Swift pickup problem

[swiftdrivr]

Okay, this is the lower rear suspension front pickup point on my Swift DB-1. When I pull in the direction of the arrows, there is visible movement of the rod-end, but no visible movement of the bolt. I have replaced the bolt and rod-end, but no change. Therefore, despite the lack of movement of the bolt, I think I have to consider that the hole is wallowed out [love to use the technical terminology] I think the next step is to hone the hole true-round again, and sleeve it. However, there isn't a lot of meat there, so the sleeve has to be very thin. Does anyone have any better idea? Is there a sleeve kit for this problem? Anyone know of someone who can make an accurate sleeve for this application?

Thanks for your thoughts

[R. Pare]

Is the rod end missing a "top hat" spacer?

[John Robinson II]

Jim
On my db6 that pick has a steel sleeve/hat assembly. Disassemble and check that, maybe install the bolt without rodend and see if you can move the bolt with needle nose pliers.

John

[swiftdrivr]

Good idea, John Thanks..

If I understand the joint mechanics correctly, the top hat and the lower hat should be in contact and absorbing the compression force, as opposed to the aluminum taking the force. I have measured and added a thin washer to the stack, which I think has achieved that, but maybe it is part of the problem.

[John Robinson II]

On mine, the bottom sleeve is longer than the oil tank so no washer is needed. Are you certain you have the hats in the rod end? Even from the compression of the bolt you should not be able to move the arm in and out, maybe up and down if no hat. How tight is the bolt? Maybe the bottom sleeve is not moving up against the hat. But then again, you are working on a 30+ year old machine.

[swiftdrivr]

I haven't taken it apart since Atlanta, [ it's not a new problem, but I thought it was solved] but I was planning on doing that in the morning. I'll know more then. If I recall, it's torqued to 12 ft/lb . BTW: there is no movement up and down

[DaveW]

Jim,

If your description is correct, there has to be slop between the bolt and the rodend. Otherwise they could not move independently. So the above analyses are likely correct. If you install tight enough fitting "hat" bushings, the slop should be gone.

However, having said that, the basic issue with that sort of arrangement is that the cast "clevis" that everything fits into is very rigid, and cannot be compressed against the hat spacers and rod end no matter how tight you tighten the bolt. This means that any useful amount of friction between the cast clevis and the hats cannot exist, because you can't pull them tightly against each other. That also means that as the parts rub against each other, the hats wear into the clevis or get sloppy in the rodend. All the usual forces are horizontal, so a horizontal slot may have developed in the clevis from wear over the years or the hats have worn horizontally. That slop is why it moves horizontally but not vertically.

So the only thing you can do is to make sure there is as little play (clearance) as possible between the hat spacers and the rodend ball ID on their outside, and likewise to the bolt on the inside. And if the bolt isn't tight in the casting's hole, you will need to sleeve the casting to eliminate the motion.

[swiftdrivr]

Thanks Dave. Looks like I am going to need a machinist with more skill and accuracy than myself, along with a steady hand and a set of fine hones.

[Garey Guzman]

It's been quite a few years since my DB1 but I seem to recall some special washer or spacer that went in there. Here's a zoomed in, old picture that kinda shows it.

[DaveW]

Thanks Dave. Looks like I am going to need a machinist with more skill and accuracy than myself, along with a steady hand and a set of fine hones.

You may be over-thinking this a bit. Standard clearances (in the 0.001-0.010" range) will work. Even if the clearances are VERY tight, but the motion can't be totally stopped by clamping, stuff will eventually wear. You just have to be conscious of the issue and replace things as they wear out.

When I was testing HD truck-sized air springs at Firestone Research on hydraulic MTS testers, I had to find a way to keep from failing the 1" dia Grade-8 bolt that connected the hydraulic ram to the test fixture.

To solve this, I had the aluminum test fixture machined to accept pressed-in bushings that were tight enough to not move under the test loads, but could be clamped tightly by the bolt tension to 1" spherical joints to prevent the bolts from seeing bending stress. It was a big PITA to change fixtures and rod-ends when needed (the insert had to be pressed out a bit to allow this), but it did eliminate the repeated bolt failures.

In the case of most racecars, this sort of thing is not normally used, since it requires too much work to make and also to replace components. So we just have to design the setup to not fail and replace/rework stuff as needed.

[swiftdrivr]

Okay, so I think it's multifactorial. The sleeve is a tiny bit loose, [shown] and the bolt is a bit more loose in the hats. I can cut out some of the motion if I can find a machinist who can make really close tolerance hats, but the sleeve seems too close to correct to re-drill it yet. I was thinking of using this stuff to minimize the sleeve-to-hole slop effect. Anyone with any knowledge about this or similar products? Other ideas?

https://www.permatex.com/products/th...eeve-retainer/

[scorp997]

make sure the bolt is an AN type and not a standard bolt. the shanked area is much longer and more accurately sized in diameter.

I have picked up some oversized shanked bolts (when Boeing surplus used to be open), that could help if you reamed out the hole slightly to fit.

[swiftdrivr]

make sure the bolt is an AN type and not a standard bolt. the shanked area is much longer and more accurately sized in diameter.

I have picked up some oversized shanked bolts (when Boeing surplus used to be open), that could help if you reamed out the hole slightly to fit.

Hey, thanks for the offer! I may need to try one and see, but unfortunately that won't fix the sleeve-to-hole issue. Let me try fixing that first, and get back to you if I can't find a machinist to make tighter hats to address the bolt to hat slop. BTW; I am using AN bolts from Pegasus, but obviously, not with oversized shanks.

[steve zemke]

Hey, thanks for the offer! I may need to try one and see, but unfortunately that won't fix the sleeve-to-hole issue. Let me try fixing that first, and get back to you if I can't find a machinist to make tighter hats to address the bolt to hat slop. BTW; I am using AN bolts from Pegasus, but obviously, not with oversized shanks.

Ream the aluminum casting to the next size that will round it up and get rid of the ovality. Then have a local machinist machine a new insert for you. I have used the sleeve retainer successfully in the past. Both parts were of the same material.

[John Robinson II]

Jim,
Based on your previous comments, I don’t think the alum casting is wallered. If it was, the insert would fall out when you removed the bolt. My first inclination is the bolt isn’t tight enough. I don’t think 12 ftlbs is enough. You might have it so that the rod end with hats just fits in and won’t move up and down, but it isn’t clamped tight enough to stop the lateral movement. With the jet nut more torque is required than a non locking nut. On my db6 the insert is very hard to move, I have to pry down to get enough clearance to easily move the arm.
Before tearing it apart, try tightening it more and maybe even smack the bolt with a hammer. A larger hammer generally fixes everything except fiberglass.

[Fred Michael]

Wasn't that designed to have an insert that the bolt threads into?
Since yours goes through with a nut on top, is there a sleeve in the top clevis leg to account for the missing or drilled out insert?

[DaveW]

If you are going to replace the sleeve (assuming there is one there) that goes through the casting at the head end of the bolt, I have a suggestion that would allow proper tightening and prevent the ass'y from "working" back and forth which causes wear and looseness:

Replace the current sleeve with a purely cylindrical thick-walled steel sleeve that is a snug press fit into the casting and extends past it slightly toward the bolt head. That way, when you tighten the bolt, it will push the sleeve into contact with the hat and the resultant friction between the parts will prevent relative motion. This will hopefully keep it from getting loose again, similar to my fix on the MTS machines.

[tstrong]

Dave W is right on (as usual). Dave Bruns worked at McDonnell Douglas before going racing full time and clevis type joints in castings were very common and putting significant bending stresses in lugs was a big no-no. The typical configuration was a hat (or T) bushing in one leg of the clevis with heavy press fit, and a straight bushing that is longer than the clevis thickness in the other leg that is a light press fit. When you fit the bearing in the center the bolt tension “slips” the straight spacer tight against the bearing and clamps up the inner race to eliminate all looseness and/or rotation between the inner race and bolt. Basically eliminates all free play (both radial and axial) at least until the axial force applied overcomes the clamp up bearing force . Also as mentioned can be a pain to take apart / reassemble but usually can be done if careful to keep everything aligned and a little bit of prying. Maybe overkill, but if you do it right now it should be good for a long time.

That said, lots of DB’s out there that just shim as close a possible with washers or spacers and work just fine based on just getting a good close fit on the holes. Oversize bolts are a great option but hard to come by. Another thought would be using an NAS bolt which has slight larger nominal diameter and tighter tolerance than AN. Several ways to go but most are going to require at least the ability to hone the holes to a close fit with whatever bolt you end up with. Todd

[swiftdrivr]

Okay, here is the pickup schematic. If I interpret this right, I need the sleeve, [in yellow], to extend out of the bottom of the hole so that all clamping force is placed on the sleeve. It seems the resulting force-path would actually result in clamping the washer under the nut, [at the top], against the oversized flange of the hat, and therefore placing the clamping force on the bellhousing material surrounding the un-sleeved hole at the top end. To me, that suggest that the upper hole needs a sleeve also, if the goal is to place the clamping force on the bolt, not the bellhousing material. That may not be the desired goal, however. The resulting force-path [with the longer sleeve] does prevent squeezing the upper flange of bellhousing material against the lower flange of bellhousing material, which I assume would crack if over-loaded, so maybe that is the appropriate goal.

[DaveW]

I think that squeezing the green t-bushing to the rodend ball and the yellow sleeve to the t-bushing on the other side of the rodend ball is the goal. So, it looks like what you've drawn there will work.

I don't think you need an upper sleeve unless neither the green t-bushing or the upper hole in the bellhousing is tight enough to locate the upper end relative to the bellhousing.

[swiftdrivr]

I think that squeezing the green t-bushing to the rodend ball and the yellow sleeve to the t-bushing on the other side of the rodend ball is the goal. So, it looks like what you've drawn there will work.

I don't think you need an upper sleeve unless neither the green t-bushing or the upper hole in the bellhousing is tight enough to locate the upper end relative to the bellhousing.

That's a relief, because it would be really difficult to drill / ream / machine and install a sleeve there. I will concentrate on finding a machinist to make some close-tolerance hats, and a longer insert!

[tstrong]

Agreed, If the upper hole is a tight fit to the bolt you don’t need a bushing/sleeve there. Also the green hat bushing doesn’t really need to be “oversized” as long at hole is tight and you have good bearing area (more is better but if you are actually making new I would make both the upper and lower the same). The key is that the two hat bushings must not touch in the center of the rodend. There must be a gap when clamped up. And that the yellow bushing is longer than the thickness of the lug so that it slips in the lug and clamps tight against the rodend without bending the lug. Todd

[tstrong]

Also you show counter bores (spotfaces) for the hat bushing to sit in (or maybe just meaning the machined surface of the casting. These are not required unless they are already there. Todd

[DaveW]

Thanks, Todd.

It's always good to have another knowledgeable person critique stuff. There are always details that get overlooked otherwise.

[swiftdrivr]

Thanks, Todd.

It's always good to have another knowledgeable person critique stuff..

Double lucky for me, that you both offered your help! Many thanks!

[Neil_Roberts]

The cartoon drawing and all of the following conclusions are exactly what I remember from my DB-1 days. The key safety detail that I have to offer is to cross drill the bolt heads big enough to pass 0.080" baling wire (McMaster 2000T81) through them. For bolts this large, regular 0.032" safety wire is just a visual indicator that the bolt moved. Those joints on my car were sloppy, so I had to re-tighten them every time the car came to a stop. The big baling wire did not stop them from loosening, but it did stop them from falling out. I recommend the same for every DB-1.

I really don't like this joint design for exactly that reason. Double shear clevis joints are the right answer for suspension pivots, but at least one side of the clevis must be thin and long enough to flex easily when the pivot bolt is tightened. This particular joint is obviously not, but it could have been with some small design refinements.

I had the privilege of working with David Bruns for a few years. He was a brilliant designer, but he had a giant blind spot when it came to mechanic-friendly design details.

[Roland V. Johnson]

Sorry, I am getting confused. Isn't the upper pickup bolt hole a blind hole that is threaded 5/16 X 18 (course thread)? Someone told me years ago the bolts were something like DC-3 rod bolts which were made in great quantity.

In comparison with other cars of that era I thought the DB-1 was extremely mechanic friendly.

[tstrong]

Roland as you know better than I originally this was a bolt that threaded into the top lug but if you go back to the very first post it appears it has been drilled out for a through bolt with nut. Either will work, but the through bolt has some advantages with self locking nuts, etc. The trade off is more parts, longer bolt, more weight.

Neil, as you mention making one leg of a clevis long and thin and ductile material is one way of allowing clamp up without over stressing the legs (look at almost all vintage car shock mounts). This however is generally not a good practice with cast parts hence the design that we have been discussing where you have a light press fit bushing in one leg of the clevis. Then the max lug stress is determined by the press fit of the spacer and is independent of the gap between the lugs or the bolt torque. Also, I’m not sure David has a blind spot for maintenance as much as an unwillingness to compromise on a design intent that was important to him, yet would make other compromises that seem out of place but were not critical to his priorities. As Roland mentioned given the tightness of packaging of the cars I don’t think they are that mechanic unfriendly but that’s just an opinion. Todd

[DaveW]

delete

[Garey Guzman]

When I had my DB1, I didn't like the way this was fastened so I took it to a machinist/welder of some repute and had this done. As I recall, steel sleeves in the bolt hole and the A-arm rod end got two reduction bushing/top hat washers. I can't recall if the machinist had to weld the tank but I know it was pressure tested to ensure it didn't leak.

Hopefully these pictures make more sense than my words.

[swiftdrivr]

The safety wire in the first picture is not firmly attached to anything. Essentially, It is just there to tell me if something moved. Drilling the head would be useful, as noted above, as I could loop the safety wire through both ends of the bolt.

additional questions:

1] Anyone know a good machinist who would take on a job this small?

2] the bearing sealer above looks like it would be a help, minimizing motion and preserving the joint. However, if the upper hole required reaming, the insert below would have to be removed [500 degrees to kill the bond, then press out.] No way to get a press on this except hammer and drift. Is the heat safe for the bellhousing? Anyone know how much force would likely be needed after the heat? Not fond of hammering on the clevis [even though JR2 seems to be in favor of hammers, as a general rule at least!]

[Roland V. Johnson]

Sorry I missed that first pix.

[tlracer]

The Van Diemen RF85-on series, which used a 'very similar' design to Swift approached this area slightly differently, but it may help:

a) They used a separate clevis mounting, rather than cast-in;
b) The rod-end has two T-bushings (or top-hat bushes as they're known on this side of the pond!) inserted in the same way
c) The bolt goes through the clevis mount and both bushings but there is no additional spacer

My understanding is that the T-bushings should be of a length that they just don't touch together in the rod-end bore. This is because otherwise the rod-end would be able to move up-down on the bushings.

The total measurement of the rod-end assembly, ie. the ball's face width + the thickness of both T-bushing flanges, is a tight clearance/slight interference fit into the clevis, to prevent the 'ears' of the latter bending inward when the nut/bolt is tightened.

I would have thought the sleeve in the Swift design would be there (in the original, thread-in-casting iteration) to take the compression loading of the bolt, preventing this from breaking the lower 'ear' of the casting. As the bolt is tightened, the sleeve and two T-bushings would be clamped together rather than simply pulling the lower 'ear' upwards and - possibly - resulting in a stress crack at the root.

Now there is a separate nut at the top, this essentially means there's a second 'ear', above the rod-end. This too will be subject to load when the nut/bolt is tightened so should really have a similar steel sleeve to resist this compression and possible stress-cracking.

[swiftdrivr]

Finally, success. I appreciate the advice and multiple offers to help, but in the end, I figured that it would be necessary to have the car on hand to get this done. So I ordered drills, and reamers, and carbide bits, and "stressproof" [I guess we'll see] cold-rolled steel, and did some reading and you-tubing. It took a few failures, and my fingers are bruised from the reamer, but after obtaining "as close as I could get it" tolerances on a new sleeve and both hats, it's now tight. Fortunately, I didn't have to ream the top hole, because I couldn't figure out a good way to accomplish that. [Lets hope I have no reason to re-visit that issue.] I shielded the lines with rubber inner-tubing when I re-assembled everything, too.

Again, many thanks for the advice, ideas, and offers of help!

[DaveW]

That's the way to go! Make it yourself and you know it's done right ! And if you have to do it again, you know what to do.

[swiftdrivr]

That's the way to go! Make it yourself and you know it's done right ! And if you have to do it again, you know what to do.

Thanks Dave. It's amazing what carbide does for my teeny, tiny TAIG lathe! I picked up a few other bits too, like how to make the cutoff tool work on an underpowered lathe, and the beauty of a four-jaw chuck, once you learn the [really simple] trick to centering the piece. I wish I'd known that when I made the rims for the model car.