Usually this is done so that the inner bush on the big side can be full length, allowing the clamp-up to put the bush into compression rather than bend the forkend lugs together.

KS... just for clarity... for a dumb-old-guy like me... request You provide an illustration of what You verbally described.

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

Hello all its not perfect but something lihe below..

There is something terribly wrong with that picture, If you have a tie rod double lug fitting attached to a spherical rod end bearing, the bushings should only clamp on the inner housing ( Ball ) of the spherical bearing, they should also provide clearance for the outer housing of the bearing to move ( Wriggle room). typically this is a detail where two control rods are joined together. This illustration shows none of that.
B.E.

You are judged not by what you know, but by what you can do.

@berkshire, right, we see the two bushings clamp on the spherical bearing along pin or bolt axis on faces of spherical bearing, picture is not best, the outer housing is not shown I agree.. but main question is why one sliding bushing body OD is smaller than flanged one?

Are you sure there are supposed to be bushings in the holes? That does not make much sense because you already have the spherical bushing. So where is rotation supposed to occur? The double lug with two different hole sizes is usually for use with a shoulder bolt. The small hole is for the threaded part of the bolt and the larger hole is for the body of the bolt. This provides for a very precise tolerance on the clearance between the spherical bushing bushing and the supporting lugs. The bushing must not be tightly clamped between the lugs, but allowed to rotate at the inner bore. The spherical part of the bushing is to allow angular alignment, not rotation.

OK... I get it... but oddly complicated design...

I am normally used to the two clevis holes being the same diameter.

The flanged-plain-bushing is usually press-fit [interference-fit] with the lug-hole and the bushing-shank protrudes slightly into the gap between the lugs [ensures clearance with the lug-face]; and...

the plain-bushing is machined for a slight clearance-fit [0.0005] with the lug-hole and is positively longer than lug-thickness to ensure the stack-up cannot crush the lugs together when assembled; and...

the IDs of the bushes allow for a close-tolerance slip-fit of the bolt; and...

the plain bushing is installed 'wet-with grease'... to allow for axial-slip/displacement of the plain bearing; and then...

the stack-up... two bushes, spherical bearing inner-race, bolt, nut and washer(s)... is torqued-tensioned together without binding the lugs together, since all binding occurs against the flanged bushing end/hat/lug and the inner-race of the spherical bearing.

DANG... RPstress already said this in far simpler terms.

NOTE.

This design methodology works well for complex parts in flight control systems with anti-friction bearings, too.

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

Right WKtaylor, all of that I clear now, but as I say the figure is not the best, bushings are protrude inside to clear bearing, etc. all good.

But still my question not answer... why OD of flange bushing > OD of slide in bushing?

It would appear that somebody has designed this to make life easier for the mechanic, The most common configuration for this type of fitting is to place a small OD washer between the inside faces of the hemispherical bearing and the inside faces of the lug/clevis, on a nominal bolt clearance hole, then insert a Plain bolt through the lug/clevis and tighten down. Then depending upon the design philosophy of the manufacturer the lug is left finger tight with a castellated nut and cotter pin , or it is clamped tight forcing the Uniball joint to rotate upon movement . The two bushings eliminate the need for the washers. I would however like to see a washer under the nut on the red side, without it the assembly would be free to float until the outer housing contacts the side of the lug.

You are judged not by what you know, but by what you can do.

Yes there was a washer under nut, sorry forgot to draw..

The installation shown is similar to what ARP 5770 specifies (figure 68), except the blue hat bushing should be installed with the flange on the inner face of the clevis lug.

I find nothing odd about this design. Every feature in the lugs has a purpose.
In fact I plan to use one like it on an upcoming project.
The only feature I missed was the differential OD, which makes sense if you can't install the through-bushing from the right side, for some reason. It can pass freely through the left-hand hole to be pressed into the other lug. If I had 1000 of these to assemble, I'd be happy to not have to flip the lug over in the fixture, because that would save a few minute on each assembly.

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

SparWeb... Also the joint is complicated from another perspective... especially where relatively small clevis joints are concerned.

Two clevis lug holes [bushed or not-bushed] would be just fine as the 'same hole diameter' for a slight clearance fit... IF the thru Bolt was a shoulder-bolt that had a grip-length slightly longer than the cross-the-outside-dimension of the clevis [and of course a washer and castellated nut and cotter-pin]. Typical shoulder Bolts...

NAS1160 BOLT, SHOULDER, HEX HEAD, CLOSE TOLERANCE CRES-A286
NAS1297 BOLT, SHOULDER - HEXAGON HEAD
NAS1298 SCREW - BRAZIER HEAD, SHOULDER
NAS1299 SCREW - 100' FLAT HEAD, SHOULDER

Plus castellated nut and a thin steel/CRES washer

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

Great ideas WKTaylor, and I do appreciate the value in clamping up on the shoulder. The NAS hardware you linked is not quite big enough for the future application I was referring to, though I have seen them used successfully as you describe. Door hinges, etc.
The application in my case will have complex geometry through with the fastenings must be made. No hope of getting the "jesus bolts" to align in a nice rectangular pattern. Interfaces need fittings that allow for misalignment during installation, yet remain very stiff when all the bolts are done up. Sorry if that sounds cryptic - trying not to hijack this thread.

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

I think the key design feature is there is a clearance gap (or should be IMHO) between the clevis lug and the rod end, so that the bolt is clamping the bushings and not clamping the clevis.

Another thought is what is the impact of clamping the clevis ? what are the working loads/stresses like ? (If low then you can get away with clamping the clevis.)

Does the joint articulate ? (then I wouldn't clamp the clevis)

Does it really matter to the joint if assembled the wrong wey ? (ie if different diameters are "only" to prevent misassembly, why is this critical ?)

Sometimes "fussy" design is required, but sometimes it's done 'cause that's what the designer knew.

another day in paradise, or is paradise one day closer ?

OK I get it... sorta...

I have a current situation where a stock shoulder bolt ['company-spec'] is altered for a special application [shortened threads for clearance] on a drawing. These 'company-spec' bolts do not have length limitations of the NAS bolts [I cited]. These bolts are used in engine mount connections in close-spacing to honeycomb fairings.

I nave been seriously contemplating making these 'drawing' bolts by altering [making from] stock NAS6312U42 or NAS6712U42 [or longer] bolts [procurement spec = NAS4003]. This is a 3/4-Dia bolt that is already certified and on-the-shelf.

In this case I would trim the shank/threads-off to the length required [to start]; and grind the end down for threads [0.6250-18UNJF-18A]; and then have the threads installed by cold-rolling and the excess thread-tips ground down [to max allowed diameter]; then have the new threads NDI'ed; and then have the thread transition area glass-bead blasted for clean-up; then have a cotter-pin-hole drilled thru the threads as required; then static test at least one to ensure clean ductile break at a reasonable strength.

Perhaps You might want to consider making shoulder bolts. IF there is no significant tensile load in the bolt, then machine the each-one.

NOTE. I've kicked-around a simpler way to do this... groove the thread transition area and then cold-roll-work the groove... then install a 'snap-ring' and then install a thin CRES shear and nut to a low/snug torque against the snap-ring.

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

Wil,
I guess I just don't have the guts to start making bolts. That said, I've made barrel-nuts before, as in personally at the lathe/mill for a day. In that case, cut threads in the nut, but hardly any tension in the joint. For your project, have you also found qualified surface finishing vendors? I'm just thinking of a very long chain of processes needed to make the bolt well enough to minimize the "pucker factor". Turn down end, thread roll, grinder touch-up, (shaker table?), NDI, blast, drill thru thread, (plating?) and pack it up to go. That bolt goes from expensive to priceless, if it's a small batch. That would be a gutsy move, even for me.

My current inspiration is Boeing BDM 1520. That is full of illustrated fitting arrangements like a designer's catalog. I'm in the "self-aligning" and "Clamping" and "fusible" category so that the transverse load can be applied to both lugs of the fork, yet the equipment must break away in case of a gear-up landing.

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

SparWeb... Your case, large dia/long fasteners for clevis joints...

Fud-4-Thot...

Ever considered using bolt [drilled threads] with grip-length exceeding clevis width... then fastening with a combination of thin solid washer + wave-spring or Belleville washer + thin solid washer + castellated Nut and cotter pin? Specify a torque on the castellated nut that is 'just enough' to get noticeable/light compression of the spring-washer before cotter-pinning.

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

Yes, thank you. I will think about the Belleville washer idea.
The bolts in the joint will probably be removed ~10 times in the life of the mission, ie, once per year to afford access for inspection.
Considering that, I will have to expect the bushings, washers and bolts to be replaced at some point. Thankfully, Bellevilles aren't so rare.
Going back to BDM1520, the reasons to avoid clamped lug fittings are made plain, and your suggestions amplify my interest in the alternatives.

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

@SparWeb, I think it makes sense, but the slide bushing can pass from left side through same size hole if it can fit the same size hole right side right?

@tbuelna, why flange of bushing inside?

Kristina Sornikova-

One consideration with the type of joint shown, a spherical rod end and clevis, is constraining (axial clamping) of the spherical bearing inner race so that any rotation in the joint occurs at the interface between the inner and outer spherical bearing race surfaces. This interface is designed for accommodating rotation in the joint. Below is an example of an aircraft control system spherical rod end and clevis joint from SAE ARP 5770.

I think the point to the inside flanges on the bushings is to protect the clevis from wear, in case the spherical bearing is thicker than the bushing wall so the sph bearing "could" rotate against the clevis leg. Other variations would replace the left flanged bush with two flanged bushes, one on the inner face and the other on the outer. To my mind I can't much point to the plain bush (making one clevis hole bigger than the other) other than assembly sequence (the same rotations happen on the other leg).

another day in paradise, or is paradise one day closer ?

tbuelna,
In figure 67 what stops the NAS538B bushing, on the right hand side of the picture, from sliding into the fork and fouling the bearing seal on the double row ball joint? It would seem that the bushing would have to be a press in fit.
B.E.

You are judged not by what you know, but by what you can do.

@Kristina,
Yes, I was noting how the bushing for the right-hand lug can pass through the left hand hole.

@tbuelna / rb1957,
The guidance in the BDM tells me that the bushing flange on the inside makes a more effective bearing surface against the inner race of the bearing, when the bolt is tightened against these bushings.
This is less of an issue for CRES bushings than it is for bronze bushings. As shown in your example above, one of the bushing flanges is completely unclamped even though the bolt is clamped to the inner bushing stack. It occurs to me that the neutral axis of the fork's two lugs are farther apart, as a result of having the bushing flanges on the inside. That may offer a slight improvement to the design.

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

Sparweb,
It is that un clamped bushing I am talking about. It that is not a press fit in the fork, it can float out of place and foul the rod end.
B.E.

You are judged not by what you know, but by what you can do.

Berkshire,
I tried comparing that lug with the types recommended in BDM1520, but BAC would put the flange on the outside.
Both have the clearance to keep the lug forks from being squeezed, just arrive at it differently.
I do agree with you that right-side flanged bushing would have to be press-fit in order to behave itself.

Tbuelna,
Can you tell us more about the specified for this joint by the SAE paper? I don't have a copy of that one.

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

SparWeb,

Here is what SAE ARP5770 states on p.84, "In Figure 69 the slip-fit bushing is retained by segment staking of the exterior side."



You can find more details of bearing/bushing staking procedures in NAS 0331.

If you don't want to pay $78 for a copy of ARP5770, with a bit of digging you can find public source aircraft mechanical control system design handbooks from NASA, DoD and FAA. They will probably provide similar design guidance for this type of joint when used for aircraft primary flight control applications.

Thank you for the NAS 0331 tip. I probably have a copy of that. Ability to stake the outer edge is sufficient reason to orient the bushing that way, so I will compare with the BDM way and see which has the advantage.

It always seems to be like this: I have a bunch of stuff in a library to refer to, and maybe it's enough, but then I find something like this, or you show me just 1 more option, and it throws up so many more questions!

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

For grins... Some ancillary documents [not mentioned in ARP5770] worth a look...

SAE
AIR1594 Plain Bearing Selection for Landing Gear Applications
AIR4094 Aircraft Flight Control Systems Descriptions
AIR5875 Methodology for Investigation of Flight Control System Anomalies

MIL-HDBK-1599 + N1 Bearings, Control System Components, and Associated Hardware Used In the Design and Construction of Aerospace Mechanical Systems and Subsystems

NOTE. I am not wild about the double bushing [plain bearing] system shown in tbuelna’s last post. The concept of having asymmetrical holes [two different sizes] in a symmetrical dual lug system tends to create a design/stress/maintenance asymmetry in the lugs at the holes. From my experience, this is NOT elegant/simplistic/durable/repairable as it looks.

NOTE. Any time a lug/clevis system is assembled like this, without means/methods to shim the single clevis lug to the center of the double clevis lugs, there will be a load eccentricity in the double clevis system... even a few thousandths of an inch toward one lug has a noticeable added/subtracted load affect; and in a highly loaded structure can induce significant secondary loading/bending that has to be accounted for.

NOTE. Chrome-plate bolts [shanks only] are often used in these type joints where 'bolt/pin shank-wear' is a potential issue within the bore of a hardened steel bearing inner-race.... leading to looseness.

NOTE. MANY years ago [1980], a friend who had worked for Boeing thru the 1960s into the early 1970s] ... and was ‘pink-slipped’ along with hundreds more engineers en-mass... walked-out with his mini design manual. He gave me his mini-BDM to photocopy ~1979. Best gift I ever received... It was loaded with structural and mechanical design insights, tips, lessons-learned, examples and general knowledge that kick-started my life my life-long quest/thirst for knowledge.

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

This a deeeep rabbit hole, Wil!

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

hey will, care to scan that ?

another day in paradise, or is paradise one day closer ?

RB,
Why would you need access to a second copy?
If you don't know what I mean, you should phone our mutual friend, "SAITAETGrad"

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

He's got Will's early BDM ?

another day in paradise, or is paradise one day closer ?

RB...

I have corresponded/exchanged data-with one E-T member... CoryPad... several years ago... thru Dave... but I am on a mission to accumulate/document my massive library and take care of injured/recovering wife... so spare time for such an endeavor is very sparse, right now. Perhaps I'll make some stuff available after retirement.

WHOAAAAA........

I just realized that I had not seen any forum entries from CoryPad in awhile. I just check his profile... he hasn't logged-in since Nov 2017!?!?! Anyone know his status?

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

Quote (rb1957)

He's got Will's early BDM ?

At least 2 versions, maybe a third vintage now.
BTW, he's back on the payroll of your company now.

Wil,
CoryPad has made many contributions that were of value to me, too. I'm another who will miss him if he's gone.
But there is always hope it's something good keeping him away... like a world tour on a cruise ship.

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

thx steve ... "a nod's as good as a wink to a blind bat" ... found it (copied it !)

another day in paradise, or is paradise one day closer ?