Automated Motorised Traverser

[tender]

Its basically the same as my 'DCC Stepper Motor Controlled Turntable' with a few slight modifications to the code and using a different stepper controller chip which is a lot quieter in operation. Theres some more detail here:

 

[Hector_A]

Hello everybody!

Please bear with me while I introduce myself. I'm from Montevideo, Uruguay, a retired land surveyor and model aircraft builder and competitor (FAI team racing F2C) for about 50 years. I'm an old hand at CAD drawing and am resurfacing my model railroading as a more sedate alternative in accordance with my present age (86). Around the 80's I built a small layout which was abandoned for several reasons in addition to my interest in pursuing team racing (both activities shared with my younger son - now also a surveyor).

 

Living now in a house with a spare room 3m00 x 2m40 (roughly 10' x 8') I've started the build of an HO scale layout based on American mining and logging railroads. The room already has some wardrobes which I already built as a solid flat base for the future layout. It was a compromise with my wife as they are needed for out-of-season clothes. The layout plan itself is loosely based on Robert Schleicher's featured in Model Railroading magazine issue from November/December 1983. This is a two level plan whose lower level is exclusively destined as a storage/fiddle yard. Albeit, the design uses linear staging plus a return loop, something I do not really fancy as it consumes a lot of track and turnouts.

 

Having seen an article in Model Railroader about a traverser for storage I came immediately hooked on the idea and proceeded to redraw that level incorporating a modest traverser only 3' long and with just 5 tracks. Needless to say I was elated to find this post and avidly read all six pages of it!

 

I have a full woodworking shop in my garage and have access to high quality drawer slides, so the building part I feel it's already solved. The big question is whether to use some kind of mechanical indexing for the track alignment or delve into the software controlled version. BTW I'll be using DCC (Digitrax).

 

Ordering an Arduino or a Raspberry Pi is no big deal. I have also been looking at Rocrail. What I would really appreciate is some sound advice on the matter in order to proceed. Will be really pleased to hear from anyone who takes the time to point me in the right direction. Thanks in advance

 

Hector

[Simond]

Hector, welcome!

 

I should start by saying I have not yet built a traverser, but I have built and programmed a turntable, and the issues and principles are similar.

 

For an HO traverser with 5 tracks, you’ll need a total movement of something in the region of 150mm.  This is convenient, as there are lots of ready-made Chinese “stepper, leadscrew and guide assemblies” on the web with this travel.  You could use two of these to support & guide your traverser deck, I would not bother with drawer slides.  

You can arrange the woodwork to suit

 

You could control them with an Arduino, I think there’s code on here to do that.  You’ll need some kind of zero-point detection, could be optical, could be as simple as a microswitch.  I’d probably fit detection at both ends, and arrange my program to confirm & adjust the straightness of the deck whenever it’s switched on, as if the deck is moved when it’s off, the controllers won’t “know” where it is.  
 

The other purchases necessary will be “H-bridges” to drive the steppers, and a suitable power supply.

 

Hope this gives food for thought, I look forward to seeing your progress

cheers

Simon

[Simond]

an example

 

https://www.ebay.co.uk/itm/283269246477?hash=item41f429ae0d:g:fJAAAOSwOd9dpXkJ

 

these were much cheaper a few years back, but they are probably still good value even now.

 

Other option is to use a toothed belt and stepper motor - the toothed belt can be arranged around pulleys to provide a parallel motion.

Edited June 23, 2022 by Simond

[DCB]

10 hours ago, Hector_A said:

Hello everybody!

.The big question is whether to use some kind of mechanical indexing for the track alignment or delve into the software controlled version. BTW I'll be using DCC (Digitrax).

Hector

 

The way I see it,..  Electronic traverse and indexing gives a much softer slower operation, much less likely to derail stock than purely mechanical grab a hold of it and pull operation, the problem is keeping the deck straight.  both ends have to move together, and that's not easy with a mechanical linkage or pairs of motors . I have a hand operated turntable indexing with a brass L section bar dropping into a hack saw cut in a rail which has worked and not been adjusted for 30 years, it's inaccessible under a station.    Making the  deck slide friction free is probably the key here, more important than how you move the deck, 

[Michael Hodgson]

I think mechanisms like that can be found in some digital scanners, so if you've got an old one, salvage may be an option though obviously from a safety point of view it will necessary to know what you're doing.

[Deev]

I managed to get one of these working, but it wasn't easy and I moved house before I could take acceptable photos of it.

 

My experience is that you need provide much more power than you think you'll need, but drawer runners are the things to use as they are fiendishly accurate.

 

I'd look at a 3d-printer driver board and NEMA17 motors.  The world of 3d printers has exploded over the last 10 years and you can buy a fully-populated motherboard that includes the Arduino, silent motor drivers (eg. TMC2208)  as well as all the connectors for things like limit switches for less than £60..  Search for Einsy RAMBO.

 

Obviously you'll need to write your own code for it (and before you ask, I've lost the code that I wrote) but it isn't actually that difficult.  You'll also need to work out a way of calibrating it, either through limit switches or through stall-monitoring through back-EMF.

Edited June 28, 2022 by Deev

[Hector_A]

Hello folks!

 

Thanks for the numerous and detailed replies. It sure helps to consider several options. Indeed, I took my time to ponder on the many alternatives and went over a load of documentation I've saved.

 

At this stage I am inclined to use a mechanical traverser with a simple indexing system that I submit for your consideration. Let me tell first that I am a fervent follower of the KISS principle and thus my proposal is drafted along this lines.

 

I agree with Deev in that good quality drawer sliders provide smooth operation at a very reasonable cost. I presently have a 40cm long pair (made in Chile) which I modified to travel in both directions. Their length is now 38cm. Have seen lots of traversers built with these (some quite larger than mine) so I feel that's the way to go. Mounted flat they ideally fit my scheme.

 

I attach two .pdf files which show the planned  two level layout. The one with the 94.5 cm long traverser features 22" radius curves throughout (albeit with easements overall). There's a second drawing of the staging level where there are two 18" radius curves which enable a 121.8 long traverser. Since these curves also have easements and will be hidden, I do not foresee problems here. The first shorter traverser has only 5 tracks with a 60mm separation. The  second option has 6 tracks with 57mm separation. Would like to hear views on track separation in this case since access to the traverser is somewhat limited.

 

BTW lines in blue represent the main line, lines in green are easements. Note that I provided for out-and-back operation from the main station if bridges are raised/lowered. Red lines show the present wardrobes already built in this room. In the drawing of the larger traverser I included a detail of the drawer slide and its modest required travel.

 

For the traverser operation I'm thinking of using micro switches in each track which will get tripped as they reach the home position, perhaps turning on a led light. This would enable me to know when a track is nearly aligned in place and then manually operate a mechanical registering shaft with a pointed tip. That, I hope, should provide the required track alignment with a minimum of fuss and parts.

 

Of course, I'm open to hear comments/suggestions as this is still just brainstorming from my part.

 

Pardon the long story and I hope to hear from your soon.

 

Cheers!K+C_Layout_Longer_Traverser-Model.pdf

K+C_Layout-Model.pdf

[Hector_A]

I failed to add that all traverser tracks will have one rail permanently powered and the micro switch (when tripped) should supply juice to the other rail, thereby enabling the selected track to operate normally.

 

Should eventually a motorised version be chosen, I'd like it to be push button operated as the one by Michael Hardwick published in the August 2018 issue of Model Railroader. Here's the link to a video of it in operation: https://www.trains.com/mrr/videos-photos/videos/layouts/layout-visits/video-push-button-controlled-traverser-for-a-staging-yard/ The code for the Arduino can be downloaded at: https://www.trains.com/mrr/how-to/arduino-control-code-for-traverser/

 

As you can see, Michael's traverser is different in that he just moves one track  to register with several others while mine is supposed to move several tracks to register with one entry and exit point. Having no experience with the Arduino code, I wonder if it could work "as is" or it requires a modification for my case.

 

I have a copy of MR issue and there's a list of parts and suppliers. Took the time to look up the prices of said parts and they add up to a modest USD 88. Quite accessible as compared with the Einsy RAMBO which is USD 125 to which you must add the NEMA 17 motor and associated parts.

[Harlequin]

Hi Hector,

Wow, that looks ambitious!

Have you checked the gradients and clearances between levels? (Remember you need "easements" in Z between grades and levels as well as those in XY between straights and curves.)

Is the bridge across the doorway fixed or opening? If opening, what's the mechanism?

 

[Hector_A]

Hello Harlequin:

 

As a retired Land Surveyor I'm well aware of vertical easements as well. I have kept most grades in the 2 to 3 percent range except on the switchbacks on the upper level.

 

There are two bridges in the doorway. The lower one is hinged to drop down and is already built. The upper (main level) will have a two track (with different grades) that will be also hinged to be lifted up. This will be more complex but I'm still far from that. Just working on the lower (hidden staging) level at present.

 

Will upload photos as soon as some track is laid so that there will be something worth to look at!

[Hector_A]

@ Deev:

 

I'll be very grateful to hear your comments on my two posts of last Sunday. I value a lot your views on this matter and have carefully read all six pages since you started it back in 2012.

 

If I eventually decide to go for a motorised version, I'd like to know where to look for parts to arrange the belt as your final version depicted. Bear in mind that I can only access US suppliers as our Customs regulations only provide for three 200 dollars yearly purchases tax-free from that source.

 

Thanks in advance for your kind help and time.

[Deev]

Hi Hector,

 

Well, I like your plans - I'm always a massive fan of complex multi-level layouts.

 

What I will say is that the traverser looks quite short.  I don't know what length trains you'll be running, but I reckoned on a traverser about 2500mm long to allow a full length HST set to be moved.  If you're running shorter rakes, then this won't be an issue.  This said, your layout might look better served using more traditional points and servo motors to avoid the obvious complexity, not to mention the cost, of scratch-building a traverser.  If you are after the technical challenge of it, then I commend you.

 

Other considerations are the amount of space in front of and behind the traverser you'd need to allow it to traverse.

 

Also, have you considered building a helix so that you have the possibility of a second traverser?  How are you going to run cabling to the traverser?  Are you going to fit sensors on it to automatically stop trains?  A cable chain might be a worthwhile investment here.

 

My final thoughts are with a traverser, to buy the most expensive and powerful mechanism you can.  Depending on the length of the traverser, and the amount of stock on it, you're liable to have a lot of weight to shift, so you'll probably need multiple NEMA17s, say four at least by my reckoning.  If I were building one, I'd use the best mechanisms I could find (8mm diameter, 8mm pitch, 4 threaded leadscrews with POM (not brass) nuts and use an Einsy Rambo control board, purely because all the hard work is done for you and it can run 4 steppers silently.  Having a noisy stepper driver would drive you insane after a while.  It would be annoying to build a traverser cheaply and then curse every time it jammed or stopped working.

[Ian_H]

4 hours ago, Deev said:

you'll probably need multiple NEMA17s, say four at least by my reckoning.  If I were building one, I'd use the best mechanisms I could find (8mm diameter, 8mm pitch, 4 threaded leadscrews with POM (not brass) nuts and use an Einsy Rambo control board, purely because all the hard work is done for you and it can run 4 steppers silently.  Having a noisy stepper driver would drive you insane after a while.  It would be annoying to build a traverser cheaply and then curse every time it jammed or stopped working.

Hi Deev,

Hope you don't mind me asking a question with regard to the 2500mm motorised traverser, I appreciate that your answer was to Hector, may I ask what a typical target price would be for all the hardware you propose?

 

I've seen a few automated motorised traverser, some are good others seem to struggle a little. Selector plates are also a consideration and one that I particularly like is the one used on Liverpool Lime Street. Unfortunately not all the videos are available on RMWEB since the changeover but I've found this on youtube, it may be of interest.

 

 

[Hector_A]

Hi Deev,

Thanks for your prompt reply and detailed comments.

 

I'm modelling american railroads based on western regions covering mining and probably some logging operations. I do not envisage long trains but still thought that my initial plan with a 94.5 cm was too short. Therefore I included another .pdf with a longer 121.5 cm traverser, which seemed more adequate. That modification implied using 18" curves (with easements) to enter and exit the fiddle yard. Did you see it?

 

I included rerailers at both entry and exit points but have not seen them being used in any of the traversers I looked in the web. Therefore, if those rerailers are discarded the yard now grows to 167.5 cm which, anyway, is the longest I can fit in this room. See attached .pdf As you can see, I also have enough space for its travel without compromising the available space for operators.

 

The original plan indeed had linear staging with lots of turnouts but I hate the idea of crawling under scenery either to operate or attend to eventual derailments. Once I saw a traverser I got hooked on the idea of having one in my layout. The physical building task is, for me, relatively easy. And all my staging would be in a single location.

 

Controlling it is quite a different matter. I still waver between manual and software control options. As I said before, I'm an advocate of simple mechanisms. The less you have there, the less things are apt to go wrong. Will keep on thinking and let you know the final outcome of my brainstorming!

 

Thanks again for the additional suggestions. Food for thought there.

 

Cheers!

 

 

K+C_Layout_Longer_Traverser-Model.pdf

[Deev]

On 08/07/2022 at 17:05, Ian_H said:

Hi Deev,

Hope you don't mind me asking a question with regard to the 2500mm motorised traverser, I appreciate that your answer was to Hector, may I ask what a typical target price would be for all the hardware you propose?

 

I've seen a few automated motorised traverser, some are good others seem to struggle a little. Selector plates are also a consideration and one that I particularly like is the one used on Liverpool Lime Street. Unfortunately not all the videos are available on RMWEB since the changeover but I've found this on youtube, it may be of interest.

 

 

Yes, I've seen that.  Everything about that layout is something else.  Simply a masterpiece.

 

Regarding the costs, hard to give a definitive figure, but here are my guesses:

 

1 x Controller (Einsy Rambo) = £70

1 x Optocoupler board = £5

4 x NEMA 17 stepper motors - 4 x £15 = £60

4 x 8mm leadscrews with POM nuts = 4 x £15 = £60

2 x pairs 600mm drawer runners - 2 x £10 = £20

Aluminium section = £30

Sheet of 15mm Plywood = £100

Cable chain = £10

200w Power supply = £20

Screws, glue, microswitches, bolts = £20

 

TOTAL = £395

 

Add to this the track and copper-clad.

 

I may point out that you could buy a lot of points and motors for £400.

[Ian_H]

7 hours ago, Deev said:

I may point out that you could buy a lot of points and motors for £400.

Yes, I totally agree that if you have the space ....... ;-)

[Hector_A]

@Deev

 

Thanks for the compliments. To be honest, mine is an adaptation of the original designed by Robert Schleicher which appeared in Model Railroading November/December 1983 and later published in a compendium of several articles by him which I found very interesting and informative. I have scanned the relevant article in case you're interested.

 

My merit, if any, is the inclusion of a traverser instead of the linear staging. I also included a Timesaver which  came from a picture in this forum flipped over and set as a secondary station (placed over the traverser). Of course, there was a general adaptation to the real dimensions of my room , the door location  and a sewage pipe coming from the upstairs bathroom. Plus I lowered radius to 22" instead of 24", something not expected to give any real problems in operation with short locomotives and cars.

 

I've been working these days in the layout lower (hidden) level which is now about 50% completed. This includes 15mm plywood subroadbead but not yet the 6mm thick cork roadbed. The traverser is installed and works very smoothly. Still missing is the mechanical indexing system (already thought out) partially constructed. As you see I went this way and, so far, are very happy with the progress and operation.

 

I like manual interaction with my layout so this seemed the right thing to do. On this venture, manually operated turnouts are also envisaged, except for hard to reach locations.

 

Will certainly include photos as soon as that level is more finished and worth looking at.

 

Hector

Schleicher-1.pdf Schleicher-2.pdf Schleicher-3.pdf

[Hector_A]

Here are some photos of my mechanical traverser. I'm very satisfied with its smooth operation and the simple, efficient, indexing system I was able to design and construct. Photos 01 and 02 show the locking mechanism respectively unlocked and locked. Photos 03 and 04 depict the 8mm dia. steel rod also in the locked and unlocked position. Note the bullet shaped end of the rod in photo 04.

In this same photo there are three pulleys which I built myself as I failed to find these in our local specialist shops. I turned the 35 mm dia. center from hardwood and cut the top and bottom caps from 3mm thick hardboard. The channel is 13 mm wide to accommodate stout plastic packing strip. They have steel tube bushings for 1/4" Allen screws tapped into the MDF table top.

Photo 05 is an underside view of the indexing aluminum angle. Each hole corresponds to a track position (six tracks). the packing strip is clamped with an aluminum piece and two 1/8" machine screws at both ends of the angle. The lower series of holes was a failed intent on using a ball catch to align the rods with the upper holes.

In photo 06 you can appreciate the final rigging and how it works. When moving the traverser the packing strip rides on the pulleys and shows an index line for each track position. This line must register with the fixed lines in a backing block. It's very simple and does not require to be finicky with the precision. It can be a couple of millimeters off and the bullet shaped rod end will engage the hole anyway and drag the whole table into perfect alignment.

Last but not least, it's a really low-cost option.

[Hector_A]

Hello everybody!

 

Thanks for the thumbs up fellows!

 

As you can see from my last post, I have gone the way of a mechanical traverser. IMO my posts are out of context in this thread and should be moved elsewhere. My problem is I don't what would be the correct placement. Will appreciate any suggestions from you or a moderator on how to properly proceed.

 

Would like to share my work and progress on the whole layout now that the roadbed for the lower, staging, hidden level has all the cork roadbed already installed.

 

Thanks in advance for your help.

 

Hector

[Simond]

Hector, you could simply start your own thread, and copy your posts over to it.  You could link it here as an alternative approach.

[Hector_A]

Hello Simond,

 

Thanks for your reply. Of course, I had already thought of doing that. The questions is under which of the many available headings should I start that new topic.

[Simond]

Ah, yes, well, I guess most folks tend to put their layout thread in the group of the scale they’re modelling in..