West Somerset Railway - Bishops Lydeard to Minehead (merged posts)

[whole diary]

[stuving]

I'm concerned about the GPS side of things.
Occasionally in incident reports there are mentions that the drivers GSM-? didn't work, or there was no radio signal in that area.
Everything works well when there is a signal.
There are many areas around the UK^▸ (United Kingdom) with either no signal or a very poor signal level.
Perhaps too much dependence on technology?

The whole point of the Rubust train Positioning System (RTPS) is that it combines several sensors so that it always has a fallback if any of them is not available.

For example, if you have an absolute positioning system like GPS, and one with no external inputs like an odometer, they complement each other. The ododmeter (counting wheel rotations) drifts due to uncertainty about the wheel radius, but only slowly so the position error remains small for many miles. GPS can be obstructed, but when you get it back is drift-free and can be used to remove the position error that's built up due to using the odometer. You can add extra sensors to cover for other kinds of outage/failure, in this case using a radar rather than an odometer, or balises.

The same method is now standard for airliners, combining GNSS (including GPS) with an inertial navigator (INS or IMU, also part of this RTPS). Again, the INS suffers from slow drift and complements the potential error modes of the GNSS (to do with integrity as much as errors). That depends on a bit of magical chicanery called a Kalman Filter, which I think is also used for RTPS.

[GBM]

Having looked up Kalman Filter (https://en.wikipedia.org/wiki/Kalman_filter)
I'm even more out of my depth!

Thank you for your reply, I thought it was complicated before I queried GPS/GRPS!

[CyclingSid]

Interesting if used in the north of Scotland when MoD have their GPS "spoofing" exercises!

[broadgage]

The West Somerset railway have restored to largely original condition a vintage "gunpowder van"
A local resident complained in the local pub about the storage of gunpowder within sight of a childs bedroom.
I explained that the vehicle was to add historical authenticity to the railway, and did NOT actually contain gunpowder. They seemed unconvinced. Despite me suggesting that the WSR have no need for gunpowder, and that if gunpowder WAS needed for some exceptional reason, that this would be be obtained, stored, and used by an licensed explosives expert and NOT left in an unattended goods van.

I wonder that they think about the petrol tanker ! That was nearby.

[broadgage]

The recently renewed Seaward level crossing was not working correctly this afternoon, so far as could be seen from the relevant webcam.

The diesel hauled 18-00 arrival into Minehead came into webcam view at about 18-10. Approached the crossing very slowly indeed and then stopped at the signal just on the London side of the crossing. Someone alighted from the cab, presumably to use the telephone.

After a few minutes, THREE out of the four barriers lowered, but the fourth one remained up. After a few more minutes the train passed over the crossing at walking speed, with one barrier still up.
One would presume that a lookout was halting the road traffic, but this was out of webcam view.

[broadgage]

Still faulty this morning.
10-00 diesel departure from Minehead held at the new semaphore signal, three only out of four level crossing barriers then lowered.
Semaphore signal remained at danger. Passed at danger as a hand signalman showed a green flag.

[IndustryInsider]

Still faulty this morning.
10-00 diesel departure from Minehead held at the new semaphore signal, three only out of four level crossing barriers then lowered.
Semaphore signal remained at danger. Passed at danger as a hand signalman showed a green flag.

Presumably the operators of heritage railways follow the same procedures as Network Rail do when level crossings fail?  From your description it sounds like it.

[broadgage]

Yes, so far as I know, heritage railways do follow the same or very similar rules as apply on the national network, in case of level crossing faults.

The Seaward Lane level crossing was operating correctly today, so far as it could be seen from the webcam at about 13-55.

[The Tall Controller]

The problem was caused by a failed hydraulic power pack in the barrier. As broadgage reports, the issue has now been fixed. I certainly hope so as I'm the signalman there tomorrow!

[broadgage]

Here is a bit of technical trivia, brain teaser.

The Seaward Lane webcam gives of a good view of the now functioning correctly level crossing and of the nearby semaphore signal.

Under certain conditions, the light in the signal appears to flicker or cycle between dim and bright, sometimes appearing to go e out for a second or even several seconds.

I believe that I know the reason for this, but would anyone else care to suggest the reason ? Noticeable right now, at nearly 17-00.

[Surrey 455]

Here is a bit of technical trivia, brain teaser.

The Seaward Lane webcam gives of a good view of the now functioning correctly level crossing and of the nearby semaphore signal.

Under certain conditions, the light in the signal appears to flicker or cycle between dim and bright, sometimes appearing to go e out for a second or even several seconds.

I believe that I know the reason for this, but would anyone else care to suggest the reason ? Noticeable right now, at nearly 17-00.

Would it be anything to do with the frame rate on the camera? In a similar way to videoing a dot matrix display which appears readable to the eye but the camera is taking perhaps 50 frames per second and on some of those frames the dots are off for a fraction of a second?

[IndustryInsider]

Yeah that’s pretty much it.  The LED light flashes on and off quickly (too quickly for the human eye to see) but if the recording device is recording at a different frame rate then it will record some of its frames when the light is on and some when it’s off.

60Hz vs 50Hz is usually the case I believe?

You might be able to stop it on your own camera by changing frame rates from 30/60fps to 25/50fps.

[paul7575]

I think the buzzword is modulation. 

But for older readers, does anyone remember the effect you often got on older TV sets when they played old black and white movies, and spoked wagon wheels would look as though they were going round backwards.  That’s the same principle, the original film frame rate, and the TV frame refresh rate would be in conflict.

Paul

[broadgage]

Yes, both the above are correct. The LED must be lit either from AC or perhaps rectified but not smoothed AC, that will result in the light flashing at twice line frequency or 100 times a second.

The webcam is probably 25 frames a second. As the mains frequency or the camera frame rate varies slightly then the light might be on or off when the camera "shutter" is open.

Only noticeable in dull daylight. In bright conditions the light is presumably lit as normal, but can not be seen as it is drowned out by the bright daylight.
At night, each frame captured by the webcam will involve the "shutter" being open for longer and spanning several flashes of the LED.

[broadgage]

BTW^▸ (by the way) the apparently flickering light was much talked about by a group of railway enthusiasts in the local pub, including both national network staff and WSR volunteers.

Not one of whom was able to deduce the correct answer.

"Oil lamp flickering due to wind"
"Oil lamp badly adjusted and flickering without any wind"
"Electric light flickering  due to bad connections"
"Insects in the lamp housing partially blocking the light"
" Some new and different meaning of the signal, now THREE aspect, green, steady red, flickering red !"

And other even more complex and improbable suggestions.
No one thought to go and view the signal in person, it was within walking distance of the pub in question.

But at least three members here quickly deduced what was happening. And others may well have known what was happening.