False Occupancy Detection?

Occupancy detection has two uses.  At a minimum it allows the monitoring of physical train progress around the layout.  It can be used for layout automation, Dispatching or Signaling.  This section only focuses on current based detection.

DCC brings some additional electrical complexity to locomotive detection process that does not occur under DC operation.  It is these additional complexities that create reliability problems especially on large layouts.   Below are various devices and how to deal with the problem one might encounter when you use them.

IT IS ASSUMED YOU READING THIS SECTION BECAUSE:

1) YOUR HAVING A DETECTION PROBLEMS THAT COULD NOT BE SOLVE BY REDUCING THE OCCUPANCY DETECTOR"S SENSITIVITY.  Any further reduction in detection sensitivity results in unreliable detection.

2) YOUR LOOKING TO AVOID PROBLEMS WHEN INSTALLING OCCUPANCY DETECTION.


COMMON PROBLEMS
If we assume there is no wiring errors or insulated rail joiners (gaps) that have failed, then the culprit is leakage current.  Leakage current is the term used to describe the flow of unintentional current due to sneak path.

There are TWO type of sneak paths.   RESISTIVE and CAPACITIVE

The problem can also be a combination of BOTH a DC and a AC leakage current component that when added together pass enough current to trip the occupancy detector.


RESISTIVE LEAKAGE CURRENT

DESCRIPTION: This problem is found in the form of oils & dirt that contain conductive particles such as metal filings and/or graphite lubrication.  The sneak paths have an accumulative effect.  It is not in one single spot but is spread over an area.  The most intense spots can be around turnout that have received graphite lubrication or equivalent.  The other less obvious is a long stretch of ballasted track that is dirty.  The best preventative solution to this problem is keep the right of way clean.  But if oil and dirt have penetrated into the ballast, there is not much you can do but redo the ballast. It is also possible that some of the material used in the ballasting or painting of the track in question may have some electrical conductivity.    All of this would be the same problem if using DC power instead of DCC power. Stated another way, it does not matter what your power source is, resistance will pass current if its DC or AC.  Does not matter.

TEST:  The two types of test for this problem.  The easiest is to use an Ohm Meter on Ohm mode.  With a scale that can read between 1K and 100K resistance values, read the resistance across the rails.   FOR THIS TEST TO WORK YOU MUST DISCONNECT ONE LEAD TO THE TRACK SUCH THAT IS HAS NO POWER.  YOU MUST VERIFY YOU HAVE NO POWER BEFORE PROCEEDING OR YOU MAY DAMAGE YOUR METER.  If the meter shows a resistance less than 20K (20,000) ohms, you have section of track with a lot of resistive leakage.

LOWER RESISTANCE VALUE = HIGHER LEAKAGE CURRENT.

SOLUTION:  Clean and vacuum the track in question completely and do a good job of it.  You looking for a problem spread over the entire track in question.  Make sure you focus on the turnout areas such as the throw bar.


CAPACITIVE LEAKAGE CURRENT.  (Only applies to DCC) 

This problem is found in how you ran your DCC track bus wiring and track feeders and how long they are.  Again this an accumulative effect.  It is not in one single spot but is spread over the wiring starting from the point of where the detector is located all the way to the end of the wires that connect to the track being monitored.  It does NOT involved the wiring starting from the booster to the input of the detector.

Do not know what capacitance is, go here:Capacitance

TEST:  There is no known test an average person can perform without advance test equipment.

MORE CAPACITANCE = MORE LEAKAGE CURRENT

SOLUTION: There are several variables and corresponding solutions for each.  They are as follows:

A) GOOD WIRING PRACTICES THAT INCREASE CAPACITANCE. 
Check to see if you have implemented one or both of the GOOD layout wiring practices that address track noise problems.  Unfortunately both of them are incompatible with current based block occupancy detection.   Why?  They both increase the capacitance.   They can be used without concern if done before (input side) of the detection device.  It is only a problem if they are done AFTER (Output side) the block detector.

A1) TWISTED WIRE: 
Problem: Wire capacitance increases if the track bus wires are twisted together.  Twisting solves many other electrical problems but using inside a detection zone creates more problems than it solves.  There is a complete discussion about this complete with diagrams found here:
Track Occupancy Current Detector vs Twisted Wiring Capacitance

Solution:  Undo the twisted wiring between the occupancy detector output and the track.

A2) RC FILTERS/SNUBBERS: 
Problem: Wire capacitance increase with the installation of an RC filter.  More specifically the capacitor in the RC filter is electrically connected between the two track bus wires.  The REC Filter solves many other electrical problems but inside a detection zone creates more problems than it solves. 

Solution Remove any and all RC filter that are placed between the occupancy detector and the track.


B) RELOCATE BLOCK DETECTOR LOCATION.
Problem:  To much wiring exist between the occupancy detector location and the track.  It has to much capacitance over it run length. 

Solution: The solution does NOT involve re-wiring.  It involves moving the location of the detector closer to the track that is being detected.  You are in effect reducing the portion of the wiring that is inside the detection.  AKA "local" block detection where the block detector is located at the track.  THIS IS THE BEST SOLUTION AND SHOULD BE THE FIRST THING YOU DO IF POSSIBLE.   THIS ADDRESSES THE ROOT CAUSE OF THE PROBLEM. 

Drawings of remote block detection is shown here:
Track Occupancy Current Detector vs Twisted Wiring Capacitance

With individual block detectors are easily remote-able. The NCE-BD20 is an example of individual block detector.  Centralized Block Detectors like the Digitrax BDL168 and CMRI DCC-OD are harder to do.   This is electrically why such products are not ideal for foolproof detection operation consistently from layout to layout.  Local block detection is guaranteed by design to be more reliable. 


C) DESENSITIZE THE CURRENT DETECTOR FROM CAPACITANCE


This section offer specific discussion on how to do this for two classes of current detection devices.  This is a more technical solution that involves electrical modification of the detection devices themselves.

1) Diode Voltage Drop detectors
BDL168 & BD4 False Detection/Noise

2) Current Sense Transformer detection
Current Sensing Coil Type (BD20, DCD-OD)

7/15/15