Cost of Third Rail Electrification

[LNW-GW Joint]

I have been trying to find some actual supporting documentation for the estimate given by Network Rail for electrifying the Borderlands line at DC (£207m) however the number just seems to appear in the Merseyside RUS saying analysis has shown it.
But there appears to be no actual documentation of how the estimate was derived.

I have a copy of the Network Rail report (003600 Borderlands, 09/07/08), which quotes £207m.
It might have come from the Welsh Government site.

Broadly:
construction cost £102m (of which 44 electrification, 29 permanent way, 6 signalling)
design/testing/contractor profit £49m
NR PM costs £20m
contingency £35m

For this figure you get a 2tph electric service and 3 new stations (Beechwood, Woodchurch, Deeside), with existing signalling immunised for DC operation.

It also shows the earlier estimates of £130m (Mott McDonald) and £34m (First Class Partnerships/Merseytravel) and how they differ.
Basically the earlier estimates were done without knowledge of the actual route assets, and mostly omitted the overheads mentioned above.

 

[pablo]

It may be of interest, that Shotton is the terminalion of the Scottish 750vDC Interconnector (running down the Irish Sea) where an inverter station is being built before the AC lines go across the Dee into Connah's Quay. Shotton is also something of a Grid cross-roads.

PS: I think the £49m mentioned above probaly includes the civils element also.

EDIT: Is it possible to tap off the incoming DC supply (before conversion) to feed the railway directly? The converter station is being built only a few hundreds of meters from the railway.

 

[Camden]

That's a rash leap of logic.

Having personally stuck steel sleepers into routes that conceivably, one day, might see third rail electrification I can sure you that the decision to do so was not to 'deliberately... ensure that third rail electrification could not take place'

The decision was taken on the basis that at that time, there was no realistic prospect of third rail electrification arriving in the foreseeable future*. The track needed renewing now, and the cost of doing it with steel sleepers was less than half the cost of concrete - I'm talking millions of pounds cheaper. There was no way the 'funders and regulators' would sanction spending extra millions on something there was no realistic chance of using. And so it has turned out, at least for the 10 years so far.

*oil prices low, serious consideration being given to removing OLE up north for reliability reasons.

On another point, steel and concrete sleepers have a nominal 40 year life.

Leaving aside any theories as to reasoning, may I ask you these questions:

1) Bearing in mind what you say above about the difference being millions, was the difference in cost between third rail incompatible steel sleepers and third rail compatible concrete sleepers more or less than the cost of ripping up the steel sleepers and installing concrete now? Bear in mind the difference between the NR estimate and the Merseytravel estimate appears to be around £100m.

2) Bearing in mind that it has been stated that any electrification programme will have to bear the cost of replacing these recently installed steel sleepers, was there any corresponding offer prior to installation of these steel sleepers to Merseytravel to allow them to try and find the funding for the difference in cost in order to upgrade to concrete, and thus future proof a key regional growth priority?

It's all very well pointing out that electrification has not gone ahead "at least for the 10 years so far" but had it cost £100m less then perhaps it might well have done - there was certainly appetite to do so, and there certainly hasn't been a paucity of funds for infrastructure when it comes to spending it in Manchester and London. As it stands, with a price tag of over £200m and those steel sleepers in the way - with their 40 year lifespan - this line's electrification has most likely been killed off for a generation. That doesn't vindicate the decision to install steel sleepers in the slightest.

 

[Bald Rick]

Leaving aside any theories as to reasoning, may I ask you these questions:

1) Bearing in mind what you say above about the difference being millions, was the difference in cost between third rail incompatible steel sleepers and third rail compatible concrete sleepers more or less than the cost of ripping up the steel sleepers and installing concrete now? Bear in mind the difference between the NR estimate and the Merseytravel estimate appears to be around £100m.

2) Bearing in mind that it has been stated that any electrification programme will have to bear the cost of replacing these recently installed steel sleepers, was there any corresponding offer prior to installation of these steel sleepers to Merseytravel to allow them to try and find the funding for the difference in cost in order to upgrade to concrete, and thus future proof a key regional growth priority?

It's all very well pointing out that electrification has not gone ahead "at least for the 10 years so far" but had it cost £100m less then perhaps it might well have done - there was certainly appetite to do so, and there certainly hasn't been a paucity of funds for infrastructure when it comes to spending it in Manchester and London. As it stands, with a price tag of over £200m and those steel sleepers in the way - with their 40 year lifespan - this line's electrification has most likely been killed off for a generation. That doesn't vindicate the decision to install steel sleepers in the slightest.

The line I put them in was not anywhere near Merseyside, so I can't answer 2).

The cost difference between steel sleeper and concrete sleeper renewal is not the sleepers. (Steel sleepers actually cost about 10% more each). The cost difference is that with concrete sleepers you have to remove the ballast underneath, and you don't have to do that for steel - the design of the steel sleeper is such that it stabilises the old ballast sufficiently (once it has been scarified). Thus you save the expense of lots of digging, spoil trains, ballast trains, labour and time. Back then, steel sleeper track renewal was less than half the cost of concrete sleeper track renewal.

So, basically, NR had to borrow less than half as much to do steel sleeper renewal than concrete. Which, at the borrowing rates applicable then, and the fact that concrete sleeper renewal is now cheaper than it was then, means that I estimate that the cumulative saving by doing steels at that time is already more than the cost of doing concrete now.

As has been said upthread, I don't believe that replacing steel sleepers is a significant element of the difference between estimates, certainly not £100m worth. That price buys you around 100 single track miles of track renewal with concrete sleepers. And Wrexham - Bidston isn't that far, even if it is were all steel, which I don't believe it is.

 

[Camden]

The line I put them in was not anywhere near Merseyside, so I can't answer 2).

The cost difference between steel sleeper and concrete sleeper renewal is not the sleepers. (Steel sleepers actually cost about 10% more each). The cost difference is that with concrete sleepers you have to remove the ballast underneath, and you don't have to do that for steel - the design of the steel sleeper is such that it stabilises the old ballast sufficiently (once it has been scarified). Thus you save the expense of lots of digging, spoil trains, ballast trains, labour and time. Back then, steel sleeper track renewal was less than half the cost of concrete sleeper track renewal.

So, basically, NR had to borrow less than half as much to do steel sleeper renewal than concrete. Which, at the borrowing rates applicable then, and the fact that concrete sleeper renewal is now cheaper than it was then, means that I estimate that the cumulative saving by doing steels at that time is already more than the cost of doing concrete now.

As has been said upthread, I don't believe that replacing steel sleepers is a significant element of the difference between estimates, certainly not £100m worth. That price buys you around 100 single track miles of track renewal with concrete sleepers. And Wrexham - Bidston isn't that far, even if it is were all steel, which I don't believe it is.

It's not the process that interests me, the cost differential between upgrading to concrete up front versus ripping up the steel ones and doing it all again does. I won't get into inflation, but needless to say the cost of doing anything rarely goes down. The cost of concrete you say was double the cost of steel, so there you have a ballpark figure of how much it would have cost to upgrade to concrete: 50% of the cost of installing concrete sleepers (in other words 100% of the cost of installing the steel sleepers).

So, since you say you know the costs of the items, how much was the project worth in total?

And the other crucial thing that interests me is whether this differential was offered to Merseytravel to fund, rather than them being lumbered with those rip up and replace costs, which you don't know the answer to.

 

[LNW-GW Joint]

It's not the process that interests me, the cost differential between upgrading to concrete up front versus ripping up the steel ones and doing it all again does. I won't get into inflation, but needless to say the cost of doing anything rarely goes down. The cost of concrete you say was double the cost of steel, so there you have a ballpark figure of how much it would have cost to upgrade to concrete: 50% of the cost of installing concrete sleepers (in other words 100% of the cost of installing the steel sleepers).

So, since you say you know the costs of the items, how much was the project worth in total?

And the other crucial thing that interests me is whether this differential was offered to Merseytravel to fund, rather than them being lumbered with those rip up and replace costs, which you don't know the answer to.

The project was never at the "get 20% off and we'll have it" stage.
They are all feasibility reports, prior to the formal GRIP process (GRIP 3 is when money gets committed).
Like with its rolling stock ambitions Merseytravel never had the money anyway, and the Welsh Government would also need to find its half.
It would have failed "value for money" tests at every stage, except possibly the short 3-mile extension from Bidston to Woodchurch.
Even then, you have to rack your brains to find an economic Merseyrail timetable for the route.
Does the Wirral system via Birkenhead North need at extra 2tph, or will it just dilute what is there already?

The steel sleepers are on the southern half of the line, which needed upgrading to maintain the heavy freight loads to Tata Steel at Dee Marsh.
Conversely, the northern half is not steel but the line is in poor condition as, over the years, it has been patched and mended for light DMU use and needs a lot of work for 3rd rail use.

 

[Camden]

The project was never at the "get 20% off and we'll have it" stage.
They are all feasibility reports, prior to the formal GRIP process (GRIP 3 is when money gets committed).
Like with its rolling stock ambitions Merseytravel never had the money anyway, and the Welsh Government would also need to find its half.
It would have failed "value for money" tests at every stage, except possibly the short 3-mile extension from Bidston to Woodchurch.
Even then, you have to rack your brains to find an economic Merseyrail timetable for the route.
Does the Wirral system via Birkenhead North need at extra 2tph, or will it just dilute what is there already?

The steel sleepers are on the southern half of the line, which needed upgrading to maintain the heavy freight loads to Tata Steel at Dee Marsh.
Conversely, the northern half is not steel but the line is in poor condition as, over the years, it has been patched and mended for light DMU use and needs a lot of work for 3rd rail use.

Obfuscating the issue in operational matters around Merseyrail doesn't answer the questions. The way I see it, the lines of Merseyrail that are actually finished are highly successful. The Bidston line certainly has potential, and there are a number of creative service patterns that the city as a whole could benefit from.

The issue remains. For how much extra could concrete sleepers have been installed instead of steel ones, and was this ever offered out to Merseytravel?

Never mind they didn't have the money, or it wasn't on the cards. Funding can be obtained if a case can be made for it (such as £300m very recently for Metrolink in Manchester to the Trafford Centre), but were they ever given the opportunity to try?

I'm seeing a short term saving of X million, versus a rip up and replace cost of Y million and an electrification project kicked well into the long grass. If people are going to try and defend that, they could at least answer the simple (but awkward) questions posed.

 

[LNW-GW Joint]

Obfuscating the issue in operational matters around Merseyrail doesn't answer the questions. The way I see it, the lines of Merseyrail that are actually finished are highly successful. The Bidston line certainly has potential, and there are a number of creative service patterns that the city as a whole could benefit from.

The issue remains. For how much extra could concrete sleepers have been installed instead of steel ones, and was this ever offered out to Merseytravel?

Never mind they didn't have the money, or it wasn't on the cards. Funding can be obtained if a case can be made for it (such as £300m very recently for Metrolink in Manchester to the Trafford Centre), but were they ever given the opportunity to try?

I'm seeing a short term saving of X million, versus a rip up and replace cost of Y million and an electrification project kicked well into the long grass. If people are going to try and defend that, they could at least answer the simple (but awkward) questions posed.

Sorry I'm such a dunce.
The paper I have doesn't give the answer you are looking for.
There is a simpler answer though - there is no money.

 

[Bald Rick]

Obfuscating the issue in operational matters around Merseyrail doesn't answer the questions. The way I see it, the lines of Merseyrail that are actually finished are highly successful. The Bidston line certainly has potential, and there are a number of creative service patterns that the city as a whole could benefit from.

The issue remains. For how much extra could concrete sleepers have been installed instead of steel ones, and was this ever offered out to Merseytravel?

Never mind they didn't have the money, or it wasn't on the cards. Funding can be obtained if a case can be made for it (such as £300m very recently for Metrolink in Manchester to the Trafford Centre), but were they ever given the opportunity to try?

I'm seeing a short term saving of X million, versus a rip up and replace cost of Y million and an electrification project kicked well into the long grass. If people are going to try and defend that, they could at least answer the simple (but awkward) questions posed.

Back in 2005, it cost on average (network average) just under £400 per metre to renew with steel sleepers, and just over £800 per metre to renew with concrete. Both 2005 prices. I don't know how much renewal in steel was done on this line in question, nor when it was done. Also I don't know if it is a single line, which widens the differential as logistics gets difficult.

In 2014, the average £ per metre of concrete renewal is, I gather, lower than £800 per metre in 2014 prices, and with the advent of the high output machines, single lines cost the same.

 

[Camden]

The other part of the maths needed then is the number of metres done.

Plus, finding out whether or not the upgrade cost to concrete was offered up to Merseytravel & partners to find the funding for, rather than the electrification project being lumbered with rip up and replace costs.

 

[pablo]

Camden, Politics and public services ooop north are rather like single cell amoeba that don't join together to form two-cell animals but do move discerenably in a very slow jellylike fashion. The result being that the area is left far behind elsewhere.

 

[contrex]

It may be of interest, that Shotton is the terminalion of the Scottish 750vDC Interconnector...

Is it possible to tap off the incoming DC supply (before conversion) to feed the railway directly? The converter station is being built only a few hundreds of meters from the railway.

Long distance HVDC links run at a much higher voltage than 750 V. If you mean the East-West Interconnector which runs from Shotton, North Wales, to Rush North Beach, County Dublin, in Ireland, that is a +- 200 kV bipole. The use of DC is purely for transmission reasons; it is not practical to use it for supply purposes, e.g. railway traction. The Scotland-Northern Ireland (the Moyle interconnector) link is a separate one, also at a very high DC voltage.

 

[pablo]

Long distance HVDC links run at a much higher voltage than 750 V. If you mean the East-West Interconnector which runs from Shotton, North Wales, to Rush North Beach, County Dublin, in Ireland, that is a +- 200 kV bipole. The use of DC is purely for transmission reasons; it is not practical to use it for supply purposes, e.g. railway traction. The Scotland-Northern Ireland (the Moyle interconnector) link is a separate one, also at a very high DC voltage.

No. I don't mean any of that. It's inovative. See here: http://www.westernhvdclink.co.uk/default.aspx

 

[contrex]

No. I don't mean any of that. It's inovative. See here: http://www.westernhvdclink.co.uk/default.aspx

Where did you get the 750 V DC figure from? It is suspiciously low (by a factor of 500 to 1,000). Wikipedia says it will be 600 kV.

 

[pablo]

Where did you get the 750 V DC figure from? It is suspiciously low (by a factor of 500 to 1,000). Wikipedia says it will be 600 kV.

IIRC it was on the Western HVDC Link website a couple of years ago. Maybe IDRC! It does seem a very low pressure to be driving 2,200MW around two 6" cables but I'm no elec eng.

Wiki is not very specific about this scheme. There is a proposal to connect to Iceland similarly for cheap geo-thermal surplus. I'll ask those in the know when I'm back at work next week.

So, what do we do? Convert DC to AC and then back to DC for traction current?

 

[contrex]

Trust me, HVDC links of any length run at hundreds of kilovolts. The purpose is to feed power into the AC National Grid. The best way of feeding any 750 V DC railway electrification would be to take AC from that same National Grid and convert it in standard substations. The proximity of any HVDC link notwithstanding.

 

[Elecman]

Correct covert to ac, transform to a suitable transmission voltage then transform and convert back to dc at the traction feeder station,t or use either a VERY big resistor or a very large dc motor/generator!!

 

[pablo]

Perhaps I've dropped an 'em' somewhere along the line.

 

[HSTEd]

One interesting thing would be that, in this case any extra conductors would better be bonded to the running rails as a ground consuctor.than being used to reinforce the conductor rail. Which is not a result I would have anticipated.

 

[AM9]

....ground consuctor....

Sounds interesting, please explain.

 

[HSTEd]

Sounds interesting, please explain.

Well a third rail DC system, unlike an AC system (25kV), has to be kept as a closed circuit as ground leakage problems will cause corrosion of metallic objects in the return path, for example rails will corrode as they leak current into the earth.
THis is the reason we cannot have steel sleepers in third rail areas.

This means the resistance of the running rails becomes a major factor, in the cae of a system where both of the running rails are available for current return standards assume a typical value of ~19mOhm per kilometre of track.
While that is, by normal standards, rather small and in the case of an AC system would be smaller than that of the overhead wiring conductor, a high conductivity aluminium/steel third rail has a resistance of only 7mOhm/km.
That means that the majority of the resistance in the circuit is actually on the 'ground' side of the load.

Let's say we have a second conductor rail equivalent conductor (so 7mOhm/km) which we can connect in parallel with either the conductor rail or the return rail, which should we do?
If we connect it in parallel with the conductor rail the resistance of the conductor side of the circuit drops to 3.5mOhm/km, but the running rail side resistance doesn't change - so the loop resistance is 22.5mOhm/km.
If, however, we connect it in parallel with the running rail the conductor rail resistance stays at 7mOhm/km but the running rail side resistance drops to just 5.1mOhm/km. So the loop resistance is only 12.1mOhm/km.
Which means the substations can be 85% further apart.

There is also another factor - while we normally, simplistically, assume that the ground conductor remains at 0V and the entire 'voltage drop' appears as a literal drop in the voltage of the 'hot' conductor. However this is not the case, some of it will appear in as a 'voltage rise' in the voltage of the ground conductor.
This is a serious problem since the running rail in a third rail system will have a voltage relative to the 'true ground', this can become a safety issue if this voltage becomes larger than about 60V as it can deliver dangerous shocks to people who happen to come into contact with the running rail for whatever reason.
The ratio of the split between voltage drop in the hot conductor and voltage rise in the ground conductor is determined by the relative resistances of the two.
If no extra conductor is employed the actually permissible voltage drop for the system is only 82V. This drops in the scenario where the extra conductor is on the 'hot side' to only 71V, whereas if the extra conductor is on the 'ground side' the voltage drop climbs to 142V - combined with the reduced loop resistance that would permit 3.25x the substation spacing than the base case with no extra conductors.

This means that it is probably required that any available conductors be added to the ground side of the system to hold down the running rail voltage rise even if that is not the optimal placement in terms of minimising the overall loop resistance.

This finding also opens the door to the idea of employing recent advances in switched mode power supplies to have devices analogous to AC 'booster' transformers to draw the current from the running rails and hold down their voltage. I will run some calculations on that after my exams I think before I post the aforementioned tract on third rail in the 21st century.

 

[snowball]

I don't think AM9 really wanted a technical explanation. I think he was just commenting on "consuctor" - for which the explanation is presumably that S is next to D.

 

[AM9]

Well a third rail DC system, unlike an AC system (25kV), has to be kept as a closed circuit as ground leakage problems will cause corrosion of metallic objects in the return path, for example rails will corrode as they leak current into the earth.
THis is the reason we cannot have steel sleepers in third rail areas.

This means the resistance of the running rails becomes a major factor, in the cae of a system where both of the running rails are available for current return standards assume a typical value of ~19mOhm per kilometre of track.
While that is, by normal standards, rather small and in the case of an AC system would be smaller than that of the overhead wiring conductor, a high conductivity aluminium/steel third rail has a resistance of only 7mOhm/km.
That means that the majority of the resistance in the circuit is actually on the 'ground' side of the load.

Let's say we have a second conductor rail equivalent conductor (so 7mOhm/km) which we can connect in parallel with either the conductor rail or the return rail, which should we do?
If we connect it in parallel with the conductor rail the resistance of the conductor side of the circuit drops to 3.5mOhm/km, but the running rail side resistance doesn't change - so the loop resistance is 22.5mOhm/km.
If, however, we connect it in parallel with the running rail the conductor rail resistance stays at 7mOhm/km but the running rail side resistance drops to just 5.1mOhm/km. So the loop resistance is only 12.1mOhm/km.
Which means the substations can be 85% further apart.

There is also another factor - while we normally, simplistically, assume that the ground conductor remains at 0V and the entire 'voltage drop' appears as a literal drop in the voltage of the 'hot' conductor. However this is not the case, some of it will appear in as a 'voltage rise' in the voltage of the ground conductor.
This is a serious problem since the running rail in a third rail system will have a voltage relative to the 'true ground', this can become a safety issue if this voltage becomes larger than about 60V as it can deliver dangerous shocks to people who happen to come into contact with the running rail for whatever reason.
The ratio of the split between voltage drop in the hot conductor and voltage rise in the ground conductor is determined by the relative resistances of the two.
If no extra conductor is employed the actually permissible voltage drop for the system is only 82V. This drops in the scenario where the extra conductor is on the 'hot side' to only 71V, whereas if the extra conductor is on the 'ground side' the voltage drop climbs to 142V - combined with the reduced loop resistance that would permit 3.25x the substation spacing than the base case with no extra conductors.

This means that it is probably required that any available conductors be added to the ground side of the system to hold down the running rail voltage rise even if that is not the optimal placement in terms of minimising the overall loop resistance.

This finding also opens the door to the idea of employing recent advances in switched mode power supplies to have devices analogous to AC 'booster' transformers to draw the current from the running rails and hold down their voltage. I will run some calculations on that after my exams I think before I post the aforementioned tract on third rail in the 21st century.

I've never thought about this before but after reading your text about -ve rise above ground makes me realise that DC 3rd rail is even more troublesome than I thought before. If the -ve conductor, (i.e. the running rails rise above 60V) that means that all grounded conducting surfaces near to the platform edge at stations must either be insulated from human conduct or tied to the variable floating DC running rails. If that is done, at some point the rest of the station's normal ac circuits must be separated from the platform edge zone.

 

[edwin_m]

I've never thought about this before but after reading your text about -ve rise above ground makes me realise that DC 3rd rail is even more troublesome than I thought before. If the -ve conductor, (i.e. the running rails rise above 60V) that means that all grounded conducting surfaces near to the platform edge at stations must either be insulated from human conduct or tied to the variable floating DC running rails. If that is done, at some point the rest of the station's normal ac circuits must be separated from the platform edge zone.

The normal approach on DC tramways, which I think also applies to DC railways, is to connect metallic objects within a certain distance of the track to the return rail via a device which normally insulates but becomes a conductor if the voltage across it exceeds the danger value of 50V. This is because these devices are also connected (usually unintentionally) to the earth (think metal bridge!) but you don't want to earth the DC return for the reasons explained above. However the system will be designed so that the 50V is only reached under fault conditions, probably only for a very short time before the breakers cut the power. In the case of the tramways this fault situation is usually the risk of the OLE falling onto a metallic object.

 

[MarkyT]

. . . where both of the running rails are available for current return standards assume a typical value of ~19mOhm per kilometre of track . . . a high conductivity aluminium/steel third rail has a resistance of only 7mOhm/km. That means that the majority of the resistance in the circuit is actually on the 'ground' side of the load.

That's quite a difference in conductor resistance figures but don't forget that in multiple track installations, all the parallel rails can be effectively bonded together at intervals, either via inductance bond centre connections where track circuits are used or by simple cable bonding with axle counters. That could lower actual resistance from any particular train location to the substation further, say to around 10mOhm/km for a double track.

The aluminium composite is impressive, but resistance figures for traditional steel conductor rail is not so good, usually taken to be 16mOhm/km, according to: http://www.rgsonline.co.uk/Railway_...ck/Railway Group Standards/GMRT1001 Iss 1.pdf

A strengthening conductor for either leg does not need to be a custom manufactured conductor rail, but might be a (perhaps?) cheaper aluminum power cable with similar low resistance instead. Either way it needs to be bonded periodically to the primary conductor to ensure a useful amount of parallelism is achieved wherever the train is relative to the substation.

The normal approach on DC tramways, which I think also applies to DC railways, is to connect metallic objects within a certain distance of the track to the return rail via a device which normally insulates but becomes a conductor if the voltage across it exceeds the danger value of 50V. This is because these devices are also connected (usually unintentionally) to the earth (think metal bridge!) but you don't want to earth the DC return for the reasons explained above. However the system will be designed so that the 50V is only reached under fault conditions, probably only for a very short time before the breakers cut the power. In the case of the tramways this fault situation is usually the risk of the OLE falling onto a metallic object.

I found this useful document: http://orr.gov.uk/__data/assets/pdf_file/0012/5070/TTGN3.pdf

From here: http://orr.gov.uk/publications/guidance/health-and-safety/tramway-technical-guidance

Voltage limiting devices as you describe are mentioned in section 8 of this document. I do not recall any similar devices being used with DC 3rd rail electrification on Southern Region lines, but it is highly unlikely any member of the public would be in the position of being able to effectively make good contact with the running rails without also being within touching distance of the much more dangerous conductor rail, or without also being earthed by the same 'probe' (i.e by the shod foot when stepping on a rail at a level crossing). In theory the car body of a train could drift up to the rail voltage which could shock anyone on a platform touching a door handle assumingthey are also making a good earth. Metal equipment cabinets were not typically connected even to a local safety earth on older schemes, although more modern schemes require this for staff safety of those working within them, and absolutely nothing, except the elements of the return current system and any track circuits, is ever connected to the running rails which are always carried on insulated pads and held in place by insulated clips.

 

[AM9]

That's quite a difference in conductor resistance figures but don't forget that in multiple track installations, all the parallel rails can be effectively bonded together at intervals, either via inductance bond centre connections where track circuits are used or by simple cable bonding with axle counters. That could lower actual resistance from any particular train location to the substation further, say to around 10mOhm/km for a double track.

The aluminium composite is impressive, but resistance figures for traditional steel conductor rail is not so good, usually taken to be 16mOhm/km, according to: http://www.rgsonline.co.uk/Railway_...ck/Railway Group Standards/GMRT1001 Iss 1.pdf

A strengthening conductor for either leg does not need to be a custom manufactured conductor rail, but might be a (perhaps?) cheaper aluminum power cable with similar low resistance instead. Either way it needs to be bonded periodically to the primary conductor to ensure a useful amount of parallelism is achieved wherever the train is relative to the substation.



I found this useful document: http://orr.gov.uk/__data/assets/pdf_file/0012/5070/TTGN3.pdf

From here: http://orr.gov.uk/publications/guidance/health-and-safety/tramway-technical-guidance

Voltage limiting devices as you describe are mentioned in section 8 of this document. I do not recall any similar devices being used with DC 3rd rail electrification on Southern Region lines, but it is highly unlikely any member of the public would be in the position of being able to effectively make good contact with the running rails without also being within touching distance of the much more dangerous conductor rail, or without also being earthed by the same 'probe' (i.e by the shod foot when stepping on a rail at a level crossing). In theory the car body of a train could drift up to the rail voltage which could shock anyone on a platform touching a door handle assumingthey are also making a good earth. Metal equipment cabinets were not typically connected even to a local safety earth on older schemes, although more modern schemes require this for staff safety of those working within them, and absolutely nothing, except the elements of the return current system and any track circuits, is ever connected to the running rails which are always carried on insulated pads and held in place by insulated clips.

I agree that it would be a rare event, but a passenger standing on a wet salted platform with leaky shoes could touch the body of the train quite easily when boarding it.

Slightly OT, but such a risk should not necessarily be an absolute nono, - I remember when I waited for trains under an umbrella at GEML stations, that if I stood somewhere near the platform edge, I could get a mild shock sensation from the ferrule of the umbrella. I imagine that the wet canopy acted as a low value capacitor to the OHLE causing a low ac current to flow.

 

[Ships]

On another point, steel and concrete sleepers have a nominal 40 year life.

Whilst the steel sleeper them selves have a decent life span, a steel sleeper renewal (ie rail, sleeper, scarify) only lasts 10-15 years. If you look at the steel sleeper renewals on the SAC in the early 2000's they are now at the end of their service life.

Steel sleepers are actually about twice as expensive as concretes, but they're cheaper to install being lighter so two men can lift and set them out, (concretes need 8 men to lift and are ideally set out using a 7 sleeper hydraulic beam, thus needing an additional RRV).

The recent renewals on Wrexham bidson where all steels as the ballast was in fair nick and the main reason for the renewal was eliminating the existing jointed track.

Regarding redrilling concrete sleepers I've always been told this isn't really a viable option (neither is installing eg47 3r every 5th (?!) sleeper).

Interestingly on the line from Wigan to Kirby line (which Mersey rail has aspired to electrify) the more recent renewals seem to be on sleepers drilled for third rail pots.
--- old post above --- --- new post below ---

The other part of the maths needed then is the number of metres done.

Plus, finding out whether or not the upgrade cost to concrete was offered up to Merseytravel & partners to find the funding for, rather than the electrification project being lumbered with rip up and replace costs.

Let me put it like this, there is money to do all the work as steel sleepers or 1/2 or less as concrete. The objective of this sort of work is component driven, eliminate jointed or other poor quality track. Unfortunately there isn't money in the track renewals to provide passive provision for aspirational electrification.

 

[Bald Rick]

Whilst the steel sleeper them selves have a decent life span, a steel sleeper renewal (ie rail, sleeper, scarify) only lasts 10-15 years. If you look at the steel sleeper renewals on the SAC in the early 2000's they are now at the end of their service life.

Settle & Carlisle is an unusual case, in that when the steels were put in there was nothing to suggest the very large increase in heavy coal traffic that materialised a couple of years later. That is what has knocked hell out of what lies beneath the sleepers.

On low tonnage, predominantly passenger routes (which the S&C was of course), steels prevent the need for any renewal work to the ballast for roughly the same duration as if concrete or wood was used.

 

[HSTEd]

That's quite a difference in conductor resistance figures but don't forget that in multiple track installations, all the parallel rails can be effectively bonded together at intervals, either via inductance bond centre connections where track circuits are used or by simple cable bonding with axle counters. That could lower actual resistance from any particular train location to the substation further, say to around 10mOhm/km for a double track.

This is of course true, but remember that Track Paralleling Huts and similar equipment will result in the conductor rail being paralleled, reducing its resistance to 3.5mOhm/km.
So you don't actually change the calculation about what fraction of the voltage drop is actually a rise in the ground voltage.

The aluminium composite is impressive, but resistance figures for traditional steel conductor rail is not so good, usually taken to be 16mOhm/km, according to: http://www.rgsonline.co.uk/Railway_...ck/Railway Group Standards/GMRT1001 Iss 1.pdf

Indeed, but since the cost of aluminium is negligible compared to the project price - about £13500/km it seems silly to use steel rails.

A strengthening conductor for either leg does not need to be a custom manufactured conductor rail, but might be a (perhaps?) cheaper aluminium power cable with similar low resistance instead. Either way it needs to be bonded periodically to the primary conductor to ensure a useful amount of parallelism is achieved wherever the train is relative to the substation.

Indeed - but I was just using it an extra rail as the reinforcing conductor because it allows you to pick a reasonable quantity of aluminium without further calculations. Otherwise you might suggest a colossal aluminium bus-bar that costs a fortune without realising it in the initial calculations.

 

[ainsworth74]

Just on talk of sleepers, I thought third rail used longer than normal sleepers (to make room to mount the conductor rail) therefore surely any scheme involves the replacement of the old sleepers so it doesn't really matter what type they were previously? Or have I misunderstood?