A-arm fabrication

[rjwoods77]

Hello all,

I worked late into the night trying to finish up a redesigned pair of a-arms for our car and had to stop/didn't finish because of lack of tools to finish the job. Here is my problem. I have a jig that holds the two frame mount points and then the balljoint mount point. Welding stuff on the ends goes fine but in the process of welding the two "legs" together the frame mount points spread from the warpage of welding. What i mean by this is when i pop the three bolts off two of them line up with the holes and the third does not. In the past we had to heat one of the legs up and use woodworkers squeeze clamps to pull the mount back into position by pulling it past the mount hole and when the part cooled it would draw back to position. So it was this dance of heat.pull,check,heat,pull,check until it was lined up. These are made on mild steel just to note. The welder is very experienced and knows what he is doing and we were careful to move the heat from welding around. I am just wondering if we are the only people who have problems with this. If I made the jig out of the same material as the a-arm would it be smart to heat treat (normalize?) the a-arm and jig in a oven to relieve the stress? When I release the bolts will the positions be where I want? Is this what people do with welded upright? I would really appreciate some input on this because it is terribly frustrating doing this process and I also don't like localized heating to get the thing to move where I want it?

[rjwoods77]

Hello all,

I worked late into the night trying to finish up a redesigned pair of a-arms for our car and had to stop/didn't finish because of lack of tools to finish the job. Here is my problem. I have a jig that holds the two frame mount points and then the balljoint mount point. Welding stuff on the ends goes fine but in the process of welding the two "legs" together the frame mount points spread from the warpage of welding. What i mean by this is when i pop the three bolts off two of them line up with the holes and the third does not. In the past we had to heat one of the legs up and use woodworkers squeeze clamps to pull the mount back into position by pulling it past the mount hole and when the part cooled it would draw back to position. So it was this dance of heat.pull,check,heat,pull,check until it was lined up. These are made on mild steel just to note. The welder is very experienced and knows what he is doing and we were careful to move the heat from welding around. I am just wondering if we are the only people who have problems with this. If I made the jig out of the same material as the a-arm would it be smart to heat treat (normalize?) the a-arm and jig in a oven to relieve the stress? When I release the bolts will the positions be where I want? Is this what people do with welded upright? I would really appreciate some input on this because it is terribly frustrating doing this process and I also don't like localized heating to get the thing to move where I want it?

[exFSAE]

Doesn't matter how good your welder is, pull happens. You melt metal, it cools, it shrinks.

The trick is as follows, and has worked great for us..

Weld it all up on your jig. It will pull. Put it back on the jig (will take some muscling to get it on there and stretch it). Take an oxy torch, and flame normalize the whole weld area. With material bein so thin it takes like a minute to soak. Goes right back to where it should be on the first go.

[Chris Clarke]

What about adding a small 1/2 tube connecting the two frame mounts? Sort of like an anti-intrusion bar used on the monocoques?

If I was making some A-Arms, I would also seriously consider having the frame mounts in-line and then making them from one piece of machined tubing, welding/stress-relieving, and then cutting out the section between the two mounts.

[rjwoods77]

Exfsae,

Obviously a bad thing to do with 4130 but works for mild steel. We will give it a try.

Chris,

Out frame mounts are inserted in the ends of the tube. Its on the other end that we have problems. Stress at the Y.

[flavorPacket]

why is normalizing 4130 a bad thing? we've been doing it for 20 years, never had any problems.

PS if you make your jig beefier, this won't happen. Our control arm jig weighs 55 lbs.

[rjwoods77]

Flav,

The jig doesnt have anything to do with the warpage. It holds everything fine until you back one of the bolts out then the part warps out. Are you saying make a similar material jig, which will end up heavy if you have steel a-arms, then throw it in the oven with the a-arm?

[fade]

2nd exFSAE's method

[Steve O]

We use 4130 and a particle board jig with 1/2" wood strips nailed in to each side to hold the steel in place. The upright side is bolted down to the jig. We pull our a-arm out and it fits on the car no problem. Heat treating sometimes gives a little bit of an issue.

Steve

[exFSAE]

Normalizing 4130 is a bad thing? No. It is not. If anything it is a better thing to do in 4130 than in 1018 or some low carbon steel. Low carbon steels won't get as brittle when welded. Higher carbon steels will. Flame normalizing will help soften up the HAZ. Even in the normalized state 4130 should still have a yield in the 50-60 ksi range if I'm not mistaken.

I can't see why using a beefier jig plate would make any difference. We used one out of 3/4" thick plate aluminum with everything bolted and/or clamped down.

The problem isn't so much constraining things and preventing them from moving, I dont think. Its that when you're welding, particularly when adding a lot of filler I think, that when that weld pool cools it shrinks. Even if everything is "fully constrained" it will be effectively preloaded.

When I welded ours.. it would be all nice and constrained when we put it on the jig, but then taking it off it would be tight and then "spring" into a slightly distorted position. Put it back in the jig, and flame normalize that whole welded are and it will relieve some of that residual stress, let the a-arm go back to where it should be, and make your welds less brittle.

[RacingManiac]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Rob Woods:
Flav,

The jig doesnt have anything to do with the warpage. It holds everything fine until you back one of the bolts out then the part warps out. Are you saying make a similar material jig, which will end up heavy if you have steel a-arms, then throw it in the oven with the a-arm? </div></BLOCKQUOTE>

Thats basically what we do to this day. Beefy steel jig, weld the a-arms in them, then we send them out for stress relieving by heat treater with the jig.

[flavorPacket]

same with us, racingmaniac, except we have an oven in house that gets hot enough to get rid of all the martensite.

And Rob, the jig definitely matters. Stresses introduced while welding will cause your hardpoints to move if they are not held correctly. I have personally seen a 3/8" thick 1018 plate dish out by 100 thou in one year's use as a control arm jig.

[rjwoods77]

Racing and flavor,

Thanks for the feedback. I think a hardcore jig and oven is the way to go. Racing, do you do this with your uprights as well? Do you have a pic of your setup for either upright or a-arm? I know I can look this up somewhere but what is your heat treat spec for 1020 DOM and 4130. I imagine it would be different then what is done for drive shafts.

[flavorPacket]

for 4130, the normalization temp is 710-720 C. The length of heating depends on a bunch of stuff, it's better to ask a pro.

[Composites Guy]

Any chance you guys would post pictures of your jigs? It would be interesting to see.

I designed a set carbon a-arms with a carbon inboard flexure which you might have seen at the last couple of competitions.... which means I haven't tried my hand at steel a-arms yet.

I am using 4130 steel, and was considering a 3/8" aluminum plate for a jig, with a brass insert where the outboard sphericals go to minimize distortion. Like Rob, I am worried that I need to normalize the steel, at least at the outboard end. I was thinking of flame normalizing the outboard end, but was worried that would cause the the arms to buckle outward as they heat/extend. Hence... the clamps on each arm. Sounds like you guys think the aluminum jig is a no-go?

http://i110.photobucket.com/albums/n114/lehigh_jack/Jig1.jpg

[rjwoods77]

Composite,

Thats the jig that we use more or less. The jig exhibited no warpage. I know someone had mentioned about bending a jig but in my experience I havent seen it. I think that would work fine. My original question was if there was a better way than putting on heat but it seems that is the answer or to build a complete steel jig and throw the whole deal in an oven.

[Fil]

We weld inch RHS together to roughly resemble the base of wishbones, turn up some spacers to take our bearing cups and stake in ends, mill holes in the rhs at the correct position, weld the spacers in, fit the tubes up, and then weld the wishbones.

We get them heat treated, but before they are sent off to the heat treater we weld on a tube between the two points so when they are stress relieved the points stay relative to one another. We have had problems with arms not comming back from heat treaters straight. obviously we cut the stick off after it comes back from teh heat treater.

Also we have done testing to see if post weld heat treatment is worth the extra effort, and the difference is scary. We did this test on a single batch of pull rods we made, and tested on our instron tensile tester, one was heat treated and the other one wasnt. The non heat treated failed well below its yield and exhibited no plastic deformation, the heat treated one failed as expected. Both were welded by our guru welder. you are crazy not to heat treat any welded part like an upright or wishbone. i will try and post photos of our upright and wishbone jigs.

[exFSAE]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Rob Woods:
Composite,

Thats the jig that we use more or less. The jig exhibited no warpage. I know someone had mentioned about bending a jig but in my experience I havent seen it. I think that would work fine. My original question was if there was a better way than putting on heat but it seems that is the answer or to build a complete steel jig and throw the whole deal in an oven. </div></BLOCKQUOTE>

Tellin ya, much easier to just flame normalize it. The fear I had with using a steel jig and everything and putting it ALL in an oven at once was with all the steel reaching recrystalization temp at once if anything else would start warping or fusing together. Warping from machining of the jig and relieving stresses there or whatever.

So the jig and flame normalizing worked great.

[RacingManiac]

Upright Jig(opposing side can fit the opposing upright):
http://www.supercars.net/pitlane/pics/70153/3315346d.jpg

A-Arm Jig:
http://www.supercars.net/pitlane/pics/70153/3315346c.jpg

For all the critical points(bearing bore and so on), the bore are made undersized, then was post machined after heat treating to the right dimension. In our a-arm's case the 2 inboard ones are used with press-fit spherical bearing and outboard side is a spherical with retaining ring, so the inboard holes are reamed post heattreat, and outboard housing are bored to size and machined with retaining ring feature to spec. Similar deal with the upright where the critical portion is the wheel bearing bore, which is bored out after heat treat. All the fasteners and fake heat treat bearings have liberal amount of high temp antiseize applied to them such that to make sure we can still take them apart afterward....

[billywight]

http://img301.imageshack.us/img301/4949/jigxs1.jpg

Never had any post welding issues when using this setup, there was very little distortion. The jig is just 6061 and the round bit rotates to various angles so one jig does all the arms. All a-arms and anti-roll bars were heat treated post welding as well. All were 4130 with 4130 welding rod and no flame normalizing.