Knuckle Coupler

[spikejones52002]

A knuckle coupler is a hing just like on you front door.

The hinge pin transfers all the load between parts.

The knuckle pin just locks the knuckle closed.

[Randy Stahl]

A knuckle coupler is a hing just like on you front door.

The hinge pin transfers all the load between parts.

The knuckle pin just locks the knuckle closed.

No , the knuckle pin keeps the knuckle from falling out , many are made of plastic nowdays. You don't need the knuckle pin to keep the knuckle closed , That's the job of the Draft pin , Lock .. etc..

[cordon]

I will try to answer the question about forces.  BTW, that was some thread, that other one.

First, a couple pictures:

http://www.answers.com/topic/coupling-railway

http://www.sdrm.org/faqs/couplers/index.html

And now a crude diagram (case-sensitive):

http://unimind.us/scirockit/COUP02.gif

I went to an old display caboose in our town today to get a close look.  What I saw is very similar to the photos at the links above.  But the photos don't show the "lock" inside the coupler body.

Please refer to my drawing.

The knuckle pivots on what I called the "PIVOT PIN."  Others have called it a knuckle pin.  The knuckle has two ridges that I named "FORCE BEARING RIDGES" that bear against mating surfaces inside the body of the coupler on the side that holds the PIVOT PIN.  You can see the ridge closest to the PIVOT PIN in one photo, and you can clearly see the one farthest from the PIVOT PIN in the other photo.  When the knuckle is completely closed, the PIVOT PIN appears to be unloaded, or loose, as others have said in this thread.

There are similar ridges on the bottom of the knuckle, so there are a total of four FORCE BEARING RIDGES.  

The "LOCK" is inside the coupler body.  You can't see it until you get down and look inside.  It is fairly large piece of steel, probably cast.  When the knuckle closes all the way, the LOCK falls down beside the inner end of the knuckle and prevents the knuckle from opening.

The "PIN" is connected to the LOCK inside the coupler body and to the lifter bar on the end of the car with a couple of chain links.  The casting of the coupler body had a square hole in the bottom directly beneath the PIN where a PIN can be installed to operate from underneath the coupler body.  When one pulls up on the PIN, the LOCK raises up so that the inner end of the knuckle is free to move, allowing the coupler to open.

Now to the forces.  I believe from what I saw that much of the force is on the FORCE BEARING RIDGES and their mating surfaces.  Significant force exists between the inner end of the knuckle and the LOCK due to the angle of the knuckle in its closed position.  There probably is some load force between the opposite coupler's knuckle and the side of the coupler body opposite the knuckle, again due to the angle of the knuckle.  These are not simple shapes, so I won't try to do an actual vector analysis.

Finally, I repeat, the PIVOT PIN or knuckle pin appears to be unloaded when the coupler is closed.

Having examined the coupler closely, I understand why most people talk about the PIN.  Really, that's all you can see.  Until I grabbed onto both the PIN and the LOCK and moved them, I could not verify that they were two separate parts. 

I hope this helps. 

 

[nbrodar]

This thread reminded me of an interesting discussion/argument that was going on in the MR forum a while back:

www.trains.com/trccs/forums/1053198/ShowPost.aspx

I wondered if anyone would bring up the unforunate thread.

As previously stated, the draft forces are transmiitted through the knuckles, via the Lock and the mating surfaces inside coupler. 

The pivot pin has nothing to do with transmitting the force.  Yes, the knuckles will still transmit the force, if the pivot pin is missing.  I've had many knuckles fall out as I cut cars apart.

I've seen alot of those plastic pivot pins, mostly broken.  They seem to wear much faster then the steel ones.

Nick

[Semper Vaporo]

Cordon -- THANK YOU!! THANK YOU!! THANK YOU!! for your WONDERFUL explanation of the "FORCE BEARING RIDGES".  Now it makes sense that the "PIVOT PIN" ("Knuckle Pin") can fall out and yet the coupler does not come apart.

Thank you also for using alternate terms to describe the parts of the coupler.  Knowing the "CORRECT" terms is one thing (and I certainly appreciate knowing the correct terms) but one has to have some point of reference to know which parts have which CORRECT terms and your drawing and alternate terms sure helped to explain it all.

I assume those "ridges" are similar to the ridges in a pair of "Channel Lock" (tm) pliers.  I can see now how the weight of the train is NOT attempting to cut that skinny little hinge pin in two between the coupler body and the knuckle (acting like a pair of scissors).  I just KNEW that little pin could not withstand the forces that must be involved, but I never could figure it out from the patent drawings.

I am not completely sure I understand how the yardman pulling on the lever on the side of the car causes the "Lock Lifter" to cause the "Lock" to move out of the way to allow the knuckle to open, but I can see why there was confusion over the use of the term "pulling the pin".  The 'Lock Lifter" sure "LOOKS" like a "PIN".

A part of the confusion is also probably because my (and I suspect most others also) only hands on view of a coupler is with toy trains that either have Kadee (tm) type couplers that do not have pivoting knuckles (the whole coupler shifts sideways in relation to the opposing coupler), or the working (pivoting knuckle) couplers of the Lionel (tm) type.

In latter type there is a "PIN" that apparently is a combination of the "Lock" and "Lock Lifter" of a real coupler.  If this pin is fully inside the coupler body it is in the way of the part of the knuckle that is inside the coupler body and it thus cannot pivot on the "Knuckle Pin".  Pull this pin out of the body and the knuckle can pivot and the coupler opens.  Because the toy couplers don't have the "force bearing ridges" in the knuckle and the mirror image ridges in the body of the coupler, the force pulling on the knuckle IS a "shear" force on the hinge pin ("Knuckle Pin"); the knuckle and the body of the coupler acting like a pair of scissors trying to cut at the top and the bottom.  I have seen toy couplers where some extraordinary force was applied (the brat next door yelled "Watch this!" and actually KICKED the gondola) and sheared the pin!

Thanks again!

 

[cordon]

I took a photo, but I was in a rush earlier and forgot it.  My apologies. 

The camera is about a foot from the knuckle and looking directly into the innards of the coupler.  I have pushed the knuckle almost to the closed position. 

You can see the ridge farthest from the knuckle pin, and you can clearly see the LOCK.  In the pic it is just on the edge of the knuckle extension.  When I pushed the knuckle a bit more, the LOCK fell down.

The LOCK has a complex shape; it doesn't have a plain rectangular cross section as I showed in my simplified sketch. 

There are variations on couplers.  Some have metal guards above, or below, or both to prevent the mated couplers from sliding out from each other vertically.  AMTRAK couplers have a triangular extension that fits into a triangular notch on the mating coupler to create a rather well-locked assembly.  That's probably why AMTRAK trains seem to stay hooked together in a crash better than freight trains do.

 

[ericsp]

Hey===Hey--- has no one wondered about the weight???? and where do you just happen to find a knuckle coupler???? where do you people live???? in a railway wrecking yard??? lets have some answers instead of more questions.

I have seen a broken coupler sitting next to tracks. It must have broken while the train was out on the road and the railroad did not care about hurrying to pick it up.

[Anonymous]

I will try to answer the question about forces.  

I welcome your effort to diagram and explain this, but it does raise some questions in my mind.  I do not believe the features you label as "force bearing ridges" are actually load carrying features.  They are simply size reductions on the knuckle mass.  Those reductions are correspondingly filled in with mass size increases of the coupler body.  These mass changes are done is steps, and are curved to match the swing of the knuckle pivot. 

Moreover, in your diagram, I think you have the features you label as force bearing ridges reversed. In actuality, the knuckle mass steps down as it approaches the lock-engaging portion, whereas your diagram suggests that the mass is stepping up to a wider mass.  So, in the true configuration, they would not be able to carry load.  They would simply pull apart as they do if you were to release the lock on a closed knuckle with the knuckle pin missing. 

If I am not mistaken, your diagram shows the basic concept of the original Janney design where the lock only prevents the knuckle from pivoting.  It does not prevent it from pulling out in line with the lengthwise direction of the train.  With this design, the knuckle pin does indeed share the load as a shear force.  The load would be shared between the knuckle pin and the lock.  Look at it this way:  In your diagram, if the knuckle were closed with the lock engaged, and if you removed the knuckle pin; the knuckle could be pulled out even though the lock is still engaged (and considering that the "force bearing ridges" would not stop it as mentioned above).

What the original Janney design has evolved into with the modern coupler is a lock that prevents direct pullout in line with the train, as opposed to preventing knuckle pivot.  With this modern design, it is the engaged lock that prevents the knuckle from pulling out if the knuckle pin is missing.  It is not the features you label as "force bearing ridges". 

The mating faces of the knuckles have a sort of reverse curve or ogee form that creates a hook effect.  This hood effect transfers the tensile, pulling force through the coupler centerline, directly to the lock.  So when the lock is engaged, there is no force that is trying to cause the knuckle to pivot.  When you release the lock, the knuckle is pulled forward, engages the knuckle pin, and pivots on it.

When couplers are mated, the force is transferred from the mating knuckle faces directly to the locks, and the locks transfer the force directly to the coupler body.  These forces are transferred on the train centerline as tension, and they do not induce any pivot force on the knuckles.  

[Modelcar]

Cordon....Sorry I seem to be so thick headed about this but which way is the camera facing.....Are you looking down vertically or into it's end....Just not sure what view we are looking at....

[tangerine-jack]

Now this is getting interesting.  Thank you cordon and Bucyrus, you both give me much to work on.  Bucyrus, where did you get your information?  Is it from a tech journal or from your own research?  I would like to study it in greater detail.  I think I am beginning to understand now the forces acting upon a coupler, but I want to be sure what I know is correct.

[Anonymous]

Now this is getting interesting.  Thank you cordon and Bucyrus, you both give me much to work on.  Bucyrus, where did you get your information?  Is it from a tech journal or from your own research?  I would like to study it in greater detail.  I think I am beginning to understand now the forces acting upon a coupler, but I want to be sure what I know is correct.

I am not an expert on all facets of coupler design, and if what I said is incorrect, I would welcome any criticism or correction.  I do not have a technical publication that details the operation of the modern coupler, however, I do have diagrams of the initial Janney design.  Prior to this thread, I had not considered the function of the knuckle pin.  I simply took it for granted (falsely) that the knuckle pin was loaded by the pull through the coupler, as is the case with the original Janney coupler. 

So my explanation is simply based on the logical conclusions that must follow from the fact that the knuckle pin bears no load of the locomotive tractive force. 

With the original Janney coupler as well as several imitators, the tractive force was transmitted to the knuckle and tried to swing the knuckle open, by pivoting it on the knuckle pivot pin.  So the knuckle amounted to a simple lever with the knuckle engagement head at one end, a kind of tailpiece or "tang" at the other end, and a pin passing through the middle.  When the knuckle was closed, a locking pin prevented the knuckle tang from swinging in rotation about the knuckle pin, and thus prevented the knuckle from opening. 

So with the original Janney, the pulling load transferred through the pivotal knuckle to the locking pin, which prevented the pivot.  And in reaction to that prevention of pivot by the locking pin, the knuckle transmitted force back to the knuckle pin.  So both the locking pin and the knuckle pivot pin shared in a force that was applied in shear to each pin.  This explanation is clearly indicated in the diagrams of the original Janney design.

Somewhere along the way to today's coupler, there was a sea change in this knuckle functionality.  Apparently the locking pin does not engage the knuckle tang in a way that prevents the knuckle from pivoting, but rather, in a way that prevents the knuckle from being pulled directly out of the coupler body in a straight line that matches the train centerline.

Picture the letter "C".  If you connect a load to one end of the "C", and pull on the other end, the "C" will simply transfer the load (in a north/south direction in relation to the letter "C") with no inducement to rotate.  This is the modern coupler.  The locking pin simply connects the load to one end of the "C", and the "C" of the mating coupler pulls on the other end. 

But the "C" sill has to pivot in order to open for connection to another "C".  So add a pivot pin through the line of the "C" half way from one end to the other.  With the locking pin locked, the force runs through the "C" in a straight line.  Thus it places no load on the pivot pin providing that the pin fits loose enough to not get involved in transferring any of the straight line pull. 

Now when you pull the locking pin, the "C" is pulled forward taking up the slop in the fit of the pivot pin.  When this slop is taken up, the "C" then pivots to open.  So the functionality of the knuckle changes from a straight-line connector to a pivotal member according to the engagement or disengagement of the locking pin.

With the original Janney design, there is this same "C" and midway pivot pin.  However, instead of the locking pin resisting a straight line pull of the "C", it resists the rotation of the "C" on the pivot pin.  So the fundamental difference between the early Janney design and today's coupler is the relationship of the locking pin to the knuckle tang.

[cordon]

Thanks for the feedback.  As I said, the coupler I investigated and photographed is on a historical caboose.  I have not yet had the opportunity to investigate a modern coupler because it would involve trespassing, which I have promised BNSF I will not do (as part of the BNSF Citizens for Rail Security program).  Maybe I can get permission to do that in a yard nearby today.  It's quite possible that modern couplers are different.

Back to the one I looked at.  First, the photo is looking from just beyond the end of the coupler along a line nearly parallel to the drawbar towards the middle of the caboose. I moved slightly to the right to get a view around the end of the knuckle. 

I couldn't put a lot of force on the coupler compared to the forces it gets in use; but, when I pulled as hard as possible after getting the LOCK to drop into the locking position, the mating FORCE BEARING RIDGES were in close contact and the PIVOT PIN was loose.  The LOCK in this case has no hook on the inner end of the knuckle.  It firmly prevents the inner end of the knuckle from moving to the right (opening the knuckle), but it appeared that the knuckle would come straight out if both the PIVOT PIN and the FORCE BEARING RIDGES were not there.  There were a few small rocks inside the coupler that prevented the LOCK from fully descending.  I could not reach them to remove them.  It's possible that they interfered with the operation.

It's also possible that the PIVOT PIN was loose due to wear and tear.

None of this information comes from reading.  I am describing only what I personally saw, touched, and photographed.

I must admit that I never looked this closely at couplers in all my years of railfanning, although I have opened and closed them and operated the lever on the car.  I used to think that the PIN and the LOCK were a single piece and that the PIVOT PIN carried the load.  That proves you're never too old to learn. 

 

 

[cordon]

Today's field exercise.  Found a boxcar, CP230388, in a warehouse in Carrollton, TX, where I could walk up and handle the coupler and take close-up pictures.

This photo shows three of the FORCE BEARING RIDGES on this coupler.  It looks as if there are marks on the knuckle ridges from contact with the coupler body.

This photo shows the lower FORCE BEARING RIDGE and the LOCK partly engaged.  These parts look just like those from the caboose that I posted earlier.

I walked into the Dallas, Garland, and Northeastern (DGNO) repair yard in Carrollton and asked if I could take closeup pictures of a coupler.  Unfortunately, the answer was a polite "No," but on the way in and out I observed two or three units.  One was disassembled.  The parts looked just like what I've described here and in my earlier posts.

This is a telephoto showing a coupler at the yard without the knuckle.  There appears to be a knuckle, or a broken part of one, on the ground to the left.  

Finally, this is the coupler on a work train dropping off long rail sections on the DGNO line south of Carrollton.

These sightings and photos are all consistent with my earlier description.  I did not see any coupler knuckles with a C-shaped tongue that would engage the LOCK, as Bucyrus described.

This has been a fascinating exercise, and I certainly will take a closer look at couplers at every opportunity in the future.  If I see anything different, I will post it.

Check out the video on the thread "Tagging Along on an NS Local."  There are couplers in action.  At one point the conductor opens the knuckle on the engine by pulling up hard on the coupler lever. 

   

 

 

[Anonymous]

cordon,

Thanks for posting the photos.  Now I see your point about the force bearing ridges, and I think it is accurate.  Although, in your diagram, you callout two sets with one set close to the knuckle pin.  I don't see that set.  I thought you were referring to that first point that the knuckle steps down from its full height.  But I think your observations are correct for the ridges on the knuckle "tang" (my word).  Are these mating ridge features on the top and bottom?  I can clearly see them on the top. 

The "C" shape I was referring to was meant to be in a schematic sense, not the actual shape.  I meant that the general structual shape of the knuckle when view from the top of the coupler is the shape of a "C". 

I am still a bit confused as to how the knuckle pin remains unloaded during the tractive pull.  I can see that the locking pin does prevent the knuckle from pivoting, contrary to what I said in the earlier post.  I know that the locking pin is impossible to pull when the coupler is under load, so I must assume that the locking pin is loaded against rotation when the tractive effort is applied.  I think I am going to have to make a field trip.

[Modelcar]

cordon:  Good coupler photographs.  Thanks for your effort and sharing.

I believe I'm going to need walk up to one to eye ball it in it's reality to see if I can see where the actual force travels through these parts....Just can't see it in photos...Now to find a safe place to do this...Sure must be one around here with all the cuts of cars I see on side / passing tracks.

[Anonymous]

Prior to this thread, I had not considered the function of the knuckle pin.  I simply took it for granted (falsely) that the knuckle pin was loaded by the pull through the coupler, as is the case with the original Janney coupler. 

With the original Janney coupler as well as several imitators, the tractive force was transmitted to the knuckle and tried to swing the knuckle open, by pivoting it on the knuckle pivot pin. 

Somewhere along the way to today's coupler, there was a sea change in this knuckle functionality.  Apparently the locking pin does not engage the knuckle tang in a way that prevents the knuckle from pivoting, but rather, in a way that prevents the knuckle from being pulled directly out of the coupler body in a straight line that matches the train centerline.

So the fundamental difference between the early Janney design and today's coupler is the relationship of the locking pin to the knuckle tang.

Well I have taken a close look and I must retract much of the above which is extracted from my post yesterday.  There has been no fundamental change in the relationship of the locking pin to the knuckle tang between today's design and the original Janney design.  In both cases, the locking pin only prevents the knuckle from pivoting.  In the modern design the locking pin does not prevent the knuckle from pulling out in line with the direction of tractive force.

The force bearing ridges that cordon calls out in his sketch are indeed located on the coupler per the sketch.  There is one set on the topside and bottomside of the knuckle tang, and another set top and bottom very near the knuckle pin.  The mating surfaces of these sets of ridges are shiny, indicating that they do bear a load.  There is also a set of force bearing ridges top and bottom of the knuckle about midway, and these ridges bear a load when the coupler is in compression.  They are also shiny.  There are also shiny and quite distressed surfaces on the coupler tang where it swings against the locking pin.

And this is the part I found most interesting:  When the knuckle is closed with the locking pin locked, and the knuckle is pulled, it does transfer a reaction force to the knuckle pin.  So the knuckle pin does indeed bear a shear force when the coupler is transmitting the pulling force of the locomotive.   So this reverses my conclusion in the first paragraph above in that what I had taken for granted originally is correct. 

Therefore, when the coupler is under the loading of pulling the train, it shares the force with the following components:

1)   A set of force bearing ridges on the top of the knuckle tang.

2)   A set of force bearing ridges on the bottom of the knuckle tang.

3)   A set of force bearing ridges on the top of the knuckle clasp body.

4)   A set of force bearing ridges on the bottom of the knuckle clasp body.

5)   The mating surfaces of the knuckle tang and the locking pin.

6)   The hinge features of the knuckle, the coupler body, and the knuckle pivot pin.

Items #1-4 prevent direct pullout of the knuckle in line with the direction of the train.

Items #5&6 prevent the knuckle from pivoting.

There are still a couple of points to ponder.  One is the question of how so many load carrying features can all carry the load simultaneously.  Maybe they don't have to.  Maybe the load shifts between the various force bearing ridges as the coupler parts move around.  I would think the load could not possibly be equalized through all of these solid mechanical stops.  The other interesting point is that, while the knuckle pivot pin does share the load, I believe a pair of couplers would still carry the load if mated together with the knuckle pivot pins missing.

 

 

[tangerine-jack]

Wow, this must be the greatest thread ever known to the history of mankind!  I feel my few brain cells expanding and multiplying, awesome!

 I would suspect, based on the photo evidence, sketches and bucyrus' logic tree that the force bearings share the load on a dynamic basis, that is to say as the coupler moves around as the train is in motion, the force of the draw is shifted to any given bearing as the need arises, with each bearing capable of handling the load individually.  The knuckle pin must share a portion of the load at one point or another, the fact that it falls out easily when not under tension is perhaps coincidence and leads to a false conclusion that it does not play a roll in keeping the train together.

I will be going to the Norfolk Southern corporate office tomorrow for a meeting; maybe I can lay my hands on some tech info or pick somebody's brain.  If I find out anything important I'll be sure to post it.

[Modelcar]

 

....Shifting the forces "around" in the coupler assembly sure seems logical with an assembly such as it is, etc....I {really know}, for ME to "see" just how that is done I will have to eye ball an assembly in it's reality....Now if I can just find the "place" to do it...Safely.  Plenty of cars sitting on sidings around the Muncie area but I'm aware of the "don'ts, etc....So we'll see.....

[Anonymous]

I have one other lingering question.  Someone mentioned that some knuckle pivot pins are made out of plastic.  I have never seen one.  From my observations on Saturday, I assume that these knuckle pivot pins bear considerable shear force when the coupler is under load.  I can imagine that the solid steel pins could bear such a load, but I would be suprised to learn that a plastic pin could provide the same shear strength as a steel pin. 

So I would like to know more about the reason for using a plastic pin and what its strength is compared to the steel pins.  It would also be interesting to learn the exact composition of the plastic pin as well as its cost compared to a steel pin.

[nbrodar]

Again...

The pivot pin bears NO force when the knuckle is closed.  The knuckle continues to preform it's job even if the pin is missing.  Much to the dismay of trainman, when they open the knuckle and it falls out.  Be there. Done that.  Have the T-Shirt.

Even when the knuckle is open the pin bears little force.  Take a look at the pic of the coupler without the knuckle.  Most of the wieght of the open knuckle is bourn by the "yoke" that the knuckle pivots in.

Plastic pins are cheaper, but don't wear as well as the steel ones.  Most of the plastic pins I've seen are broken.

Nick

[Anonymous]

Again...

The pivot pin bears NO force when the knuckle is closed.  The knuckle continues to preform it's job even if the pin is missing.  Much to the dismay of trainman, when they open the knuckle and it falls out.  Be there. Done that.  Have the T-Shirt.

Even when the knuckle is open the pin bears little force.  Take a look at the pic of the coupler without the knuckle.  Most of the wieght of the open knuckle is bourn by the "yoke" that the knuckle pivots in.

Plastic pins are cheaper, but don't wear as well as the steel ones.  Most of the plastic pins I've seen are broken.

Nick

Nick,

I understand your point about the knuckle falling out when you pull the locking pin and the knuckle pin is missing.  And I hear what you are saying about the knuckle pin not bearing any load while the coupler is under pull.  However, I cannot see how the second point is possible.  If you pull on one end of the knuckle that is hinged by a pin about midway, and the opposite end of the knuckle is being prevented from rotating by the locking pin, how can there not be a reaction force transmitted back to the hinge pin?

On one that I looked at, the knuckle was closed and locked, and the knuckle pin was loose with a very sloppy fit.  When I pulled on the locked knuckle, it came forward while rotating to the left toward the knuckle pin.  This lose action of the locked knuckle  bottomed out on the knuckle pin, thus making the pin tight with no slop.  Pulling on the knuckle obviously put pressure on the knuckle pin.

I can see how it might be possible for the coupler to be locked and loaded with the knuckle pin missing, without the knuckle pulling out.  In the case of a pull-loaded, locked knuckle with the pivot pin missing, I would think that the knuckle rotation tendency might be stopped by either the wall of its coupler body behind the knuckle, or by the static face (or "thumb") of the mating coupler.  The knuckle would simply rotate a little further than it would if the pin were in place. 

But if the pin is in place, I cannot see how it could escape being loaded in shear when the coupler is pulling a load.  The patent on the plastic pins says their functional benefit is that they bend under load, so they don't break as easy as steel pins.  

 

 

 

[cordon]

Hello, again,

I've been back to the caboose (in Frisco, TX).  Plus, I took a look at the coupler on a brand new MWAX hopper.  As far as I can see, the couplers on the caboose look just like the one on the hopper, except the ones on the caboose do not have the "platform" that prevents the knuckle on the facing coupler from sliding out downwards when they are coupled together.  Therefore, I have decided to disassemble one of the couplers on the caboose to get better pictures.  Don't worry, it's on city property, not railroad property.

I'm not going to try to lift that thing myself, so that job will have to wait until I can get some help. 

I also verified that the coupler has something inside that pushes the knuckle open when one lifts the lever vigorously.  We saw that happen in the video of the NS ridealong.  One of the contenders on that "other" thread called it a "knuckle thrower."  If I can get the knuckle out, I'll be able to see how that works. 

Meanwhile, I've marked up two of my pictures to identify the mating surfaces.

 

 

Looking at both pictures, I believe that, when the coupler is closed, ridge #1 gets trapped behind boss (for lack of a better word) #1A, ridge #2 gets stuck behind boss #2A, ridge #3 gets stuck behind boss #3A, and ridge #4, which you can't see in one of the pictures, gets stuck behind boss #4A. 

I think it's the 1-1A and 4-4A combinations that back up the knuckle pin with heavy metal while the knuckle is fully closed.

Interestingly enough, the historic locomotive here has an older coupler design.  It has no ridges or bosses.  The knuckle extension hooks around a piece of metal inside.  I think in this one the knuckle pin takes at least side-to-side forces.

   

 

[AnthonyV]

Great discussion guys.

We can get a sense as to how much force the pin can transmit.  It appears that the pin is loaded in double shear, what makes it a pretty straightforward calculation.  What is the diameter of a typical pin, what type of steel is used, and what is the rated force for a coupler?

For example, assuming an ultimate shear stress of the steel of 100,000 psi, a 1" diameter pin could transmit about 150,000 lb before shearing.  However, working stresses in practice are much lower due to fatique considerations, impact loads, etc.  As a result, actual working stresses are propably in the range of 25,000 psi, which would reduce the allowable load on the pin to only around 40,000 lbs.

This is much lower than the maximum coupler force, which is at least 250,000 lbs.

The result of this simple analysis suggests that the coupler load is not transmitted through the pin.

Thanks

Anthony V.

 

[Rodney Beck]

the knuckles are rated for 390,000 foot pounds it all depends on how much they are cracked.

 

Rodney

[Anonymous]

Great discussion guys.

We can get a sense as to how much force the pin can transmit.  It appears that the pin is loaded in double shear, what makes it a pretty straightforward calculation.  What is the diameter of a typical pin, what type of steel is used, and what is the rated force for a coupler?

For example, assuming an ultimate shear stress of the steel of 100,000 psi, a 1" diameter pin could transmit about 150,000 lb before shearing.  However, working stresses in practice are much lower due to fatique considerations, impact loads, etc.  As a result, actual working stresses are propably in the range of 25,000 psi, which would reduce the allowable load on the pin to only around 40,000 lbs.

This is much lower than the maximum coupler force, which is at least 250,000 lbs.

The result of this simple analysis suggests that the coupler load is not transmitted through the pin.

Thanks

Anthony V.

 

AnthonyV,

Cordon has done a great job of photographing and diagramming the coupler, but it seems that the one unresolved question is whether or not the knuckle pivot pin bears any force when the coupler is mated, locked, and pulling a train.  In a few posts up, Nick explains that the pin does not bear a load, and refers to an alternate type of pivot pin made from plastic.  I reply that I looked at a coupler where the pivot pin does appear to bear a load when the closed, locked knuckle is pulled, as it would be when pulling a train.  I am not refuting Nick's explanation, but I cannot reconcile it with what I observed.

But in the middle of last night, something occurred to me.  The four sets of force bearing ridges that cordon calls out in the photos follow a circular pattern as segments of an arc that is concentric to the pivot centerline of the knuckle.  The top and bottom sets of ridges that are very close to the knuckle pivot pin are performing the same function as the pivot pin.  Each set consists of a circular boss rotating in a circular bore. 

There is one difference between the mated circular features near that pivot pin, and the combination of the pivot pin passing through the bores of the knuckle and the coupler body.  That difference is that, in the mated circular features, the circular bore is not a complete circle.  It is a segment of an arc.  It cannot be a complete circle because there would be no way to separate the knuckle from the coupler housing if it were.  There would also be no way to assemble it. 

When viewing a coupler from the knuckle end, the train pulling force on a closed and locked knuckle would tend to rotate the knuckle to bear against the right side of the pivot pin.  In other words, the outer end or engaging mass of the knuckle would move from right to left.  From one of the photos showing the knuckle, it appears that the arc segment of the circular bore is positioned to bear the direction of this right-to-left loading.  So if the knuckle pin had a lose enough fit through its bore in the knuckle, the mating features of the circular boss and arc segment bore would bear all of the knuckle reaction side force, leaving the pivot pin completely free of any shear load. 

However, as the mating features of the circular boss and arc segment bore wear, the pivot pin will begin to pick up some of the knuckle reaction side force.  I speculate that the coupler I inspected had suffered sufficient wear to cause this loading of the pin, which I did observe.  I also speculate that this is what causes pivot pin breakage of both steel pins, and especially plastic pins.

This also explains the rationale of the patent for the plastic pins, which is that they will bend without breaking, as opposed to steel pins that break without bending.  If the pin is not intended to bear a load as the coupler is designed, but begins to bear a load as the coupler wears, it happens as a forcing surface gradually approaches the side of the pin during the wear period.  So once that forcing surface begins to load the pin, if the pin could move further away from the forcing surface even slightly, it would completely relieve itself from the load. 

Therefore, a plastic pin could bend to accommodate this gradual encroachment of the forcing surface, and avoid loading, whereas a steel pin will not bend, thus quickly assuming the full load of the encroaching forcing surface.  Apparently the plastic pins are often forced to bend to a limit where they break.  And apparently, the steel pins are strong enough to carry the load of the forcing surface as it encroaches during the wear cycle, at least during the early development of that loading. 

Apparently the steel pins are not strong enough to carry this shear load during repeated cycles, once it has developed to its maximum force.  When I say that the steel pins would not bend, I only mean as a comparison to the plastic pins.  Like the plastic pins, the steel pins would bend to graduate their assumption of the shear load, but just not as much as the plastic pins would bend.  In other words, the steel pins would assume far more load than the plastic pins for any amount of bend applied equally to both types of pins. 

[nbrodar]

We've got a great discussion going...

Looking at Cordon's pictures, most of the force on the front part of the knuckle is borne by the ridge labeled 4A.   There is a corresponding depression in the knuckle.   While I don't have the physics to back up my statement, that the pivot pins bears no force, I do have experience.

1.) The knuckle continues to transmit the force if the pin is missing.
2.) It is possible to remove the pin when the knuckles are in tension.

However, Bucyrus makes some valid points also.   The vast majority of broken steel pins I've seen, have broken in the middle.   Not at the head, which is where I would expect if failure was simply rotational wear.   This would lead on to believe that the pin may bear some side to side force as the couplers slew back and forth.

Most of the failed plastic pins I've seen, have broken at the head. 

Nick

[Anonymous]

Thanks Nick,

I know of a C&NW caboose on display near here that I will take a look at.

[Anonymous]

Well I took a close look at both couplers on C&NW caboose #11142.  Both couplers had knuckle pivot pins with no cotter pin retainer to retain them against upward pullout.  The following observations apply to both couplers:

With the knuckle closed and locked, it was lose enough that I could wiggle it around and find a position where I could freely lift the pivot pin.  If I then lifted the pivot pin six inches and let it go, it would drop all the way back down until its head bottomed out.

If I repeated lifting and letting go of the pivot pin, and pulled out on the closed and locked knuckle as the pivot pin was falling, the knuckle pin bore would grab the pin and stop it in mid drop.  If I let the pin drop back down to normal position, and pulled out on the knuckle, I could see that it made contact with the pivot pin as that pin moved slightly to the left.  And when the pin moved in response to pulling out on the closed and locked knuckle, I could not move the pivot pin.

Then I pulled the pivot pin completely out and pulled out on the closed and locked knuckle.  The knuckle shifted slightly to the left and stopped as the force bearing ridges near the knuckle pin came into contact.  It appears as though those ridges would carry the load and keep the knuckle in its closed and locked position with the pivot pin missing.

When pulling out on the closed and locked knuckle and looking down the open bore for the pivot pin with that pin removed, I saw that the bore of the knuckle and the upper hinge bore of the coupler housing were not concentric.  The two bores were misaligned with the knuckle bore about 1/8" too far to the left.  The pivot pin could not be passed down through these misaligned bores.

From these observations, in the case of both of these couplers, I conclude the following:

1)   When a pulling load is applied to the closed and locked knuckle, it shifts to the left and bears against the pivot pin, transferring a substantial portion of the pulling force to the pivot pin as a shear load. 

2)   If the pivot pin were missing, under the same load pulling circumstances, the knuckle would remain in the proper position, shift to the left, and transfer its loading against the force bearing ridges near the knuckle pin bores.

[cordon]

Hello, Folks,

Lots of info here: http://www.mcconway.com/

Some good drawings there, including the knuckle thrower.  I haven't figured out yet how that works. 

McConway & Torley are located in Pittsburgh, PA.  According to the web site, the Director of Engineering is Joe Gagliardino ([e-mail]).  I suggest an e-mail to him, or even a visit to the plant if one of you lives close enough, to get authoritative information on the forces in a coupler.

I also suggest that only one of us send an e-mail and then share the info with the rest to avoid sending multiple identical requests.

Any volunteers? 

 

[Anonymous]

cordon,

I would be willing to contact him or you could.  I would put a little thought into the question before posing it to him to get it as simple as possible in order to set the stage for the answer which may not be so simple.  There are some questions about how the knuckle thower works.  And then there is the main question about the force transfer within the assembly.  Obviously the locking pin or "lock" bears a load by preventing the knuckle from pivoting while transmitting the train pulling force.

Beyond that, the question concerns the load bearing functions of the four sets of "force bearing ridges" and the knuckle pivot pin.  It would also be helpful to learn the proper terminolgy of the "force bearing ridges" and other related features of the assembly if they have specific nomenclature.