Defect Detector

[louisnash]

I had my scanner at work on Tuesday and was listening to the defect detectors call out "No Defects".

But the funniest thing happened. Where our shop is located, we are between 2 defect detectors. They both went off at the same time, due to the volume of rail traffic we have.

One cancelled out the other on the frequency I was listening to. What happens if there is an actual defect and while they are transmitting it is cancelled out by the one that may not have a defect. Will the one with the defect supersede the one without?

I did not have a clue this could happen, but I am assuming that the trains were just in those locations at the same time. It was kind of odd to me to hear that.

Thanks,
Brian (KY)

[ChuckCobleigh]

Assuming that the engine is much closer to the detector that is working on that train, then the signal from that detector will probably be significantly stronger than from the other detector. Even if they were using AM, the closer detector would over-ride the farther one. If the detector radio uses FM, then the closer detector will definitely be the only one heard. It is my understanding that defect detectors are about 20 miles apart mostly, which means that the difference in received signal power (active versus other detector) would be nearly 20 decibels at the head end of a 10,000-foot train. This is a big enough difference (100:1 in power) so that even an AM system would not be rendered unintelligible. FM is more immune, requiring a much lower signal-to-jammer ratio, which in practical terms means that the cab receiver wouldn't even know the other defect detector transmitted.

When a receiver is located between two transmitters on the same frequency, if the transmitter powers are approximately the same and the distances are about the same, then they will step on each other in some significant manner, wiping out both transmissions. That is probably what is happening at your shop. For instance, if the two transmitters are 20 miles apart, if you are on a direct line between the transmitters, you could expect an AM wipeout if you were 5 to 15 miles away from one transmitter, or 8-12 miles for FM. These numbers would vary depending on terrain and frequency, of course, but the general concepts are in the ball park.

The saving grace, here, from the system design point of view, is that the head end of the train is always going to be relatively close to the detector transmitter and relatively far from the next detector. Nature takes care of the rest.

[chad thomas]

Plus the detector that catches a defect will "talk" alot longer than the one that didn't.

[mudchicken]

...and an awful lot of them will clutter up a dispatchers VDT screen reporting back to the DS that a defect has been encountered.(at least on BNSF)

[mvlandsw]

I have heard the detectors for two different tracks at the same location trying to talk at the same time. It gets confusing. If the report is not heard clearly the rules require a stop to inspect the entire train.

[ChuckCobleigh]

QUOTE: Originally posted by mvlandsw

I have heard the detectors for two different tracks at the same location trying to talk at the same time. It gets confusing. If the report is not heard clearly the rules require a stop to inspect the entire train.

That doesn't have to happen.

System design could include a receiver with each transmitter to detect when somebody else is talking. (Actually, receiver is not entirely accurate; only that part of what we call a receiver that detects the presence of an RF signal is really necessary.) If another carrier is on, the transmission is held off until the other carrier goes away. It's a pretty simple concept (called generally "carrier sense") and can also include something called collision detection as a part of an intelligent retry protocol.

Even better is using one transmitter for the two closely adjacent defect detectors and eliminating the overlap condition, or at least routing a token line between the close detectors so that only one talks at a time.

I'm kind of surprised that this particular situation hasn't been dealt with already, if in practice it does pose a problem.

[chad thomas]

"collisions" are not really a problem because when a defect detector catches a problem it transmits significantly longer.

example non defect

SP detector milepost one one one point one no defects no defects

With defect

SP detector milepost one one one point one stop your train stop your train defect axle number one one one on left side stop your train stop your train....SP detector milepost one one one point one stop your train stop your train defect axle number one one one on left side stop your train stop your train

[tree68]

Haven't actually seen them, but there are defect detectors on both EB and WB tracks of the CSX Chicago Line just west of Syracuse. I can imagine both getting hit at the same time (haven't heard it happen, though). Unless they share a transmitter, they would indeed interfere with each other. I may know someone I can ask. MC probably knows, too, at least for other locations.

[Anonymous]

On the BNSF you can 'dial up' a detector for a rebroadcast of it's latest message.
The detectors usually have a fairly limited range, for example detectors at MP 3.4, 8.6, 16.3 on the Creston sub generally don't bother each other, but sometimes they will 'step on' transmissions between a train and the disapatcher, or train to train if they're all close enough to the dectector.

My personal favorite dectector message, as a 115 car coal empty with 2 sd70s crosses it on a cold December morning:

BNSF detector - milepost 16.3 - no defects repeat, no defects - total axels 8 -
train speed 38 mph - temperature 165 degrees - out

[:D]

[csxengineer98]

as long as their is no BEEP BEEP BEEP...and you will hear as soon as the radio clutter clears up.... you just go on to the next detector... but one that is closer will overpower one that is down the road some miles... so its not as big an issue when your on the engin as it is when your sitting in your car listing to a scanner....
csx engineer

[Anonymous]

What happen when the detector announces a defect. Does the engineer stop the train then? Always been curious about that.

Larry

[chad thomas]

Yes, the train is stopped. Then the conductor goes back to inspect the axle indicated (and a few in both directions if no defect is found). When the car with the problem (hotbox) is located they then move at restricted speed to a suitable place to set the car out.

It isn't always like that. Sometimes special circumstances dictate taking other actions.

[CShaveRR]

In UP's Special instructions, there's a whole matrix showing what actions are to be taken when a defect is reported, depending on the nature of the defect, the nature of the detector, and the nature of the train. Generally, the trains do not stop immediately, but should at least clear the detector (but there are exceptions to that, too!).

[Anonymous]

Thanks for the info!

[Anonymous]

I don't know if this was ever posted befor, but laws state that a detector must be placed every 40 miles, but most railroads place on every 20. So, if there is a problem with the transmission, or a broken detector, they can just wait till they pass the next detector (with out inspecting/stoping).

[louisnash]

QUOTE: Originally posted by Brak710101

I don't know if this was ever posted befor, but laws state that a detector must be placed every 40 miles, but most railroads place on every 20. So, if there is a problem with the transmission, or a broken detector, they can just wait till they pass the next detector (with out inspecting/stoping).

I know that the ones that I listen to are at 14.7 and 20.0. That's not that far apart and it's only single track.

Is there a reason that NS does not call out total axles on the DD? I know that CSX will call out total axles on the CC Sub in Ky.

I have never heard one call out a defect on either RR.

So what happens on NS when there is a defect? Will it call out the axle count then.

Thanks for all the replies to the thread.
Brian (KY)

[mvlandsw]

QUOTE: Originally posted by ChuckCobleigh

QUOTE: Originally posted by mvlandsw

I have heard the detectors for two different tracks at the same location trying to talk at the same time. It gets confusing. If the report is not heard clearly the rules require a stop to inspect the entire train.

That doesn't have to happen.

System design could include a receiver with each transmitter to detect when somebody else is talking. (Actually, receiver is not entirely accurate; only that part of what we call a receiver that detects the presence of an RF signal is really necessary.) If another carrier is on, the transmission is held off until the other carrier goes away. It's a pretty simple concept (called generally "carrier sense") and can also include something called collision detection as a part of an intelligent retry protocol.

Even better is using one transmitter for the two closely adjacent defect detectors and eliminating the overlap condition, or at least routing a token line between the close detectors so that only one talks at a time.

I'm kind of surprised that this particular situation hasn't been dealt with already, if in practice it does pose a problem.

They are designed to work that way but something must have malfunctioned to let them both talk at the same time. If someone is transmitting the detector will usually wait before sending its message.

[Anonymous]

At the 5.5 mile mark of the CSX, they have 2 defect detectors, and can be heard at the same time when two trains go over them. I have listened to them both on 160.230 frequency of the CSX. There have been times I would be going toward the town of Dolton and be stopped by either the Indiana Harbor Belt RR, or the CSX or both, and listened to the frequencies sound off.
By the way that too is another great place to go out and watch the railroad action going through there, considering ther are 3 rail yards and the lead to a fourth yard of the IC/GT/CN.

ralph zimmer [e-mail] (new email address)