[PRCo] Re: MetroRail Signalling

[Schneider Fred]

General Railway Signal first developed an automatic train operation  
concept in the 1960s and demonstrated it on the 42nd Street Crosstown  
subway line of the New York City Transit Authority.   The concept  
allowed a train to run itself.

The first full fledged system of that nature was the Port Authority  
Transit Corporation's line from Philadelphia to Lindenwold, New  
Jersey which opened late in 1968.   Trains simply run themselves,  
stopping at each station.   When the motorman recloses the doors, the  
train will accelerate out ot the station and stop the next station.    
The speed commands displayed to the ATO (automatic train operation)  
system are also displayed to the operator who can run manually if he  
choose.   Protocol on PATCO at least under Bob Korach and Bill  
Vigrass was for every operator to make at least one trip every day in  
road manual at their option.   They could run as many trips in manual  
or as many in ATO mode as the wanted as long as they satisfied the  
rule that least one at day be in manual in order to maintain their  
skill level.

BART, on the other hand, was set up to allow road manual at only a  
maximum speed of 25 mph and manual operation could only be done if  
officially approved by the dispatching staff.   Obviously if someone  
approved manual operation, dropping from a norm of 70 mph and a  
maximum of 80 mph in ATO to a maximum of 25 mph in manual would  
certainly constipate the system in short order.   It normally isn't  
done.   Cars are normally run manually only in yards.   By the way, I  
did once have an opportunity to run a BART train in yard manual, a  
privilege which was set up for me by the late Harre Demoro.

I have not followed the schemes that are used for equipment detection  
within signal blocks today.   However, it should be clear that  
whatever means are used to detect a train running in ATO (automatic  
train operation) should also preempt manual operation under ATC  
(automatic train control).   I have put out some questions on line to  
some people whom I think could answer and have gotten no feedback ...  
perhaps indicating that I need to be more specific or obtain a larger  
hammer.   :<)

On Jun 28, 2009, at 2:58 PM, Phillip Clark Campbell wrote:

> Mr.Alan Schneider;
>
> Thank you very much for this report.  Thank you to the many  
> contributors on
> this subject.  The news reports  'seem'  well done and I was able  
> to glean
> much information from them;  comments from contributors underscored  
> these
> observations and brought much clarity to this disaster.  The  
> postings have been
> most productive  --  'mostly.'
>
> I have changed the subject line and have questions to seek a  
> greater understanding
> of the signaling system.  I found the URL for the included article:
> http://www.fredericknewspost.com/sections/news/display.htm? 
> StoryID=91910
> because I wanted a closer look at the included photo:
> http://www.fredericknewspost.com/photos/09/06/26/91910_large.jpg
>
> It is already established that  'computers'  handle the signals  
> through  'ATC;'  while I
> understand the concept I am not familiar with the details.  My  
> reference is the US&S
> signals as used on PRC interurbans.  Rails were gapped and  
> insulated;  an impedance
> bond allowed ground return around the insulated rails while keeping  
> the signal circuit
> current within the block.
>
> 'Parallel'  construction is evident in the MetroRail photo but the  
> rails are not gapped and
> insulated are they.  Please note the dual set of cables between the  
> rails in each track
> that arc through a half circle before attachment to each of the  
> rails;  this  'suggests'  an
> insulated gap in the rails which, again, is not clearly evident.   
> There are far more cables
> between the rails than the parallel US&S impedance bond.
>
> Also noteworthy is the large gap in the 600-volt(?) 3rd-rail.  As  
> the cabling between the rails
> of each track seems identical and the 3rd-rail is gapped on both  
> tracks I am  'assuming'  that
> such a gap in the 3rd-rail exists wherever such  'signal cables'   
> exist.   Is this true?  Why is
> it necessary to gap the power supply at signal cabling locations?   
> Why aren't the running
> rails gapped, or are they?  I assume each truck has a 3rd-rail shoe  
> on each side;  it then
> 'seems'  that current is never disconnected from the train as the  
> gap  'seems'  shorter than
> truck spacing, or is it?  The right hand track has ties that extend  
> further than standard
> 'apparently'  to accommodate 3rd-rail in the gap area;  the same  
> may be true of the other
> track but the ties could be covered by the ballast.
>
> If someone is able to give some clarity to the computer style ATC  
> signals circuits it will
> be most appreciated.  Thank you to one and all for such efforts.
>
>
>
>  Phil
> Without  a   'coast'   but  not  a   'cause.'
> --  --
> "If thou wouldst rule well, thou must rule for God,
> and to do that, thou must be ruled by Him ...
> Those who will not be governed by God...
> ........will be ruled by Tyrants."
> William  Penn,  founder  of  Pennsylvania
>
>
>
>
>
> ________________________________
> From: "alschneider2 at juno.com" <alschneider2 at juno.com>
> To: etennyson at cox.net
> Cc: rejmhj at netzero.net; j_swindler at hotmail.com; pcc_sr at yahoo.com;  
> pittsburgh-railways at dementia.org; jack.may at americomm.net; gary- 
> elaine at comcast.net; nawdry at bga.com; crvlkotula at aol.com;  
> philgcraig204 at yahoo.com; billvigrass at verizon.net;  
> bob.dietrich1 at verizon.net; bbente at bellsouth.net;  
> trams2 at comcast.net; allmanr at verizon.net; shadow at dementia.org;  
> akftrain at aol.com; miklosfrank at comcast.net;  
> jaurelius at centurytel.net; russell.jackson at stvinc.com
> Sent: Sunday, June 28, 2009 10:27:36 AM
> Subject: Re: METRORAIL DISASTER-DEFFERED MAINENANCE?
>
>
> FrederickNewsPost.com
>
> Source: Metro circuits were malfunctioning for five days
> Originally published June 26, 2009
>
>
> By Marge Neal
>
>
>
>
>  Photo by Associated Press
>
>
> Workers install new fencing Wednesday in Washington, at the site  
> where two Metro trains crashed Monday.
>
>
>
>  Two track circuits in the area of Monday's deadly Metro crash  
> alternately reported the track as vacant and then occupied almost  
> constantly from June 17, when a part of a signaling device was  
> replaced, until the time of the crash, according to a Metro source.
> The circuits appeared to have intermittently lost track of the  
> trains involved in the crash of two Metro trains that killed nine  
> people, according to Metro documents obtained by The Frederick News- 
> Post.
> The documents show what is known as "flip-flopping," with the  
> circuit alternately reporting the track as "occupied" and then  
> "vacant." The report for track circuit number B2-304, where the  
> stopped train was located, starts at 4:45 p.m. and ends at 5:38  
> p.m., about 40 minutes after the crash.
> Throughout the time period covered by the report, the circuit shows  
> the track either occupied or vacant as often as every couple of  
> seconds.
> At 4:57:03 p.m., about the time of the crash, the report shows the  
> track to be vacant until 5:37:44, when it reports the track is  
> occupied. Less than a minute later, the report again states the  
> track was vacant, even though train wreckage was on the track. One  
> second after that, the track was reportedly occupied again. The  
> circuit flip-flopped again before finally showing the track to be  
> occupied at 5:38:40.
> On June 17, a part of the track signaling device known as a Wee-Z  
> Bond was replaced in the area of the crash, according to a Metro  
> source who spoke on the condition of anonymity. The Frederick News- 
> Post agreed to the request because of the compelling nature of the  
> information.
> The two track circuits involved in the crash -- B2-304 and B2-312  
> -- share the bond that was replaced. The component feeds the  
> receiver for one and gets the signal from the transmitter for the  
> other.
> Both circuits continued to alternately report the track as vacant  
> and then occupied almost constantly from June 17, when the part was  
> replaced, to the time of the crash, according to the Metro source.
> "Without seeing the schematic drawings of the (Wee-Z Bond), I  
> cannot say it failed or contributed to this crash," Najm Meshkati,  
> professor of civil and environmental engineering at the University  
> of Southern California, said Thursday. "I can say the system did  
> not get back to its normal operating mode after the replacement was  
> made."
> The replacement part did not perform its intended purpose, Meshkati  
> said. The part could have been defective, or it could have needed  
> to be recalibrated or reset to work properly, he said.
> Metro, he said, needs to take a "hard look" at its systems safety  
> practices and philosophy.
> "It is important not to blame," he said. "It's important to find  
> the root cause of the problem and change the safety culture of the  
> organization."
> The Frederick News-Post has also learned that Matthew L. Matyuf,  
> superintendent of the Automatic Train Control division, was  
> relieved of his duties after the crash.
> "I can confirm that the ATC superintendent has been temporarily  
> assigned to a special project," Lisa Farbstein, Metro's chief  
> spokeswoman, said Thursday. "It happened post-incident, and I do  
> not know what the project is."
> Wednesday night, investigators from the National Transportation  
> Safety Board conducted tests using a similar train at the crash  
> site, according to a press release from the agency. When the train  
> was stopped at the same location as the stopped train that was hit,  
> the train control system lost detection of the test train.
> Investigators are continuing to examine trackside circuits and  
> train control system data to understand how the train control  
> system functioned on the day of the crash, according to the release.
> Recorder data have been recovered by the struck train. The striking  
> train did not have any onboard accident data recorders, according  
> to NTSB.
> The operator of the striking train, Jeanice McMillan, was killed in  
> the impact. NTSB investigators found metal-to-metal compression  
> streak marks on both rails of the track for about 125 feet ending  
> near the point of impact, consistent with heavy braking, the  
> release states.
> McMillan apparently pressed a "mushroom" button to activate an  
> emergency brake.
> The operator of the struck train was interviewed Thursday morning,  
> according to the release. He told investigators that he operated  
> his train in manual mode during his entire shift that afternoon. He  
> said he had stopped to wait for a train that was stopped in front  
> of him when his train was struck.
> NTSB said investigators will continue to document evidence at the  
> accident scene and intend to conduct sight distance tests using  
> trains similar to those involved in the crash.
>
>
>
>
>