Sub-frame connectors... Please build them correctly!

-
moparmat2000 said:
wow what a poop storm this started.
Click to expand...

Bahahahaahahah.... Isnt that the truth....

Ok, So not to proud to stand down when corrected by the masses, Ill say that my stating that the "floating" connectors are next to pointless was overstated.

What I will stick to, and whether some choose to believe it or not, is that it absolutly is Stronger and better to tie it to the floor. The more surface attached to the floor the better, but all of the methods talked about that attach to the floor from the connector are better than leaving them floating. Whether it be tied with gussets, stitch welded, etc....

The thin gauge sheet metal of the floor with all its indention intentionally placed and the manner in which the load is transferred to it, makes it far stronger than most realize. On a very normal basis sheet metal is used to build many structures and in many cases has no heavier gauge materials used. Just good engineering is used in choices for shapes and load transfer.

I have obviously been publicly stoned and banished to the corner for my comments about the floating connectors, but I still stand by this as it is indeed tried and true to be better by attaching it to the floor as well.

That said, I understand entirely the other factors that go into choosing how you attach them. I myself will not like the idea of taking out the carpet and insulation in order to weld the connectors in, in the manner I prefer. Not to mention Ill be the guy doing it in my drive way under the car with a mig welder. LOL Chances are, that Ill stitch weld a set that are contoured to the floor and not weld them 100%

Now as I have said before.... I hope nobody got "butt hurt" over this whole thing as I like to keep the peace myself! LOL

Happy motoring!
 
I'm in the process of building a set of sub frame connectors. I used the page from bigblockdart.com, and built them. I didn't realize that I would have to notch the floors to get the 3"x2" tube to fit flat. I don't want to notch them, so I'm going to opt for a smaller tube, with a thicker wall and weld them to a 3x2 box that fits around the rears.

Some good points on both sides here for sure. I think I will build some gussets or some plates to attach the floors in a few place as well.

Any thoughts?

Will post pics along the way.
 
I think you should of used a poll to really
up the responses.

For me, I'll be using floor welded.
 
Some good points on both sides here for sure. I think I will build some gussets or some plates to attach the floors in a few place as well.

Any thoughts?

Will post pics along the way.[/QUOTE]


i thought about boxing my 2x3 tube to the floor as well now that i have the car on a rotisserie. if i template this out i will probably go the same gage steel as the original framerails, and have right angle tabs where it meets the floor to transition load, instead of butting a thick piece of steel to a thin gage floorpan.

just my .02
Matt

its not that bad, its only one hundred dollars.
 
For what it's worth. Not more than a few years ago, I saw a test done on the net somewhere. It was essentially a static load test to evaluate body torsion. If I remember correctly, the rear frame rails were anchored to the test bed at about the axle location. A fulcrum was built of steel and the K-frame rested on it. The apparatus was elevated such that the tires were not resting on the floor. The techs then began hanging weights from one corner of the front frame rails. They measured frame rail deviation between right and left sides with every weight that was applied.

They did the first test with a unibody car without frame rail connectors. The second test was with front-to-back frame rail connectors (welded). The third test was with connectors also welded to the floor sheet metal.

They graphed the results. The amount of twist in a normal unibody was ridiculous. (Twisting the body absorbs energy from the drive train.) Second best results came from the end-to-end connectors - the results were dramatic. The third test with the welded-to-floor connectors showed a further improvement.

I just wish I had the sense to save that test at the time.


Here is an interesting thread by some road racers:

http://www.pro-touring.com/threads/23278-What-do-sub-frame-connectors-really-do
 
FISHBREATH said:
For what it's worth. Not more than a few years ago, I saw a test done on the net somewhere. It was essentially a static load test to evaluate body torsion. If I remember correctly, the rear frame rails were anchored to the test bed at about the axle location. A fulcrum was built of steel and the K-frame rested on it. The apparatus was elevated such that the tires were not resting on the floor. The techs then began hanging weights from one corner of the front frame rails. They measured frame rail deviation between right and left sides with every weight that was applied.

They did the first test with a unibody car without frame rail connectors. The second test was with front-to-back frame rail connectors (welded). The third test was with connectors also welded to the floor sheet metal.

They graphed the results. The amount of twist in a normal unibody was ridiculous. (Twisting the body absorbs energy from the drive train.) Second best results came from the end-to-end connectors - the results were dramatic. The third test with the welded-to-floor connectors showed a further improvement.

I just wish I had the sense to save that test at the time.


Here is an interesting thread by some road racers:

http://www.pro-touring.com/threads/23278-What-do-sub-frame-connectors-really-do
Click to expand...


I would love to see actual scientific evidence of one connector being better, since this thread really has nothing but opinions and wild *** guesses posted so far. Even if it's only a single type of test, which most should know wouldn't be definitive in all situations. I know XV did some testing awhile back when they were showing how much their products stiffened the Mopar uni-body chassis. I know they had some pretty fancy charts and graphs on the subject, but I don't recall them testing other types of connectors. And of course, they weren't exactly free from bias, since they were trying to sell their connectors, which I seem to remember were the weld to floor type.

Regardless, any frame connector is better than none. The idea that non-floor welded subframe's aren't good, or aren't the "right" way to do it is just completely asinine. I'm personally not convinced that the floor welded types are always stronger. I'm sure in some cases they are- the mopar performance "bolt in" connectors, even if welded in at the ends, are pretty small in cross sectional area and wall thickness. And even they're a big improvemnt over stock, there's plenty of people on this very site that are more than happy with the MP connectors. But most folks now are using 3x2" tube with a fairly thick wall. Not to mention other real world scenario's, like some people only stitch welding the weld-to-floor connectors because of poor fit, poor welding, or simply not wanting to do that much welding upside down under your car while eating welding slag for hours with your car up on jackstands if a lift or rotisserie is out of your budget.

The best way to install frame connectors is the way you can do it. If that means buying the little Mopar Performance bolt-in connectors, bolting them in at home and having them welded in at a muffler shop because that's your only option, then that's still a huge improvement over nothing. And its just as "correct" as any other way to add subframe connectors.
 
I remember seeing that vid of torsional stressing.
Only thing I could find:

[ame="http://www.youtube.com/watch?v=g7pfNBrrP88"]US CAR TOOL 1970-1974 Mopar E-Body Cuda/Challenger Uni-Body Rigidity Test - YouTube[/ame]
 
I agree 100% with you 72 - any frame connector is better than none.

Thanks for the video, longarm, that test was very informative and was similar to the test I saw some time ago.

I like to think of a car body as a box, a box with a longer length dimension than width and height dimensions. Of course, due to many considerations, car bodies are not built as true boxes, but rather as much modified boxes.

Consider a flat plate of sheet metal with a length four times its width, pretty flexible along the longitudinal axis. Now, weld together some rectangular cross section tube steel together with the same dimensions (WxL) and weld it to the sheet metal. That will substantially stiffen the sheet metal along the same axis, but it will still flex somewhat with adequate force acting on a corner. Welding an additional identical piece of sheet metal to the top of the tube steel with further stiffen the structure. Other than internal cross bracing, we are now out of options.

Now, consider taking the original single piece of sheet metal with the one tube steel frame attached and making another identical construction. Take some more tube steel, say with length of about 1/8 of the length of the construction, and weld them vertically to the perimeter tube steel. Then invert your second tube steel-sheet metal construct and weld it to the new vertical tubes. And, lastly, weld new sheet metal around the perimeter and onto the vertical tubes to make a completely enclosed box. A typical car has at least four vertical planes of stiffening bracing - the radiator support, the firewall, cross-bracing over the rear axle area, and the rear panel behind the trunk. These vertical planes of stiffening greatly reduce longitudinal flexion by transferring loads. In structural engineering, this is called a bridge truss in three dimensions.

This is essentially a car body and we have dramatically improved the strength and torsional resistance of the structure. Along the same lines, it can be seen that increasing the height dimension of the structure (or reducing the length/width dimensions) will stiffen the structure. Therefore, a cube shape would eventually be the strongest such box structure for torsional stability (length, width, and height being identical). Of course, this is impractical for a car, unless you have one of those South Korean shoeboxes on wheels.

My bottom line is that, with cars, we must work with the hand that was dealt. An around-town grocery-getter will really not need additional body stiffening. A drag racer will need more stiffening. A road racer will need even more (they do much better with all four tires on the ground at any one time). The more stiffening, the better.
 
HemiDenny said:
pretty interesting.....

like to see the same test WITH the K ....and doors in place.

wonder how much weight that 50lb crankshaft LEVERED out front like that is simulating?
Click to expand...

I remember seeing one major structural failure at N.Y. National Speedway (1/4 mile track) back in the mid 1970s. Some guy was racing a mid-sixties 'vette. He cranked the revs and dumped the clutch on the "green." The front end had lifted with one side lifting higher and higher than the other - then CRACK! The body broke in half about mid way between the front and rear wheels. This happened on launch and he skidded to a stop within about 100 feet. I don't know if it is true or not, but somebody in the pits said that 'vettes had "x-frames" and that the owner had cut circles out of the frame to reduce weight, but, again, that was hearsay.
 
FISHBREATH said:
I remember seeing one major structural failure at N.Y. National Speedway (1/4 mile track) back in the mid 1970s. Some guy was racing a mid-sixties 'vette. He cranked the revs and dumped the clutch on the "green." The front end had lifted with one side lifting higher and higher than the other - then CRACK! The body broke in half about mid way between the front and rear wheels. This happened on launch and he skidded to a stop within about 100 feet. I don't know if it is true or not, but somebody in the pits said that 'vettes had "x-frames" and that the owner had cut circles out of the frame to reduce weight, but, again, that was hearsay.
Click to expand...

Maybe the owner should have laser cut the top of the frame off and welded it to the floor pans... you know, for strength. #-o
 
jrlegacy23 said:
Maybe the owner should have laser cut the top of the frame off and welded it to the floor pans... you know, for strength. #-o
Click to expand...
..

doh.jpg
 
Oh no. Engineers?






Here's the problem: Someone used the word "correctly".
That opens Pandora's Box. And then she spreads her legs.
(Same problem with the word "best").
Does anyone have a rotisserie that I can borrow to do it the "best" way?
All welding should ideally done down. My pans would have to be tied from the top to be best.
Does anyone weld up by choice?
I used Auto Rust Technicians connectors. They were good but they lacked flanges.
Eye Ball Engineering told me that they would be "better" with flanges like these to spread the load.
(spread the legs?)
 

Attachments

  • engineer.png
    8.1 KB · Views: 615
  • connectorsfrontend1.jpg
    33.3 KB · Views: 640
FISHBREATH said:
A road racer will need even more (they do much better with all four tires on the ground at any one time). The more stiffening, the better.
Click to expand...


While you comment is true.....sometimes it is better to allow SOME flex (usually torsional) to help keep everything planted. My current race car (non-Mopar) is stiff enough it tends to lift the front tire during the hairpins (not that I am using it anyway). Using common engineering sense (if there is such a thing), if you have a fixed plane (or 4 wheels in this case) and you lift one corner, another corner (or two) will also lift.

Many years ago we had better results on the asphalt ovals with a car that would twist than was overly stiff. We could get the power down much quicker off the corners than the others. I'm just saying there is a fine line to all of this, if you want stiff, fill the frame rail connectors with Viagra! :cheers:
 
Hyper-Performance said:
While you comment is true.....sometimes it is better to allow SOME flex (usually torsional) to help keep everything planted. My current race car (non-Mopar) is stiff enough it tends to lift the front tire during the hairpins (not that I am using it anyway). Using common engineering sense (if there is such a thing), if you have a fixed plane (or 4 wheels in this case) and you lift one corner, another corner (or two) will also lift.

Many years ago we had better results on the asphalt ovals with a car that would twist than was overly stiff. We could get the power down much quicker off the corners than the others. I'm just saying there is a fine line to all of this, if you want stiff, fill the frame rail connectors with Viagra! :cheers:
Click to expand...
done it both ways, welded -"not viagra" lol, if you don`t think connectors-welded inthe floor are stronger than just tieing the rails together you are wrong! some is better than nothing, that`s for sure. but the more things that are tied together the stiffer she gets! common sense here guys. just look at the roll cage in a prostocker, they ain`t tieing everything together for the hell of it! agree w/ the original poster all the way on this!--bob
 
famous bob said:
done it both ways, welded -"not viagra" lol, if you don`t think connectors-welded inthe floor are stronger than just tieing the rails together you are wrong! some is better than nothing, that`s for sure. but the more things that are tied together the stiffer she gets! common sense here guys. just look at the roll cage in a prostocker, they ain`t tieing everything together for the hell of it! agree w/ the original poster all the way on this!--bob
Click to expand...

Common sense in this instance doesn't have the background in structural engineering methods needed to perform an adequate analysis on the situation. If common sense could get someone through a complete stress/strain analysis on a structure as complicated as a modified car chassis then the world would be a very different place.

If you'd like to start calling people wrong, perhaps you should provide actual data. I don't think the answer is as clear cut as you would like to think. Welding to the floor provides obvious benefits, but when you start comparing different wall thicknesses, different sized gusset plates, different cross sectional areas, and welds that contain two very different gauge thicknesses, I don't think "common sense" considers enough variables.

The "best" way is the way that you can get it done. Any subframe connector is better than none, and I'd hazard a guess that 99% of the people on here won't ever push their cars hard enough to begin to be able to tell the difference between the two different types of connectors, assuming there's a significant difference to begin with. Running around on BFG T/A's certainly won't tell you. Take a look at the cars running at the autocross and on the road race tracks- what kind of connectors to they have? The hotchkis challenger? Yeah, it uses hotchkis subframe connectors, which are tubular and not welded to the floor. The Hotchkis taxi? Yeah, same non-floor tied connectors.

Are you that fast? Does it matter?

1970_Hotchkis_Challenger_B_by_TheCarloos.jpg


1970-plymouth-satellite-taxi-track.jpg
 
PoisonDart74 said:
The whole thread should have ended on this note 4 pages ago..
Click to expand...
BINGO!


But I also wanted to clear up my comment......most of it was said with tongue in cheek. From a pure racing (road, oval, drag) and engineering perspective, the more rigid you can make the chassis, the more consitant your performance will be. The trick is setting up the suspension to work with your less forgiving chassis. I agree with welding the rail to the pan, I think you should even seam weld every joint that is currently spot welded.

It is all a matter of return of investment. You have to ask yourself what you really want, what you really need, and is it worth the time effort and cost to do it.

And the Viagra is only good for the sprint racers, it doesn't hold up for the enduros! ~lol~
 
4+ hours or more is a sprint race? You're more man than I am, and my name is Rod!
 
Strange that the video in post #82 reported none of the readings on the dial indicators. They just said to view how the masking tape moved in the shaky video. You could kind of see. BTW, these weren't sub-frame connectors, but inner fender braces. Any bracing helps, but all adds weight. You have to balance stiffness with weight. A Sherman tank is very stiff. Same with rear ends and brakes. You can add giant C-body disk brakes, but lose some benefits of a light A-body car.
 
You must log in or register to reply here.
-

Back
Top