Traverser - The Journey

Traverser 1.0

My first attempt turned out very good for a rookie.  I built the table to slide on some standard cabinet drawer slides.  I used a lead screw and a stepper motor to move the table.  I programmed an Arduino to interface with a touchscreen called a Nextion to control how to move the table and the Arduino drove the stepper motor using realistic acceleration.  Images on the Nextion showed the status of the system, including whether tracks were lined up.

Results

It worked as intended.  Upon startup, the table would find its “home” position and the control panel would confirm that.  I could push graphical buttons on the panel and the table would move to the correct position to line up the tracks.  I also had additional “pages” on the control panel to allow me to dial in the correct stop positions and speeds for acceleration.

Why wasn’t it good enough?  The unit could not produce repeatable positioning movements because 1) the drawer slides have too much play and 2) I did not achieve a rigid enough connection between the the lead screw and the tabletop, which introduced even more play.  These problems resulted in too much “slop” to produce precise positioning.  Since the track alignments must be almost perfect, if the mechanism can’t move away and then move back to the exact same spot, there’s no point in having an automated movement mechanism.

Model of Traverser 1.0


Traverser 2.0

The August 2018 edition of Model Railroader has an article titled, “Push-Button Staging Solution” by Michael Hardwick in which the author describes how he built his traverser.  Here’s what I learned:

  1. Use bearing blocks riding on chrome rails instead of drawer slides.
  2. Use “match drilling” to insure the chrome rails are perfectly aligned horizontally and vertically.
  3. Use a timing belt, pulley and idler, parts normally used on 3D printers, to drive the movement instead of a lead screw.
  4. Include mechanisms that allow for fine tuning track alignment during final installation.

The author’s design still relied on a stepper motor and an Arduino controller so I figured my cool control panel could work without any changes.

Additional online research led me to conclude that it would be best to include a short section of  stationary exit track as part of the overall unit.  In Traverser 1.0, I found that aligning to the existing layout’s trackage was incredibly finnicky.

Lastly, many online folks suggested building a solid, square frame to which all the other components would be attached.

Summary of the changes I made from my first attempt:

  1. I built a perfectly square frame to hold the unit.  I could build and test the entire unit at the workbench and move it and be assured that everything would remain aligned.
  2. The frame included a section of stationary exit track at the end which could easily be mated to the end of the layout’s exit point.  This allowed me to precisely line up the movable tracks with the exit track at the workbench.
  3. I replaced the drawer slides with bearing blocks on round chrome rods.
  4. I match drilled holes in the front and back pieces of the frame into which the chrome rods would fit perfectly.
  5. I used the same stepper motor as before but, per the author’s instructions, I installed a pulley gear on it, installed a matching idler pulley and attached the 2 with a notched timing belt.  I attached the timing belt to the tabletop with aluminum L-angle.
  6. I routed in slots at various places to slide components for fine adjustments.
  7. Per the author’s suggestion, I built the tabletop with 3 layers.  The middle layer is foam that can be compressed using screws, to allow fine vertical alignment of the tracks to the exit track.

Results

The good news:

  1. The frame was a huge improvement.  Having a rigid structure helped to get things aligned and kept them that way.
  2. The bearing blocks on chrome rails were WAY better than drawer slides.  Attaching them to the bottom of the tabletop was easy.
  3. The various mechanisms employed to allow for finer adjustments worked as intended allowing me to dial in things very nicely.  In particular, I was able to vertically line up the tabletop’s tracks with the exit track so that cars ran very smoothly over the small rail gap.
  4. The stationary exit track was a HUGE win, allowing me to align the tracks and be assured that things would be okay once installed.
  5. The transport mechanism using the 2 pulleys and timing belt was an improvement over the lead screw technique.
  6. I was able to completely re-use my Arudino-based controller and graphical control panel with no changes.

As good as the transport mechanism was, there was still a perceptible amount of play when you pushed in on the table.  In a perfect world, when the stepper motor was engaged, you should not be able to move the table without a significant effort.  Because of the tighter fit of the bearing blocks on the chrome rail, the movement was much less than when using drawer slides and it seemed that the table “snapped back” to its original place but there’s no way to measure that easily.

Model of Traverser 2.0

Traverser 3.0

I wanted to achieve a more perfect transport mechanism that was super stable and moved only when the stepper motor said to.  The 3D printing world provided an answer:  V-slot.  V-slot is extruded aluminum with running parallel edges into which hard plastic roller wheels run.

The wheels are attached to a gantry plate that is driven by a lead screw, which is driven by a stepper motor.  The idea of using a lead screw initially worried me.  However, research showed that typically a lead screw provides better precision and more rigidity than a timing belt.  I concluded that my poor results in Traverser 1.0 were not because of the lead screw itself but how I had transmitted its movement up to the tabletop:  too much slop there.  Also, Traverser 1.0 used sloppy drawer slides.

My hope was that the tabletop, being firmly attached to a gantry that could not pivot, would be more likely to remain rigid at all times.

I was lucky to find a kit that includes all the parts to build a self-contained transport mechanism which appeared to be rock solid but it required some modifications to my frame.

Summary of the changes I made from Traverser 2.0:

  1. Replaced the timing belt/pulley method of transport with the V-slot-based kit provided by a vendor.
  2. Modified the dimensions of the frame.
  3. Moved the chrome rails further out towards the ends of the table to provide even more stability.
  4. Added some additional features to allow more fine tuning of alignments.
  5. Increased stability of the attachment of the tabletop to the movement mechanism.

Results

  1. Achieved excellent repeatability of hitting the track stops.
  2. Once the stepper motor was engaged, it was very hard to move the tabletop at all, producing better stability.
  3. Virtually no sideplay.
  4. Was able to fine tune the various alignments to achieve very smooth train operation across the tracks.
  5. Overall, a very solid operating table.

Model of Traverser 3.0