Quick structural question

[teamfour]

I need to fab a subframe component that is made from 1", .083 wall mild steel. The component will have a 2" segment, then a 14" straight dogleg of about 10 degrees, then back to a 2" straight segment (angle in illus is over 10 degrees, but you get the idea).

The segment will be part of a subframe that is under significant stress. Is it better to make the dogleg using a tubing bender? Or cut the tube and weld it together to make the dogleg?

Hope this makes sense.

[RobLav]

Not enough information provided. Which way are the stresses?, etc? Why have you determined the material already?

What other alternatives (besides tubing) are there?

.083" wall is fairly thick. I made my engine to chassis subframe out of 7/8" 18 gauge mild and .050 4130.

[jerry freeman]

If it's round I would bend it. If it's square you will probably have to cut and weld it because most people don't have 1" square tube benders. My 2 cents only.

Jerry

[teamfour]

OK, I finally had time to redraw the concept. The areas in red are in question; specifically where the tube jogs. This is a design originated by Jay Novak and he has assured me there is a ton of stress on the "hbar" assembly. His setup does not necessitate the jog required on mine.

[Billy Wight]

Just looking at this quickly, I wouldn't imagine you would have a problem with either method. I would suggest going with bent tubing rather over welding as it will be a slightly stronger design although harder to fabricate... One thing that jumps out; why are you using such a small tubing diameter? Going with a larger diameter (1 1/8" or bigger even) and thinner tube will result in a much stiffer structure with a minimal weight penalty as you have a lot of areas loaded in bending. It seems that Jay might be the one to ask though as he has a lot of experience with this type of car. Hope this helps-

[teamfour]

Thanks Billy. Jay is commenting on the design over on another forum. He has basically told me NOT to bend or weld the tubes; they need to be straight due to the lateral forces at play.

[teamfour]

Revised again.

[RussMcB]

I did this quicky picture before I read the later posts.

[RobLav]

At first glance, in Lee's (TeamFour) last drawing, both horizontal tubes might be redundant. Russ's drawing eliminates the upper horizontal tube.

But without the top horizontal tube, there might be a heckuva side load on the rod ends. As it is, there will be a side load on those forward rod ends anyway, even with the top horizontal tube in place. I'd build it so those rod ends are parallel to the longitudinal axis of the car.

It looks similar to a De Dion rear suspension. No camber change? No rear ARB needed either. It'd be interesting to analyze the twisting of that structure under load to one rear wheel and not the other. There is some ARB effect there. Can you use only one rear shock? How much anti-squat is there?

The design sure is different from what we are used to in FC/F1000.

[teamfour]

Rob and Russ, thanks for the comments. Yeah, the design is different from F1000, but I wanted thoughts from the experienced builders in F1000 without a bias towards the typical F500 setup. The design is actually emulated from Jay Novak's setup in F500.

I could use one "shock", but as you may know, we (F500) are limited to using rubber pucks for the suspension. Total axle travel is probably 2-3 inches. So, I need all the "spring" I can get.

[Billy Wight]

Lee-

The revised design looks much better. How much adjustment are you planning on using the rod-ends for? Or they just in place as bearings? If just bearings, you might want to consider spherical bearings rather than rod-ends as it will result in a stronger and lighter setup. If you're planning on doing adjustments with the rod-ends, keep in mind that adjusting both of them out or in will change the width of their mounting points and they won't fit into the mounting tabs (you should be able to adjust one in and one out to keep the same distance between points).

With the new design there are a fewer tubes in bending, and the stiffness will increase simply because of that, but I would still recomend going with a larger diameter tube with a thinned cross section.

[teamfour]

Billy, I may have to increase the tube size since Jay Novak has recommended using 1/2" heims for the two frame attachments. The rod ends will be used to square the axle in the chassis (remember we are talking solid axle here).

[Dragnmorad]

with the straight tube design, isnt there a way to weld on a very short legth of threaded tub at the corner points that go to the frame so they are at a 90 degree angle to the frame so that the adjustment work a little easier?

[Dragnmorad]

this is what I suggest so that your adjustments dont sress thier mounts as you adjust them as the angled ones would put alot of stress if they needed any amount of adjustment.

[RobLav]

Dragn's suggestion is what I was saying. Good drawing.

[Art Smith]

Lee-

here's a few thoughts from someone that knows nothing about F500 cars and may be not understanding your illustration. the illustration suggests to me you're using the item being designed for lateral location of a solid axle rear end.

if that's the case your structural question is a good place to start and a number of very good suggestions on triangulation of the structure have been provided. before getting to the point of the post, larger OD's and thinner walls will serve you well in this application; may even want to consider three diameter gussets to protect the welds.

the use of both rod ends and/or spherical bearing is a concern from two perspectives: 1.) rod ends and spherical bearings have relatively low axial load capability; and 2.) both rod ends and spherical bearings are not a good "stiction" solution for the transverse load being applied. a simple free body diagram suggests to me the resistance (reaction of) to the transverse load is the real design challenge; particularly given the concern for the strength of the truss. two small tapered roller bearings (just like the old solution used in front hubs before the advent of cartridge bearings) on both sides should deal nicely with the lateral load. the bearing housings in the truss assembly could be machined co-axial to eliminate the need for adjustment or adjustability could be bolted onto one or both sides of the truss. two rod ends or spherical bearing mounted at an angle (ie: with the axis of their bore not normal to the direction of travel) sounds like a never ending set-up/maintenance adventure !!!

if you "have to" use rod ends or spherical bearings, it would be my recommendation to use hardware sized for the transverse load with margin for impact loading. someone else will have to help with the stiction challenge....................

Art
artesmith@earthlink.net