40 replies to this topic

[mariner]

After we all spent ages telling Wolf he would take forever to do his GF chassis I began to wonder if maybe he is right and it is possible to do one in reasonable timeframe.

I have sketched out ( rather poorly) a possible process using plywood as the core with GF skins for strength. It is designed to be within the scope of an amateur and require no expensive tooling or moulds.

Basically most monocoques have a series of boxes which could be made in GF/plywood core. A typical two-seater , mid - engined car will have five chassis boxes with four or five lateral bulkheads
[



The process is to make each box as a separate plywood box using fillets at the corners and with access holes at each end. The box is made " dry" i.e no GF so any errors etc. only result in wasting the plywood. Once the box is built dry it is laid up with GF on the inside and outside to create a GF/ ply/ GF sandwich. The inner GF is laid through the access holes in each end. The box is turned over during lay up so all GF layering is done horiziontally and downwards.

The cross- section is like this


The process to vacuum bag the box is different to normal. An internal bag (balloon) is put inside the box via one of the access holes , the other is blanked off. Then a second vacuum bag is placed over the whole box on the outside. As shown in the next sketch the inner balloon has a sealed tube with a pressure pipe coming in but also inside the outer vacuum bag which has its own seal with a pipe from a vacuum pump.



The point of this arrangement is to balance the pressure inside and outside the box/GF sandwich. That way no great compressive or expansion load is placed on the box structure but considerable GF laminate consolidation pressure can be used. In a perfect world one pump would pull air out of the vacuum bag and pump it into the inner pressure balloon. In reality I think two pumps and some simple valves would work.

In this way a series of GF/ plywood core boxes is built up. Then the bulkheads ( metal,honeycomb or GF ) are placed between the boxes with adhesive and bolted together. The bolting is both structural and a means of clamping the joining adhesive.

This gives you a series of single curvature boxes with the laminate thickness and strength tailored exactly to need.

To achieve a double curvature box teh same basic technique is use but with an extra step. A series of temporary formers is inserted in the box before the sides are fitted and stringers are laid along these formers. Then diagonal ply planking is attached to the stringers to give the 3D curve shape. Once this is done the formers are removed before closing the box and laying up the GF as described earlier.

As here



I think this approach could manufacture a GF composite chassis with some 3D curvature with very little specialist equipment and fairly low cost. As no jigs are needed the build time should be kept as quick as posisible.

Anyway there is the idea for critiquing here

Edited by mariner, 05 March 2012 - 00:42.

[ray b]

plywood sucks as a core
it can rot or expand or shrink due to moisture changes

airex or kleg-cel foam or other brands of core foam are far longer lasting and stable
lighter too

[Wolf]

Ray, it might not be the greatest core material available, but if the rot and moisture were such big issue, it would have been abandoned in marine uses a long ago... And as that recent thread showed, resin material might be much more a matter of concern than core (mind you, I've seen poyesther resin boats showing virtually no osmosis after decades in sea water, so I tend to believe such things to be somewhat more related to void content).

Mariner, for your one pump arrangement to work it seems that outer bag would have to contain the exact amount of air that need be pumped into inner bag, but somehow I get the feeling that two (or three?) valves might solve the problem with one pump... BTW I do not wish to draw parallels to that wet-layup of mine, but both would require on-demand delivery of 'prepreg', and I've seen DIY plans for fabric impregnator...

[mariner]

As Wolf says ply has been used in boats for a long time. It can rot but concluding that prevents it from being a chassis material is a bit like saying steel rusts so you can't use that in a chassis.

Clearly there are better core materials than ply but they don't have the ability of be self-supporting like a ply core structure so would not be much use for this process which aims at eliminating all the cost, complexity and time of mold making.

in terms of the question raised by Wolf on balancing the external vacuum and internal pressure I have thought about it and I think two pumps , one vacuum and one pressure are needed. The problem will be keeping the balance of the external vacuum crushing force and the internal blowing out force in balance automatically during the long curing period.

What might work is a balance beam with a weighted probe on one end which touches the box surface and goes up and down if that surfaces flexes. The other end has micro switches posiitoned above and below it with an adjustable gap.

The lower switch is connected to the vacuum pump motor so if the vacuum leaks off the box bulges under pressure raising the probe and pushing the other end of the beam into the lower switch. This turns on the vacuum pump and the probe moves down until the switch is disconnected. The pressure switch works the same way By adjusting the gap between the beam and the switches rapid cycling could be avoided.

[cheapracer]

Speaking of sandwich layering, the very first time I used fg I made my own skateboard at around 13 or 14, oh wait, the second time, the first time I thought the powered stuff in the plastic bag was the hardener (it's the thickener ).

Anyway 2 days later and throwing a dripping pile of wood and unhardened (but thick!) resin into the rubbish, back to the hardware store and 2 more 'Fibreglass Repair Kits" later, I used layers of single ply, mat, single ply etc. - 3 layers .....

..worked out well, rode that rather flexible board for months and only stopped 'cause I got a "Bahne Superflex" for Xmas (with Chicago trucks, Rollersport Mk IV urethane wheels and loose bearings) - that was a $50 skateboard in the mid '70's, hooo weee!

Oops, sorry, I'll just pop over to The Nostalgia Forum ...

Edited by cheapracer, 05 March 2012 - 13:41.

[ray b]

other then super cheap home built small boats ply has been abandoned mostly

the other ply boat building system is west epoxy saturated cold molded
that results in a single unit ply for the whole boat

as a long time live aboard boat owner I have owned both types
and I have seen more rotted ply on other peoples boats
they both rot as does fiber glass covered ply due to small cracks
and the needed holes for attachments and pass threw stuff
my west system 31ft cat had rot that is why I sold it cheap

ply can also warp or delaminate esp with vibration and stress


almost all ply has voids
claimed void free ply called AA-aircraft grade is super expensive
and foam core maybe cheaper [I have not priced it recently]

foam core can be used as a mold it is rigid like ply
in fact it needs to be cut in to blocks to curve in a boat hull

do it right carbon fiber foam core and epoxy
for a one off car the cost will not be that bad
and the life span far longer

[cheapracer]

my west system 31ft cat had rot that is why I sold it cheap

how old

was it

Ray

??

[ray b]

80's something home built in south Africa
wharram polyconcept with a race F-27 mast and sail rig
sailed to fla by the owner/builder
I sold her in 2006

[cheapracer]

80's something home built in south Africa
wharram polyconcept with a race F-27 mast and sail rig
sailed to fla by the owner/builder
I sold her in 2006

plenty of

cars that

go to the rubbish because

of rust that don't

last 20 years

[mariner]

I would reckon a twenty year lif ( 1980's to 2006) sin't bad . Cars have a 20 yr life made of steel

[desmo]

Cars should last far more than 20 years before corrosion becomes a structural issue.

[cheapracer]

Cars should last far more than 20 years before corrosion becomes a structural issue.

Sure but it becomes a roadworthy issue and a 'is it worth repairing' issue with all the outer skin and other panels before then.

Relevant to this thread ...

http://www.3dengineers.co.uk/

Edited by cheapracer, 05 March 2012 - 18:35.

[mariner]

I went to the UK Retro Race show for the first time recently and it was noticable how big digitzing has become in classic car restoring in the UK.

One German vendor of 3D digitzing equipment had though it worthwhile to set up a stand.

A guy on the Crosswhaite and Gardner stand told me that they use digitizing for most of their recasting of old race car blocks and heads etc. They can cast you a complete MB pre war straight 8 block and are one THE world experts on such things. So if they reckon digitzing is the way to go that is a good recommendation I guess.

The C and G guy said it had really brought the cost of redoing blocks etc down.

Nice to see the latest technolgy and old skills being used side by side.

In another example of digitizing helping cut costs on classics I was talking last Silverstone classics to a guy enginering the Nissan Group C car that still runs in historics. A wonderful and fearsome beast . When I asked hime about body spares ( having just seem a Porsche 962 get shunted) he said no spare body but they digitised the whole car before running it so they can make replica body parts as and when needed without factory molds.

[Lee Nicolle]

look at Jay Lenos site. They have made quite a lot of items using 3D equipment.
But the equipment is very expensive to buy and to make the items. Only good for the very well heeled.
But if you have it you can ressurect a lot of very interresting cars and parts.

[Lee Nicolle]

Sure but it becomes a roadworthy issue and a 'is it worth repairing' issue with all the outer skin and other panels before then.

Relevant to this thread ...

http://www.3dengineers.co.uk/

I have seen a lot of fibreglass boats that the skins have gone 'pulpy' as the water has got in the fibreglass.One was only about 5 years old. and that is without timber in them at all. Mostly ski boats and runabouts.

[gruntguru]

in terms of the question raised by Wolf on balancing the external vacuum and internal pressure I have thought about it and I think two pumps , one vacuum and one pressure are needed. The problem will be keeping the balance of the external vacuum crushing force and the internal blowing out force in balance automatically during the long curing period.

Inflate the pressure bag to loosely fit the void. Vacuum bag the whole thing. Job done. The air bag will automatically expand as vacuum is applied ouside it.

[Wolf]

Lee, osmosis, I think it's called- epoxy should be quite resistant to it, but polyesther not so (with a caveat I've mentioned in earlier post)...

[Kelpiecross]

[quote name='mariner' date='Mar 5 2012, 11:30' post='5568887']

I hate to say it - but I don't think your scheme is hugely practical.

How are you going to join the boxes?
I think any ideas about fibreglassing inside a box through an "access hole" would be very awkward and messy and verging on the impossible.
Same applies to vacuum bags etc. - not really practical.
All the above is considering that you are looking for a simpler and quicker method of fibreglass construction.

I think it would be possible to make up a plywood chassis in one piece (like your drawing or like the similar Marcos chassis) and then encapsulate it in fibreglass - reinforcing the inside of the box areas (that you can get at) with 'glass as you proceed. But you still would need a body shell.

Maybe you should consider more radical schemes like making the chassis out of expanded polystyrene, covering it with 'glass and then dissolving the foam with acetone (or something similar). Possibly you could incorporate the outer shell and chassis in one unit.

I think a lot depends on whether you are trying to build a car to drive or are more interested in the building and development of new construction methods (I am in the second group).
If you want a car to drive - spaceframe etc.

People have been attempting for years to come up with methods of 'glass construction that avoid the buck/female mould methods but without any real improvements.

[Greg Locock]

Maybe you should consider more radical schemes like making the chassis out of expanded polystyrene, covering it with 'glass and then dissolving the foam with acetone (or something similar). Possibly you could incorporate the outer shell and chassis in one unit.

That's exactly how we built the chassis for our solar cars from 95 onwards, and is more or less how surfboards have been built since the year dot.

The only difference to your suggestion is we leave the foam in place.

Having been through the agony of making female moulds for the exterior a few too many times I'm a big advocate of the surfboard technique for one offs.

[NeilR]

core materials need to be stable and stiff enough to transmit the forces between the skins. Polystyrene is a poor core material - it is an open cell foam with limited strength. XPS foam is better, but there is a good reason why the other core materials were invented. Ply and glass would rapidly become too heavy.

[ray b]

forget the inside bag

no matter what core is used

build the box with sides and bottom
and apply the glass/CF WITHOUT the top in place
[or whatever side gets the access hole]
glass the top inside before attaching it to the box
seam tape the box with strips of F/G thru the access hole
then do the out side of the box bag plan

for the tape strips wet the glass out side the box
on a flat surface
and apply the wet strips to the work
that way is far eazyer to control resin density and complete wetting
then applying dry strips and wetting in place
work upside down so gravity helps the strips stay in place

use minimum hardener and use heat to cure the work
that gives both more time to work and a stronger finished part
litebulbs,[heatlamp or old style incandescent] hairdryer or heater can be a source of heat
a cardboard box is good to retain the heat near the curing F/G
you do not need an oven or real high heat 150 f is good
too much hardener or heat weakens the work
hot dry weather is best time to do glass work with polyester
rain or high humidity is a very bad time for glass work
as is nighttime
fumes are NASTY work outside or with very good ventilation

[mariner]

The comments are all interesting. Everybody seems to agree that a full female mold GF structure is not economically viable for one offs - having done some bodywork with GF I can only agree!

I know my suggested process is a liittle odd but I did give it some thought.

The biggest design problem with GF structures is IMHO getting the loads into the thin skin from points like engine and suspension mounting bolts, and keeping their location accurate.

I think that almost inevitably ends up with bulkheads per Wolf's plan and since you want to have all the suspension points accurate vs each other it makes sense to fabricate a bulkhead covering all the mounting points, ideally with machined locations. Its most likely that you will end up with four or five transverse bulkheads. Rear, roll bar, front suspension rear and front suspension front.

That why I went for individual GF boxes rarher than a single homogenous shell. You have got to connect the stresed skin frame to the 4 bulkheads so its easiest to have three boxes front to rear - front foot box, cockpit centre and rear engine bay legs. Each box is of a managable size and if you screw one up you haven't wasted an entire chassis.

Also once you have the bulkheads and boxes you can do accurate final assembly on a surface table before bolting and gluing the boxes to the bulkheads.

In terms of laying up the GF inside the ply box I think that you could do a lot of lay up before closing off the last side and so minimise the tricky bits but I agree its a possible downside.

Foam is a well used approach , my concerns are accuracy based on my own , inept, use of it. I like ply because of its workabilty , wide choice of thickness and strength. Also any build up of multi layers on foam requires either a bulge spoiling the bodyshell surface or very accurate sanding to get the outer surface flat. With ply you can add mutliple GF layers inside and not spoil the surface look.

I am not claiming any revolution, I'm just one of the "how could I make that " brigade

Just to add I have read Ray B' s post and think that's very good advice.

Edited by mariner, 06 March 2012 - 16:01.

[Wolf]

Mariner, I still think it's relatively easy to do as I suggested- and the mouldless construction using sandwich foam core (I was inclined to use something like Divinycel H-80) as base for layup seems reasonable, and proven, method... And once outer skin is completed, it would be reasonably easy task to place inserts in core before finishing it with inner skin layup.

[cheapracer]

The comments are all interesting. Everybody seems to agree that a full female mold GF structure is not economically viable for one offs - having done some bodywork with GF I can only agree!

Not me, bucks are easy to make IMO, use any old junk laying around (literally), sheetmetal or ply for perfectly smooth surfaces in a flash, bicycle mudguards cut in half lengthwise for wheelarch flare edges, woven rope for curved surfaces etc etc....throw some bog over it and make your female molds.

Make a one-off without molds and have an accident and you start over again - and you can't sell any to your mates.

The amount of sanding, sanding, sanding, sanding, sanding I see guys doing one-offs puts me right off.

[cheapracer]

I have posted this before but seems relevant here not just how to strengthen molds in a simple light and convenient way but a lightweight method to strengthen FG panels as well ....

Just wanted to show the expanding foam (comes in a spray can) strengthing ribs. 2 piece mold because I have a 100+ degree angle that you need to unbolt to remove the workpiece. You make your basic mold over the buck and then spray lines of the foam on, wait a few hours or the next day for it to expand and dry then simply glass over the foam.

[mariner]

wolf, you may be right.

After my last post I remembered the DH Mosquito build process. As everybody knows it was the "wooden wonder" etc and was used to justify the Marcos chassis.

It was also a marvel of production engineering, I live near the Mosquito museum in the UK and have been fortunate enough to have a tour.

The mosquito has made with the fuselage split vertically like a model plane. The outer and inner skins of each half were formed over concrete molds using diagonal ply planks fro 3D shaping.

Then the outer skin was placed down and the balsa core and all the hard point blocks were laid in, then the inner skin was placed on top to complete the composite fuselage.

the neat thing was that all the wiring, plumbing and fitments like radios were installed in th separate halves which only got joined at the end. This avoided workers struggling inside the confined fuselage and speeded up production.

Smart design all round.

[Tony Matthews]

'Aeromodeller' magazine used to have very interesting advertisements for 'Soloarbo' basa wood, often in the form of descriptions of the production, grading, treatment and use of balsa in non-modelling situations. The Mosquito was mentioned, and I think that the half-fuselarges were microwaved to cure the adhesive, it being said that this was one of the first uses of microwaves for such purposes. This could be brain fade on my part, but I'll try to find an old copy ...

[GBarclay]

80's something home built in south Africa
wharram polyconcept with a race F-27 mast and sail rig
sailed to fla by the owner/builder
I sold her in 2006

Ray

As someone who has pretty good experience in the South African boat building and sailing scene in the late 80's, I think you may have been mislead about your catamaran being built using the West System.

South Africa was the subject of international boycotts from the 70's through the early 90's and it was virtually impossible to get imported, quality epoxy, vinyl or poly resins. Imported from the USA - forget about it, impossible. While some ICI stuff was available, almost without exception cost forced us to use south african resins. The quality of the poly (and epoxy) resins was atrocious, and only in the late 80's did the vinylester resins approach a level that I would have felt comfortable using. It would have been inconceivable that a "home builder" would have had access to, or could have afforded, a product like the West System. It was also well known, that plywood being marketed as suitable for boat building, was most definitely not "marine ply". Legions of SA "homebuilts" littering the slips in Fort Lauderdale, Miami and Palm Beach in the early 90's were testament to the poor build quality of SA yachts of that era.

At least it was not a ferro boat.

Perhaps built in the manner prescribed by the Gougeon brothers, but built using the West System, sorry, I think you were mislead.

The South African boat building industry has struggled long and hard to overcome the stigma of those boats built in those times, and there are some yards that build a quality product today.

my apologies for the thread side track

I absolutely agree that a decent cross linked foam core is way more suitable for a core than the plywood suggested by the OP

[Wolf]

Heh, there was also Mallite- I think early McLaren F1 cars had tubs made with it. ISTR that it was balsa core in Al sandwich.

[Tony Matthews]

Heh, there was also Mallite- I think early McLaren F1 cars had tubs made with it. ISTR that it was balsa core in Al sandwich.

And used for floor panels in early 747s I think, the balsa was end-grain.

[mariner]

'Aeromodeller' magazine used to have very interesting advertisements for 'Soloarbo' basa wood, often in the form of descriptions of the production, grading, treatment and use of balsa in non-modelling situations. The Mosquito was mentioned, and I think that the half-fuselarges were microwaved to cure the adhesive, it being said that this was one of the first uses of microwaves for such purposes. This could be brain fade on my part, but I'll try to find an old copy ...

No brain fade , Ony but brain wave ( sorry).

The Mosquito process did use heat to speed up the glue setting . IIRC it was applied whilst on the concrete molds. One of these is at the MuseumIi think.

home page (which doesn't do it justice) http://www.dehavilla...o.uk/index.html

[NeilR]

Some of the issues above with hardner ratios etc will not be an issue with this project if you use epoxy resin as the ratios are fixed. In a similar manner venal ester resin is sensitive to off ratios so you cannot vary things by much. If you are going to use a high grade core you would have to consider a high perf resin, otherwise you are wasting the core materials. As an owner of aFG car that is now 45 years old I urge you to use epoxy resin. If you do it will also ease some of the secondary bonding issues as the resin is just a much better glue. Mind you the fumes and toxicity is much higher and you have to use different cloth etc. personally I prefer to use epoxy resin as you have more time and I find it easier to use. The foam bracing cheaply shows can be done with ' d' rope etc

[ray b]

And used for floor panels in early 747s I think, the balsa was end-grain.

DONOT USE BALSA

very good wood for model aircraft
NOT FOR CORES BALSA IS A FAIL FOR CORES
it rots like old fiats rust
instantly and in all the worse places

[ray b]

GBarclay

you maybe correct I didnot think of the boycots
but the boat was epoxy encapsulated whatever the brand or origin
west system is a semi-generic term but yes they do make a better resin and system

a most places the epoxy held up well but rubbing flexing or stress leads cracking and rot
esp in lite cats and polycon's are flexiest of cats

at least the boat did get out of SA and to fla so mission accomplished
by those who built it
and the poor boat had a ruff life here
it got tossed upside down on an island by hurricane andrew
anyway I had fun with it
and unlike many boats I made a buck on it too

[ray b]

I like wood and really really hate working with FG
and like working epoxy even less

if you must use ply only use AA-aircraft grade
or make you own with separate wood plys and epoxy
and only use epoxy to laminate the fibers to the ply
ie the west system
as that is much stronger and less prone to de-lamination
as I have personally pulled lots of FG/poly off plywood
it is really really hard to get FG to bond to ply
esp new ply
it is an art very few have mastered

foam is stronger lighter cheaper and lasts better
it can be had in sheets just as rigid as plywood
with better bonding esp to cheaper polyester vinalester
and far more consistent then plywood quality even top graded plywood
some types/brands can be heated to bend in shapes plywood will not too

Edited by ray b, 06 March 2012 - 23:13.

[NeilR]

That is true. You can make a female mould, vacuum downthe PVC foam and heat it to then take the shape of the mould. Note I have not done this personally! The ting about using a closed cell foam, epoxy and say a triaxial cloth is that we have just tripled the cost of the tub.
Can I suggest that instead of designing the structure as one that performs a structural role only, you consider the occupant safety as an equal first priority. This could very well change the overall design quite a bit. You may want to encapsulate large foam blocks near the occupants. Composites do not do well in crashes past the initial impact given that they have a fracture failure mode and the energy absorption for secondary impacts is not that good.

[Wolf]

Neil, cloths like 7781 aren't that expensive (I remember seeing it priced at 1U$D* few years back, when carbon fiber was over 15U$D, albeit on a spl offer), while at the same time I believe it should be manufactured to certain standards (DoD or FAA)... I've even seen laminate properties listed in MIL handbook.

* per yard of 48" roll IIRC, same as CF price I listed

Edited by Wolf, 07 March 2012 - 00:04.

[Kelpiecross]

core materials need to be stable and stiff enough to transmit the forces between the skins. Polystyrene is a poor core material - it is an open cell foam with limited strength. XPS foam is better, but there is a good reason why the other core materials were invented. Ply and glass would rapidly become too heavy.

I was suggesting expanded polystyrene foam only in the sense of using it as a material for a male mould to be removed later so its strength would not matter too much.
I thought this was interesting:
http://www.rqriley.com/frp-foam.htm

I am not sure if this is a great idea but it may give you a few ideas. You can get the sheets of urethane at insulation suppliers - and not just one inch thick - up to six inches or so. In the article they 'glass the foam on both sides but you could make the outer skin thicker and remove the foam afterwards - possibly by grit blasting of some sort to give a "traditional" type of 'glass shell.
The article quotes very quick building times of 3 to 4 days etc. - slightly unrealistic I would have thought.

Although the traditional female mould method is very labour-intensive it really probably is the best way to make 'glass bodies etc.

[Greg Locock]

DONOT USE BALSA

very good wood for model aircraft
NOT FOR CORES BALSA IS A FAIL FOR CORES
it rots like old fiats rust
instantly and in all the worse places

My old Australian built boat Dry Reach
A quarter tonner
was built from GRP over endgrain balsa
It is 23 years old
It was recently sold for more than the previous owner paid for it.
The hull is in excellent condition.

http://www.out-of-th...s/DryReach2.jpg

oh and here we are mid race, must be serious, we're wearing our uniforms and no beers in sight even tho we're running downwind ! http://www.histoired...DryReach_01.jpg

Edited by Greg Locock, 07 March 2012 - 00:42.

[ray b]

glad you had such a good experience with balsa
looks like a nice fast boat

I have seen and passed on many a rotten deck with balsa cores
I have never seen a true core foam built by a pro break down short of fire
sure home built with insulation foam and other disasters happen
but nothing like the name builders who have balsa failures
with decks that are like trampolines with balsa rot are too common
balsa is a run away deal breaker for me

btw you can't use Styrofoam with polyester resins it melts the foam
and styro is a poor core anyway
some people get away with heavy wax or paint barrier coats for cheap molds

[NeilR]

Neil, cloths like 7781 aren't that expensive (I remember seeing it priced at 1U$D* few years back, when carbon fiber was over 15U$D, albeit on a spl offer), while at the same time I believe it should be manufactured to certain standards (DoD or FAA)... I've even seen laminate properties listed in MIL handbook.

* per yard of 48" roll IIRC, same as CF price I listed

Cloth is always cheaper by a full roll, but check the prices where you are. Locally the market is small and not all things in catalogs are available.