Tubular control arms discussion

[AngeloX19]

As the title states, I’d like to start a discussion, or solicit opinions on tubular arms with spherical bearings or rod ends. This is a project that I’d like to do this winter if I get around to it. I’m a decent fabricator, mostly self taught and I think I’ve done enough research on the web and various sources, but I’d like some confirmation. The goal is two fold, replace the rubber bushings, and shed some weight. I recall a member here stating that “suspension failures must not be tolerated” so that’s also a consideration on whether or not I just replace bushings with bearings.

By the looks of it, material choices seem to be either 4130 or mild steel DOM with sufficient thickness. Most of the aftermarket arms that I’ve been able to find specs for are for the muscle car segment, so they are generally 1.25 od x .093 or .125 wall DOM. Having never welded 4130, I’d like to use DOM. My thoughts for the rear arms are: 1” od x .093 wall or sizing up slightly from there, 1.25 x .093. Considering my X’s weight of 1640 lbs fueled and with me, there’s no reason why I shouldn’t be able to use the 1” od tube. Please correct me if I’m wrong. As for ends, spherical bearing or rod end 5/8” or metric sized to the oe bolts. For the ball joint, I’d like to use a rod end, likely ¾” and a bolt on a drilled out upright. I would think that something in the range of 5-8 lbs per arm saving is possible.

In regards to the fronts, I can either follow the PBS route of female rod end on the trailing link or use a ¾” x .065 tube with a male end. And follow their path for the control arm as well or use 1” x .093 tube with suitable ends. The weight savings in the front may not justify the hassle of fabricating new arms. I’m only estimated about 3 lbs total for the sizes I’m proposing.

I've have some photos on my pc at home that I'll post up when I get the chance, of an X that was for sale a while back, hopefully the former owner can chime in then. I welcome all opinions and info. Thanks.

 

[eflexter]

LOGIC

Weight savings is always good, but it seems the most logical reason to fab up control arms would be to give you suspension geometries that would be useful to you in a specific form of race.

Do you turn both directions? Is your car the stock weight and width?

You can spend your life learning about suspension geometry, so it's not a trivial subject. But at least, if you can make your control arm adjustable, you can take the trial-and-error shortcut.

It may not save weight, but it would give you an edge to be able to set up the car for a specific track.

Just having the equipment to properly measure caster, camber and toe is unusual, so do your homework before plowing into this.

 

[Black-Tooth]

Ya know, Angelo... I know the winters...

... are probably long and tough in your neck of the woods, you also probably have the skills, tools, materials, and where-with-all to do this thing, the product you come up with will probably be better, lighter, and more robust and adjustable... and also probably because of all these things, not cost a whole bunch in dollars, Canadian or otherwise, and you have all this TIME to do it in...

So I said all that... and then there is this 5X SCCA Autocross Champ (Steve Hoelscher) that used a relatively STOCK suspension with cut down STOCK springs and a lot of finagalling (sp) and experimentation... (which would be included in your proiject also) ... and I wonder if your project is gonna be really worth all the effort?

That will be up to you to decide and I'll certainly respect your decision... I will leave you with this last thought though (if applicable) when it comes to weight savings... Steve said hisself that he was amazed at how much better his car performed after he lost 40 pounds on a diet.

HA!

Hopefully you will get some of the scientific answers you are probably looking for... But I would be honored if you considered some of my REAL WORLD comments also.

 

[AngeloX19]

Eddie, I've thought about all your comments already. As I said, the main goal, replace the bushings and ball joints with bearings/rod ends. 2nd goal, if it's feasable is shed weight. At ~1470 lbs or so wet, dropping weight becomes addictive.

Tony, I always respect your opinion. And at 165 lbs, there's not much more of me to give up. Besides, I like fabricating and the thought process involved. I'd rather be in a cold garage, than be a couch potato.

Anyways, here's a few photos. I may opt to keep the fronts relatively stock, but this is along the lines of what I'd like to do for the rear arms.

 

[Black-Tooth]

Angelo... thanks for the kind words...

... and that is one hell of a sweet looking set up.

OK... so lopping off some unused body parts is NOT an option for you... HA! I just wanted to possibly give you another perspective JUST IN CASE ya hadn't considered it...

Now... don't let Bob Brown or Peter R see these fotos, especially if ya PLATE those swing arms... These things being BOLT-ON parts and completely "reversable" will drive these guys bananas!

Good luck with your conversion and do keep tabs on what the net LOSS in weight will be, along with providing us with your in-progress fotos!

 

[Steve Hoelscher]

Tubular Arms

Tony, as usual is not far from the mark but, for racing, lighter/stiffer is better. I know enough about the X's control arms to know the rears flex more than I would like for their weight and the fronts are way heavier than necessary.

The reason for going tubular is simple: Well designed and fabricated control arms are lighter, stronger and provide for more accurate adjustment. I am doing exactly this on my MR2. My X1/9 ran in a class that prohibited the modification or I would have done it as well.

But to Tony's point, there is no reason for making some kind of exotic design as the standard control arm is effective. You only need to improve its shortcomings, which are entirely due to the car's original design criteria: cost, ease of manufacture, packaging.

 

[Rupunzell]

Tubular A arms are fine, many race cars have them. They can lower unsprung weight which is one of those places where weight really matters.

Given the choice, the preference would be 4130 Seamless tubing welded with oxweld CMS 32 rod..This is an old and proven way of making tubular structures. DOM is OK, but seamless is better(personally, I don't use DOM for anything). Cost for what you're doing is not an issue and seamless 4130 tubing is not difficult to get.

Also.. Go to a larger diameter and thinner tubing. Strength and rigidity comes from diameter more than wall thickness. Heat treating post fab is a good thing.

The threaded fitting should be machined parts that are compatible with welding to 4130. Don't just weld a nut on the tube and use that as the threaded fitting, this just won't do.

Use top quality rod ends.. with fiber reinforced Teflon liners, if possible get the ones that are staked to reduce slop between the ball and outer housing. Good ones come from Spec Line, NMB... They are NOT going to be cheap (don't be surprise to discover spending $100+ per rod end) and the garden variety rod ends can easily ruin the possible gains due to bearing slop or worst a failure. Basically these are junk and not useful for real race cars..

Rod end boots to keep dirt and etc out of them are available, use them.

Use the proper AN bolts with the proper grip lenght. There must be no threads in the shear loaded section of the joint. A proper conical spacer, lock nut and washers to finish up a proper joint.

Bernice

As the title states, I’d like to start a discussion, or solicit opinions on tubular arms with spherical bearings or rod ends. This is a project that I’d like to do this winter if I get around to it. I’m a decent fabricator, mostly self taught and I think I’ve done enough research on the web and various sources, but I’d like some confirmation. The goal is two fold, replace the rubber bushings, and shed some weight. I recall a member here stating that “suspension failures must not be tolerated” so that’s also a consideration on whether or not I just replace bushings with bearings.

By the looks of it, material choices seem to be either 4130 or mild steel DOM with sufficient thickness. Most of the aftermarket arms that I’ve been able to find specs for are for the muscle car segment, so they are generally 1.25 od x .093 or .125 wall DOM. Having never welded 4130, I’d like to use DOM. My thoughts for the rear arms are: 1” od x .093 wall or sizing up slightly from there, 1.25 x .093. Considering my X’s weight of 1640 lbs fueled and with me, there’s no reason why I shouldn’t be able to use the 1” od tube. Please correct me if I’m wrong. As for ends, spherical bearing or rod end 5/8” or metric sized to the oe bolts. For the ball joint, I’d like to use a rod end, likely ¾” and a bolt on a drilled out upright. I would think that something in the range of 5-8 lbs per arm saving is possible.

In regards to the fronts, I can either follow the PBS route of female rod end on the trailing link or use a ¾” x .065 tube with a male end. And follow their path for the control arm as well or use 1” x .093 tube with suitable ends. The weight savings in the front may not justify the hassle of fabricating new arms. I’m only estimated about 3 lbs total for the sizes I’m proposing.

I've have some photos on my pc at home that I'll post up when I get the chance, of an X that was for sale a while back, hopefully the former owner can chime in then. I welcome all opinions and info. Thanks.

 

[bikesandcars]

Tony, as usual is not far from the mark but, for racing, lighter/stiffer is better. I know enough about the X's control arms to know the rears flex more than I would like for their weight and the fronts are way heavier than necessary.

I once saw a team spend $10,000 turning a bushing-suspensioned radical (sports racer with a motorcycle engine) into a spherical ball-end suspension (with new control arms). The car was slower.

Stiffer is sometimes better, but not always, Airplane wings for example.

Yes, the stock bushings / control arm may have some flex, but without detained analysis and testing you can't say if that flex is helping or hurting.

Of course, weight removal is always good.

I think the advantage of making your own arms is not in the performance area. If you jig it up and make 3x sets at once, you will have plenty of quality spares that don't have the ball joint replacement issues of stock, which is worth it in a racing environment.

Personally, I"d go with rubber bushings on the car end and sphericals on the upright, its much kinder on the car if you have a little rubber to buffer wheel contact.

Regarding 4130 vs mild steel... (Bernice, did you mean ERW vs DOM?) nothing is "free"... yes 4130 has equivalent ultimate strength for thinner sizes, but heat treating and proper welding becomes much more important. Plus its more expensive. My vote would be slightly thicker mild steel DOM for the first set.

If you have a lathe you can spin your own threaded ball-end receptacles from bar stock, which is what I would do if I had a lathe, or you can order these online from race suppliers.

Day racing has those threaded rods (for rear toe) pretty cheap.

Remember, don't defeat yourself making it perfect. just make a set to stock dimensions and see how it goes from there.

 

[Steve Hoelscher]

Yes, the stock bushings / control arm may have some flex, but without detained analysis and testing you can't say if that flex is helping or hurting.

Interesting point of view. After spending a few minutes considering this, I can't think of a reason you would want the control arms to flex. Of course, everything flexes.

The stock, stamped steel rear control arms flex more than is desirable under loads that exceed what you would see for a street car. Control of the tire's contact patch is everything in generating grip. So reducing the control arm's flex gives you more consistent control of the wheel and the tire's contact patch.

And Bernice has the right idea. I like 4130 for control arms (as opposed to roll cages) as it is stiff and strong. A larger tube with a thinner wall is better. First, it is stiffer for a given weight and will bend/break easier in an accident and protect other, more expensive, items.

Replacing the bushings with spherical bearings is an excellent upgrade on the X. This is because the rear control arm bushings do not pivot in the same axis. This requires the rubber to deflect for the suspension to move through its range of travel. This deflection causes binding and therefore friction. It is common knowledge that reducing friction in the suspension pivots increases grip. The spherical bearings will allow the control arm to move throughout its range of travel without binding.

The front control arms have a similar problem to the rears. The radius arm causes the control arm to rotate and it moves. There are two solutions here. You can make the control arm with a spherical bearing at the pickup to eliminate the bushing deflection as the control arm moves vertically or you can add a spherical bearing where the radius arm attaches to the control arm thus eliminating hard attachment and thus the rotation. Not to mention the cast front control arm and solid rod radius arms are very heavy.

 

[bikesandcars]

After spending a few minutes considering this, I can't think of a reason you would want the control arms to flex.

Common sense would dictate that the stiffer you make the control arm the more control you have, and therefore you can set the car up more exactly, allowing more ultimate grip..... I'm just saying that my experience indicates isn't always the case...There are too many variables without testing... and there are tradeoffs (sometimes negative)... I would of course defer to someone with direct racing experience in a x19 replacing stock suspension with these arms and having some imperical or seat-of-the-pants data that the car handled better and was more reliable because of that change.

The suspension is a system with individual components having their flex and natural frequencies, that flex can sometimes help a suspension absorb bumps and irregularities that can be beneficial to traction, and easier on the car. Example: rally cars will often use rubber bushings to help absorb impacts.

Rubber bushings can provide somewhat of a progressive effect in the suspension.

It is common knowledge that reducing friction in the suspension pivots increases grip.

I agree that you would probably realize a benefit in handling by going to spherical mounts and eliminating suspension friction if you accordingly match and tune the shocks and springs to that upgrade. If you are using triple adjustable shocks you could better fine-tune your fast and slow bump if the suspension is solid and had lower friction (allowing it to adapt quicker to road imperfections). I'm just saying you must also accept the tradoffs of a harsher ride, and more abuse on the car, and doing stiffer arms with spherical ends may not result in significantly improved lap times

 

[AngeloX19]

The front control arms have a similar problem to the rears. The radius arm causes the control arm to rotate and it moves. There are two solutions here. You can make the control arm with a spherical bearing at the pickup to eliminate the bushing deflection as the control arm moves vertically or you can add a spherical bearing where the radius arm attaches to the control arm thus eliminating hard attachment and thus the rotation. Not to mention the cast front control arm and solid rod radius arms are very heavy.

Question, if I use rod ends on both the inboard pickup and at the front of the trailing link, there's no reason why I couldn't weld the tube for the link and the control arm together?

I've also noticed that there are threaded 4130 fittings pre made for various thread sizes, but none have a tube applications are bigger than 1"x.065. Assuming I opt out of the DOM, does anyone feel confident on this both for the front and rear arms, heat treated off course.

If not, I'm currently thinking .875 x .065 for the trailing links. 1.125 x .065 for the front arms, and 1.25 x .065 for the rear arms. Anyone have opinions? Am I on the right track?

 

[Steve Hoelscher]

Question, if I use rod ends on both the inboard pickup and at the front of the trailing link, there's no reason why I couldn't weld the tube for the link and the control arm together?

Sure. The OE connection is basically solid. This effectively makes the control arm and radius arm an A-arm anyway. The down side of welding is the lack of adjustment.

The preferred configuration would be to use a joint (rod-end/spherical bearing) at the junction of the control arm and the radius arm. This allows for caster adjustment through the radius arm without affecting the location of the body mounted end of the radius arm.

I've also noticed that there are threaded 4130 fittings pre made for various thread sizes, but none have a tube applications are bigger than 1"x.065. Assuming I opt out of the DOM, does anyone feel confident on this both for the front and rear arms, heat treated off course.

If not, I'm currently thinking .875 x .065 for the trailing links. 1.125 x .065 for the front arms, and 1.25 x .065 for the rear arms. Anyone have opinions? Am I on the right track?

I would be a little nervous using 1"x.065 for the control arms. 1.25x.065 would make me happy but the 1.125 is probably OK.

 

[AngeloX19]

Thanks thus far, but question. Your saying its better to have a rod end where the radius rod meets the control arm, for castor adjustment. Would it be the same if the rod end were at the front of the radius rod where the mount is ala PBS? I'm not seeing why one is better.

And 1.250 tube is safe, are you refering to DOM or 4130? TIA.

BTW, did you get my PM a couple of weeks back?

 

[Dr.Auto]

I like this thread.... I've been thinking of doing this for a while as I only have one set of rear control arms with GOOD balljoints. Once they wear out I will have a reason to fab some control arms up.

Was wondering if it was possible to get a more detailed picture of the rear inboard pickup setup?

I'm asking bacause if you're going through all the trouble of making you're own rear control arms why not also redesighn the pickup points closer inboard? something like this?
http://farm1.static.flickr.com/51/188742850_590d35ae59_o.jpg

 

[Steve Hoelscher]

Thanks thus far, but question. Your saying its better to have a rod end where the radius rod meets the control arm, for castor adjustment. Would it be the same if the rod end were at the front of the radius rod where the mount is ala PBS? I'm not seeing why one is better.

You can certainly have the adjustment at the forward end of the rod where it mounts to the body. But you need the joint at the control arm to allow the caster movement. If joint is welded then the control arm/radius arm become an A-arm. And if you then move the ball joint (located at the top of the "A") forward/backward to adjust caster, then the end of the radius arm where it attaches to the body has lateral movement. Not good.

To visualize this; make one foot of the "A" a fixed location. This is the control arm pivot. The move the top of the "A" (the ball joint) and see what happens to the other foot of the "A". it travels in an arch laterally to the other foot. To prevent this from happening, you need the top of the "A" to have a joint so it can allow the other foot to remain in its lateral location without binding.

And 1.250 tube is safe, are you refering to DOM or 4130? TIA.

BTW, did you get my PM a couple of weeks back?

I like the 4130 for suspension parts.

 

[AngeloX19]

Now I understand. Up to this point I was assuming that as you make castor adjustment and lengthen/shorten the radius rod, the ball joint wouldn't scribe an arc. It would just move along the axis of the radius rod. So following my crude schematic, I would want to fabricate something along the lines of what's below. Without a bearing at "a", "b" would want to arc with the castor change.

 

[Brody]

ball joints here too

I too have been playing around with this idea of fabing up some rear arms. I don't like the lack of rear ball joints to replace 'em when they wear out. I took an old set and started hacking at the rivets (the ones that just don't give up the ghost easily). Idea being to get the arm down to "shell" stage for use in making a jig.

Long term project as I've got waaay too much else going on these days:hrmph: I like the welding/tubing discussion though! Tons of good ideas for me to start buying some metal.

I'd also like to 'size' the lower area to accept a commonally available ball joint for us here in the states. My old Isuzu Amigo used a sandwiched/captured ball joint that wedged in between an upper and lower metal portion of the a-arm. Very strong setup that I'd like to emulate.

Not thinking of balljoints though (street use, daily driver). Kinda surprised someone hasn't jumped all over this in the past. I'd buy two sets if they were available!

-Brody

 

[Dr.Auto]

So Steve....
Just curious considering you're knowledge and experiance... What's you're opinion on changing the geometry of the Macpherson strut setup by fabbing the control arms longer and moving the inboard pichups closer to the center line of the car? I mean.... putting these control arms on the car puts you in a Mod class anyways right? so if you're going to do it why not improove the desighn to make it that much better? I also suspect that you'd be able to run a softer spring rate to get the same results too....

 

[Rupunzell]

The front / rear ball joint cold be replaced by ones made by

Howe Racing.
http://www.howeracing.com/c-510-howe-precision-ball-joints.aspx

Baker Precision has NMB rod ends. They are good to discuss your bearing needs with also:
http://www.bakerprecision.com/

Specline Spherical Bearings:
http://www.specline.com/Products.htm

As mentioned before, 4130 seamless tubing for the control arms welded with Oxweld CMS32. This tubing was originally used in WW-II for tubular aircraft structures and has found it's way into many applications since then. If heat treated, sent it out to a reputable heat treat shop that has been doing this for a long time. Wire a sample piece to the finished part and make sure to drill a small air relief hole (#60 is fine) in a non-critical area of the sealed tube or the part will get all out of shape due to trapped air expansion when heated. BREAK ALL SHARP EDGES !!!. Any sharp edge post heat treat will become rock hard and will become a stress crack failure waiting to happen. Heat treat or not, Break/deburr all sharp edges is a must do.

Heat treat to not more than 125Ksi. As delivered, 4130 seamless runs about 90Ksi.

1018 steel Drawn Over mandrel (DOM) tubing runs about 60Ksi.. or seamless 4130 is a bit stronger as delivered, no heat treat.

In all cases, miter all tube joints as required and do machine threaded tube end fittings as required for a weld compatible, 1018 for DOM, 4130 for Seamless.

Thread depth on the threaded tube ends should be no less than 1.5 thread diameters ( 2 thread diameters perfered).

Some ref info on materials:
http://www.aerospacemetals.com/steelalloys.html


Bernice

I like this thread.... I've been thinking of doing this for a while as I only have one set of rear control arms with GOOD balljoints. Once they wear out I will have a reason to fab some control arms up.

 

[Steve Hoelscher]

You got it.

That's it.