|
|
| Re: South Ferry Inner Loop Station. (1385617) | |||
 |
|||
| Home > SubChat | |||
|
[ Read Responses | Post a New Response | Return to the Index ] |
|
||
|
Re: South Ferry Inner Loop Station. |
|
|
Posted by Stephen Bauman on Mon Feb 15 12:17:16 2016, in response to Re: South Ferry Inner Loop Station., posted by Jace on Mon Feb 15 10:12:10 2016. The more hardware, the more higher the failure rates.You need to read up on fault tolerant networks. Here's a quick synopsis regarding relays. Let's differentiate between faults and failures. A fault is a malfunction in an individual component relay. A failure is the relay network producing a false result. Suppose an individual relay had a closing failure probability of pc. That is the relay contacts should close but don't. If the coils and contacts were wired in parallel, then both relays would have to exhibit a fault before the network of both relays would exhibit a failure. N.B. pc2 < pc for all values of pc < 1. The same paradigm applies to a contact opening failure, only that the coils are wired in parallel and the contacts are wired in series. This is an example wherein redundant components ensure reliability. And you still didn't say how a flexible door system would need to be designed. The message structure would permit opening/closing individual door leaves in each of the cars on the train. Message length is severely limited on LonWorks. This reduces the number of different messages that can be sent. That's probably why the NTT designers did not increase the messages beyond the bare minimum to make the trains operate. One way or another, the train would have to know where it is 100% of the time At this point there are T/O's and C/R's assigned to each train that do that. Axle tachometers then track progress. That's a problem. The tolerance for wheel diameter is 1/8 inch out of 34 inches. That's 0.37% tolerance. That translates to a 19.4 ft error for every mile traveled. Any dead reckoning system to measure distance needs to be reset at intervals that are frequent enough to keep accumulated error within a desired range. I'd say is minimal gain (one station that hasn't been used in 40 years). Let me give you another instance where being able to operate individual doors could be used. The L train station platforms are 536 feet long. They could accommodate 540 ft long trains, with the front and rear overhanging the station by 2 feet. If the front and rear doors did not open, there would be no danger to passengers of missing the platform when they got off the train. This would increase the effective service on the L train by 12.5% or 3 more trains during peak am service. That's a lot cheaper than adding the CBTC equipment on those 3 extra train sets and the recurring cost of 3 crews to operate them. The bucking you get now has to do with the way the propulsion system is designed. Agreed. The propulsion system was not designed to obey specific speed/acceleration commands. That's easy to implement with modern motor controllers (available since the 1970's.) So far as I know, the R142's use open loop motor controllers. Most motor controllers use closed loop control to ensure tight performance specs. |
|
|
|