Sunday, June 30, 2013

June Progress

Family issues remained this month with Dad being transferred from hospital to a nursing home and sadly he will not go home.  Thanks to those that have contacted me regularly to "have a chat".
Slow progress again this month as my wife completed her rehab and we drove to the country (850 km round trip) to make sure her unit was clean and stocked for her return. I also spent nine days away whilst going to Traralgon (165 km from Melbourne towards Sydney) see Peter's Blog. .

I needed something to take my mind of family matters (oh writing computer code used to do that for me a few decades ago!) so it was time for my first Launchpad project - "Helix Train Movement Sensor and Display".

My son has helped me put the final light fittings on the ceiling and now for the switching and dimmer bit.

First Full Operating Session - Cancelled

Invitations to the full operating session on 26th June were issued however with all the family issues I would not be able to do the event justice.  I reluctantly decided to cancel the event which will be rescheduled when I have the time to do all these things I was going to cram into the last month.....

I'm thinking sometime before Christmas now but will depend on how the family stuff goes....

Ceiling Lighting

Light fittings over Onka Valley and Loco arrival in place.

Command Station Consisting

Don Moyses and I have prepare a video using the Roco wireless MultiMausPro controller to utilize command station consisting.  Once consisted the consist locos are available individually whilst within consist.  Still o be tried (but expected to work) is command station consisting of sets of loco sets previously consisted through the use of CV19.

Take a look....

Helix Train Movement Sensor and Display

Most of you who know me understand that my engineering design and project management background dictates that I go through the normal steps of documentation Concept => Design => Pilot => Testing => Final Implementation.   I couldn't help mentioning it Tom Barbalett!

The Concept

 The "Block Occupied" LED is operated from the Block Watcher that will provide the signalling logic input.


I am utilizing the Texas Instruments MSP430 Launch Pad (MSP-EXP430G)

The coding

The last time I did any serious coding was back in the 70's using Fortran V.  It was a pleasant surprise that the general structure of C++ was similar albeit the syntax was different in some areas.  After reading the "Google'd Tutorials" which brought back memories of days of Uni lecturers and tutes....  The stuff provided at Terry Terrance's Blog site also gave some great background info.  I have done an article for Terry's site which will appear in due course.

I'll put a copy of the program at the bottom of this post but don't cut and paste as the formatting stuff gets copied and does not work in the compiler!

Thanks to Toni Ryan for providing ideas and then feedback on my code writing.

Putting the theory into practice

The 5v Power supply utilizing a 9v battery and a LM7805 regulator.


A piece of pine 70 mm x 19 mm for the length of track with sensors placed just longer than a 40' box car apart.  I have never "wire wrapped" so off to GoodLuckBuy again for wire wrapping wire and a wire wrapping tool. I must say it was easier to do than I thought.  The only requirement was some magnifier glasses in order to "see" what I was trying to do.

Sensors were angled down and were missing the floor level of the car

Cardboard packer behind the first one to make vertical - the others are still angled incorrectly.



There are differences between the sensors - all were setup the same but sensor 4 did not register the box car.

Attempting to get all sensors set up similarly.

Light colour cars work great....

Dark colours not so good.

I only have one "all black" consist which luckily will not travel up the helix - maybe a plus!

Mountings for the helix

The mounting to be used in the helix raises the sensor higher and can be positioned adjacent to the track for testing prior to final fixing.. Double sided tape then the mounting screw added.


Wiring the prototype

Wiring the TI Launchpad was easier than I thought.

After loading my first program the Launchpad did what the tutorial said so I "stepped" through the program but all my LED's came on (see photo below) with no optical sensor inputs - I had the test set for "high" but the sensors are normally "high" and go low when the sense something (the trusty multimeter reliably informed me of this!).

Change the test from "high" to "low" and he presto it worked fine.

Change the time the LED's stay on and ....

For those that may be interested in the Code that makes it work it is available here... 

Final Implementation

This remains to be completed it will need two boards as there are ten sensors required over the 100 or so feet of track.  I have started the construction and I have a 12V regulated supply available close-by that feeds the double slip indication circuitry. You can see the progress and notes below..

Dave Loman idea

tutorial 4

I have made one mistake though - when checking the 3V voltage it was high - around 4,2V so I went back and added extra diodes after the 5V regulator!  Then I found this...

"1N4148 not giving correct voltage drop 

Do you have any load on the circuit? – Wilhelmsen Aug 22 '12 at 21:20

Unless you consider the multimeter a load, nope. I think that's the problem. – capcom Aug 22 '12 at "

I guess some mods and further testing is still required!

That's it for this month.

Thanks to those that continue to help me achieve my ambitions and hello again to all my followers!

Till the end of July.


No comments:

Post a Comment