Our new ticketing site is now live! Using either this or the original site (both powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!
Interesting - the circuitry we were given shows a second voltage-free contact on the lamp-proving relay going to the monitoring equipment in the supervising box. with no further current sensing. I'm unclear how you could further sense the current in a trackside Location cabinet or relay room - it would have to be done on the 100V supply at about a quarter the current - plus whatever the transformer takes on its own. Is it possible this sensing was developed after our 1970s heads were put into use?
The volt-free contact that you mention is the first-filament fail alarm to warn the techs, so that they can replace the bulb (hopefully before the backup filament fails):
The common bulb of the past few decades is the SL35, with two independent filaments each rated at 24 watts. There’s a small relay in series with the “main” filament: if that filament fails, the relay drops out and changes over to the “auxiliary“ filament. That then triggers an alarm to the local technicians that one of a group of signals is working on its ‘auxiliary“ filament and needs attention.
The LAMP proving (as opposed to the first-filament proving) is done in the trackside equipment case. As I explained in post #21, with a relay interlocking, this would be done with a current-sensing relay (BR Spec 941 if I remember correctly), wired in series with the 110v AC signal feed. This relay has a small transformer/rectifier built into the same case as the relay. With SSI, the signal circuit is wired back to an input on the signal module, where it is fed through a resistor, and the voltage drop across this resistor is detected.
Lamp proving was in use long before the SL35 signal heads were developed.
The LAMP proving (as opposed to the first-filament proving) is done in the trackside equipment case. As I explained in post #21, with a relay interlocking, this would be done with a current-sensing relay (BR Spec 941 if I remember correctly), wired in series with the 110v AC signal feed. This relay has a small transformer/rectifier built into the same case as the relay. With SSI, the signal circuit is wired back to an input on the signal module, where it is fed through a resistor, and the voltage drop across this resistor is detected. Lamp proving was in use long before the SL35 signal heads were developed.
BR spec 941A for AC lamp proving, according to this comprehensive list of relay types supplied by Mors Smitt. https://www.morssmitt.com/uploads/files/page/bro-q-style-relays-v2-6.pdf
In SSI, ISTR lamp proving is activated by enabling current sensing on one or both of the output current return terminations. That is achieved by adding certain configuration straps in the TFM plug coupler.
In SSI, ISTR lamp proving is activated by enabling current sensing on one or both of the output current return terminations. That is achieved by adding certain configuration straps in the TFM plug coupler.
Each SSI signal module actually has 4 lamp-proving current-return inputs (Current Paths):
CP1 (for SL35)
CP2 (for 2 * 60w lamps)
CP3 (for a traditional "feather" route indicator)
CP4 (for SL35)
However, you can only use one out of CP2, 3 or 4, so only two can be used on any one module (CP1 and either CP2, 3 or 4).
As you say, each input is configured for either its standard current-proving value, or for a non-standard value using an external resistor, using the configuration straps on the plug-coupler.
Gentlemen, thanks very much for your comments; one continues to learn!
To Bald Rick in particular, the extra wiring was to feed the 5W tail-lamp bulbs without resorting to using the 100/12V transformers, whose condition was unknown to us; the signals had been in a Network Rail store for an unknown length of time before they were donated to us. In addition this was one of our earliest demos (2011) and we were rather concerned about having 100V around out of doors. The board with all the non-standard wiring contains three small relays switching the bulbs as required. These boards have not enjoyed living in a signal head. In addition the light-weight signal cable running in ducting nearby to link the relays to the sequencer (sited in the box) that runs the display has been subject to rodent attack on occasions.
So I'm currently engaged in rewiring the whole lot with more substantial cables back to the sequencer which is now located near the signals in a proper Location Cabinet and have also done away with the relay boards by modifying the sequencer to drive the bulbs directly.
We have also found that the 5W tail-lamp bulbs with a smaller filament than the proper SL35s can only be seen from a very narrow angle of view, so they are to be replaced with 3.5W LED bulbs as part of the rewiring.
I hasten to add that all our colour light signals are sited well away from the nearby MML so there is no chance of us accidently misleading drivers!
We have also found that the 5W tail-lamp bulbs with a smaller filament than the proper SL35s can only be seen from a very narrow angle of view, so they are to be replaced with 3.5W LED bulbs as part of the rewiring.
Railway signals do have a very narrow angle of view anyway, certainly compared to traffic lights. The signal lens usually has a special segment for close-up viewing by drivers stopped at the signal. Some older signal heads had extra bulbs mounted on the side of the head (nicknamed "pigs ears") for drivers stopped at the signal.
Thinking about your earlier comment about the wiring diagram you were given not showing external proving. Most (if not all) signal heads include a wiring diagram, pasted to the inside of the head. If this is the diagram you are referring to, it would only show the internal head wiring.
If the lamps were AC driven, you could put a capacitor in series with the rectifier that supplies the LEDs themselves.
Would cause them to have absolutely appaling power factor though, but should draw enough current to trip relays without dispersing huge amounts of extra power.
LEDs run best when cold for both efficiency and lifetime. While large chips are available it is usual to make up a light using lots of smaller elements often in series parallel strings. This enables the elements to be spread out over a heat sink surface (with possibly fins on the back) to keep devices cool; although this would not be good as filament lamps replacement because of the optics going from a near point source to a lit surface plane. However elements can also be obtained with lenses already attached to give a relatively narrow beam and of course you will use LEDs of the required colour so there is no need for either a front lens or coloured glass.
Having several parallel strings eliminates the need for a backup as the failure of one string just reduces the light level. (I have seen this happen on road traffic lights where there is a black line across one of the lights where a sting has failed.) I do wonder if signals use circuits to detect if a single string goes open circuit and so alert maintenance.
The improvements in power saving must be massive as long life filament bulbs are notoriously ineficient and you generate light at all colours then throw away most with a colour filter. You may achieve a power saving of 20 times. This could have useful implications when considering things such as backup power supplies such as generators or battery's.
The improvements in power saving must be massive as long life filament bulbs are notoriously ineficient and you generate light at all colours then throw away most with a colour filter. You may achieve a power saving of 20 times. This could have useful implications when considering things such as backup power supplies such as generators or battery's.
It was posted above that a typical filament signal lamp is 24 watts. At any time the railway has two or three of those illuminated per mile, say 100 watts which would be costing about 2p per hour in energy charges. If the power saving with LEDs is proportionate to what it is with domestic lamps then it would be about 1.5p per mile. Both figures are somewhat inflated by losses in their power supplies
A typical street light is at least 50W with LEDs, the sodium technology that used to be used ranged from 120W to 1kW, and there will be twenty or more of them per mile of illuminated road. So the power saving from substituting LEDs on railway signaling is essentially peanuts - replacing a couple of street (or station platform) lights with the latest LEDs will save more energy than replacing signals on a mile of railway.
As someone pointed out above, the real benefit of LEDs is their longer life meaning less cost and safety hazard from having to send people trackside to replace them. The lower failure rate also means some extra safety benefit as it's less likely a driver will encounter a black signal or have to be talked past one - either of which has a small chance of causing an accident.
Railway signals do have a very narrow angle of view anyway, certainly compared to traffic lights. The signal lens usually has a special segment for close-up viewing by drivers stopped at the signal. Some older signal heads had extra bulbs mounted on the side of the head (nicknamed "pigs ears") for drivers stopped at the signal.
Thinking about your earlier comment about the wiring diagram you were given not showing external proving. Most (if not all) signal heads include a wiring diagram, pasted to the inside of the head. If this is the diagram you are referring to, it would only show the internal head wiring.
One problem we found with the narrow angle of view, while prototypical, was that a group of half a dozen or more visitors mostly got only a partial view of the aspects.
The wiring diagram inside the heads was our only ref until I got hold of some of the ISRE 'Green Books' - but the Signal Control Circuits book doesn't appear to mention the LOC-located external monitoring, hence my lack of awareness of it.
Having several parallel strings eliminates the need for a backup as the failure of one string just reduces the light level. (I have seen this happen on road traffic lights where there is a black line across one of the lights where a sting has failed.) I do wonder if signals use circuits to detect if a single string goes open circuit and so alert maintenance.
The improvements in power saving must be massive as long life filament bulbs are notoriously ineficient and you generate light at all colours then throw away most with a colour filter. You may achieve a power saving of 20 times. This could have useful implications when considering things such as backup power supplies such as generators or battery's.
As mentioned previously, most LED signal heads are designed as a near like-for-like replacement for SL35 heads, so there is no power saving. The one area where they differ is that there is no equivalent of the first-filament failure monitoring. So there is nothing to alert maintenance if one string fails (assuming that they are wired up as parallel strings).
== Doublepost prevention - post automatically merged: ==
One problem we found with the narrow angle of view, while prototypical, was that a group of half a dozen or more visitors mostly got only a partial view of the aspects.
The wiring diagram inside the heads was our only ref until I got hold of some of the ISRE 'Green Books' - but the Signal Control Circuits book doesn't appear to mention the LOC-located external monitoring, hence my lack of awareness of it.
Which of the Green Books do you have? I have two that are relevant:
No.7 Signal Control Circuits. My version (priced 3/6d) is the Third Edition Revised 1968, but the content is obviously much, much older than that. Books 7 and 8 (Typical Selection Circuits) do seem to correspond very closely with pre-war Westinghouse miniature lever-frame practice. It is obvious that the author thought there was nothing wrong with signals using non-proved single filament bulbs, for example. The section on Lamps for Colour-Light Signals shows the signal being indicated by an indication lamp wired in series with the main signal lamp, as was standard practice with Westinghouse miniature lever-frames.
No.15 Circuits for Colour Light Signalling. My version (priced 6/- 30p) was published in 1969, and largely superceded No.7. It was obviously written when the SL35 bulb was relatively new, and the SL17 was still in widespread use. It has some wondrous circuits where both filament and lamp-proving are done external to the head (figs 15 and 16), that I have never seen used. However, figure 17 shows what I have always known as the standard SL35 wiring.
As mentioned previously, most LED signal heads are designed as a near like-for-like replacement for SL35 heads, so there is no power saving. The one area where they differ is that there is no equivalent of the first-filament failure monitoring. So there is nothing to alert maintenance if one string fails (assuming that they are wired up as parallel strings).
When the Dorman LED heads came out, the first production examples in UK, the company also produced a special light output measuring device to ascertain if the array was ok or dimming below acceptable levels. I'd guess you'd want to test them all periodically, but at fairly long intervals, and especially as they get older. The latest LEDs are supposed to have >10years of expected service life, zero maintenance, self-cleaning lenses. It's a world away from the endless proactive and reactive changing of filament lamps of old.
Which of the Green Books do you have? I have two that are relevant:
No.7 Signal Control Circuits. My version (priced 3/6d) is the Third Edition Revised 1968, but the content is obviously much, much older than that. Books 7 and 8 (Typical Selection Circuits) do seem to correspond very closely with pre-war Westinghouse miniature lever-frame practice. It is obvious that the author thought there was nothing wrong with signals using non-proved single filament bulbs, for example. The section on Lamps for Colour-Light Signals shows the signal being indicated by an indication lamp wired in series with the main signal lamp, as was standard practice with Westinghouse miniature lever-frames.
No.15 Circuits for Colour Light Signalling. My version (priced 6/- 30p) was published in 1969, and largely superceded No.7. It was obviously written when the SL35 bulb was relatively new, and the SL17 was still in widespread use. It has some wondrous circuits where both filament and lamp-proving are done external to the head (figs 15 and 16), that I have never seen used. However, figure 17 shows what I have always known as the standard SL35 wiring.
I've got No. 7, No. 9 and No. 11 in a single 2007 reprint, and an older copy of No. 7 from 1968 - which precedes No. 15 published in 1969 and so doesn't mention it! Otherwise I might have tried to get a copy. Thanks for drawing my attention to it.
The Howells Mk3 SL35 LED Light Engine is the direct replacement for a Classic SL35/SL35LL filament lamp in a standard signal head. It’s not quite a direct replacement, as it doesn’t use the bayonet contacts and requires a wiring alteration (though, once that is made, changing the LE is a quick process) and sometimes a slight voltage alteration. The changeover relay is replaced with a dummy module, and the Filament Failure indication circuit is no longer operational.
The lamps are coloured as well, so are aspect specific, which simply makes the colour bolder to the viewer.
We are aware of an issue with emails from the Forum to Microsoft-based email accounts (hotmail/outlook/live.com email addresses). This is being looked into currently, thanks for your patience meanwhile.
