Future of the St Albans Abbey line

[mr_jrt]

But you would still have a 45min service unless a passing loop as The Planner says. The line can only survive if patronage rises. A loop at Bricket Wood (think Penrhyn style) looks like the best solution, would allow a 30 min service, (possible squeezing it down to a 20 min service in the peak. If the passengers cvome, then maybe the other things could follow. As it is the line will just drift on like a longer version of the Romford-Upminster operation.

...as I said previously, the drastic options only make sense once the line already has at least 2tph (if not 3tph or so), so I suspect we're actually in slight agreement.

The line is 16 minutes end to end. That means with a 0-minute turnaround it's 32 minutes to get there and back. Add in a rough 5 minute turnaround at each end and you get the magic 42 minute figure with a few minutes of slack. Do away with the turnaround at Watford (i.e. as a through service) and you get down to 37 minutes to do Watford-St Albans-Watford. Without a loop you could have two units on the line, but have one wait in the platform at St Albans so it can be released as soon as the next train arrives into the other platform, and you would then get it down to the base 32 minutes, abet you end up with stock sitting idle longer at St Albans than if you built the passing loop.

I suspect if you operated metro-style stock with better acceleration (read: LO 378s), then you would be able to shave a few more minutes off as well, getting you below the magic 30 minute mark.

 

[34D]

If end to end was non stop, how much quicker than 16 mins would it be?

Could an hourly clockface with one being all stops and the other being fast?

Alternatively, what about WFJ-SAA 16 mins, 4 turn, SAA-WFJ 16 mins, 4 turn, WFJ-GSN 6 mins, 4 turn, GSN-WFJ 6 mins, 4 turn (all stops).

Would just hourly be better than every 45?

 

[AM9]

If end to end was non stop, how much quicker than 16 mins would it be?

Could an hourly clockface with one being all stops and the other being fast?

Alternatively, what about WFJ-SAA 16 mins, 4 turn, SAA-WFJ 16 mins, 4 turn, WFJ-GSN 6 mins, 4 turn, GSN-WFJ 6 mins, 4 turn (all stops).

Would just hourly be better than every 45?

You'd still need to pay for a full-time train and driver/guard.
--- old post above --- --- new post below ---

...as I said previously, the drastic options only make sense once the line already has at least 2tph (if not 3tph or so), so I suspect we're actually in slight agreement.

The line is 16 minutes end to end. That means with a 0-minute turnaround it's 32 minutes to get there and back. Add in a rough 5 minute turnaround at each end and you get the magic 42 minute figure with a few minutes of slack. Do away with the turnaround at Watford (i.e. as a through service) and you get down to 37 minutes to do Watford-St Albans-Watford. Without a loop you could have two units on the line, but have one wait in the platform at St Albans so it can be released as soon as the next train arrives into the other platform, and you would then get it down to the base 32 minutes, abet you end up with stock sitting idle longer at St Albans than if you built the passing loop.

That would require points, a length of track and a new platform at St Albans. Given that the line would then need 'token' working, the costs of the above would be at least the same as the preferred solution, i.e. a passing loop at Bricket Wood which is:

a) near enough halfway down the line
and
b) probably the greatest potential for generating commuter traffic.​

 

[NSEFAN]

Could linespeeds be raised if the ABCL at Watford North was converted to full barrier? I know this might upset the locals, so a bridge would probably be needed to prevent people jumping it! When I used the line last, it seemed to take several minutes to operate the barriers, and the road seemed busy enough that a full crossing would be more appropriate.

 

[mr_jrt]

That would require points, a length of track and a new platform at St Albans. Given that the line would then need 'token' working, the costs of the above would be at least the same as the preferred solution, i.e. a passing loop at Bricket Wood which is:

a) near enough halfway down the line
and
b) probably the greatest potential for generating commuter traffic.​

Would it be the same though? Reinstating the disused platform at St Albans would create an island platform, so you would save on the lifts and footbridge, etc. not to mention you would only need one set of points, not two.

...and would the token operation be simpler (or required at all?) if you could only depart when another train was in the other terminal platform? ( I suspect you would still want it, but would it be as needed? )

Don't get me wrong, I think the loop is the ideal 1st step solution, but it's always good to have options.

...actually, scratch that. Just clocked that if we're talking first-step options then a bay at St Albans is useless without a second platform at Watford Junction as well.

 

[Mojo]

We can, it is just that a) no one wants to pay for it, if HCC were that desperate for it why haven't they coughed up at least in part, and b) no one thinks it is value for money or worth it.

c) the prices suggested by Network Rail for the work were sky-high

 

[Bald Rick]

c) the prices suggested by Network Rail for the work were sky-high

And d) no one else came up with a viable plan for doing it more cheaply

 

[RichardN]

Would it be the same though? Reinstating the disused platform at St Albans would create an island platform, so you would save on the lifts and footbridge, etc. not to mention you would only need one set of points, not two.

...and would the token operation be simpler (or required at all?) if you could only depart when another train was in the other terminal platform? ( I suspect you would still want it, but would it be as needed? )

Don't get me wrong, I think the loop is the ideal 1st step solution, but it's always good to have options.

...actually, scratch that. Just clocked that if we're talking first-step options then a bay at St Albans is useless without a second platform at Watford Junction as well.

I understand the proposal was for a Penryn style loop, which doesn't need a footbridge either. See https://en.wikipedia.org/wiki/Penryn_railway_station

The loop could also feature train operated points, whereas adding an extra platform at St Albans would mean having worked points.

 

[jon0844]

Clever design. Makes you wonder why it needs to cost in the millions to implement.

 

[DynamicSpirit]

I suspect one reason is that the whole line is currently being run as one long siding - hence no signalling required north of Watford, but putting in a passing loop requires signalling to be added.

I suspect another reason is that the costing is being done on the basis of the line being kept open throughout, so limited possessions etc. That could be significant in the context of this thread because obviously any costing for converting to a busway will be done on the basis of the line being closed for a year or so while the work is done, so there are no expenses associated with the limited possessions. If I'm correct, that could mean you end up with costings being compared that aren't like-for-like comparisons. (Obviously that's hypothetical at the moment, since I doubt anyone's attempted a costing for converting to a busway, but if Herts CC do push further with this crazy scheme, that could be something to look out for).

 

[jon0844]

The signalling is probably the key issue, and one I wasn't aware of (and most other people too). I would think closing the line to do it all in one hit would make sense, operating buses for the duration.

Hopefully users would accept this, especially if it made the difference between happening and not happening.

 

[Searchlight]

How exactly would a Guided Busway work, given the rail formation is single track? Surely, the conversion costs would cost many millions. and, for what, a much slower service?

 

[misterredmist]

perhaps NR and Herts County Council should do a joint promotion with "Spielplatz" to get more punters on the service ?

anyway, on a more serious note, I will be utilising the service to visit my Aunt in
Bricket Wood, but the 45 min timetable is ridiculous !

 

[MarkyT]

I understand the proposal was for a Penryn style loop, which doesn't need a footbridge either. See https://en.wikipedia.org/wiki/Penryn_railway_station

The loop could also feature train operated points, whereas adding an extra platform at St Albans would mean having worked points.

Train-operated hydro-pneumatic points as originally provided for RETB and Central Wales type NSKT remote crossing loops are no longer manufactured or supported. Most, if not all by now, have been removed and replaced by conventional point machines or clamp locks. The answer for signalling a Penryn-style heavy rail loop on Watford - St. Albans is definitely not tokens, but conventional axle counter based TCB using the 'modular signalling' series of components. It would be 7 additional SEUs for a loop (yes that's right a Penryn loop can also save 1 SEU over a conventional loop which requires 8 SEUs minimum). The line is currently OTNS worked, that is one train with no staff, and a series of track circuits and treadles at the Watford end has to be operated in a precise sequence to clear a train out of the single line section (rather like one end of a tokenless block system). OTNS is used:

1. Because it requires no continuous train detection throughout the length of the line. That's not such a huge problem with today's axle counters as sensors are only required at extremities so they eliminate most of the intermediate equipment (multi-sectional track circuits as each TC is length limited to typically around a kilometre).

2. Becasue with today's standards the alternative of TCB needs a signal at any reversing point, i.e. at the St. Alban's Abbey terminus for starting back towards Watford.

So my scheme would have Axle Counter TCB between Watford and the new Bricket Wood passing loop and OTNS beyond, hence avoiding the additional starting signal, or indeed any other changes, at St. Albans. This is almost exactly the same configuration as the Falmouth branch in Cornwall (except Falmouth has some extra key release arrangements for a ground frame at the terminus).

I may need an additional starting signal (1 SEU) at the St. Albans end of the Watford platform if that is currently within the OTNS section.

If we work on the basis of 4 SEUs/£1 million, thats £2m for signalling and perhaps another £1m for track and platform extension work.

Could a £3m infrastructure scheme fly for the Abbey Flyer?

 

[Bald Rick]

It would be a lot more than that.

Power supply for the signalling (and presumably points heating) at Bricket Wood. A few miles of cable from Watford, in proper troughing route. Workstation alterations and new interlocking at Rugby. So double the signalling costs.

2 S&C units will be over £1m on their own. Presumably a couple of hundred metres of plain line, which will have to have all new foundation and formation will be the same again.

120metres+ of platform extension will be another million when you include lighting, CIS, PA, signage, etc.

Then there is the cost of land itself, and the cost of getting to the point where you can buy it (consents process) and environmental protection. Easily another million (there's some expensive back gardens).

Electrification - at least another 8 masts and a wire run, even assuming no switching that will be the best part of a million.

So that's £9m before allowing for any civil engineering, design costs or contingency. Penryn was about £8m 7 years ago.

Ps, Compensation to LM for cancelling the service to facilitate would be peanuts in this, certainly less than 6 figures.

 

[MarkyT]

Power supply for the signalling (and presumably points heating) at Bricket Wood. A few miles of cable from Watford, in proper troughing route. Workstation alterations and new interlocking at Rugby. So double the signalling costs.

SEUs were supposed to include most of that (except the points heating), but I have no idea what the current going unit rate is on a typical small project today I admit. NR was supposed to be reducing the price of SEUs wasn't it? Hopefully there should be space for a few signalling TFM addresses on the existing interlockings and data links for the Watford area, so not an entire new cubicle perhaps. Ideally a scheme such as this could also be timed to coincide with planned renewal for such equipment so overheads might be minimised.

2 S&C units will be over £1m on their own. Presumably a couple of hundred metres of plain line, which will have to have all new foundation and formation will be the same again.

Oops completely forgot the plain track and formation.

120metres+ of platform extension will be another million when you include lighting, CIS, PA, signage, etc.

Yes and no doubt would require complete update of the existing platform section at the same time. I can see that adding up.

Then there is the cost of land itself, and the cost of getting to the point where you can buy it (consents process) and environmental protection. Easily another million (there's some expensive back gardens).

If the extension is Watford side the adjoining land is a builder's merchant. I wonder if it is or was leased railway land (probably sold off for a quick buck by BRB residuary a few years ago though!).

So that's £9m before allowing for any civil engineering, design costs or contingency. Penryn was about £8m 7 years ago.

OK thanks for that benchmarking.

 

[Bald Rick]

Ideally a scheme such as this could also be timed to coincide with planned renewal for such equipment so overheads might be minimised.

Renewed last year, so just the 39 years to wait!

The SEU rate is of course an average, based on medium to large schemes. This is a tiddler, and there are dis-economies of scale. 3 miles of cabling and troughing route costs the same for 7 SEUs as it does for 70.

 

[MarkyT]

Renewed last year, so just the 39 years to wait!

Excellent planning! It is to be hoped then that the project engineers sensibly allocated a few spare signal IDs and TFM addresses for additional facilties in the area as they are supposed to do. At least any scheme would be starting from a good baseline with modern equipment, good records for the area etc.

 

[DynamicSpirit]

The problem with access to Sainsbury is that the platform is on the northern side of the line and the southern side is occupied by a large'ish office complex, so it would require a new footbridge or for passengers to walk 50m to Hollywell Hill and then through the office complex as at present.
To provide a 'short'cut to Prospect Road would require about 140m of pathway alongside the track from the unused end of the platform to the end of St Julians Road cul-de-sac. Not the sort of access favoured today although Sainsbury could help fund that with a footbridge as it would provide access into their car park. Were that to be the case, then there would be a need to prevent Sainsbury's becoming a free commuter car park.
Access to buses out of town is good however as local, Watford and Green Line buses all serve the adjacent stop in Holywell Hill. It is not so convenient to catch buses northwards into town, more than 200m away, crossing Holywell Hill by pelican crossing as well as two side road entrances.

Thanks for the info. So it sounds like there's a few things that could be looked at there. The inconvenient Northbound bus stop presumably affects almost all people who might make bus/train connecting journeys, since if someone uses the convenient Southbound bus stop on their outward journey, then they'll need to use the Northbound stop on their return journey, and vice versa. What are the possibilities for relocating the stop or providing and additional one, along with making it easier for pedestrians to cross the road?

Regarding access to Sainsburies, and perhaps the office complex, a footbridge was what I was thinking of. Good point about wanting to avoid it becoming a commuter car park - though the other side to that is perhaps, if a small area near the station was turned into a paying all-day car park (with easy access over the footbridge) then not only would more commuters be attracted to the line, but Sainsburies might make a small profit from the charges (especially as some of those commuters would end up shopping there at the end of the day). I don't know whether the practicalities of that would work in this particular case, but it seems to me that's exactly the kind of small-scale but innovative thinking that's required to make the line more useful, and get usage up to the point that more expensive interventions start to look more worthwhile.

 

[jon0844]

Having discovered that a footbridge (with ramps that extend for what seems like miles) in Cheshunt cost nearly £2m, I suspect that idea isn't going to happen.

 

[Tetchytyke]

If we can't manage a single passing loop, I don't hold out much hope for the future of the railway.

The passing loop would be simple. Paying for it would not be.

The simple fact of the matter is that all this talk of flyovers and flyunders at Watford Junction is trying to fix a problem nobody knew they had. People who live in St Albans have no need or desire to commute to Aylesbury or Rickmansworth by rail. They are commuting to London. And even if the line was connected to Overground or the Met they wouldn't use it all the way into London; as it stands, Overground takes almost an hour to Euston compared to less than 20 minutes on London Midland. London commuters will already take the Thameslink train, and money would be better spent on that route, to be perfectly honest.

I'd like to see a few quick wins. If you can extend platform 11 at Watford Junction (WFJ) to take 8-cars (which won't be cheap, but should be straightforward) you could put the peak Watford Flyers back to St Albans Abbey, either using SDO or joining units at WFJ. That would give Garston a better service, which is the only place on the line where anybody lives. But beyond that you'd be struggling to ever justify the cost of the major infrastructure required to improve the service frequency: there are simply not that many people who are in that much of a hurry to travel between Watford and St Albans.

I don't see the line closing, it doesn't cost much to operate it and it is popular enough to a) justify its existence and b) have a proper connection, at great cost, installed to the WCML. But the line is what it is: a small self-contained branch line that will never be anything more important. Transferring operation to Overground might not be such a bad thing, but on the same basis as the Romford shuttle not as an all-singing extension of the DC lines.

 

[DynamicSpirit]

It would be a lot more than that.

Power supply for the signalling (and presumably points heating) at Bricket Wood. A few miles of cable from Watford, in proper troughing route.

Out of interest, does the passing loop have to be at Bricket Wood? Presumably, to make a half-hourly timetable work, the only requirement is that neither end of the line should be more than 11-12 mins from the loop? It looks from Google maps like there's plenty of countryside between Garston and Bricket Wood, where buying land would presumably be cheaper (and less distance to lay the cables). And that would mean no platform extensions required. I'm guessing though that the other side to that is that a passing loop not in a station might need to be longer to keep the timetable robust.

Even more countryside north of Bricket Wood, but that obviously would mean further to lay the cables.

Workstation alterations and new interlocking at Rugby. So double the signalling costs.

I have no knowledge of signalling technology, but that statement surprises me, since it seems to imply that the signalling at Rugby has been built with no capacity for improvements to any of the track under its control. Is that the case or have I misunderstood? And if that is the case, is that normal practice when building signalling technology?

 

[Bald Rick]

All major new signalling projects have spare capacity in their system architecture, and Watford / Rugby will be no different.

There is however a difference between providing spare capacity in the architecture and actually making use of it.

There would need to be interlocking alterations, which means design, data prep, checking, approval, pre testing, rehearsal, testing and commissioning. There are not many people who do this, and they all get charged out by the signalling contractors at eye watering daily rates. Ditto for the changes to the workstation screen layouts.

 

[MarkyT]

I wouldn't put the loop on plain line between stations, difficult to regulate that properly avoiding extra stops and if you want to plan to keep the trains moving then they've got to be fairly lengthy. Bricket Wood is pretty good really if you build the new loop formation and platfrom on the side away from the houses at the Watford end. You only affect the adjoining builder's merchant yard and would be largely shielded by trees.

Another thought evolves . . .

Taking the two platforms at one end idea mentioned earlier, why not have these at the Watford end, requiring only one new turnout, 3 new signals and a distant (could be a fixed board).

Shorten the journey time by closing BOTH How Wood and Park Street stations and replacing then with one new station at the intermediate bridge (about 1/4 mile from each of the old stations). Another measure could be to move At. Albans Abbey bufferstops and platform about 300m towards Watford. That could halp some local pedestrian connectivity as well connecting both the station and the adjacent shopping area to the residential street St Julians Road (only do this if absolutely necessary for journey time though as it entails yet more station construction).

De-electrify or mothball existing wires, so as to avoid new catenary work at new Watford platform.

Employ Vivarail D-Train diesel microfleet of two three-car units with local maintenance employed.

Hence the whole branch would get a half hourly service.

Some assumptions:
No more than 14 minute journey time each way.
2 minutes turnround at St Albans Abbey.
O minutes at Watford with the next train departing just as the previous arrival has cleared the incoming turnout.
New platform track laid on car park behind existing platform, widened or renewed as neccessary.
Assume car park tarmac forms sufficiently secure and well drained base for new track without significant work.

Some savings over Bricket Wood loop option:
Long signalling cable route saved
4 SEUs saved
1 turnout saved
OHLE work saved

The main extra costs are clearly the additional new How Wood platform.
(The Watford reconstruction I've considered equivalent to the Bricket Wood platfrom work)

Other observations.
Same number of trains and crew as Bricket Wood option, and despite seemingly inefficient long breaks during Watford layover, that actually makes scheduling official breaks very easy.

 

[Bletchleyite]

Employ Vivarail D-Train diesel microfleet of two three-car units with local maintenance employed.

You'd want three so there was a spare. The reliability of Bedford-Bletchley, which has only two, is atrocious.

 

[mr_jrt]

Still don't see why you can't settle for a basic token-based setup as a first step to reduce costs rather than gold-plating things.

 

[MarkyT]

Still don't see why you can't settle for a basic token-based setup as a first step to reduce costs rather than gold-plating things.

Tokens are not cheap, at least not anymore. You've still got conventional point machines with conventional train detection at a passing loop because the old train operated loop points are no longer available. And then you need TPWS equipment with your stop boards, and indicator lights to show TPWS is working, which, along with your points indicators, might as well be signals). You need power for all this and a good secure cable route between the token machines at either end of the section for the circuits. On top of that the machines themselves are probably now completely alien to local S&T maintenance staff who otherwise are trained to deal all modern equipment. Skills, Spares, Extra Training, Token balancing, etc . . .

http://www.vmslimited.co.uk/cls.htm

Axle counter based TCB with the simple lightweight colour light signals from (for instance) VMS would be far more suitable for such a minor low-speed route. As an LED this signal is very reliable and low maintenance, and for the extremely rare occasions you do need to access the head, you take a possession and fold the unit down to work at ground level rather than the structure requiring modern safety access ladders to fully fenced high-level platforms so you can work up there. Those traditional measures increase the size and weight of conventional structures significantly and increase the size and cost of their foundations.

 

[AM9]

Thanks for the info. So it sounds like there's a few things that could be looked at there. The inconvenient Northbound bus stop presumably affects almost all people who might make bus/train connecting journeys, since if someone uses the convenient Southbound bus stop on their outward journey, then they'll need to use the Northbound stop on their return journey, and vice versa. What are the possibilities for relocating the stop or providing and additional one, along with making it easier for pedestrians to cross the road?

The northbound stop could usefully be relocated just south of the gap opposite the station entrance as Holywell Hill is 2 x 2 lane dual carriageway at that point. This would have another benefit in that it would give easy pedestrian access to both the Abbey Theatre and the new leisure centre. Note: Google Maps just shows the old swimming pool further into the park but Bing Maps shows it in construction; - it also labels the Abbey Theatre as Abbey View Golf Course, but heigh-ho!

Regarding access to Sainsburies, and perhaps the office complex, a footbridge was what I was thinking of. Good point about wanting to avoid it becoming a commuter car park - though the other side to that is perhaps, if a small area near the station was turned into a paying all-day car park (with easy access over the footbridge) then not only would more commuters be attracted to the line, but Sainsburies might make a small profit from the charges (especially as some of those commuters would end up shopping there at the end of the day). I don't know whether the practicalities of that would work in this particular case, but it seems to me that's exactly the kind of small-scale but innovative thinking that's required to make the line more useful, and get usage up to the point that more expensive interventions start to look more worthwhile.

That seems OK. Sainsbury has asked for a two storey extension to the non-retail part of their store in recent years but I don't beleive that has been built. There may have been objections from some of the residents on the North side of the line, but there is a fair amount of screening from the trees on that side so presumably a footbridge could be accepted, giving the residents of the Prospect Road area easy pedestrian access to Sainsbury's store. I think it is already part of a CPZ.
There is some regular patronage from the Buses there, from the south as despite the 321 service running to Watford, it can get delayed at M25 J19A roundabout, Garston and the A41 Dome roundabout such that the nominal 35 mins from SAA to WFJ can take up to 50 mins, especially when there are problems on the M1 or M25.

 

[Bevan Price]

Another thought evolves . . .

Taking the two platforms at one end idea mentioned earlier, why not have these at the Watford end, requiring only one new turnout, 3 new signals and a distant (could be a fixed board).

Shorten the journey time by closing BOTH How Wood and Park Street stations and replacing then with one new station at the intermediate bridge (about 1/4 mile from each of the old stations). Another measure could be to move At. Albans Abbey bufferstops and platform about 300m towards Watford. That could halp some local pedestrian connectivity as well connecting both the station and the adjacent shopping area to the residential street St Julians Road (only do this if absolutely necessary for journey time though as it entails yet more station construction).

De-electrify or mothball existing wires, so as to avoid new catenary work at new Watford platform.

Employ Vivarail D-Train diesel microfleet of two three-car units with local maintenance employed.

Hence the whole branch would get a half hourly service.

Some assumptions:
No more than 14 minute journey time each way.
2 minutes turnround at St Albans Abbey.
O minutes at Watford with the next train departing just as the previous arrival has cleared the incoming turnout.
New platform track laid on car park behind existing platform, widened or renewed as neccessary.
Assume car park tarmac forms sufficiently secure and well drained base for new track without significant work.

.

14 minutes by dmu is probably impracticable. Watford Jn - St. Albans Abbey time by dmu in 1964 was 17 minutes - with two fewer stops - Garston & How Wood did not exist then; (although trains were mostly timed at 15 minutes in the reverse direction.) .

Train service in 1964 was every 20 minutes at peak hours, and on saturday afternoons, reducing to every 40 minutes at other times, including sunday afternoons.

In 1977 timetable, with Garston open, trains were allowed 17 minutes in each direction. Frequency was irregular, with gaps varying between 40 & 81 minutes (Watford Jn departures) - and the sunday service had gone.

And don't forget that, back then, they did not have compulsory running brake tests after each reversal; you might save about ¼ to ½ minute each way if there were no running brake tests now.

 

[DynamicSpirit]

All major new signalling projects have spare capacity in their system architecture, and Watford / Rugby will be no different.

There is however a difference between providing spare capacity in the architecture and actually making use of it.

OK ta for the explanation

There would need to be interlocking alterations, which means design, data prep, checking, approval, pre testing, rehearsal, testing and commissioning. There are not many people who do this, and they all get charged out by the signalling contractors at eye watering daily rates. Ditto for the changes to the workstation screen layouts.

Really? Where can I get trained up to become one of these people?

(:joke: in case it's not obvious. And oh bugger what's the code for the joke smiley)