A working RPO in N-Scale

[VonRyan]

I'm sure many of you have seen the object of my inspiration.

I have always been fascinated by the American Flyer operating RPO ever since I was a child of only 2 years old, and as I got older the car was still as fascinating, but I began to have a better understanding of how it works.
About a month ago I had the opportunity to observe such a car once more on a S-Gauge modular-layout. I watched the car go through its routine of catching the mail-bag while spitting out the one it previously caught.

I was so busy observing the action that at first I did not look closer to see just how it worked.
Finally, I saw it. There was a slightly raised outside 3rd-rail and a contact shoe on the lead truck of the RPO. When the shoe hit the rail, the hook swung out to grab the hanging bag of mail, and as soon as the shoe left the 3rd-rail, the door immediately closed. Now, since it all happens so fast on account of the speeds those S-gauge trains were going, I could not tell exactly if it was the outswing of the hook that caused the other bag to be thrown out from the inside, or if it was the new bag forcing the other bag out.

It would appear that there is some kind of solenoid inside the RPO that has a spring to force the actuating arm back after the solenoid looses power. Going off of this mental picture, I began to plot some ideas for how to cram a similar mechanism inside a Micro Trains RPO, but without it becoming too toy-like.

The obvious perameters are that the car must not only pick up the mail, but it must also be able to spit out a previously caught mail bag.

Looking at the size of the door on the Micro Trains RPO, it is apparent that the clearances are going to be tight. The width of the door only allows the catch hook to swing out so far, and consequently it cannot be as far as the molded on catch-hook. My intentions are not to modify the prototype integrity of the car for the sake of this tidbit of animation. I'm not certain of the prototype for the MT RPO but unless there was a prototype version with a larger door, I plan not to modify it as much as possible.

As for the internal mechanism, I have begun constructing an new internal floor from .005" brass sheet with a tounge to go through the door that has already been cut out. The car's cast metal weight has also been removed, orginally to allow for better running, but given the work ahead of me, it is best to remove it because plenty of weight will come from the brass interior framing.
The actuating solenoid is planned to be an old Arnold Rapido 16v side-mount turnout solenoid, specifically the smaller kind that could be easily removed from the turnout to make it a hand-throw.
I have hoarded these little gems for just such a project.

The plan is to make a sort of cattle-shute for the swing arm so that the mail-bag being discharged can do so easily. I am hopping that I only need the common wire and one directional wire to make the hook swing out, and have a spring to push the solenoid arm back.
The main issue with this idea is that I will have to find a spring strong enough to push the solenoid arm back, but weak enough that the solenoid can overcome its force when it becomes energized.

Since the RPO will be predominately operated on DCC N-Trak layouts, utilizing track power does not appear to be an option. Instead I will need to have two pick-up shoes.
The idea is to simulate water-pans and have the Red-Line's pan be actually two strips of styrene channel lined with Beryllium-Copper. The Yellow and Blue lines will have a single piece of channel that simply will get a layer of gloss medium.
The RPO's pick-shoes are invisioned to be two bent strips of Phosphor-Bronze with just enough spring to make good contact, but flexible enough to be able to glide over diverging rails of turnouts.

The catch-hook will actually have an internal partner at 90 degrees from it. It will be a lower gate type arm that will push the other mail-bag out of the RPO.


So far, I have cut the new brass floor peice and bent the tounge to create a platform that is flush with the outside of the car. I am going through my stash of solenoids to determine which has the smoothest movement. If none is noticeably better, I will have to disassemble one and coax it into being smooth.

Aside from some progress photos that I still need to take, I am also going to try and work on creating a drawing in MS Paint to help those reading along to visualize how everything will come together.



-Cody F.

[peteski]

Cody,
will you get this done before or after you finish the wagontop boxcar castings? 

I shouldn't be saying this to you - I'm a king of unfinished projects! 

[bbussey]

Since you're planning on running the animated RPO on DCC trackage, wouldn't an accessory decoder to control the solenoid be a viable option?

[VonRyan]

Cody,
will you get this done before or after you finish the wagontop boxcar castings? 

I shouldn't be saying this to you - I'm a king of unfinished projects! 

Probably after. I'll consider it "finished" when it near-flawlessly operates.
It'll be done before the BLI M1a 

If I dared to list all the project I have started, even if just a single piece of something, I might be close behind you.

Luckily the double-door wagontops are a production project, so my partner will be helping with the process and bugging me to get my end of the work completed.

Since you're planning on running the animated RPO on DCC trackage, wouldn't an accessory decoder to control the solenoid be a viable option?

I did consider that as an option, but that would mean it wouldn't be able to operate if the layout were to use DC power. Although the N-Trak club I belong to doesn't run DC as often as they used to, having that added degree of flexibility is always a good idea, plus, I don't own a DCC throttle and the throttles available for club members to use are just UT4Rs.




-Cody F.

[peteski]

UT4Rs still have headlight switch, correct?  Then you could assign F0 to the solenoid control function in the decoder. But operating that on DC would not work (unless you got one of those erzatz pseudo-DCC controller which provide more functionality in DCC locos running on DC).  I think that MRC makes one of those.

[VonRyan]

I could do something like that, but it would effect the pulling engine.
The overall intent is to have the function be automatic (though a switch will be installed to allow me to prevent the solenoid from receiving power if the module with the mail-stand is not in the layout.


-Cody F.

[VonRyan]

As promised, some photos.

Just some iPhone grabs taken at my workbench.


It is hard to differentiate the remnants of the door from the interior, but later photos clarify this.


The new brass floor, taken from the bottom to show the fold-over of the tongue.


The RPO with the brass floor sitting in place, showing how the fold-over sits almost flush with the side of the car.


This is the modified roof. The window glazing that was attached has been replaced with slices of the clear plastic insert from a MT box.
To recreate the frosted effect, the back of the acetate was sanded with 1000-grit sandpaper.


A better, and more comprehensive view of the interior of the RPO showing the acetate and the new brass floor.



Two different views of the Arnold-Rapido turnout solenoid.



-Cody F.

[peteski]

Cody,
Couple of things to consider.  If that solenoid stays powered up for more than couple of seconds, it will melt.  Also, the solenoid action is rather abrupt and the mass of the rapidly moving core might actually jostle the entire car.

Have you considered using memory wire?  It would require less power, it could withstand being powered up for extended periods of time, it would be much more compact, and lastly, its action would be much gentler and slower than a turnout solenoid.  It will also automatically retract when the power is removed. Its motion is probably also much easier adjustable than when using a solenoid.

I seem to recall someone on TRW using that method for animating semaphore arms.

[VonRyan]

Reading the instructions (I kept a few of them, I started with 20 of them) it says that there is a protection switch inside that prevents it from melting, plus the casing is metal. I actually have a couple that aren't complete and there is indeed some form of switch there (they call it a 'commutator' to "switch the power", but it is really just a simplistic slide-switch).
The action isn't as jolting as it could be, but since it is parallel to the direction of travel I would think that the action of the solenoid wouldn't be detrimental.

I recall the memory wire you talk about, but its shrinkage/expansion factor is much too small for the catch-hook, and it appears as if it retracts too slow compared to a spring pushing on the solenoid, or especially if I was to add a third contact to apply power to retract the arm rather than a spring.


-Cody F.

[peteski]

Reading the instructions (I kept a few of them, I started with 20 of them) it says that there is a protection switch inside that prevents it from melting, plus the casing is metal. I actually have a couple that aren't complete and there is indeed some form of switch there (they call it a 'commutator' to "switch the power", but it is really just a simplistic slide-switch).
The action isn't as jolting as it could be, but since it is parallel to the direction of travel I would think that the action of the solenoid wouldn't be detrimental.

I recall the memory wire you talk about, but its shrinkage/expansion factor is much too small for the catch-hook, and it appears as if it retracts too slow compared to a spring pushing on the solenoid, or especially if I was to add a third contact to apply power to retract the arm rather than a spring.


-Cody F.

Yes, if the Arnold switch machine has those contacts built-in then it will be safe to use.  But if you use a spring to retract the plunger then that will defeat the function of that switch. After the solenoid is powered up and it moves past the point of the power being cut off by the switch, the spring will immediately pull the plunger back to the point that the switch will energize the solenoid again. Then the cycle will repeat, in rapid succession. You will end up with a vibrator or a buzzer. Similar to the old fashion doorbells.

As far as the slow motion of the memory wire-powered hook goes, wouldn't that be more realistic than a snap action of the solenoid?  Or you are going for a true emulation of the American Flyer car?  The amount of motion generated by the memory wire can be controlled using a fulcrum with the hook on one end, and memory wire on the other end.

[VonRyan]

I'm not seeking to recreate the American Flyer car in a "model of a model" sense, but in a sense of keeping its reliability and simplicity.
The point of keeping the speed of the hook in retraction is mostly to prevent the bag from falling short of the car and/or getting jammed in the doorway.

The spring won't defeat the purpose of the switch because the spring will be just weak enough that when energized the solenoid can compress it, but strong enough to quickly make the arm return once the contact shoe leaves the simulate "water pan". Of course, if I cannot find a suitable spring, then I will have to add a third pick-up shoe and "water-pan" to electrically retract the solenoid.

Although not 100% correct by American prototypes, if this same mechanism were to be transferred to a UK prototype mail-car it would be more correct since in the UK prototype the force of the mail bag hitting the catch-net would cause the whole assembly to retract, the attendant then further retracting the assembly into the car to be secured.



-Cody F.

[dougnelson]

Hi Cody:

An interesting project! 

FYI: once you perfect the working RPO on the MTL car, you can apply it to the Hell Gate Models prototypical PRR BM70k.  It comes with the doors already open.  We will have a small run of these cars (and the B60b) soon.

Doug Nelson

[VonRyan]

Hi Cody:

An interesting project! 

FYI: once you perfect the working RPO on the MTL car, you can apply it to the Hell Gate Models prototypical PRR BM70k.  It comes with the doors already open.  We will have a small run of these cars (and the B60b) soon.

Doug Nelson

Looks like I have no choice then. 

Any car with a wider mail-door would be far easier as the clearance tolerances would allow for the mail-pole to be farther back from the edge of the tracks. As is it looks like I will have to make the mail pole rotatable so that it doesn't possibly effect other passing trains.


-Cody F.

[nkalanaga]

Some of the mail poles could be rotated for that very reason - to allow wider freight loads, and in some cases large steam locomotives, to clear them.

[VonRyan]

Some of the mail poles could be rotated for that very reason - to allow wider freight loads, and in some cases large steam locomotives, to clear them.

I know it was done with UK mail poles, but I couldn't find definitive answers in regards to our own practices.

Of course... A mail pole would never be next to water pans... The spray would make for some soggy mail.