daveyrace wrote:Just so you know. If it is of any help.
From what I remember from studying crash testing, the problem with a road car in a frontal crash is that the engine does not deform.
If the crash structure in-front of the engine cannot absorb or dissipate the energy then the engine is effectively forced back into the passenger compartment.
In the case of a road car the crash structure is basically the two metal struts either side that the engine is mounted to. If you hit square on then both of these can act together to deform and absorb the impact. If you impact just on one side then only one of these structures does the job, obviously not as well as two. However, if you hit a pole/tree in the middle, you do not use these crash structures and it simply hits the engine, which will not deform(much) and transfer the energy to the point behind it, the passenger compartment/bulkhead.
This is where bigger cars do better on crash tests as they have more room in this area for the engine to move without hitting anything.
Basically an engine block directly in-front of the passengers is a bit of a problem for the designers when trying to perform well in a variety of crash scenarios. But it gives more room for all the rubbish we pile our cars up with in the back.
This is why they test for frontal offset in NCAP, but interestingly they do not test against a frontal pole impact(I suspect because all cars would perform very badly).
Now a road car without an engine in the front is far easier to design to take an impact well, across the whole area, as it is a veritable free for all for the designers.
An example that springs to mind is the Mclaren F1 which did pretty well in crash tests at the time despite having a relatively short bonnet.
http://youtu.be/DJjf8W22xWI
Or a smart car, with its engine underneath at the back.
http://youtu.be/mnI-LiKCtuE
Now if you could have a reasonable area in-front of the engine to build a crash structure and provide a path around the engine for the energy to be absorbed rather than through it, it would be fine. Plus you might want the driver to have a gap between their feet and the engine.
I think you may well end up with an rather long vehicle once this is done. And some frustrated designers/engineers trying to work out how to balance the thing 
.
There is a quote I recall from a talk I attended given by a crash tester from NCAP which went along the lines of. "The only way to make all cars 'safe' would be to give them 20 foot long bonnets".
Now most road cars are made from folded sheet metal which under impact absorbs the impact by bending.
http://i.imgur.com/1WdpOup.png?1
The image above shows that the crash structure bends, using up the crash energy. However, Only the parts of the metal that bend are doing any work, the unbent sections between do nothing.
Carbon composites are your friend. 
http://i.imgur.com/guSdN0U.png?1
Carbon structures when impacted will dust up and fracture. Far more of the material will do work in this instance in a much shorter distance.
So it may well be possible to produce a front engine F1 car, but in my opinion it would be far better to use the engine as a load bearing chassis member out the back than a problem to work around in the front.
Having said all that I would love to see some front engined cars razzing round a track with the rear engined ones. 
Thank you for your comment and for the interesting images.
I completely agree with you and with most of the differences bettween a road car and a F1 car.
Considering the "roadster" shape I imagined, the adsorbing structure that in rear engine F1 is placed in front of the front axle would be unchanged, so (being the total mass constant, and the section of the deformable volume about the same or greater) safety would be at least the same.
Other advantages would be:
[*]Bigger distance from the nose to the helmet > it would be possible to install thicker tyres barriers
[*]Longer sidepods (actually this could be realised even with a rear engine if rules would be changed)
[*]Smaller mass (= smaller inertial force) that the surviving cell would withstand in case of frontal impact (this is not the main problem at the moment - but it was in the past, see M. Donnelly's crash - but it could help).
In the title I wrote about an impact adsorbing structure, but this is quite difficult to explain my idea, so I'm going to prepare a simple drawing and I'll be beak later
For the moment, consider that new V6 engines (the thermal units at least) are very compact (less than 500mm I think).
EDIT: Hope this drawing could help
