OK I heard something today from a builder I have the utmost respect for that set me back. He said:

For me I have never heard of anyone stitch welding any car that did not feel it benefited the chassis and in a situation where we are putting 3 times the stress and load on the chassis than it was designed for I would think it would be a benifet.

Yes it is impossible to get all the seam sealer out but isn't a crappy weld better than no weld?

 

When you weld, you do heat the surrounding metal, which has a tendency to weaken it a bit. As such, the new weld has to do double duty, both in stiffening the area being stitch welded, and hold the two panels together. Done right, it can add a lot of strength to the panels of the unibody. Done poorly, and it will weaken them and the welds will eventually crack/fail.........



Bill S.

PS: I know my limitations, and structural welding is one of those items I prefer to leave to the professionals for my peace of mind.

 

First off, I am going to go ahead and do it to my car just because of the fact that I don't know enough to support my argument and it's better to spend a couple days now then find out later that I should have done it. Because at that point, it's a much bigger job.

But with that said. I truly do believe that with proper supports (strut tower brace, cowl support, x-brace under the car, roll bars and door bars) that the difference in the end will be small. I just see a lot of discussion about it and it looks pretty clear. The people that have spent the time to do it defend their decision and the people that didn't defend theirs. The only way that I (and most) would be convinced one way or the other would be to take a car that is built (v8 conversion and all the supports I listed above) and run it for a season on the track. Then do all the seam welding and run another. Then compare results. Even then, people would argue the extra season of experience.

I just see pictures like this (directly from one of Fylin Miata's builds) and really question it.



This picture kind of proves my doubts about the whole "seam welding" theory and how people do things for no real reason at all. Look closely. They seam welded a seam that doesn't do anything (white circle). It's a seam that was there to support a piece that they had already cut off. If they wanted to weld something there, they should have removed the piece of useless left over metal and seam welded the real seam hidden under it. (on a side note, I wish my welds looked that good).

 

Most of the homemade seam welds have not looked so great. If I were to do mine, I would weld some 1" strips of metal over the factory seams so that they could be welded to clean metal. I also thought of replacing the factory seam sealer with some structural panel adhesive.

 

The welds that WolfGT pictured appear to be TIG welded which will produce much nicer welds due to more amperage control.

 

I cut and pasted the top post into an email and sent it to a guy that is a professional welder in the US Navy. This is his response:

He has a point, there is a risk of that, but the metal is not going to be weaker just less ductile. Because of the heat required to melt everything together welds are a different molecular structure than the base metals around it. The weld itself will always have a higher tensile (tension) strength and be more susceptible to breaking by bending. But being more brittle just means it won't bend as much; it's actually a higher tensile strength than it was before.

The way you avoid that problem is by doing exactly what you did - lots of little tiny welds instead of one big long bead. That way, and with metal that thin, you don't heat up the base metal so much that you make it brittle.


Full disclosure said professional welder in the US Navy is my son.

 

Wolf has a great point that it is nearly impossible to know if it makes any real difference at all and I doubt many of us can weld that good. And no question properly gusseted braces and supports would reduce the benefit of welding the chassis.

Even so I think it would be safe to say that any unibody car that has seen as many pot holes as most of our cars have would be less sloppy after welding it. Keith at FM said this about his car here:

[Chassis welding is] "Definitely worthwhile. For example, on the 200,000 mile shell for my Targa car, you could flex the windshield frame by hand. After some seam welding (not stitch welding), you couldn't do this anymore."

From what I found reading on the subject last year it looked like about 5 for welding to 1 against it and if the good folks at Flyin Miata feel it is worth the effort that is good enough for me.

 

I know this isn't really on topic, but I think it bears asking here:

Why aren't doublers/triplers/etc. being used?

I'm in the aviation industry, and when we have an area that needs reinforcement we typically add another layer of metal with sealant in-between the plates and then rivet them together. It provides an exceptionally strong structure with a minimal space compromise and you suffer fairly minimal weight gains.

The sealant keeps corrosion at bay since there's no aerobic activity. I'm not talking about pop rivets here, but solid rivets, nailed in wet (with sealant).

It would seem that doing this across the firewall, and across the rear vertical area (behind the seats) would provide an optimal backbone for some internal I-beams to run from front to rear to box the car without having to resort to bulkier door bars and a roll bar.

Has anyone done this?

 

We add a layer at the back corners where the gussets come out of the engine bay but that is about all that is usually done. Most guys weld up the existing supports like behind the seats just to firm things up on a high mile tired chassis. Part of is the slippery slop of while I'm here with the welder out I may as well...

Yes everyone puts floor bracing in but you know triangles make the best structures so that's where the door bars and tower braces come into play. Roll bars are for roll over protection as lots of tracks won't allow converts to run with out them. They add far more weight then benefit to chassis stiffness IMO.

I'm sure you have seen the frame rail caps most guys use but there are other options. Click me: My DIY Frame Rail Solution - ClubRoadster.net

Jim and Jenna build diary, Miata LS1 Conversion: Todays adventure - welding upside down on tin foil

It's probably less effective than a door bar but if you want a less obtrusive support you can tie the end of the frame rail to the diff frame to box it in like so: http://jimsmiata.blogspot.com/2013/0...back-half.html

 

Seam welding an entire chassis is a GIANT pain in the ***... to do it properly you need to soda blast or acid dip the entire chassis. Otherwise you are constantly setting things on fire (paint on the other side, interior bits) or are contaminating the weld with paint, seam sealer and other gunk.



One of our fabricators here has built pro level road race chassis, where every ounce of insulation and seam sealers can be removed (it helps to start with a real Body in White) and seam welding the unibody is commonly done. Even then it takes a body dolly and weeks of work to make this very small addition to torsional rigidity. A lot of work for a very small gain.



Of course there are times when a little structural addition and even seam sealing makes sense. We work on a lot of cars in our shop, and do chassis prep and cage work, so we try to learn what matters where. The BMW E36 chassis above has some known weak areas that get reinforce with plate and/or seam sealing.



There are areas on the NA/NB Miata that seem to benefit from some extra attention with the welder. Two of our staff have Miatas and one of those has built a lot of these, including a fully seam welded chassis for SCCA Prepared class racing. The problem he ran into was after weeks of stripping and welding he went to bolt it together and... the tub was bent. It was off on the diagonal measurements, and the now stiffen tub wouldn't ever be able to be pulled back into shape. He wanted to seam weld that tub because a previous build he raced for 2 years in an autocross class was bending like a pretzel, and it was raced on 140 treadwear street tires!



Some of the bigger gains seem to be in the front upper frame horns, and the bolt-on solutions from BossFrog, GarageStar, or V8roadsters look like a good addition to most Miatas. Something like that should be a requirement for all V8 Miatas, in my opinion.



And it doesn't have to be bolt-on... welding is even better. We've done similar reinforcements with some 1.25" square tubing and a little plate steel. A little effort can go a long way in this area. But with most of the V8 Miata swaps people have to cut up and remove parts of the tubs in some structural areas...



And many of the swaps do away with the PowerPlant Frame. So where do you make up for that? Subframe connectors are a good idea, and a little bit of stiffness can be found in a decently braced bolt-in roll bar, too.



Anyway, just some thoughts. I don't claim to be a Miata expert, and we've been cutting away bits and pieces of our NB LS1 swap candidate for a while, the whole time knowing we're going to have to add more structural rigidity somewhere, somehow. And with 285mm Hoosier A6 tires at all 4 corners we will be putting some lateral and braking loads through this chassis, too...

 

This is a good subject to develop. For most V8 swaps like has been mentioned, we take more out at the firewall/frame rail joining than some realize, I'm sure there is strength lost there. It'd be interesting to see what can be done to put it back. I really like the welded in strengthening you guys did under the fender.

One thought that came to mind is what can be done for over the motor mounting. I've seen the stock MSM strut tower brace used, so there must be some space available. Something like the viper power plant brace may be interesting. A subframe brace that bolted to the back of the cowl area was available for the miata, I wonder if something like this would do anything for the chassis strength up front?

http://www.carsimg.org/wallpapers/20...2-900x1600.jpg

 

^^ Exactly! Anything you can do to tie things together at 45 degree angles makes a very strong box.

I think tieing the corners to the frame like this made a huge difference in cowl shake


Don't look at the welds!


Of course tieing the corners of the frame together too.


And the rear sub fame to the frame


Makes a nice box


There was more I wanted to do but got tired of it. For more ideas there is a great thread here.
[NA] More testosterone than my ride can handle. (LS1 conversion) - MX-5 Miata Forum

 

I think for me, a big driver is what tire you want to run... If I stay 15" wheels, you can still fit a hoosier 275 without tons of surgery. I am finding the only real alternative is as most have mentioned, 285/295-30-18s. Lots of surgery, but room for brakes. If you have to conduct a lot of surgery, you are already half way there to building in bracing.

** I can now see some photos, but 90% of stuff is blocked on my computer at work. If I use a work computer I get more access. That thread at CR.net is clutch. A few of his braces around the towers, and especially what he did on the subframe I may incorporate. It looks nice. I'm going to keep it simple, though, I don't want to go nuts as I am not a chassis engineer. But I think the front forward of the firewall needs doing.

A second thought is I have to clearance this intake, so I may cut the front core supports and replace them with a bolt in solution for easier maintenance. If you've worked on an Audi A6/A8, something like that idea.

 

You got that right! The guy on CR net is an excellent fabricator and I swiped the cowl corner brace idea from him. Made a lot of sense to me.

I do a fair amount of autox with mine so wanted things really solid. The lot on Sioux falls is a bit ruff and autox puts loads on things in ways that never happen on the street. Not to mention the occasional cone impact...

 

You didn't by chance do any before/after testing or use scales to test frame twist before, did you? I probably won't do it myself, either, but just asking. I've seen some good methods of just in the garage testing of things like this. The only concern is crash zones, but I guess you can throw most of that out anyway on these cars, since the motor/subframe change it a bit.

Nick

 

LOL yeah the crumple zones are long gone in my car!

There is a guy in a thread somewhere that did the testing you are talking about on an NA but off the top of my head I don't remember where. I think it's on turbo Miata.net.

What I can tell you for sure is I have a friend with the same year Miata as mine and no bracing and it's easy to see the difference. He is on slightly firmer than stock springs with KYB shocks and his feels considerably softer in the chassis than my car does. It's easy to see the cowl shake over stuff like rail road tracks especially if you hit them at a slight angle. One little test we found out just by chance. If you jack up one side at the rear of the rocker you can feel his door drag when it opens and mine doesn't do that. Either side will do the same thing.

Sure there is a ton of variables and differences in the car but with more weight and much stiffer suspension my cowl should shake more then his on rail road tracks and it doesn't. So to me there is no question the seam welding and braces are effective. I hate the weight it adds but sure love the effect.