survive flight into terrain (which is what the study is modelling in my opinion) is not one that i am looking for in a VFR only ultralight - just my thoughts

 

 

[kgwilson]

Been mainly on the computer designing or tending another personal item the last few days, but did manage to get the diagonals and some horizontal braces in.Also got a test firewall and a test rear floor cut, might get them on tomorrow or the day after.

 

Was whimpering in the corner so I let him outside for a while ...

Looks fantastic. I can see its ancestor in my frame which with the steel subframe probably weighs double what yours weighs

 

 

 

[bexrbetter]

From the looks it's a plumb square up-and-down firewall rather than slanting rearwards at the bottom to avoid digging the fuselage in after the engine comes off in a crash?

 

The impact of the firewall into the soil is a very important factor in the crash worthiness of an aircraft.

Well you Guys had better get to the Vans Forums and tell the owner's of the world's most popular aircraft kit to stop flying immediately for their own sakes - and whatever you do please don't look at a Zenith, also one of the world's most popular kits, with it's obtuse angle 'shovel nose' ....

 

But ok, if you want me to address this seriously, the firewall is perpendicular to the ground but; A/ the lower tubes extend away from it downwards and B/ the very strong engine mount primary tubes actually come up out of the lower tubes on an angle that might satisfy you guys, as described here ....

 

 

 

[bexrbetter]

And with a metal airframe - even one with very strong edges as per the Bex design - t

The corners are intentionally built strong for the purpose of distributing impact loads.

 

There are a number of safety ideas going in, but a bit early to talk about those.

 

 

[Kiwi303]

Can we see the hot babe in the small bikini?

 

 

[bexrbetter]

Looks fantastic. I can see its ancestor in my frame which with the steel subframe probably weighs double what yours weighs

Yes, anyone who asks me I am happy to tell them I think the Morgan has a great foundation and has inspired a few ideas through this build.

 

Of course all thanks to the KR2S ..

 

 

Darn, I over cropped it, sorry.

 

 

[SDQDI]

Can we see the hot babe in the small bikini?

I did a little hacking and found the image of babe in a bikini, you can thank me later:thumb up:

 

 

 

[rgmwa]

No wonder he cropped it.

 

rgmwa

 

 

[Marty_d]

I did a little hacking and found the image of babe in a bikini, you can thank me later:thumb up:

[ATTACH=full]44026[/ATTACH]

Well that keeps the farm boys happy.

 

 

[Kiwi303]

I did a little hacking and found the image of babe in a bikini, you can thank me later:thumb up:

And here she is all hot and nekkid! (NSFW, you'll drool on the office computer)

 

 

 

[Deskpilot]

Must'ave over cooked her as she's now under ground.

 

 

[AussieB1rd]

Some extra crackle pls.

 

 

 

[bexrbetter]

So the firewall eventually went on with a few very minor mods needed later ...

 

You can see where the engine mount attachment tubes will come out (the 4 square holes of course) ..

 

 

 

 

[bexrbetter]

So I've decided to try a wing stub out rather than a feed through wing.

 

The best picture I can find to describe that is of the KR2, something, but not the same, like this ..

 

 

It makes it easier and cheaper to develop landing gear and a folding wing on the end rather than a removable wing. It should make the Pax area stronger and safer too.

 

I have some thoughts and just fiddling in 3D these couple of days with them.

 

 

[pylon500]

It was a bit slow off the mark, but you're starting to make the right decisions.

 

A centre section is always better from weight and strength point of view.

 

It gives you the ability of wider U/C, somewhere closer to the CofG to put fuel (that's NOT in the cabin, lower stress wing attach points, and somewhere to do control linkages that are not buried under a seat.

 

Some other 'predictions' of your learning curve...;

 

Don't use an all flying tail, (or do it properly if you do...)

 

Don't use 'Telemorse' cables for primary control systems, I know they're tempting and look easy, but they make an aircraft fly like crap!

 

Don't create control systems with bolts just drilled through bits of tube, use ball joints where you can.

 

Use a trim tab, not bungees!!!!

 

 

 

[bexrbetter]

So anyway, yesterday I drew up another couple of main sideplates and crossmember support plate, had them cut this morning and managed to mostly get it in today less some verticals and some rivets. I had to add a dropdown to carry the floor (red arrow) and make it in 2 pieces to save myself stripping the frame down again to get it in but otherwise would be a single piece crossmember support plate.

 

All in all sits pretty well ...

 

 

 

 

 

[Head in the clouds]

The general idea of your carry-through is OK but I'm a bit concerned about some of the details.

 

I realise you've just put a few pop rivets in to hold things together at this stage but I trust you'll be solid riveting it in the final assembly?

 

You've removed a lot of the web material at the top and bottom with multiple lightening holes in the rivet zone. I wonder whether you'd be better off leaving that material there because, apart from saving the cost of the laser cutting, it adds to the spar cap value which at this stage looks as if it's way too light.

 

A quick calc of your spar caps - the spars look to be about 180-200mm high so the spar cap centroids are about 150mm vertically apart. I'd guess you'll have an effective wingspan of about 8m and your centre section is about 2m so each bolt-on wing will be about 3m span-wise. Consequently the half-span of each wing panel is about 1500mm providing a leverage/moment ratio of 1500:150 at the spar caps = 10:1. If your all-up weight is around 600kg then each wing panel will be carrying about 3m/8m of the total weight, minus the panel's own weight and any fuel the wing might contain = (3/8x600) - 50 = 175kg. At 6G that becomes 175x6 = 1050kg. At the ratio of 10:1 that means the spar caps must yield at more than 1050 x 10 = 10,500kg = 10.5T in either compression or tension.

 

The spar caps shown in the pictures appear to be just architectural grade (5 series aly), which apart from having poor fatigue characteristics, has a yield strength of only around 28kpsi/190MPa = 12.7T/sqin = 19.7kg/sqmm. Therefore you need 10500/19.7 = 532sqmm of cross-sectional area at each spar cap. Well that's the theory anyway. In fact you need more than that because you need to allow for break-out values for the bolts in less-than-ideal bolting arrangements - as they invariably are when joining spar caps of angle material - and the yield strength of aluminium in compression is about 80% of the tensile yield strength, so cantilever wings typically have larger top spar caps than lower ones. Then you need to consider the buckling issues as the top cap tries to move towards the lower cap under normal upright flight loads (exacerbated by any dihedral you might incorporate). That requires a significant web stiffener at the wing attach point. Perhaps you plan to add that but your outer web lightening hole has removed most of the material where you need to attach it.

 

Currently your spar caps appear to be something like 30x2mm angle providing a cross-sectional area of around 120sqmm which is about 4-5 times less than required. If you used 6061T6 (yield 276MPa/40kpsi) you'd need 375sqmm of cross-sectional area, or about three times as much as you appear to have used. In similar builds my carry-through caps were made from Alcan/Capral/Alcoa 2in x 1/4in 6061T6 structural angle (structural angle has a radius in the internal corner which helps resist buckling), which has a cross-sectional area of approx 600sqmm and a yield value of just over 17T. By the time bearing strength and compressive strength is taken into account it works out just about perfect for this class of aircraft.

 

The web - in conjunction with the caps described above - and after practical testing - I found that I needed a 3mm thick web, solid riveted with 3/16 rivets at 25mm c/c (Type 2117-T4 Hard 26kpsi shear, not Type 1100 Soft 12kpsi shear, and most certainly not any kind of pop/blind rivet (around 6kpsi shear), or even commercial grade aly solid rivet which normally test at around the same as Type 1100). Also - the largest lightening holes that didn't cause a significant strength loss (in buckling) were small enough to be pointless, so I just used a solid web. Additionally - web stiffeners were needed at c/c spacing no more than the height of the web, and half that spacing at the outboard ends i.e. stiffeners about every 200mm along the web and the last 2-3 of them at about 100mm spacing.

 

Another note on spar web lightening holes - it's important to keep the load-path direction in mind -

 

In the part of the wing that is cantilevered - i.e. the outboard wing panel - the load direction in the web is at 45 degrees, consequently you must have your lightening holes sufficiently far enough apart (and/or small enough) to allow a 45 degree load path between them - and it's beneficial to flange the lightening holes to provide those load-paths with stiffened 'edges' which significantly helps to resist localised buckling and to stabilise the entire spar.

 

In the spar centre-section the load path is quite different. When both wings are lifting/loaded equally (or close to equally, as they usually are) then there is a small amount of 45 degree tension in the web as the lower (in upright flight) cap tries to stretch and the upper cap tries to compress, but the far greater load is that of the upper cap trying to move toward (or apart from) the lower cap. So keeping the web from buckling under compression is the major issue, hence web stiffeners are crucial and lightening holes don't play much of a part in that area.

 

It's very easy to underestimate the immense loads that must be resolved by the spar carry-through.

 

Rear spar - at this stage you don't appear to have provided for it but do keep in mind that it's better to have a full carry-through than to join stubs outside the fuselage. This is to resolve uneven drag loads i.e. lozengeing of the wing/fuselage junction. And to achieve that you'll need substantial triangulation structure between the main and rear carry-throughs which in this configuration can be made from flanged sheet material and skins (rather than extrusions) to form a tough monocoque seat-base structure - hint ... consider the control runs while designing it ;-)

 

One last thought - earlier you spoke of your firewall. I've seen practical demonstrations of an engine bay fire. Even an ignited bad fuel spray or hot oil spray (912s have external oil lines) doesn't cause too concerning a fire. That is - until you add airflow of 60kts or so, at which time it becomes a huge blowtorch. An aly firewall of 3mm thickness was shown to burn through in less than 15 seconds - then that fire is blasting straight into the cabin. After seeing that demo was the last time I used an aly bulkhead instead of a proper firewall. Firewalls must be made from stainless steel, and though stainless is a heavy material you only need 0.5mm thick stainless sheet and that is less than the weight of even 1.6mm aly sheet. And 0.5mm stainless will not melt in event of an engine-bay fire.

 

 

[Marty_d]

Firewalls must be made from stainless steel

Hi HITC,

 

My 701 plans call for the firewall to be made of 26 guage (0.457mm or 0.018") galvanized steel. Is it mandatory or highly advisable to change this to S/S, or is galv an acceptable substitute?

 

 

[Head in the clouds]

Hi HITC,My 701 plans call for the firewall to be made of 26 guage (0.457mm or 0.018") galvanized steel. Is it mandatory or highly advisable to change this to S/S, or is galv an acceptable substitute?

Galvanized steel sheet, or more usually zinc plated steel sheet is also acceptable though I think the stainless has a slightly higher melting point. The effect of a bit of zinc burning off and breathing the gases therefrom probably wouldn't be much of a concern while you were dealing with the more pressing aspects of an inflight fire.

 

In fact galv sheet would be a better option when the rest of the airframe is aluminium because in event of any electrolysis between the two the galv would be sacrificial whereas between stainless and aly, the aly becomes sacrificial.

 

On a welded steel airframe like DooMaw, for example, you would probably choose stainless rather than galvanized steel sheet, though the stainless must never be welded to the chromoly frame because doing so causes localised brittleness of the 4130.

 

 

[Kyle Communications]

The savannah has a zinc or gal plated steel firewall supplied in the kit

 

 

[IBob]

The savannah has a zinc or gal plated steel firewall supplied in the kit

That saves me from digging it out to have a look.............

 

 

[bexrbetter]

The spar caps shown in the pictures appear to be just architectural grade (5 series aly), .

Yeah, Nah, I don't use any low grade materials in my builds and your figures, while no doubt spot on for the materials and sizes you have guessed, aren't the right guesses, besides the structure being incomplete - one of the problems with showing stages.

 

Thanks and I'm sure your intentions are well placed but, as many a girl has said too me, way to soon.

 

The firewall info is appreciated, I had intended to use a 25mm thick aluminium sandwich with fireproof material in the middle, I may review that especially as stainless is common and cheap locally.

 

 

[Nobody]

Yeah, Nah, I don't use any low grade materials in my builds and your figures, while no doubt spot on for the materials and sizes you have guessed, aren't the right guesses, besides the structure being incomplete - one of the problems with showing stages.Thanks and I'm sure your intentions are well placed but, as many a girl has said too me, way to soon.

 

The firewall info is appreciated, I had intended to use a 25mm thick aluminium sandwich with fireproof material in the middle, I may review that especially as stainless is common and cheap locally.

For the firewall also consider the rivets used to hold it on. There is no point having a SS firewall if you use aluminium rivets. You can either protect the with proseal ( or even better 3M FIREBARRIER 2000+) or use steel ones. In the overall scheme of things the extra weight of the steel rivets is small.