kennyrayandersen
|
| posted on 14/12/09 at 09:17 AM |
|
|
quote:
Originally posted by Benonymous
So is this the way it's done Kenny?
The bulkhead is blue and made of steel tube. It bolts into hard points in the tub where the red cylinders (bolts) are. This way the load is fed into
the outer skin of the tub and not into the rear plate of the tub. If they were bolted into the rear plate I agree, the load on the rear plate would
be trying to pull it out of the back of the tub.
That’s getting better. What you’ve drawn, I assume, presupposes that there is an axial member at the top of the tub. Also, the unsupported kink
about mid tub will cause the upper tube to go into bending. Also, the torsional stiffness would need to be provided by the center tunnel since it
won’t be supported by the sidewall.
What I was thinking a little more like is a box section on EITHER side of the tub – that being met by a bulkhead in the rear (and a corresponding one
at the front of course) which has the aft frame member pretty much drawn like you have it. The actual attach point though would have to be a fitting
that has fasteners in the plane of the aft bulkhead as well as a hard point (tension fitting) to the corner of the tub. I’m swamped at work or I’d
try and draw at least a sketch in word or PowerPoint. Maybe in another month when my contract ends and I’m unemployed (not to worry, I put a little
in the bank and there are jobs out there if you are willing to go to them (might have to do a little traveling!))!
The OP in the other chassis design thread has the aft frame narrower than the ‘tub’ (and the tub wasn’t torsionally rigid). This caused the load to
be dumped in away from the corner which put the aft inboard/outboard tubes in bending. If you had a box section on either side of the tub, you could
put your fitting that was tying the frame to the tub in-line with the inboard wall – you would still have a good load path and would avoid bending
that way.
|
|
|
|
|
ceebmoj
|
| posted on 17/12/09 at 12:46 PM |
|
|
quote:
Originally posted by aitch
i would use glass cloth and epoxy resins, slightly more expensive but will transfer stresses in a nore prediatale way with far less material and
therefore weight. the tecniques are all very well doccumented, take a look at this wing spar which demonstrates just how strong a seemingly thin
walled and skinned box section can be, these spars are rated at over 9g loading http://www.cozybuilders.org/chapters/chap14.html heres a link to the
firewall http://www.cozybuilders.org/chapters/chap15.html and to the fusalage construction http://www.cozybuilders.org/chapters/chap06_1.html in order
to be able to bolt on in this case the engine mount the firewall or bulhead usses plywood instead of a foam core, the forces as you can imagine in
attaching a 6 cylinder rotax enfine are going to be huge and similar meathods should transfer well to a car tub my concerns are that to interconnect
steel subframes to composite tun with GRP the weight will probably end up in excess of that using one meathod of construction or the other. i will be
looking into all three meathods, all composite, all steel and combination of the two, as i plan to add a body it may be that i go for steel and gain
experience biulding a composite body shell (less critical but wil gain me experience) before a second all composite build aitch
That is an interesting build diary you have linked to are there any more picks of the fire wall construction? All I can see is that the ply wood witch
is covered in GRP has a metal skin applied. Another are that I would be interested in seeing in more detail if it exists is how the frame that carry
the engine is joined to the tub.
|
|
|
kennyrayandersen
|
| posted on 17/12/09 at 04:58 PM |
|
|
It's interesting, but the pictures are crap for actual detail. The key is that they mentioned hard points. This meaning in the analytical
aerospace world means that that bulkhead would be backed up with something to give it stiffness and to provide someplace for the load to go.
Just think about it. Why would you put a punch load (out of plane no less) into a flat panel? Where is the support. That thing would be flexing big
time every time you put a load on it. The biggest mistake designers make (and I'm talking about professionals with degrees), is that they just
don't think about the load paths.
I'm not trying to belittle anyone by any means. What I'm suggesting is that even for trained people laying out this stuff it is not easy.
For the backyard mechanic with little experience or training, it can be disastrous, or worse yet deadly.
What Beanonymous posted is certainly getting closer. Just think about if you were a loa, where would you go?
|
|
|
aitch
|
| posted on 17/12/09 at 05:09 PM |
|
|
quote:
Originally posted by ceebmoj
quote:
Originally posted by aitch
i would use glass cloth and epoxy resins, slightly more expensive but will transfer stresses in a nore prediatale way with far less material and
therefore weight. the tecniques are all very well doccumented, take a look at this wing spar which demonstrates just how strong a seemingly thin
walled and skinned box section can be, these spars are rated at over 9g loading http://www.cozybuilders.org/chapters/chap14.html heres a link to the
firewall http://www.cozybuilders.org/chapters/chap15.html and to the fusalage construction http://www.cozybuilders.org/chapters/chap06_1.html in order
to be able to bolt on in this case the engine mount the firewall or bulhead usses plywood instead of a foam core, the forces as you can imagine in
attaching a 6 cylinder rotax enfine are going to be huge and similar meathods should transfer well to a car tub my concerns are that to interconnect
steel subframes to composite tun with GRP the weight will probably end up in excess of that using one meathod of construction or the other. i will be
looking into all three meathods, all composite, all steel and combination of the two, as i plan to add a body it may be that i go for steel and gain
experience biulding a composite body shell (less critical but wil gain me experience) before a second all composite build aitch
That is an interesting build diary you have linked to are there any more picks of the fire wall construction? All I can see is that the ply wood witch
is covered in GRP has a metal skin applied. Another are that I would be interested in seeing in more detail if it exists is how the frame that carry
the engine is joined to the tub.
thts just one of very very very many build diaries, if you do a search on cozy mk IV, Velocity, Berkut, and the list jut of this aircrafy type goes
on, within the tub's there are box sections at the top edges, a central box section a bit like a transmission tunel in a car the seat back
triangulates the top of the bulkhead to the bottom of the tub floor, when i get the chance ill post links to some other diaries that show some more
detail..
aitch
|
|
|
kennyrayandersen
|
| posted on 17/12/09 at 11:55 PM |
|
|
pics:
http://www.advancedcompositetech.net/images/engine.png
http://www.berkut13.com/engmnt.jpg
There are also several shots in the general buld area:
http://images.search.yahoo.com/images/view?back=http%3A%2F%2Fimages.search.yahoo.com%2Fsearch%2Fimages%3Fp%3Dengine%2Binstall%2Bplane%26ei%3Dutf-8%26y
%3DSearch&w=400&h=341&imgurl=www.berkut13.com%2Fengmnt.jpg&rurl=http%3A%2F%2Fwww.berkut13.com%2Fberkut24.htm&size=26k&name=eng
mnt+jpg&p=engine+install+plane&oid=fa55df4c80dee306&fr2=&no=18&tt=70&sigr=1143qmko5&sigi=10ris8g5n&sigb=12lpp642s
http://www.zenithair.com/design/912-firewall.gif
http://www.fly-aerotrek.com/photos/tech/912uls-engine.jpg
You’ll notice in the several pictures that the [typical] 4 mounting points are mounted outboard as far as practical on the bulkhead very near the
skin. What you don’t see is any reinforcement or backup structure that might (or should) be included.
What I did find while looking for pics though was very enlightening and apropos, I ran across this Vans aircraft safety bulletin. It was interesting
to say the least:
http://www.vansaircraft.com/pdf/sb08-6-1.pdf
Pages 5 and 7-10 are particularly helpful in demonstrating what I’m talking about. What we have here is a woefully inadequate design that they fix by
slapping on a doubler. First, the problem occurred for the exact reasons I was mentioning; which is load was being dumped (reacted) by flat-plate
bending in the bulkhead. This is a terrible load path to begin with and it really, IMO, should be backed up by a little angle fitting in the corner
behind the bulkhead. A tension fitting would receive the load from the engine mount and then pass it along to the outer skins. It would weigh mere
ounces and the design would be far more robust. You can fix it the way they did, but I don’t think they are making it a particularly robust design
that way and they are likely to have problems with it in the future. I’ve always fancied the idea of building a plane and generally Vans has a decent
reputation, but after seeing these photos, I’d take a HARD look at the one I wanted to build before making that leap (unfortunately not many buyers
are going to have the technical expertise to evaluate the structural integrity of what they are buying and will rely on the manufacturer to do that
for them…).
The whole problem would have been avoided in the first place had they properly backed up the attachment (i.e. their ‘hard-point’ was actually a hard
point!) like I was yammering on about in the earlier posts. This also demonstrates what I was saying about even professionals (at at least people
professing to be such) fairly frequently have trouble executing.
Now, let’s get back to cars. The fore and aft sub-frames that the OP is talking about mounting to a tub are going to be quite similar to these and if
those points are not backed up with a little structure, you can expect the same [crappy] results. IMO, there is no problem to do it; it just has to
be done with a little forethought and common sense. You can do it by heavily reinforcing the bulkhead, or by making a light-weight tension fitting,
but you must do something. These days you can get on-line, upload your solid model and have someone machine it for you relatively inexpensively.
There is not a good reason to do a half-assed job of it just because you don’t have the full shop (you should note though that lacking the full shop
you would have to have the FULL budget).
I know I only mostly work on fighter aircraft and helicopters, so I have some learning to do about cars. Nevertheless, the principles are universal
and if understood and applied properly it will result in a safe and robust design. In the end, we are back to where we started eh? Stick with steel
unless you want to play the Man of La Mancha.
|
|
|
kennyrayandersen
|
| posted on 17/12/09 at 11:57 PM |
|
|
Now here is a nice looking engine attach:
http://www.flycorvair.com/jan04601h.jpg
right out to the edge (where the skin can get the load and where the bulkhead won't get the crap knocked out of it. Also note the overlapping
stiffeners make a natural spot for the load to be dumped into (guess I'll have to take another look at that Zenair).
[Edited on 18/12/09 by kennyrayandersen]
[Edited on 18/12/09 by kennyrayandersen]
|
|
|
kennyrayandersen
|
| posted on 18/12/09 at 10:48 PM |
|
|
The following is a response to someone looking to build a monocoque-type chassis in another yahoo group I frequent. The responder lists some sources
that would be applicable to trying to do a composite tub (or other monocoque construction). There is probably some good information there as the
sources sound like they are relatively authoritative. I you , or anybody else wants to do this type of construction (composite or not), they should,
at a minimum, do a a little 'light' reading:
Woody,
Earlier this month you posted you were looking into producing trikes and wanted
advice on designing monocoques.
I was browsing in a bookshop last week and came across a book "How to build a
motorcycle engined racecar" by Tony Pashley, published in 2008 by Veloce in the
UK. The author built several hillclimb racers using tubing and then alloy
honeycomb sheets. The builds were covered by articles in Racecar Engineering
magazine. The book incorporates these articles and expands on that info. I had a
quick skim and it had some useful info that might help you. There are several
other books you may have or know about but I will mention them here for other
members as well who may be thinking on the same lines.
"Competition car composites: A practical guide" by Bryan O'Rourke and "The race
car chassis HP1540" by Forbes Aird. I have his earlier work called Race Car
Chassis - Design and Construction (1997) in my library and found it to be very
good. Composites can make a lighter, stiffer and stronger chassis than a
comparable steel tube chassis. The downside is cost and practicality. They are
difficult to make - e.g.drilling accurately in a honeycomb where the drill runs
off and even harder to repair. They also require a lot more design skill and
knowledge than a simple tube frame. But we need builders who can push the
envelope so best of luck if you follow this route. Have you considered using
timber plywood and epoxy resin. This is very suitable for one-off or small
production runs. The Marcos sportscar designed by Frank Costin in 1960 and the
Mosquito bomber in WW11 were built using this method. Costin used Aerolite 308
two pack resin glue which is still sold today as Selleys High Strength glue. But
there have been great advances in epoxy resins and glues. I have been working on
a timber monocoque and am using an aerospace glue called Hysol. It is used to
stick the skins to airliners. Mind you at >$100/litre it is not cheap. If you
want to study Costin's designs you could look for a copy of "Flying on Four
Wheels" by Denis Ortenberger. This is an excellent study of Costin and his work.
You might also look at the West epoxy system used in boatbuilding. The epoxy is
called Bote Cote and is used with plywood or balsa to create amazingly strong
composites.
I know a few members are looking at making several trikes. If you are, a 'must
read' book IMHO is one by Walter Korff called "Designing Tomorow's cars: from
concept, step by step, to detailed design" Published by M-C Publications in
1980. (there are 6 copies on Amazon). I found a copy in our State Library
several years ago and photocopied it there as you can not borrow it. He is/was
an excellent engineer who consulted to the large auto firms like Chrysler and
also built a range of small cars including trikes. Robert Q Riley's efforts
remind me of Korff's work. Many of his design views appear in later works by
other authors. E.g. Korff is the oldest reference I have found to the 'circle of
traction' concept in suspension design. More importantly, he has a whole section
in the book on what you have to do to move from a concept car like a running
prototype - essentially what we are trying to build - to production. It is
daunting reading !!
I shall suggest to our Moderator Joe that we create a page where members can
post book titles they have found helpful. This could act as a resource and
greatly assist our many new members, indeed all our members, to complete their
builds.
Cheers
Grant Campbell
|
|
|
ceebmoj
|
| posted on 22/12/09 at 08:34 AM |
|
|
witch forum is the post from?
|
|
|
kennyrayandersen
|
| posted on 22/12/09 at 09:22 AM |
|
|
http://autos.groups.yahoo.com/group/3WheelerReplicas/message/4745
which is a bit odd actually as most of the guys are doing something rather Morgan like, though one guy did a Messerschmitt Record Breaker
|
|
|
Sam_68
|
| posted on 9/1/10 at 12:48 PM |
|
|
Sorry - late to the party, I know: I don't visit the Locost forums often, these days, so I've only just come across this thread.
quote:
Originally posted by djtom
Westfield made a composite chassis car a few years ago. FW400 I think it was called. 400kg but OMG expensive.
I own one of the three FW400's left in this country. The chassis is a very nice piece of work, designed by an ex-Lotus F1 Chief Designer by the
name of Martin Ogilvie and manufactured by ex-Lotus composite guys. The cars that were built (there were only 6) were 'pre-production'
models and as such the monocoque is made up of lots of flat panels and simple mouldings bonded together. If they'd continued production, the
intention was to simplify it into fewer mouldings.
As such it's pretty much a half-way house between a 'proper' moulded and autoclaved tub and the sort of routed-and-folded flat-panel
tub being discussed above.
If anyone wants further details or photographs (or to take a look at the car), give me a shout.
With regards to the routed-and-folded approach:
quote:
Originally posted by StevieB
A monocoque se7en using this technique should be fairly simple (famous last words...)...
Martin Ogilvie pioneered the bonded honeycomb panel approach with the Lotus 87 F1 car.
He built an electric 'Westfield' Seven called the Wisper (Westfield In Structural Plastics Electric Roadster) that uses the
routed-and-folded approach. He was kind enough to talk me through the process in some detail when I took the FW over to see him a couple of years ago,
and (the way he tells it - and bearing in mind he's a talented and experienced F1 designer, not an amateur in a shed), it really is as simple
and straightforward as it sounds. He reckoned that once you've got the design, the actual routing, folding and bonding of the main tub structure
was basically a day's work.
There's an article on the car >>here<<, which includes a couple of
photographs where the chassis tub can be seen, but unfortunately it majors on the elecvtric power rather than the chassis construction.
As others have already identified on this thread, the bit that needs more thought and care is the feeding of loads into the tub. The FW400 (and, I
think) the Wisper use clever threaded 'top hat' bobbins that screw into each other from either side of the panel, being bonded in place as
they do so. You'd need quite a few of these machining up, which would add to the cost of the tub, but it's a simple and elegant
solution.... provided you have enough structural knowledge to place them correctly.
|
|
|
Paul_C
|
| posted on 11/1/10 at 12:29 PM |
|
|
An interesting read is 'Motor Vehicle Structures: Concepts and Fundamentals' by Jason C. Brown, A. John Robertson and Stan T. Serpento,
Butterworth Heinemann. It helps in identifying loads and load paths in a monocoque.
I have used Google SketchUp to draw up my guess of the Costin Amigo structure and produced a 1/10 th scale cardboard model which helps to visualise
the load paths.
I have also perfomed a crude analysis of the torsional stiffness using the LISA-Finite Element Technologies software, available at
www.lisa-fet.com.
From what I saw on a Plane is Born on the telly where they appeared to bolt directly through a tube with no re-inforcement to prevent crushing. And
also where the undercarriage didn't have any proper suspension it's probably best not to use Kit Planes as best design practice.
|
|
|
ceebmoj
|
| posted on 15/1/10 at 09:11 AM |
|
|
link that some one else has posted up of the design of a formula student car. witch has a hunycome chassis
http://www.locostbuilders.co.uk/viewthread.php?tid=128266&page=1#pid1077629
|
|
|
SPYDER
|
| posted on 25/2/10 at 07:41 PM |
|
|
aitch you have U2U.
Geoff.
|
|
|
