unintended uncoupling and riveting

A weak knuckle spring would not cause unintentional coupler opening; it would cause the knuckle not to open when the uncouple mechanism is activated. What needs replacing is the spring that holds the armature assembly.

The Brakeman's Riveter is recommended for many light riveting repairs.  A press would be needed for heavy duty repairs.

I doubt the Harbor Freight press would have the proper tools for our needs. The press itself may be useful, but the proper tools would have to be purchased in addition.

Here are the proper presses and tools for repairing our trains.

Larry

I am with TrainLarry on the Brakeman’s Riveter. That is what I use for my truck work.  Here is a link to the Hobby Horse catalog.

http://www.hobbyhorseproducts....HorseCatalog2015.pdf

It explains what is needed.  I now buy most of my press tools from Jeff Kane, The Train Tender.   He sells Hobby Horse tools and makes a few of his own.   If you go to a bench press, the arbor press (Harbor Freight) and the Riveting Press ( Hobby Horse and Chicago Rivet) are different designs,  The arbor press drives the ram with a gear and has a constant mechanical advantage. The rivet press drives the ram with a cam, which allows for an increase in mechanical advantage right at the end of the stroke when the rivet is being headed. 

There are two major different kinds of couplers, one with a magnetic coil around the shank and one that uses a magnet between the rails to uncouple.  On both there are two springs. One to keep the coupler closed and one to open it. On the coil coupler there is a pin inside the shank that the magnetic pulls back to uncouple. The pin is held forward with a spring to lock the knuckle. That spring can be a problem and replacing it may solve your problem.  To get the locking pin out, you have to remove the knuckle rivet, which will allow the knuckle and knuckle spring to come out. Then the locking pin can be removed and behind that is the spring that keeps the coupler closed. When reassembling all this I would recommend replacing the rivet and both springs.

The other style has a pin that comes up from under the coupler. It has a flat spring that holds the pin up in place locking the knuckle shut. There are also the thumbtacks style couplers that use a coil spring in this application. These flat springs can be a problem and can be replaced by removing a rivet under the truck.  The knuckle does not need to come out to replace this spring, but the truck has to come all apart. The flat spring can be replaced with two flat springs, which will reduce the chance the coupler will uncouple when it is not supposed to, but will require more force to get it to uncouple. Another cause of this style of truck having under desired uncoupling is the vertical lock pin is loose and does not stay vertical.  There is also a press tool for retightening these vertical pins.  All these tools and parts are available from Jeff Kane. 

I need to redefine my uncoupling problem and then ask for additional comment. I have found that my uncoupling issues stem from a momentary vertical mis-alignment of the two coupled cars. I don't mean that one coupler lifts out of the other (but staying closed) but rather that somehow this momentary vertical displacement actually releases (opens) one of the couplers.  Because it is repeatable and i can make it happen at specific places in the layout, I was able to determine that shims were needed below the tracks and /or trestles to keep it from happening, particularly at points where the slope of a grade changes.  So I corrected that issue to reduce the incidence of uncoupling in MOST cases.  But I still have a problem with 1 or 2 MTH Lo-V subway cars that use the pin type couplers. I do not understand the exact mechanism that triggers the accidental release (opening) of these couplers.   Manually manipulating the coupled cars does not cause them to uncouple and the coupling seems very secure.  But something happens while actually running as these move up and down even SLIGHTLY relative to each other that then triggers the release. Surely I am no the only one who has seen this.    I have a theory that the up/down movement of the couplers relative to each other forces one car's coupler to downwardly trip the finger tab release of the other, but I don't fully understand HOW.   Need some comment on this. For now , I just wire the tabs closed.

Gary,  interesting.  My son was telling me that he's seeing the same issue with the unintended uncoupling.  I won't be able to experiment until i get back down to the layout in GA in Aug.  However, this is what i believe he's seeing:  Consist - aprox 20 cars.  Occurring at diff positions in the consist.  Frequently on hills. Not sure if it's at the apex, but think it's troughtout the grade.  Grades 2% and 2.5%.  Yes, up & down movement between couplers.  Seems to be occurring on cars where the uncoupling release tab (don't know the tech term but it's the tab you'd manually push down with your finger to uncouple the cars)  is plastic and the plastic is slightly deformed and is not holding the coupler closed when u get some vibration.  I've wanted to switch out the entire assembly (trucks w coupler) to all metal parts and then try to reproduce the problem within the same place in the consist and on the layout, but it will be months before i get back to the layout. Yes, i've looked at the rivets involved for switching out this plastic tab to a new one or a metal one and said " this will be ugly", so the riveting tool info here was interesting.  Hopefully, my son will jump in on here and elaborate on what he's seeing.     

I suggest taking a side view photo of where this is happening on your layout as a car passes over a trouble spot. Post the photo here. 

Sometimes the pin that holds the coupler closed "walks" down as you run until it releases.  It appears that when it's under tension it's fine, but during a momentary slack time from a track imperfection, it moves a little.  After a few of those, the coupler pops open.  It's not always that easy to figure out why that happens, or how to correct it at times.

Menard's cars had a unique issue in that the way the trucks and couplers were built, if the uncoupling arm got bent slightly internally, the coupler would uncouple when there was any slack that pushed against the coupler.  I posted a thread on fixing these when I got my first bunch of Menard's cars.

David Johnston, thank you for the link to the Hobby Horse catalog.  All of the couplers that I need to repair are the ones that use the flat springs and pin (no coil).  It is not clear to me what the minimum set of tools are that I would need to service these. Perhaps their HH-1007?  Do I need their HH-1024 as well?  Since I am not a "repair shop" I really don't want to invest hundreds of dollars in tooling.  But it appears that even if I elected to purchase a complete new truck (with coupler) I still could not avoid riveting the truck onto the chassis, so some minimal investment in tools is necessary. 

The basic riveting tools are DC 1008 to DC 1012.  I do not use the HH1024.  To remove knuckle rivets I use a small pair of side cutters and a 3/64 pin punch. The pin punch is a gun smithing tool.  The HH 1007 is the basic set of tools for the press if you are not going to do wheel pressing. If you are going to buy the press, the HH 1007 would set you up for 90% of the jobs you would ever run into. If you are going to change knuckle springs, and many need to be changed, you will also need good NON MAGNETIC tweeters to handle the tiny springs.  With the work you are planning you could use the Chicago Rivet (Lionel) press. It a takes about half as much space as the Hobby Horse press, which was designed to also work on standard gauge.  Both presses come up on eBay.  Both presses will need to be bolted down to a bench. The front edge needs to hang over the edge of the bench to allowing adjusting the lower tool.  When pressing the handle hangs over the front of the press when you are pushing hardest on it.  If the press is not bolted down it can flip off the front edge of the bench. I also have a desk lamp sitting right next to the press so I have enough light to see the work. 

Hobby Horse has a book (green cover), about 25 pages, on how to use the press.  They put out the same book for Lionel, which has a orange cover.  I found this book very helpful, especially when pressing wheels.  When setting rivets, the lower tool is adjusted up and down so the rivet is fully set just as the cam goes over center.  I find the easiest way to adjust this lower tool is to set the part to be riveted in the press before taking the old rivet out as it is at the correct finished length.   Many of the pre and post war rivets were nickel plated brass.  Today most rivets are nickel plated steel. The brass rivets are much easier to set. I am always looking for new old stock rivets. When selecting a rivet to use, I go through the rivets with a magnetic and select from the rivets left behind.  

The most difficult part of riveting is getting the old rivet out. I drill them if I can, but once the rivet starts to spin you have go to the Dremel grinding tool.  When grinding out a rivet, care has to be taken so not to get the rivet  so hot it melts any plastic around it.  Most small rivets can be removed with side cutters.  It only takes a couple of rivets to get the feel of riveting. Then it is easy. 

Gary,

As others have said, your uncoupling problem has nothing to do with the knuckle or the knuckle spring. Replacing them would serve no purpose in correcting your problem. Read John's and Tom SuperO's posts again.

It is the pin that enters the coupler from the bottom (connecting to the pull-down tab) that is the problem. It is weak springing action holding this pin in place that is the problem. The configuration of this springing mechanism is different for different style and different manufacturer's trucks.

The plastic ones are the most problematic, but the metal ones can also suffer from weak springing action. Sometimes the only remedy short of replacing trucks is wrapping a small wire around the coupler shank as you have done.

Some have used a small dental rubber band around the coupler shank (the type meant for dental braces). Sometimes this is enough to hold that pin in place to prevent uncoupling, but still have enough elasticity to allow for normal uncoupling.

Jim

Jim thank you for your observations.  While I have been dealing with this problem intermittently for a while, only in the past few weeks have I tried to investigate the actual mechanical cause.  What remains baffling is that in an "off the track" scenario , it is virtually impossible to manipulate the coupling to spontaneously open, and yet in the running configuration, it does!  If the spring were weak, I would expect it to cause the pin to sag, and possibly open, but observing it statically, the pin is fully pushed into the locking position.  One of my theories is that the end of the flat spring partially intrudes into the coupling gap. So if the two cars vertically mis-align due to track height anomalies, that protruding edge can get caught by a downward movement of the mating car's coupler.  I cannot make that happen when manipulating such a simulation by hand, but it's really the ONLY conclusion I can reach based on observation of a running train. I might try grinding down that protruding edge to see if that helps.  Mine are all metal, by the way, not plastic.  In the locations where I have this problem, the mating cars are rarely intentionally uncoupled, so yes, it might just make sense to use the rubber band or wiring to keep them closed, and not spend myself silly trying to "fix" it.

FWIW, I still say one likely scenario is the pin simply walking down and eventually uncoupling the car.  I have actually seen this running over a bumpy Fastrack loop, I could watch the coupler slowly work it's way open.

John, I have not discounted your observation, so I need to acknowledge it.  But mine tend to always occur at the same spot on the layout, whereas your observation would probably occur at more random locations.  So I have focused on vertical track alignment first and resolved the worst of that with shims.  Then I began to question why specific couplers were more troublesome than others. 

Gary, if it's walking up slowly and then you come to a bump in the track, that's a likely spot for it to decouple.  FWIW, my current test track is Fastrack just laying with no attachment to the surface, and the joints tend to be somewhat rough.  It's a great test for couplers as I've seen a few that pop going around the loop.

Gary, I agree. It's very difficult to re-produce "off the track".  Some interesting theories and observations guys.  I'm sure we'll "pin" this down soon.  Wish i was near the layout. Dave Johnston - thanks for the info on the tools. 

Went to the site:  Very cool.  So, approx,  for the HH107 tool kit ($145) , the HH 1,000 press ($285), and the Repair book ($40) we're looking at around $500 and change.   Hmm - dental rubber bands would be free if son pays for grandkids braces. 

Jeff Kane is fixing me up with a used 566-1 truck,  a tool to remove the truck pivot rivet, and a stud to fasten the replacement truck without reverting to riveting. All for about $37. Actual parts were about $5, but the rivet removal tool was $29.50 and of course I can reuse that next time.  This was for a 1958 era Lionel 6464 freight car. My other problem child is a set of 4 MTH Lo-V subway cars.  3 of them (motorized lead car excluded) exhibit spontaneous de-coupling on both ends, so I'm convinced MTH made all of those coil springs that keep the couplers locked too weak.  I just put rubber bands on all of them since the subway set is never uncoupled in normal operation.