Why don't Underground trains have swing-plug doors?

Doors are one of the major causes of delay.

Need to keep them as simple as possible.

I'd assume anything that opened outwards would be an absolute no-no on the underground now in case they hit tunnel walls in an accident or passengers packed onto platforms during normal service. I've watched plug doors open and they're much slower than the sliding doors so any increase in capacity gained would be lost due to increased dwell times and less trains per hour. I'd also have thought that a straight sliding door would be far more reliable and hard wearing than a plug door. The heavy and frequent use required by the underground would dictate the most reliable method.

The way I'd explain what I observed is that sliding doors have two simple operations, slide open and slide shut. A plug door adds swinging outwards to the opening and closing action thus making them slower. Multiply this over the course of many station stops and the time does add up. In trains where stops aren't as frequent and aerodynamics at speeds approaching, or over, 100mph are of more importance, a flush fitting door is of greater value. Underground trains rarely exceed 60mph and then only on the fastest sections of line so the drag of the doors isn't a significant disadvantage.

In the death, lots of intelligent, very experienced and well paid engineers have concluded that these doors are the best design for the purpose of mass transit in London. Although I'm not qualified in any way shape or form in the subject, from my perspective and from what I've seen, I agree with them.

Best to you!

Rich'

maxym - join us on SWT and watch those plug doors open and close, stop watch in hand. The studies for TLK showed that for a 45 sec dwell time, nearly half was consumed in plug door opening times. BTW airsmoothing is not really an issue for trains running at tube speeds.

To be fair Desiros are notorious for having slow doors, this applies to the Heathrow Connect ones too. (Even if the nice loud slam when they close gives the impression that the doors are very solid )

Also these type of doors are also more likely to go wrong, it happened the other day on a Hounslow service on SWT and we got into Waterloo 10 late after being on time until Wandsworth Town where the problems started.

The trams have swing plug doors, and quite often during the morning when they are packed they will slide to almost closed and then when they start to swing inwards, they will find themselves pushing against a person, and unable to move and so move back to open. Everyone breathes in, and pushes in towards the middle of the vestibule and the door finally closes. This happening on the tube would be quite a problem, not least because on the tube there would be significantly less room to squash into. Also, the argument about operation time is quite significant, the new desiro cities have sliding pocket doors as opposed to plug doors for the 28tph through the core. I doubt that you could make a plug door operate in the same time as sliding door, without it slamming shut at high speed, which could be a bit of an issue in terms of whacking someone in the face, or back of head &c.

Nice to see I've generated some lively debate. Very interesting. Swing-plug doors on the Chiltern line don't hang about when opening or closing... but I guess they could give you a nasty whack if you're in their way...

Yes, I recall hearing that no manufacturer of plug doors could guarantee the reliability that LU rightly demand.

For a metro operation like the Underground, dwell time has the most significant impact on journey times and service, so reliability of the train is vital.
As an example, an S7 on a Circle line journey from Hammersmith to Edgware Road:
7 cars, each with 6 door leafs per side. The doors on the train will be opened and closed 36 times each in that ~75 minute journey (if I have counted correctly). So that is 7 × 6 × 36 × 2 = 3024 door movements per journey or 2419 door movements per hour.
Compare that to a SWT class 450 with 4 leafs per side on a four-car unit. On a Waterloo to Basingstoke journey (eg the 15:12 tomorrow) a trian composed of 2 4-car units will call at 12 stations, so the doors will be opened and closed 11 times each. The sum is therefore 4 × 4 x 2 × 2 × 11 = 704 door movements on a journey of approximately the same length.

So the S stock has approximately 2419 door movements per hour, whereas the Class 450 has about 563 - a little more than 4 times fewer. Generally speaking Underground trains have a shorter layover between trips and will be in service for longer each day than NR trains, so you can see that even a small difference in either the failure rate or the speed of operation will make a huge difference.

To the Quizmaster, you're not comparing like with like - a four-car train v a seven-car one. If you compared an S7 to a notional 12-car class 450 the figures would be different... And you have to be careful with that sort of argument: for maximum reliability you'd want one set of doors (or even one door, D stock-style), or even a one-car train. The point about frequency of stops is well made though.

To the crusty, that Siemens for you. As I suggested earlier, Bombardier do rapid opening and closing really well.

I was actually comparing a 7-car S stock with a train composed of 2 4-car 450 units because I was comparing real-world operations. I don't have access to any set working details for any operator, but if anyone does it would be interesting to get a comparison of the typical daily workload of the door sets.

One other thing that I neglected to mention is headways and flexibility. A delay of 2 minutes on the Underground is far more significant in terms of reliability of the service than the same delay would be almost everywhere on the mainline network. The mainline network also has a significantly greater number of places a failed train can be worked around or put out of the way than the Underground does.

The people who are saying that they are not using plug doors because they are too unreliable and/or too slow are actually not answering the question asked.

They are answering the question: "Why don't underground trains use the same plug doors as mainline trains/buses/trams?"

Making plug doors work fast enough and with sufficient reliability for underground operation is simply an engineering problem and is highly unlikely to be insoluble.

The real question is: Would the advantage of plug doors be great enough to warrant the cost or development and production of suitable examples?

And the answer is evidently: No.

It's also, of course, possible that the safety elf has had some input in the decision.

The main advantage is that, like slam doors, they open away from the carriage side and therefore there are no door pockets to take up valuable interior space. (On more recent tube stocks - 1992 on - some space is saved by having no outer skin to the door pocket, so the doors slide on the outside of the car, but this means the body side (and therefore the interior) is unable to be the maximum width allowed by the loading gauge.)

As the OP said, they also allow smooth bodysides, which improves the aerodynamics.

The downsides are greater complexity and that, if stuck in the open position, a plug door may exceed the loading gauge.

I'm not sure that aerodynamics is that effective at underground service speeds, but the loading gauge issue is significant to me.

To answer the question I would suggest that money is a key factor!

In the open air, maybe not, but with the tight clearances in tube tunnels they will be more significant.