Stiffer is always better so that the suspension can do it's job more effectively and so that you can put power to the ground. Even on modern WRC cars they still stitch weld when they prep the chassis.

Downsides is increased NVH and increased cost of repair. When the seams are all welded together you can no longer separate body panels after a crash as easily as you can with a production car. Other downside is cost as it takes many hours to grind to bare metal, weld a seam, do it slowly to avoid overheating and warping the body shell and panels, and then repainting to avoid corrosion.

If you have the time and a MIG welder, it's worth it. It will make an unbelievable difference in the way the car handles.

Thanks for your response.

I realise it would be of some benefit I just wasn't sure if it would make enough difference to be worth the effort.

Difficulty repairing it is a good point I hadn't considered, I'll just have to make sure I don't bend it . I've got a new car coming next week which will be a complete strip down so I think I'll give it a go. I have MIG and TIG, I hadn't really thought about the MIG, I was going to TIG it without adding filler as much as possible.

I thought I'd run a 6mm weld with 30mm between them on less stressed areas and maybe 6mm with 15mm gaps in more stressed areas what do you think ?

From what I know about welding, in this case TIG is definitely a waste of time or the wrong application. The difference in filler added will be minimal, and even with a factory new chassis there might be too much contamination to get TIG to work properly. TIG is best used in specialty applications and is more difficult and requires more skill and time to weld. MIG is more than adequate for stitch welding, which should be more economical and a better use of time on heavy gauge steel panels.

Unless you're welding on an NSX or something that is an aluminum body...

You will often find paint, primer and sealer in between the panels you are trying to weld together. Getting enough of this cleaned out to bust out clean TIG welds consistently is nearly impossible and not worth the time.
You still want to remove as much foreign material as possible with MIG but it will bust through a little contamination much better.

I recall reading a paper prepared as a university dissertation where the student took a Rover SD1 bodyshell and fitted scaffold tubes in place of the axles to measure twist before and after

1 the fitting and removal of a(iirc 6 point bolt in)roll cage
2 seam welding
3 welding of fillets between cage and A & B pillars

In a nutshell the results were
Roll cage - no measurable difference
Seam welding - no measurable difference
Fillets - OHH Yes! this is the key to increasing torsional rigidity

That doesn't surprise me, the idea of throwing away unused stiffness potential in the structure for the sake of fewer spotwelds would not go down very well. Spot welds aren't free, but they are cheap.

But, in the bad old days, the hit rate of spotwelding was probably only 60% or so, so even though more welds were used, if you happen to be unlucky you could be missing several welds in a location that was crucial for stiffness.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

The only real way to do the job correctly is to remove all the paint and sealants prior to welding.

Most WRC cars start life as a bodyshell taken from a production line. They are then thermally stripped in a pyrolysis oven prior to be chemically cleaned and passivated.

Once they are welded and re-sealed they are then re-painted.

There ahs been much debate about the value of seam welding for many years and things have changed.

If you look back to the days of Jim Clark and the Mk1 Lotus Cortina you will see cars which waved front wheels in the air and most of this was due to twist in the shell.

The floor was attached to the inner sill in 6 places and when you removed the caulking you could see daylight.

A 'modern' Historic spec Mk1 Lotus doesn't lift wheels and this is due to improved bodyshell stiffness. (Welding + cage)

Bodies did get better as monocoque assembly techniques improved but then robot assembly and side assembly techniques reduced the number of welds.

If a student measured a Rover SDI (Which is about as rigid as a jelly) with and without a 6 point cage and couldn't find a difference I hope her failed to graduate.

Even the most basic structural analysis will show a clear benefit in terms of both twist and bending.

By improving stiffness in all planes and improving the rigidity of the suspension pick up points you must improve the behaviour of the car in terms of consistency. The basic handling will always be the basic handling but stiffening the structure will eliminate glitches and odd behaviour at the limits.

Unless you have an accident you will also preserve the life of the bodyshell.

It’s difficult to actually weld a seam –and the spot welds do a reasonable job as long as they hold. Since the seams need to be clean for welding, brazing would seem to be the better go. If you can get braze into the seam (doubtful) it would be a bonus. But brazing makes an excellent fillet if done purposefully.
Roll cages can help a bit if you can tie them in at points near the suspension input to the chassis. However, to do much with chassis torsion stiffness the cage should be tied in with the roof structure –often against the rules.

For what it's worth, and this is an engineering forum - welds parallel to the longitudinal (parallel to the length) axis in a vehicle will be the most highly stressed in torsional loading and should be treated first. Also, welds farthest from the centerline of the vehicle will have the most "leverage" to resist torsion and should be considered.
If your chassis has torsional strength, bending strength will (generally) follow.

This is what I had in mind;

http://pdfpiw.uspto.gov/.piw?Docid=06505990&ho...

Um - The reason a wheel is hanging in the air on a corner is because the body is stiff enough to hold it there.
That happens when the roll stiffness on one end (usually non-driven) is much higher than the other, and the shell is stiff enough to transmit all that torsion.
right?

Jay Maechtlen
http://www.laserpubs.com/techcomm

In case you actually want to do this I did some research a while back and unless you have access to that pyrolysis oven like the wrc guys this is likely the next best thing for mere humans.

http://www.rallyanarchy.com/phorum/read.php?5,6465



It's used to fog liquid nitrogen, you freeze all the undercoat and filler then chip it out. I've never used this myself but I have used dry ice and the difference between this method and solvents and scraping is unbelievable. If you are doing an entire car at strange angles dry ice would be tricky to use effectively and you use it up much more quickly than you might expect.

With the dry ice on odd angles, I crush the dry ice in a bucket and make a slurry by adding isopropyl alcohol. Significantly better results, but decreased working time.