Magnetic uncoupling options for Atlas Code 55

[bbussey]

Brian, I wanted to pick your brain a little further.  I have been experimenting with two flat magnets which are 3/4 by 3/4.  When installed this creates a magnet which is 1.5 inches by 3/4 inches with the 1.5 inch dimension being at 90 degrees to the track.  Cars uncouple fine.  The problem I continue to have is when I go to pull cars from the siding.  The first three cars come out fine at slow speed but the fourth car uncouples as it cross the magnet.  I have tried different combinations of cars but the problem remains.  I know that you stated you were using 1/2 by 1/2 inch magnets, and I wonder if this reduces the strength of the magnetic field sufficiently to alleviate this problem.  Can you tell me if you have experienced anything similar. 

Two other points of note:  If I add another layer of cork so that the track is now 3/16's above the magnet, the problem is resolved, ie the cars still uncouple when stopped on the magnet but don't uncouple when pulled across it.  This would strongly indicate that the magnet is too strong for what I am trying to achieve.   Second point is that I am using truck mounted micro trains couplers.  I notice that by the time I reach the fourth car in a string, the tension is gone and I am getting some backward and forward movement in the car.  This movement contributes to the cars ability to uncouple. 

I realize that a simple solution would be to lower the magnets into the benchwork, however, this is a retrofit scenario and lifting track and cork to router out holes 3/16 inches deep would be a pain in the $#@.  If I know that you are not encountering the same problem, then, I will move away from my local supplier and try the KJ Magnetics that you recommended.

Appreciate any info that you can provide.

That's why I went with the 1/8" thick magnets and made the footprint as small as possible along with keeping the functionality reliable.  You didn't say how thick your magnets are, but try to go as thin as you can.  That gives you some more leeway.

MTL couplers were a fantastic invention nearly 50 years ago, but that damn oscillating spring can be a nuisance.  The original-design 1023 has the best visible profile of all the MTL couplers, but the spring is behind the pivot post which causes these exact problems.  The truck-mounted couplers are designed similarly.  I strongly recommend installing body-mounted couplers using the 1015 or 1016 (or the 2001/2004 or 2019 for low-mount) as they have the spring in front of the post, and installing FVM or BLMA metal wheels in the MTL (or other manufacturer brands of plastic) trucks.  There is no oscillation pulling a freight consist when the springs are in front of the post, and having the axle points and truck frames being dis-similar materials generates less friction than the MTL wheels.  I also add one MTL axle retaining spring on all my cabooses, which would not apply in this scenario but helps overall.  Try these equipment modifications before ripping your magnets out, and I believe you will eliminate your unwanted uncoupling.

[shark_jj]

Brian, you are confirming what I expected that the problem lies in the couplers.  I am going to try another strength of magnetic to see if that offers a solution.  I have found a scientific supply company only 30 miles away from me (Indigo.com) that supplies Universities and Government Agencies and stocks rare earth magnets.  They are in Waterloo, Ontario, so I don't have to worry about Customs, etc,   I have ordered 12.5mm by 12.5mm by 1mm (1/2in by 1/2in by 1/16in).  I have also order 25mm by 12.5 mm by 2.5 mm (1inch by 1/2in by 1/8in).  Checking the specs the Field Strength (Gauss Rating) of my current magnets is 1151.  The 12.5 mm squares is 926 and the 1 in by 1/2 in is 1791.  Based on the ratings I am assuming that the large magnets will have the same problem, but since that size worked for you, I am willing to give it a try.  The 1/2 inch square magnet is weaker, so it will be interesting to see if it still uncouples the cars, but allows them to roll over top of it without uncoupling. 

I understand your point about body mounting the couplers, however, I am currently building a medium size layout 14X20 with 98 turnouts, so I have to be very careful in prioritizing tasks that need to be done if I am to make progress.  Putting body mounts on around 600 freight cars is down the list.

[bbussey]

Make sure the poles on those 1/16" magnets are on the long edges (and not the short edges or the faces) before buying.  But yes, that should weaken the field enough to prevent coupler oscillation from causing unwanted uncouplings.

All of your freight equipment is MTL?  Because all other manufacturers' equipment with operating knuckle couplers either uses a design without springs, or uses the MTL design with the spring in front of the pivot post.  So this only is an issue with MTL truck-mount-coupler equipment.

Also, on those 1/2 x 1/2 x 1/16 magnets (provided the poles are in the proper locations), you should be able to create a 4x4 arrangement for an overall 1 x 1 x 1/16, which increases the footprint to a more workable area.  MTL between-the-rails magnets are at least 2 inches long, which I think is excessive.   A 1" footprint is adequate.

[shark_jj]

good suggestions Brian, I hadn't really looked at other car types, I had just grabbed four available cars from the layout and they were MT's.  I experimented again with four Atlas Cars with metal wheels, and then a combination of Atlas and ESM X58's.  It was interesting to see the effect the magnet has on the metal wheels.  With four Atlas cars, the uncoupling interestingly takes place on the first car, and the last three pull through ok, but you definitely see the effect of the metal wheels.  By adding an ESM car to the front of the consist with 3 Atlas cars I don't encounter this uncoupling.  Not sure what all of this means, as it is a pretty small sampling and nothing has been done to the cars, they are right out of the box.  I expect to receive the new magnets in the next couple of days, and it will be interesting to see if a slightly weaker magnet makes a difference.  Once I receive the new magnets I'll test again and let you know what I find.

[mark.hinds]

good suggestions Brian, I hadn't really looked at other car types, I had just grabbed four available cars from the layout and they were MT's.  I experimented again with four Atlas Cars with metal wheels, and then a combination of Atlas and ESM X58's.  It was interesting to see the effect the magnet has on the metal wheels.  With four Atlas cars, the uncoupling interestingly takes place on the first car, and the last three pull through ok, but you definitely see the effect of the metal wheels.  By adding an ESM car to the front of the consist with 3 Atlas cars I don't encounter this uncoupling.  Not sure what all of this means, as it is a pretty small sampling and nothing has been done to the cars, they are right out of the box.  I expect to receive the new magnets in the next couple of days, and it will be interesting to see if a slightly weaker magnet makes a difference.  Once I receive the new magnets I'll test again and let you know what I find.

The automatic magnetic coupling problem has a lot of variables, and IMHO you would be well advised to eliminate some of these variables initially at least; in particular:

1)  Stick with MT couplers only, until you get things working.  Then if you must, experiment with adding additional types to your working setup, one at a time. 
2)  Avoid using cars with ferrous materials in them, as these will introduce disruptive lateral accelerations near uncoupling magnets.  For example, the new Fox Valley horizontal ribbed box cars have a steel weight, and the new Atlas PS-1s have steel in the wheelsets..  If you must use these cars, then replace the offending weights / wheelsets with non-ferrous equivalents.  For wheelsets, I have successfully used MT plastic, and Fox Valley metal (which are non-ferrous). 
3)  Adjust your equipment, one engine/car at a time, and test each piece on a test setup.  The test setup should be a section of track, with uncoupling magnet, at eye level, under bright lighting.  Then you can examine any malfunctions under magnification, and see exactly what is happening. 
4)  Note that the newer MT coupler designs (aka "RDA")  has/had a flaw which may inhibit reliable uncoupling.  It certainly does using their under-track #308 uncoupling magnet, which I use on my layout.  See my post on this issue in the following thread (starts with "As additional input..."):  https://www.therailwire.net/forum/index.php?topic=26118.msg264373#msg264373

Mark H. 

[nickelplate759]

Exactly the way you have it on the right, if you envision your red and blue to each be a 1" x ½" magnet.  The two magnets attract each other, and the N/S poles at the joint are negated.

I tried 1/2" wide, it wasn't reliable.  The 1" x 1" magnet works perfectly, especially given that the K&J magnets are far more powerful than the magnets MTL uses for their above-ties uncoupler.

I'm using them mostly for motive-power changes on New Haven passenger consists.  But I also prefer the shorter magnets for delayed uncoupling.  Two-inch-long magnets make it more difficult to manipulate shorter cars (such as 33' twin hoppers) without additional uncouplings between adjacent cars.

I'm trying this - with the same K&S magnets, and having a lot of trouble getting it to work well.  The magnets are under the ties, and further recessed by 1/16" of styrene.  I'm using it in a freight classification yard.

The magnets are strong enough to pull cars to them just on the basis of attracting the trip pins (no steel axles or screws).   Some cars uncouple every time they are pulled over the magnets, others have more trouble - it seems to be based on coupler style, but something else as well.  I can't get MT 1027 to uncouple hardly at all, for example.  It seems unrelated to car type.

Before I go further (like bury the magnets in ballast, I thought I'd ask for help.

[nkalanaga]

Part of the problem is that many of today's trucks roll much more freely than those 40+ years ago.  Back then, the drag from the axles in the sideframes would keep the car from being pulled into the magnet.  Today's cars can be pulled quite a distance.

They'll also find grades where the track was thought to be level!

[bbussey]

I'm trying this - with the same K&S magnets, and having a lot of trouble getting it to work well.  The magnets are under the ties, and further recessed by 1/16" of styrene.  I'm using it in a freight classification yard.

The magnets are strong enough to pull cars to them just on the basis of attracting the trip pins (no steel axles or screws).   Some cars uncouple every time they are pulled over the magnets, others have more trouble - it seems to be based on coupler style, but something else as well.  I can't get MT 1027 to uncouple hardly at all, for example.  It seems unrelated to car type.

Before I go further (like bury the magnets in ballast, I thought I'd ask for help.

I've modified my procedure since 2014.  I get more consistency using K&J Magnetics part# BX081 (1" x 1/2" x 1/16") and locating them adjacent to the track rather than directly underneath.  I place one magnet on each side of the track parallel to the rails, buried 1/8" down and covered with a 1" x 1/2" x .060" piece of styrene to make it level with the cork top.   I haven't had any issues with any style of coupler, KD/MTL or otherwise.

It's also easier to remove or relocate the magnets when they are adjacent to the track rather than underneath it.

And, as Mark suggested four years ago, swap out any steel wheels, screws or weights for brass or zinc equivalents.  The magnets are strong enough to stop cars, even buried in the cork off to the side. 
 

[Lemosteam]

It might be the orientation of the magnets.

See this https://www.therailwire.net/forum/index.php?topic=45101.0

[bbussey]

Also true.  When I get a batch of magnets in, I test one beforehand to confirm which face must be facing up and then use a Sharpie to mark each top face with a “T” so they are installed in the proper orientation.

[Maletrain]

Now that we have had the discussion about how hard it is to find the "right" configuration of permanent magnets to reliably uncouple and reliably not uncouple cars with various coupler types, does somebody have some detailed guidance on how to make electromagnetic uncouplers?  I understand the basics, but would like some guidance on how many turns of what gauge wire to use with what power supply to get an uncoupler that will reliably uncouple cars without catching fire or blowing the power supply.

I know there is a commercial version that was available, but those are out-of-production and expensive, compared to a home-grown approach.

It should not be difficult to make one with some magnet wire and some bolts and pieces of steel strap from the hardware store, if we have a good design to use.  The commercial version seems overly complex to install, with its ends "needing" to protrude through the roadbed.  I would think that it should be as simple as making an electromagnet that mimics one of the under-ties permanent magnets that has been shown to work here.  Just make some coupler slack and push a button when you want to uncouple, and pull the train across without pushing the button when you want the train to not uncouple.  Seems a lot easier to make and install some of these than to replace all of those non-MTL couplers, ferrous parts, and MTL couplers with the springs on the wrong side of the pivot pin.

[bbussey]

I don't know if there's one that can be built small enough to drop in easily and powerful enough if it is small.  Pretty sure also that it would have to be installed under the track rather than adjacent.  I like the option of being able to install (or remove) hidden magnets anywhere along the right-of-way without disturbing the track.

[DKS]

An alternative to electromagnetic uncouplers: permanent magnet uncouplers that drop out:



From: http://davidksmith.com/modeling/article-3.htm

[nickelplate759]

I've modified my procedure since 2014.  I get more consistency using K&J Magnetics part# BX081 (1" x 1/2" x 1/16") and locating them adjacent to the track rather than directly underneath.  I place one magnet on each side of the track parallel to the rails, buried 1/8" down and covered with a 1" x 1/2" x .060" piece of styrene to make it level with the cork top.   I haven't had any issues with any style of coupler, KD/MTL or otherwise.
...

I just tried this variation - it's definitely better.  I've come to the conclusion that part of my problem is also inconsistency in coupler trip pins.
Thanks.

[bbussey]

Glad that the revised placement is working for you.  One thing I haven't tried, now that the magnets are on the sides, is a magnet narrower than 1/2" so that it doesn't affect neighboring tracks.  This isn't an issue with industry sidings, but in yard areas it might be.  I am not at the point where I have to worry about it, but eventually I will experiment further.