Dublin Metrolink - future routes for next Metrolink

bk wrote: »

strassenwo!f, to be VERY clear, you suggested more trams to Stephens Green. The NTA are still saying only 24 TPH to Stephens Green, but the possibility of 30TPH to Charlemont with a turn back platforms at Charlemont to link to the Metro there. Transfer people onto Metro at that location.

bk wrote: »

That is not at all what you were saying. You were saying that you can run more then 24 TPH on street running sections to Stephens Green, TII are still saying 24TPH is still the maximum for street running sections, but are suggesting an alternative of 30 TPH for a fully segregated section.

bk wrote: »

Note that a turnback facility at Charelemont will most likely require extensive CPO's and bulldozers around that area. It will likely be as expensive as the Green line tie in. But it does kick the can further down the line so to speak.

bk wrote: »

The other issue is that 30TPH through Dunville Avenue, etc. will almost effectively close it, with barely any time at all for cars and pedestrians to cross. IMO it will end up worse for them them then the over/underpass option.

salmocab wrote: »

Just on Dunville and the trams effectively closing the crossing, I don’t think it will be as bad as some think but certainly will make a bit of a balls of that crossing. It’s only a one way at a time crossing currently. From a northbound tram at the station it takes about 14/15 seconds from green to the tram passing. This will give probably 35/40 seconds on average for cars to cross I’d imagine that would get maybe 6/7 cars through. Obviously this will be variable based on the next tram along the other direction but it will mean some cars getting through followed by a wait of probably a couple of minutes at least until your direction gets the green again. The rush hours might be a massive pain especially for people trying to get off the residential roads that run parallel to the tracks. During the day it will be fine with less traffic and trams.

strassenwo!f wrote:

Ive just gone through Dunville ave a few minutes ago and timed the crossing, it took 14 seconds from the light changing to the back of the tram passing the crossing and presumably another 1/2 secs for the light to change again. That’s conservatively 15 secs per tram if anyone cares to work out the maths. It was a 55m tram I’m pretty sure.
That's very much in line with my calculation above that it should take just under 12 seconds for a 55m tram to get across the crossing. Add in 2 seconds or so that the driver needs to check that there are no messers near the crossing and you get to the 14 seconds you saw. Another second or two for the lights to change and you get to, let's say, for the sake of argument, 20 seconds for the whole operation.

I've checked the throughput at that location several times, both on the website and by being there, and it is clear that the current throughput is 20 tph at peak times, in the morning peak heading into town. A tram, on average every 3 minutes. As I also mentioned above, my local traffic lights (on a two-way road, a single lane in each direction) gets on average 10 cars across the junction, in each direction (20 in total).

Given that there are fewer trams going against the peak direction, it should assuredly be very easy to set the traffic signals to favour trams travelling in the peak diection of travel, and trams going in the other direction may have to wait a short time. Thus in the morning peak they would prioritise northbound traffic at this junction and southbound traffic might have to wait a short time before crossing. This should not be a problem - it happens at many junctions in the city that a tram has to wait a short time.

At Dunville Avenue, at present, the above figures should give, on average, 2 minutes and 40 seconds to cross before the next tram cycle. Allowing 1 minute and 15 seconds for each direction, with a 10 second gap for the change, that equates to around 60 cars going across the crossing during each tram cycle, assuming the figures for my local crossing are pretty much replicated (and I can't see why they wouldn't be). 20 such cycles per hour (given a tram throughput of 20 tph) should theoretically allow 1,200 cars across that crossing per peak hour.

There are probably going to be people who 'just need to pop in for a second' to Morton's, at peak times, to pick up the week's supply of arugula and coconut milk, and park awkwardly outside the shop, screwing the whole thing up.

But even at just half the above figure (600 cars per hour), is there really such a level of demand to cross at Dunville Avenue at peak times?

If the LUAS Green line were to go to the maximum throughput which is currently allowable for tram systems in Ireland, 24 tph (a tram every 2.5 minutes), the corresponding figures would be a tram cycle of 150 seconds, a traffic cycle of 130 seconds, and a theoretical throughput of 960 cars per hour across that crossing at peak times. (If we factor in an occasional necessity to take the 4x4 to get some Manuka honey in Morton's, at peak times, and 'just park outside for a second', we can probably reduce that to a practical capacity of around 500 cars per hour).

Interestingly, in relation to this junction at Dunville Avenue, an increase of tram throughput to around a tram every 2 minutes - as seen in several European cities - would not appear to have a disastrously negative effect on the local road traffic at peak times. It would, of course, require a change to the current transport rules in Ireland - but that's easily done (much easier than the recent abortion change, or the civil partnership change, in the country's big rule book).

The next step for Dublin, from the current 20 tph to the maximum currently allowed (24 tph), would then be to a tram cycle of 120 seconds, which would probably be too short for travel in both directions at the Dunville Avenue junction. Thus you could dispense with the two-way traffic mentioned above, saving 10 seconds on the proposed changeover across that crossing.

Thus, a tram cycle of 2 minutes (120 seconds) would equate to 100 seconds of car traffic throughput of around 40 cars in just one direction on each cycle. The first cycle would be from one direction, the next would be from the other, and with 30 cycles per hour that equates to 600 cars across that crossing.

That would involve people waiting to cross the crossing, in some cases for about 4 minutes, but that is really quite small compared to what other Dubliners have to do to get across a junction. People travelling in cars across the DART line, for example, would often face much longer waits.

And it would probably be necessary at that stage to tighten up on the peak-time demand for quinoa-infused avocado.