Not necessarily. A firm I worked for tested them once a month. On one occasion the Facilities Manager thought it would be smart to put the batteries on boost charge afterwards. Then he forgot and they boiled dry.
A few days later there was a real power cut and the genny failed to start leaving entire complex without any power.
The battery chargers for the railway standby generators that I have seen are fully automatic types. Two chargers and batteries for the control system and two chargers and batteries for the starter motor. All four chargers are fed from the load side so that the batteries recharge from both the mains and from the generator supply.
Surely the ROC must have a UPS with the batteries capable of keeping it going for a few hours should the generator fail to start. I suppose the problems come when track circuits, axle counters etc lose their local supply. Are signal heads locally powered with standby batteries?
The UPS battery capacity is only normally rated to supply the load between the time of the loss of the mains (DNO) supply and the time the standby generator takes to start and get up to speed, plus a "safety" margin.
Track circuits and signals and the control equipment (remote from the signalling centre) do not have any kind of local/individual battery back-up. Some points may have a battery to reduce the load on the power network when they operate, others may not use any battery.
Axle counters do normally have a battery backed supply so that they don't loose the axle count during a short duration loss of power.
At Three Bridges ROC, in the event of power failure there is a Battery UPS that catches the loss of supply and then gives the huge Diesel Generator time to spool up and settle down before the UPS hands supply duties over to the generator (a matter of a few minutes). When I was a Radio Tech I was tasked with babysitting the GSMR system, whilst we did a power outage test. There were other Telecoms techs present and a couple of members of the D&P to ensure if it didn't work as designed then we could get it back on the DNO supply and reboot any systems that fell over. It worked fine when we tested it.
Yes, at one time there was actually a written instruction that S&T should be on site when a "crash test" was to be conducted by the D&P. I think it was in the general instructions to S&T staff.
The flip side is that it can take a certain scale before redundancy is worth it. Does every individual signalbox have generator backup? Whereas with something of the scale of the ROC you absolutely will put in UPS, generator backup, redundant grid feeds etc.
No. PSB and signalling centres (including ROCs) are required to have a standby generator, but it's optional for traditional signal boxes. Where a signal box has not been modernised or is not considered important enough, it will not have a standby generator.
Some signal boxes that have been modernised, may have both a standby generator and a UPS, with the UPS only powering for example computer based interlocking such as SSI equipment.
Having previously worked in Facilities Management, incidents like this massively trigger me. I can regale people with numerous tales of poorly designed power failure systems and processes, poorly maintained UPS and generator systems and power system interlocks that haven’t ever been tested in anger. I’m just thinking of some poor maintenance tech, the FM and his mates at the moment trying to get a stable power supply back on whilst the world and his wife are shouting down the telephone at them.
Been there. Total loss of signalling on part of the South Wales main line due to part of the 650V AC network having failed. Every time I picked up my multimeter to try to see what if any voltage there was, my company mobile ' rang... Someone else wanting to know what was going on... After this happened the third time, I put the 'phone on the floor and whacked it so that it slid to my workmate (it was a floor in a building).
Switching from a good balanced 3 phase from the generator to an unbalanced mains (e.g. 2 good + 1 struggling phase e.g. lower volts and unusual waveform) is good recipe for stuff going wrong. [I've seen this one before when an EA contractor repairing a flood defence damaged a DNO cable causing issues on just a single phase, it took a lot of effort to work out what was going on. The net result was 3 phase equipment playing up at times of day then the local loads were high]
Experience - presumably with single phase supply? Single phase is much simple to deal with when thing have gone wrong!
The control circuitry that is part of the standby generator system includes under voltage detection for each incoming phase. If one phase is lost, it treats it as a complete loss of the DNO supply.
This is another reason why the transfer back to the mains DNO supply is done manually. No point in switching back to a DNO supply that can't support the load. That just causes even more problems. An unstable or intermittent supply will upset both computer based interlocking and signalling systems and relay based interlocking and signalling systems. Drivers and signallers are not keen on signals flashing red, black, red, black etc... And that can cause fuses to blow and MCBs to trip out. Or cause actual signalling equipment to fail.
Not just a business case, but also inspire a sense of urgency.
I've noticed how quickly problems caused by points and signals failures are resolved by NR, to reduce train delay as much as possible. The delay minutes counter (and the costs associated with it against NR) whizzing round certainly seems to motivate speedy responses and trains quickly on the move again.
When that internal financial motivation has gone perhaps we'll just have to wait on trains while the convenient, but slower, approach is adopted.
No, it's never that simple. When I joined the BR S&T department in the late 1980s, the mantra was to do everything possible to keep trains moving as long as it was safe to do so and to minimise delays.
Privatisation upset that. We were instructed not to use our own initiative but to just report any problems that we found rather than trying to fix them on the same shift. In other words, we were only to do the work that we had been tasked with. The manager would put the reported defects in the job bank.
Unfortunately many times, by the time the job got planned, I would arrive on site to see some shiny new equipment or that the problem had already been fixed. At some point between being reported and the work being planned, the problem had escalated into a failure (often causing train delays) and hence the problem had been fixed before I turned up to do the planned work...
Network Rail has since introduced risk based maintenance, as in we do less maintenance and less often to save money if the powers that be think the risk of a train delaying failure is low (which they do unless the equipment has a history of train delaying failures). But of course, this is a risk, as sometimes the equipment does fail...
Because less maintenance is being carried out, there is now less maintenance staff. So when it does go wrong, there are less staff available to fix the problems...
And compared to Railtrack, there are less MOMs (Mobile Operation Managers) to manually operate points, work as pilots (single line working), act as handsignallers or work signal boxes.
Oh, nearly forgot, Network Rail managers have a delays budget. If the number of delays is within the budget, there is less pressure on. Conversely, if the delays are far exceeding the budget, they may hold a special meeting to try to find the underlying cause.
