Friday’s Headlines: Rainy and Cold Edition

OK, so maybe it won’t be a gorgeous weekend, but at least the kids have off from school on Monday.

Ponder that as you take in today’s headlines:

By all measures, the paired protected lanes on 43rd and Skillman avenues in Sunnyside are doing great. So come celebrate them on Nov. 10 or protest them on Nov. 18, when the bizarrely named “Queens Streets for All” complains that the lanes have hurt business and endangered the public (when, in fact, they have done the opposite). (Queens Chronicle)
Meanwhile, Bike Snob put it succinctly.
Stand clear of the changing announcements. (NY Times)
Another Queens pedestrian was killed by a driver. (NYDN)
The MTA is trying something new to get trains to move faster. (NY1)
The guy who used to run Citi Bike is now working for Virgin’s hyperloop. (The Verge)
Friend of Streetsblog Joel Epstein points out why the MTA’s “Adopt a Station” program is going so badly. (Crain’s)
The great Neil Demause takes a deep and exhaustive dive on Amazon’s proposed move to Long Island City — and finds that there are a lot of unanswered questions. (Old man editor’s note: Hey, kids, this is a perfect example of a diligent reporter taking the extra day to get a lot more details than everyone else had.) (Gothamist)
Meanwhile, the Times also looks into the broader implications of the Amazon economy.
And, finally, a California woman threw a birthday party on the J train. (NY Post)

Joe R.

How about just removing the timers, except the ones protecting sharp curves? It defies logic to have timers on some dead straight express runs, but the MTA does exactly that. Also, restore the trains to their original performance specs.

I also don’t know how a timer can “go out of calibration”. A cheap quartz watch has an accuracy of 15 seconds per month, or 0.0006%. It shouldn’t be hard for timers to hold calibration to, say, 0.1 mph. I’d love to know exactly what they’re using to time out the interval. It certainly isn’t a quartz oscillator if the timers are this badly out of calibration. Maybe it’s a 1955 era Mickey Mouse watch, although even that could probably keep time to within a few tenths of a percent.
fdtutf

input it into the control circuitry

HAHAHAHA

You don’t know how the signals work in New York, do you? For the most part, they’re not computerized. (The only exception is the CBTC that’s active on the L and coming on the 7 and the Queens Boulevard line.) I don’t specifically know what kind of timers the MTA uses in the subway, but it’s probably something electromechanical.
Joe R.

The signals themselves are probably controlled by relays but any type of time delay should be electronic. You don’t necessarily need computers or microcontrollers. A simple quartz oscillator and logic can do the job. The key is what is actually doing the timing? If they’re using something mechanical in 2018 those responsible should have their head handed to them. The way the MTA does things if you didn’t know any better you would still think it was 1950 (and I’m being charitable here).
fdtutf

The key is what is actually doing the timing? If they’re using something mechanical in 2018 those responsible should have their head handed to them. The way the MTA does things if you didn’t know any better you would still think it was 1950 (and I’m being charitable here).

No disagreement here. My point was simply that I think the current situation won’t allow quick fixes. Fully agree that this is an urgent problem requiring a solution.
Larry Littlefield

The Economist says the tech giants are creating a “kill zone” around them that stifles startups and entrepreneurship.

https://www.economist.com/briefing/2018/09/01/silicon-valley-is-changing-and-its-lead-over-other-tech-hubs-narrowing

Subsidize today’s firms and lose tomorrow’s, then find that today becomes yesterday and tomorrow becomes today. How are Xerox and Kodak doing up in Rochester?

You want the next Amazon.
sbauman

I also don’t know how a timer can “go out of calibration”. A cheap quartz watch has an accuracy of 15 seconds per month, or 0.0006%. It shouldn’t be hard for timers to hold calibration to, say, 0.1 mph. I’d love to know exactly what they’re using to time out the interval. It certainly isn’t a quartz oscillator if the timers are this badly out of calibration.

They are using adjustable time delay relays which use an RC circuit to establish the delay. They don’t even have a 555 type voltage comparator so the pull-in time is supply voltage dependent.

Let’s say there’s a 500 foot section that’s got a 20 mph timer. This means the timer should be set at 17.01 seconds. A 0.1 mph speed difference represents a 0.5% difference. Translated to the timer it’s a 0.0855 second time difference. Even a 1 mph difference translates to less than a second difference at the timer. Factor in the installer’s reaction time. It’s fairly difficult to make that fine adjustment, when the time standard is a mechanical watch with a sweep second hand.
Fool

It is someone’s job to not maintain those timers!
Fool

These are a different signaling system.

The MTA currently has… 4 different kinds of signaling systems operating.
Larry Littlefield

Perhaps these folks?

https://www.dnainfo.com/new-york/20120820/new-york-city/mta-workers-charged-with-faking-signal-inspections/

https://www.wnyc.org/story/284800-union-says-ny-subway-inspection-fraud-arrests-miss-the-mark-big-perpetrators-are-getting-off/

But remember this. The big news here isn’t what happened, it’s some measure of accountability for what happened. You don’t have that elsewhere in the NY public sector, or even elsewhere in the MTA.

There seems to be a pitchfork and torches mob out for the subway since its failings was reported on a year ago. But I can tell you there is worse elsewhere.
fdtutf

What are “these”?
Fool

1) The actual control signalling for the subway is the “old” block type signalling. That system is universally deployed across the system.
2) These timers that trigger emergency stop are a completely separate signalling infrastructure that operates on top of the block type system.
3) Then there is the bluetooth beacon based “countdown” signalling system. Completely separate from the above.
4) CBTC that has been installed on the L and the 7.

Edit: Actually 5 signalling systems. Forgot about the Time Square Shuttle automation. (Before the TWU arson’d it.)
sbauman

2) These timers that trigger emergency stop are a completely separate signalling infrastructure that operates on top of the block type system.

You’re talking about the wheel detector (WD) signals. Their timing duration is set by the time for successive wheels to pass them. WD signals are the exception rather than the rule for the number of timers in the system.

3) Then there is the bluetooth beacon based “countdown” signalling system. Completely separate from the above.

This isn’t connected to the signal system nor is it used for traffic control (collision avoidance) or train supervision (ensuring trains adhere to their schedules).
bolwerk

You’d think our region would be better about taking advantage of Silicon Valley’s price squeeze. Manhattan might be as expensive, but much of the rest of the boroughs are not. We have several points in our favor:

– quietly perhaps, but a robust base of technology companies and research institutions (and even a history of tech innovation) – so there is a talent pool here already, and has been for generations

– most major Silicon Valley companies have at least New York satellite offices

– a finance industry that could probably be leveraged toward venture capital

– we’re three time zones closer to to critical customer bases in dozens of other rich countries, and within hundreds of miles of large customers between Virginia and Maine

– we have some lock-in with networking companies that need to be here because of how communication infrastructure developed

On the downside, it’s a bit hard for us to make the case that our public transportation system is world class, and public transportation does appeal to tech people. But at least the Hudson Valley and Long Island have room for tinkerers, including a declining industrial building stock.
Joe R.

Ugh, then it’s as bad as I thought. They should have replaced the RC circuit with a quartz oscillator decades ago. It’s not like any of this is some newfangled, unproven technology. Hopefully someone will fill in Byford on the heart of the problem, and he’ll order changes which prevent it from ever happening again.
Joe R.

Part of the reason for the mob gunning for the subway are the revelations of just how inefficient and outdated many of the MTA’s practices and technology are. I understand the time and expense involved with wholesale replacement of the antiquated block system. I don’t understand using RC circuits for timing in 2018. I learned in school that you *don’t* use those in applications where being off 20% or more can cause problems. And yet the MTA is using them to time something where tenths of a percent matter. That’s not even getting into the fact most of the timers are not needed for safety or other reasons. They just represent yet another thing in a complex system which can fail and delay trains. It’s almost as if the MTA is actively trying to drive away customers so its employees can get paid for doing nothing all day.
sbauman

Hopefully someone will fill in Byford on the heart of the problem, and he’ll order changes which prevent it from ever happening again.

I’m not hopeful. Mr. Byford does not have a technical background to independently analyze the situation. He appears to rely on vendor input that masquerades as technical expertise. Cost reduction and providing quantitative performance data aren’t in a vendor’s vocabulary.
fdtutf

(2) is an add-on to (1).
(3) is not a signaling system.

So there are two signaling systems in the subway now. As you point out, the automated system used on the 42nd Street Shuttle no longer exists.