I measured the front frame flex on my custom 2"x 3" 11 ga tube frame while on the lift with the body on and V6 engine/ 5 spd trans, front and rear suspension and tires installed. This car should weigh about 2500 lb finished. The wheelbase is 114" and there is 2 shallow angle joints in the 4.5" kickup in the front frame. I measured the lift by jacking the front crossmember and measuring the vertical deflection when frame just begins to rise off the front lift point (36" from front lift point). I guess this means that all of this deflection is in the front 36" I get 7/16" deflection. Seems like a lot but shouldn't bother the body assembly because this is a fenderless car. Maybe I should gusset the 2 shallow angle joints in the front frame. I suppose once the car is on the ground driving the 7/16" up flex may not increase much more going over bumps ??
What are you using for clutch linkage? I had a model A frame from Total Performance which would flex enough to bind the linkage using stock chevy linkage. Remember that the original frames were designed for rough roads and flexed a lot to compensate, even though these frames are now made with tubing they still have no stiffening K member.
You're worrying too much about nothing, to build a flex free frame you'll have to overbuild a monster. The body helps with rigidity, the engine, drivetrain helps, driving, you'll never notice any minor flex, its steel, steel moves, put her together and run it hard.
Almost a half inch of flex, in just the front half of the frame? Way too much for me. Ida built the rails out of at least a 4" tall tube. Maybe you can rescue it with gussets, or maybe an X member can have legs reaching forward and stiffen it up some.
The car is on my lift, so it is supported by the frame near the back end of the rockers and supported by the frame near the front end of the rockers. I make the measurement just as the front support gets a little loose (can't yet see between the lift pad and frame). My lift pads do have 1/2" rubber cushions though. Right now the car is very light, just an empty body shell with engine/ trans, suspension and tire weight. The body is not yet bolted tight to the frame I am building, but the flex all seems from the front rocker to the crossmember. I have sub framed several cars with 2x4 x 11ga tubing but thought this car was so light that 2x 3 11 ga would be OK.
2" x 3" tubing will flex a lot more then 2" x 4" tubing will. The 4 1/2" kick up isn't helping either. Any gussets you can add will help a little. A body with a floor will help a lot, if that body has a roof, it will help more, if you tie the radiator support to the firewall, that will help even more. I don't believe you can eliminate all the flex, but I also don't believe it will make much difference at the end of the day. I had a 2" x 3" tube frame under my 35 Dodge sedan. If I was doing serious hard cornering, I could feel the flex the the chassis, but we drove that car 77,000 miles without an issue. Gene
I don't really understand how the OP is lifting, checking twist, or what the body is. I sort of think two post lift but body has rockers so is what?. I'd like a picture...
In my head i see a rectangle (frame) with relatively centered supports (lift arms) and the corners being lifted. With nothing on any of the opposite corners. I’ve built a few frames here and performed surgery and straightened out a bunch of messes including alignment of body panels. Every single one of them sat completely different on frame stands than on its suspension. More than that the arrangement or location of the stands will change it too. Some convertibles will change horribly right before your eyes yet on the suspension they are fine for 6 or 7 decades and 200,000 miles Routinely I lift some large structures that are very rigid with no discernible flex at all. It’s very easy to get one corner higher than the other 3 with central support and no support under the other corners. The frame here may very well be flimsy and flexible flyer type or it could be great but the method being used and described for determining that isn’t the correct method. A full cage on this or any other frame will eliminate most flex. A really nice hoop support even just a bolted down sheet metal cowl structure will add a bunch of support
I have had US built convertibles that if you pick it up with the top down, the top will not close and doors will not open when lifted. Jack up a vehicle on one side under the cowl and observe how the door on that side fits and opens. Ever notice how airplane wings will sag? I have a model T frame. I can easily pick it up by myself and twist it considerably. About the same dimensions as yours and thickness but not boxed. When you pick up a simple ladder style frame on a lift like that you are adding a stress to it that it doesn’t experience when it sits normally on the wheels. Think about the gap between the cab and bed on a truck when picked up like that. It gets bigger when lifted. Even on a frame built much heavier than yours.
I built the frame for my Morris Minor gasser out of 2x3 1/8 inch wall tubing it had an 87 inch wheelbase and 5 crossmembers plus a 10 point roll cage . It had less than a 1/16inch flex measured from corner to corner. It also has a double frame rail from the front kick up to the rear kick up. For a street rod you should be fine Add some more tubing if you are worried about it.
Many factory chassis’s are noticeably flexible some types more so then others.. Perimeter frames like GM has made are remarkably flexible and so are the more stout boxed railed Ford Torino chassis’s.. I kind of view them as front and rear subframes with subframe connectors.. The lack of any real structural center crossmembers or inner bracing are probably the culprit but the bodies were used as part of the structural components tying it together.. Really soft oem spring suspensions didn’t stress out the luxury boats of the day much..
Might be important to note that a lot of these frames need to be straightened or cracks repaired from all that flexing. Also, door sag is pretty common. I don't know much, but my assumption is more that the A was built cheaply more than being built for the road conditions. That assumption is based on seeing frames from the same era as the A. A 1930 Pontiac frame is considerably more reinforced. Same goes for Chevy, Buick, Cadillac (albeit a larger car). As well, there's reason why from 32 and on the frames got more reinforcement. It wasn't that the roads improved drastically over the 4 years of Model A (when everyone else was already running more bracing)... more power, more comfort, more abilities, and so on. To the OP, the material size is pretty common that you are using for the car's weight. Can't compare it apples to apples on a stock A wheelbase because you're 10" longer. Do you have any detail pics?
Early Ford suspensions were able to rock laterally on their spring shackles up to about 3½° of roll, without significantly loading the springs. What kind of suspension are we talking about here? The forces tending to twist a frame due to road inputs are significant, more often than not exceeding first-gear driveshaft torque by a fair amount. It's a serial-spring problem: think of the frame as a spring acting on any of the suspension springs in series. To reduce frame deflection, either make the frame stiffer or the springs softer. And, of course, vice versa, should you want to do that for any reason. Softer springs mean more roll, but that can be counteracted by higher roll centres – which may bring other problems. Making the frame stiffer is likely to be fruitless unless a lot of superstructure is added, as in @southcross2631 's Morris above. You could argue that it defeats the advantage early cars have that the body is for all intents and purposes non-structural and therefore within the capability of ordinary people to make, and make light, but that's another debate. The other way is to interconnect the front and rear suspensions (or at least the front and rear roll-resisting means) as was done on the Citroën 2CV, the '55-'56 Torsion-Level Packards, and the BMC ADO14/15/16/17/61/67/71 (and 1964-1969 Huffaker MG Liquid Suspension Special Indy car.) I've been arguing for suspension interconnection on hot rods for years. It really needs to be a very early design consideration, though.
With no pics, I can only imagine what you've got but as far as the tubing goes, I would have gone to 7 ga. A little heavier and a heck of a lot more rigid. These guys that talk about how the original frames were built to flex aren't remembering that the roads were a lot rougher and the cars didn't go as fast. The model T suspension wasn't too far removed from horse drawn wagon suspension.
I don't think I can show pics. I started a post as a build a few months ago, got a lot of comments, but got booted off in a few days as NOT TRADITIONAL. Could be a 1958 Volvo 444 fenderless, tube frame, S10 2.8L V6, 5 spd, Mustang II front suspension, 4 bar parallel link. I do appreciate the help that a wayward soul such as myself am getting.
Maybe gussets like these cardboard templates might help. The flex occurs when jacking the front crossmember to the point that the frame rises above the front lift support up with clearance.
The tape shows the distant between frame at rest (hanging in air) . The 7/16" gap is the frame flex from hanging down to raising off the front lift point.
I don't know if you have any cross-members? A stout 'K' or 'X' member will take away a lot of that torsional flex and improve handling in the long term. Down here centre X-members are mandatory for that reason. Engineers sometimes requires frames, depending on modifications, to undergo torsional and beaming tests to ensure compliance. Insofar as your welded joints are concerned I'd suggest fishplating both sides of joint with adequate welding as well as using plug welds on fishplates. Butt welded joints are the weak link in the chain and more prone to failure.
My eyesight has gotten pretty bad, what you see are second pass welds that were set too hot and weld was literally dropping off to the floor. All the frame welds will be ground smooth and add a small flitch plate (about playing card size) to lap over the joints and get some fresh weld away from the HAZ (heat-affected zone) .
I never leave butt welds only, I always add a small flitch plate (inside & outside of frame weld. I have always used 2x4 11 ga. I think 2x3 11 ga is biting me on the butt My 57 BelAir has lots of front droop when on the lift, due to the thin frame, less than 11 ga.
All full frame cars and trucks without a cage will do the same. Unibody stuff is different. Card gussets, yes but I’d probably reconfigure and turn into the the fish plates as well as the gussets and do both jobs with one plate. Then a small fish plate on the other side. Once again though,, if that thing were on the suspension the majority of the (now hanging cantilevered) front weight would be on the tires. It’s a quite different leverage and load arrangement from reality on the road and what you’re measuring now. A whole bunch of that 7/16 is just the leverage that’s the weight of the engine is putting into that tube frame rail
I decided to plasma cut 2x3 tubing in half lengthwise and scab the extra 2x1.5 on top of the front 2 frame sections where my flex is occurring. I will also fish plate over the butt joints.