occupancy detection alternatives

One of the MRH sponsors, Iowa Scaled Engineering, offers infrared (IR) occupancy detectors [ link].  Other companies (e.g., http://www.azatrax.com) offer similar solutions.

If you go the Arduino microcontroller route, there are inexpensive ultrasonic range detectors that can also be used.

BDL168

Onewolf

I'd strongly recommend the BDL168 over the BD4.  It will report directly to LocoNet and no problems with the SuperSonic decoders.

Dave

I will second the use of the

I will second the use of the BDL168 over the BD4. The BD4 will do the job, but not alone (could be wrong on that). I believe that the BD4 has to be piggy-backed onto either a DS64 (switch controller) or an SE8C (this is how I am using the ones I have. They do not truly report the way that the BDL does, the reason that at some point down the road I will be pulling them out and replacing with the BDL168. 

Another option (wish it was around when I started buying all my stuff) is the RRCirkits product line. They have come a long way on their components and to be honest I like the way they are used more than Digitrax. 

Local support? What does your LHS stock do you have other close friends that have used any products that are willing to help or get and give advise. This is important, while you may have plenty of experience it is always good to have someone that you trust to "bounce ideas" back and forth with.

Good Question

It think the BDL168 is more flexible...

Hi Onewolf,

I'm currently installing detection & signaling on our layout and am using the BDL168 for detection along with the SE8C for signals & turnout control through a standalone LocoNet.

I think the most flexible option for yourself is the use of the BDL over several BD4's, the BDL will sense 16 detection blocks over 4 zones from the one board.  Because you mention you want to detect the staging tracks, the turnout ladder, and the mainline loop, it will not take long to use up the 16 detection blocks.

The flexibility of the BDL could also allow you to break down the turnout ladder and the staging tracks into several blocks.  This would allow you to drill a train through the ladder and once clear of say half the ladder length, pull another train out of staging.  Conversely if you have the staging tracks blocked in half, thirds or even quarters you can place shorter locals in them and have the ability to double up you consists in one staging track, if you have each staging track as one block you will loose this ability.

So you could use two detection zones to give you 8 detection blocks for the staging tracks, one zone of 4 detection blocks for the turnout ladders, and one zone of 4 detection zones for the mainline loop.  The one zone for the mainline loop will work perfectly as you will need a auto reverser in this section, and these devices (be it a Digitrax or something like a PSX-AR) will need to be wired into one of the 4 zones only.

Considering your going with all Digitrax for your DCC command System, I'd stick with the BDL & SE8C for your detection & signaling too.

Jas...

Thanks for all the great

Thanks for all the great information so far.  It sounds like Digitrax BDL168s are the way for me to go. 

Question: is there limitation on the reliable overall length of each occupancy detection section?  Each of my 10 staging tracks are approx 240" long. Would it be ok for each of those staging tracks to be single detection section? I do not plan to stage multiple trains in a single staging track because my goal is to run 200"-230" long trains.

The Helix will be a double track (37.5" and 40" radius) with 4 1/2 'turns' and each full turn is 235"(inside track) and 250" (outside track). I was planning to have each turn for each track to be an occupancy zone + 1 zone each at top/bottom for connecting turnout clearance into/from the inside track.  So, 4 (full turns outside) + 4 (full turns inside) + 1 (1/2 turn outside) + 1 (1/2 turn inside) + 2 (connecting turnout clearance) = 12 sections for the helix. Which leaves 4 'extra' on the BDL168 for refinements.  Does that seem reasonable?

Thanks.

edit: added Helix information

Question 1 yes, I would, if

Question 1 yes, I would, if possible place the BDL as near to the center as possible. The length would then be cut in half. 

Question 2 OK again. As you build the helix add feeders for the flex track and run all the feeders to one side of the helix. then run a bus line for each detection zone up the side of the helix to the feeders. Your overall run of bus will be drastically shorter doing it that way. I have seen some run the bus up the entire length of the helix, but really that just seams to be un-needed wire. 

Are you planning on using a computer program, ie JMRI or RR&Co to monitor these tracks or are you just using LEDs fed from the boards to the fascia? Either way is fine, but using one of the programs on the market will allow you to do some additional fun things. 

How are you planning on throwing all the turnouts for the layout. tortoise? 

Are you planning on using a

I am not (currently) planning to use computer to monitor/control layout. I plan to use LEDs on fascia to indicate occupancy status. 

I plan to use Tortoises and either NCE Switch-its or DCCSpecialties Wabbits for turnout control with fascia toggle switch and indicator LED.

One wolf

One detector in the middle will tell you the track is occupied. The next thing that you would need to know is are any of the clearance points fouled? You could do this by adding more detectors one at each end of the track at the clearance point, when it is clear the train is in far enough to not be fouling the other tracks and in the case of double ended yards is the front end still safely tucked in. There are of course other ways besides detectors to monitor the ends of the tracks. An electric eye at each end to show an alarm if a train is blocking a clearance point. A security camera at each end that can also provide an over all view of each end and may aid in identifying which train is where.

There is also the possibility that the MKII eyeball method is used if the operators can see their respective trains as they are putting them into the tracks. If that is the case no other detectors would be needed, unless some type of remote or computer control would be of interest to you.

In addition to the occupancy

In addition to the occupancy detectors, if you have the room why don't you build "herniated" helixes; so one level, lets say level three comes outside of the helix and has one passing track, this area scenicked.  This method was illustrated in the 2013 MRP (but don't hold me to that).

Waiting for trains in a helix can be tedious and with the herniated option, you have the train put in an appearance roughly half way through which reduces wait time psychologically and gives you more visible mainline.

Detection on staging tracks

From what you said, it sounds like the staging tracks would not be visible.  This would indicate that you would need two systems, (1) block detection to indicate the presence of a train and (2) location detection to indicate that the train was clear of switches or not possibly fouling another track.  Also you indicated that you were interested in led indication on the fascia.  This can be done this using Bruce Chubb's detectors (optimized detector OD using diodes, or  DCCOD for dcc).  The DCCOD uses a transformer to indicate current flow.  Bruce can supply the boards and transformer at a reasonable cost and you can do the rest of the component installation which is very easy.  They have a transistor output to light an LED (or other Devices) when the track is occupied. This will detect a powered locomotive and lighted cars, but will not tell you where the train is on the staging track.  For this you can use a simple IR LED and phototransistor circuit with 2 resistors, a transistor and a led to show when the train has past the switches or has reached the end of the track.  This has worked very well on my layout  and two other layouts, including our club.  The ultimate train locator would be to use a security camera and tv screen.  At one time Harbor Freight had a system with two cameras and a monitor for around $50 that several of my friends have used that also show when the train is moving and how fast (so you resist the temptation to increase the speed if you don't see anything happening)  If you want more info about Bruce Chubb's detectors or the IR detector circuit let me know.  These two circuits don't have all the bells that some of the commercial circuits but they do the the job that you need very well.

Bill Feairheller

Onewolf have you considered signals?

And the reason I ask is the comment below:

Because there is a likely hood later on you may use JMRI to monitor your staging tracks and possibly the layout, going buy what you have posted.  Another flexible option of the BDL's is the ability to wire up a fascia panel for your staging occupancy using LED's direct from the BDL board.

Then later on if your want to use JMRI to monitor via a dispatcher you have the option already there.  I'd strongly suggest that if your going to stick with Digitrax like you mentioned earlier you do that, you virtually have a complete plug & play system ready to go.  And if you decided to add signaling at a letter date you are 90% of the way there for hardware already.

As for your turnout control I would seriously consider again staying with Digitrax and using the SE8C, the one board will allow you to control Tortoise Switch Machines, have push button fascia controls (and via computer later on) with LED's and operate signals at a later date too.  The SE8C has the ability to operate 8 - Tortoise Machines, and at around just under $100 US is a very versatile board with a lot of flexibility.

One of the best features of the whole Digitrax system is that it plugs in easily and it works, buy adding other types of systems and hardware you run the risk of compatibility issues, and having too adapt to them to the Digitrax system.

Just some food for thought from someone who is installing it as we speak...

Jas...

Again, Thanks for all the great information

 Assumption #1) I have no idea what I’m talking about. 

Assumption #2) I will have a ‘resistor wheelset’ on the end of EVERY train (cabooses or passenger cars).

Based on the crude drawing above, this is my 'theory'.

I create 10 occupancy sections and each section has an ‘occupied’ status LED on the fascia.

1. Westbound mainline (~6 ft)
2. Eastbound mainline (~6 ft)
3. Front Ladder
4. Rear Ladder
5. Return Loop
6. Staging track 1
7. Staging track 2
8. Staging track 3
9. Staging track 4
10. Staging track 5

Rules Of Engagement:  The mainline and front/rear ladder sections must be kept clear unless a train is actively traversing the mainline/ladder sections.

Scenario: A train needs to go into staging track #3 for ‘storage’

Turnouts are set so train will enter Staging Track 3

As a train enters the return loop area the ‘appropriate’ mainline will be occupied (based on my prototype this would be Eastbound mainline). (East Mainline ON)

Eventually train reaches ladder section (Front Ladder ON)

Train travels through ladder and enters Staging Track 3 (Staging 3 ON)

Eventually end of train leaves Eastbound Mainline (East Mainline OFF)

Eventually end of train leaves Front Ladder (Front Ladder OFF)

Stop train before it enters Rear Ladder (if overshot then reverse as necessary until Rear Ladder is OFF)

Is this a reasonable design/dream?

Thanks.

DIY solution

If all you ask is to light some fascia LEDS then perhaps this little gem from Rob Paisley's site is right for you.

http://home.cogeco.ca/~rpaisley4/OptoBOD.html

+1 for Rob Paisley

Dear Alan,

+1 reccomendation for Rob Paisley's "Model Railroad and Misc Electronics" website
http://home.cogeco.ca/~rpaisley4/CircuitIndex.html

No matter if you're running analog or DCC, this site has a metric boxcar load of good ideas which give great "launch pad" starters into base-level electronics...

Happy Modelling,
Aim to Improve,
Prof Klyzlr

Add a "Stop" section

I would add a stop section at the end of each staging track.

As soon as this is occupied the train has to stop.
That is much better then try and error (reversing)