Amtrak delays 2023 H2 - 2024

Amtrak Unlimited Discussion Forum

Help Support Amtrak Unlimited Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
These incidents are a dispatcher's worse nightmare. Remember that the involved dispatchers cannot just stack up regular trains. They have to quickly make a clear path for wire repair trains to get to the location(s) of downed wire. In this case probably one on each of both tracks. Does anyone know where wire repair trains are based. In this case one or more might have been working on the Keystone route. If that was the case that wire train would have to button up its work there and then get clear signals all the way to Newark.
 
These incidents are a dispatcher's worse nightmare. Remember that the involved dispatchers cannot just stack up regular trains. They have to quickly make a clear path for wire repair trains to get to the location(s) of downed wire. In this case probably one on each of both tracks. Does anyone know where wire repair trains are based. In this case one or more might have been working on the Keystone route. If that was the case that wire train would have to button up its work there and then get clear signals all the way to Newark.

I would hope Amtrak doesn’t have all its wire repair assets on the Keystone. Then again, this is Amtrak we are talking about. 🤪

I suspect maybe Wilmington has equipment staged there. Hopefully someone more in the know will have an answer.
 
The thinking was that the track replacement on the Keystone included wire replacements. Was the PHL wire train there yesterday?? Do not know if wire replacement includes Constant Tension installation but hope so as this incident shows the vulnerability of PRR style CAT.
 
It is odd that due to traction power failure Raritan Valley Line is canceled. It does not even use any traction power. It is a diesel operation.
Apparently there were two places where there were issues - damaged catenary reported at Swift interlocking in Kearney and some unspecified damage at Hunter interlocking in Newark.

Also apparently there was loss of signal power between Midway and Bergen, so that would explain the cancellation of Raritan Valley service into Newark until signal power was restored.

Here is a report on it from Trains Mag:

https://www.trains.com/trn/news-rev...commuter-service-between-new-york-and-philly/

Oddly it mentions issues "east of NYP" which is at odds with almost everything else I have read. I think they meant to say "west" as Kearney and Newark are "west of NYP". They do not mention the signal power outage.

Also, apparently no pantograph entanglement with catenary was involved since there appears to be no report of some train stuck at Kearney with a broken panto, and an Amtrak spokesman is quoted saying that they don't know why the lines came down. So this may not be a variable vs. constant tension catenary issue. Fatigue failure of some sort maybe?
 
Last edited:
According to the New York Times, "an overhead wire that transmits traffic signals fell and struck a cable that provides electrical power to trains," which caused a "blowout" (which seems like an odd word to use without elaboration).
 
These incidents are a dispatcher's worse nightmare. Remember that the involved dispatchers cannot just stack up regular trains. They have to quickly make a clear path for wire repair trains to get to the location(s) of downed wire. In this case probably one on each of both tracks. Does anyone know where wire repair trains are based. In this case one or more might have been working on the Keystone route. If that was the case that wire train would have to button up its work there and then get clear signals all the way to Newark.

I would guess there are wire trains at the Adams Yard. I can't imagine critically important MOW vehicles being so far away. There may be other things strategically scattered around such as at Colonia switch and Durant switch.

Apparently there were two places where there were issues - damaged catenary reported at Swift interlocking in Kearney and some unspecified damage at Hunter interlocking in Newark.

Also apparently there was loss of signal power between Midway and Bergen, so that would explain the cancellation of Raritan Valley service into Newark until signal power was restored.

Here is a report on it from Trains Mag:

https://www.trains.com/trn/news-rev...commuter-service-between-new-york-and-philly/

Oddly it mentions issues "east of NYP" which is at odds with almost everything else I have read. I think they meant to say "west" as Kearney and Newark are "west of NYP". They do not mention the signal power outage.

Also, apparently no pantograph entanglement with catenary was involved since there appears to be no report of some train stuck at Kearney with a broken panto, and an Amtrak spokesman is quoted saying that they don't know why the lines came down. So this may not be a variable vs. constant tension catenary issue. Fatigue failure of some sort maybe?
I'd guess Bergen and Midway are isolation points for signal power. That Bergen is one of them makes sense because signal power is 91 2/3 Hz where there's DC (third rail) which ends right at the tunnel portals, and 100 Hz elsewhere, if old PRR info still applies.

"Fixed termination" is the standard term I guess. The two terms (constant tension vs. fixed termination) describe two different aspects: termination method (construction) vs. tension variability (advantage/effect).

Did you hear of the issues in the tunnels?
Friend of mine works in Amtrak maintaince, hes at the tunnel, he says the 1 is cleared but need inspection first before open up while other one where that wire is on the tracks and confirmed its about 3 car long, thats how the wire is on the ground -OneDisastrous998 (Reddit 7:58PM, May 22)


According to the New York Times, "an overhead wire that transmits traffic signals fell and struck a cable that provides electrical power to trains," which caused a "blowout" (which seems like an odd word to use without elaboration).
The internet is a series of tubes

I think they mean signal power (6.2kV) made contact with traction power (12kV) which has obviously bad consequences. Who knows where they got their information, they're "journalists" who never enumerate sources. Signal control and status is probably buried.
 
According to the New York Times, "an overhead wire that transmits traffic signals fell and struck a cable that provides electrical power to trains," which caused a "blowout" (which seems like an odd word to use without elaboration).

So, another incident of overhead utility line(s) falling onto the CAT shorting out the CAT and maybe bringing both tracks down as well as the signal 4400 V power line? Some of the infrastructure funds should be used IMO to put all utility lines under the tracks. Wonder who allowed the traffic signal line easement over the tracks? PRR, PC, CR, or Amtrak??

Blowout?? Maybe a transformer or CB was damaged?? Who knows?
 
According to the New York Times, "an overhead wire that transmits traffic signals fell and struck a cable that provides electrical power to trains," which caused a "blowout" (which seems like an odd word to use without elaboration).
Well, I think "overload" is probably more accurate terminology. A super high voltage surge from the train power line on the low voltage "code line" certainly wouldn't do the electronics on it any good.
 
I would guess there are wire trains at the Adams Yard. I can't imagine critically important MOW vehicles being so far away. There may be other things strategically scattered around such as at Colonia switch and Durant switch.


I'd guess Bergen and Midway are isolation points for signal power. That Bergen is one of them makes sense because signal power is 91 2/3 Hz where there's DC (third rail) which ends right at the tunnel portals, and 100 Hz elsewhere, if old PRR info still applies.

"Fixed termination" is the standard term I guess. The two terms (constant tension vs. fixed termination) describe two different aspects: termination method (construction) vs. tension variability (advantage/effect).

Did you hear of the issues in the tunnels?
Friend of mine works in Amtrak maintaince, hes at the tunnel, he says the 1 is cleared but need inspection first before open up while other one where that wire is on the tracks and confirmed its about 3 car long, thats how the wire is on the ground -OneDisastrous998 (Reddit 7:58PM, May 22)



The internet is a series of tubes

I think they mean signal power (6.2kV) made contact with traction power (12kV) which has obviously bad consequences. Who knows where they got their information, they're "journalists" who never enumerate sources. Signal control and status is probably buried.
The "blowout" term came directly from an Amtrak spokesperson, says USA Today. How about the NYT headline though? "Why Amtrak Is to Blame for a Hellish Night for New Jersey Commuters." The article under it is a decent summary and analysis of the state of the NEC. New Jersey's governor sent an angry letter to Amtrak, perhaps prompting the depth of the NYT story. It has one main source quoted for its analysis, Thomas K. Wright, who was on his way to Princeton when he got stranded, and is chair of the Regional Plan Association. That's indeed its real name, covering NY-NJ-CT.

The aging senator who said the internet is a series of tubes got a bum rap. It's not a bad metaphor. P=VI is described by water pipes in every textbook (voltage = diameter, current = current). Transistor logic can be modeled by water in clear plastic pipes with T-shaped valves actuated by water flow, in a classroom experiment, though I don't know how often it's done. If I recall there's a YouTube(!) of it. The senator just forgot to say it was a metaphor.
 
Yesterday Thursday 19, 91, 97 all took delays of over 1 hour due to overhead power problems at WAS. speculation but maybe problems with CAT on the lower level preventing ACS64s from being removed??- As far as regionals unknown?

05/23/24 6:45pm EDT
_ Mechanical Issues: As of 6:42 pm ET due to overhead power
_ issues Train 91 is currently stopped in Washington (WAS).
_ Updates to come as more information becomes available.
_
_ 05/23/24 6:57pm EDT
_ Service Resumption: As of 6:54 pm ET Train 91 has departed
_ Washington (WAS) and is currently operating approximately
_ 3 hours and 45 minutes late.
 
Yesterday Thursday 19, 91, 97 all took delays of over 1 hour due to overhead power problems at WAS. speculation but maybe problems with CAT on the lower level preventing ACS64s from being removed??- As far as regionals unknown?

05/23/24 6:45pm EDT
_ Mechanical Issues: As of 6:42 pm ET due to overhead power
_ issues Train 91 is currently stopped in Washington (WAS).
_ Updates to come as more information becomes available.
_
_ 05/23/24 6:57pm EDT
_ Service Resumption: As of 6:54 pm ET Train 91 has departed
_ Washington (WAS) and is currently operating approximately
_ 3 hours and 45 minutes late.
If you look at the run of 91 from NYP to WAS, it departed on time. Took a little more than 1 hour delay departing Newark. Then too additional 1 hour plus delays between Wilmington and Baltimore and then between Baltimore and Washinton. The latter may have been mostly at Washington after arrival. I am not sure about that.
 
According to the New York Times, "an overhead wire that transmits traffic signals fell and struck a cable that provides electrical power to trains," which caused a "blowout" (which seems like an odd word to use without elaboration).
What are "traffic signals"? Are these railroad signals, or road traffic signals (connecting or synchronizing stop lights on roads that cross the tracks)? I know there are are no grade crossings in the NEC in New Jersey, but there are many road bridges and underpasses and almost all of them have power, telephone and coax cables paralleling the streets. Some of them must also carry traffic light synchronization wires as well.

Who owns and is responsible for the "traffic signals" wire?

It sounds like they have no idea why the wire fell. Age and slow deterioration? Wind? Did someone or something strike a pole? Vandalism? It certainly wasn't snow or ice on an 80+ degree day in late May!
 
What broke is a signal power supply cable and it apparently came in contact with a traction power cable causing the so called blowout. As discussed a few post up in this thread, for a short while it knocked out all signal power between Midway (Monmouth Jct) all the way to Bergen (just outside the Hudson Tubes entrance in NJ. That is the section closest to NYP of the 100Hz signal power cable. The frequency of signal power in the tunnels and Penn Station is 90 something Hz (I forget the exact number) and was not affected. In addition it also knocked out traction power over a shorter segment. The incident of cable breakage apparently took place in the vicinity of Swift interlocking in Kearney.

The other incident in the tunnel mentioned in the article on a different day was caused by an NJ train pulling down a bunch of catenary in one of the tunnels, and as mentioned, is unrelated to this. Of late NJT has been pulling down a lot of catenary at a disturbingly frequent rate. I don't know what that is all about.
 
Last edited:
What broke is a signal power supply cable and it apparently came in contact with a traction power cable causing the so called blowout. As discussed a few post up in this thread, for a short while it knocked out all signal power between Midway (Monmouth Jct) all the way to Bergen (just outside the Hudson Tubes entrance in NJ. That is the section closest to NYP of the 100Hz signal power cable. The frequency of signal power in the tunnels and Penn Station is 90 something Hz (I forget the exact number) and was not affected. In addition it also knocked out traction power over a shorter segment. The incident of cable breakage apparently took place in the vicinity of Swift interlocking in Kearney.

The other incident in the tunnel mentioned in the article on a different day was caused by an NJ train pulling down a bunch of catenary in one of the tunnels, and as mentioned, is unrelated to this. Of late NJT has been pulling down a lot of catenary at a disturbingly frequent rate. I don't know what that is all about.
Do the catenary and signal line share poles? Is it possible that NJT pulled down or tugged on the catenary, causing the pole to shake, and in turn causing the signal line to fall onto the power line?

It would be ironic if the NJ governor fired off his nastygram to Amtrak when NJT actually caused the problem.
 
Do the catenary and signal line share poles? Is it possible that NJT pulled down or tugged on the catenary, causing the pole to shake, and in turn causing the signal line to fall onto the power line?

It would be ironic if the NJ governor fired off his nastygram to Amtrak when NJT actually caused the problem.
NJ Governor's nastygram also takes a little bit of freedom of speech with the $100 Million number (NJT contributed more like $84 million for infrastructure in FY23).

The catenary supports, feeder power lines and signal power lines share the same pole. However I have not seen any report of catenary pull down on the the date and time of this incident. Of course that does not mean it did not happen, but Amtrak spokesman is on record saying they don;t know what brought the cables down.
 
The national electrical code (NEC) for outdoor pole installations has the highest eclectic voltage on the top with descending voltages on down. So, for lines on Amtrak poles you would have the 169.0 kV 25 hZ on top, then the 12.0 kV CAT power, Next the 4400(?) V signal power, then the signal control wires (if installed and not buried), then TV cable lines (probably none), then telephone cable lines (probably not 105 V ring and 48 V talk). Even any utility lines crossing the tracks will follow the same protocol except of course they do have to be above the CAT.

With this in mind not enough info as to what fell on what?

EDIT: The one exception I know of is RRs. RR telephone and telegraph lines were installed 1st on the pole lines. Then either DC signal control or signaling and CTC + 2 or 3 power lines for transformers. were installed usually on a new bottom cross arm. AC signaling such as SOU RR had was installed on a different pole line on other side of tracks from telephone and telegraph lines to prevent interference with communications.
 
Last edited:
The national electrical code (NEC) for outdoor pole installations has the highest eclectic voltage on the top with descending voltages on down. So, for lines on Amtrak poles you would have the 169.0 kV 25 hZ on top, then the 12.0 kV CAT power, Next the 4400(?) V signal power, then the signal control wires (if installed and not buried), then TV cable lines (probably none), then telephone cable lines (probably not 105 V ring and 48 V talk). Even any utility lines crossing the tracks will follow the same protocol except of course they do have to be above the CAT.

With this in mind not enough info as to what fell on what?

EDIT: The one exception I know of is RRs. RR telephone and telegraph lines were installed 1st on the pole lines. Then either DC signal control or signaling and CTC + 2 or 3 power lines for transformers. were installed usually on a new bottom cross arm. AC signaling such as SOU RR had was installed on a different pole line on other side of tracks from telephone and telegraph lines to prevent interference with communications.

On NEC it is 6.9kV 91.67 Hz. signal power.

From the description of the outcome my guess is that a high tension power cable fell on the Signal power line knocking it out for the entire zone Midway to Bergen. But it is just a marginally informed guess..
 
The national electrical code (NEC) for outdoor pole installations has the highest eclectic voltage on the top with descending voltages on down. So, for lines on Amtrak poles you would have the 169.0 kV 25 hZ on top, then the 12.0 kV CAT power, Next the 4400(?) V signal power, then the signal control wires (if installed and not buried), then TV cable lines (probably none), then telephone cable lines (probably not 105 V ring and 48 V talk). Even any utility lines crossing the tracks will follow the same protocol except of course they do have to be above the CAT.

With this in mind not enough info as to what fell on what?

EDIT: The one exception I know of is RRs. RR telephone and telegraph lines were installed 1st on the pole lines. Then either DC signal control or signaling and CTC + 2 or 3 power lines for transformers. were installed usually on a new bottom cross arm. AC signaling such as SOU RR had was installed on a different pole line on other side of tracks from telephone and telegraph lines to prevent interference with communications.

On NEC it is 6.9kV 91.67 Hz. signal power.

From the description of the outcome my guess is that a high tension power cable fell on the Signal power line knocking it out for the entire zone Midway to Bergen. But it is just a marginally informed guess..
https://web.archive.org/web/20110723013215/http://www.amtrakengineer.net/AMT2111505.pdf
Page 21 from AMT-2

It says that signal power is 2.3kV, 91 2/3 Hz between Hack (Hackensack Substation [42] which is located at Bergen, the tunnel portals) and Gate and 6.9kV, 100 Hz for an unspecified section, likely the entire 25Hz system from Hack to Washington excluding Zoo-Paoli.

Signal wires (at least in PRR design) should be slightly above the catenary but on the side. Check page 26 for a diagram. 138kV lines are the highest. It's actually split-phase so each conductor is 69kV to ground.

Communication lines are probably buried.



The other incident in the tunnel mentioned in the article on a different day was caused by an NJ train pulling down a bunch of catenary in one of the tunnels, and as mentioned, is unrelated to this. Of late NJT has been pulling down a lot of catenary at a disturbingly frequent rate. I don't know what that is all about.
But the meltdown appeared to be unrelated to a problem on Tuesday morning with wires in a tunnel under the Hudson that led to delays of up to 60 minutes.

There was apparently a separate incident on Wednesday regarding catenary wires in the tunnel(s) as well, at least according to the live-update Reddit post from OneDisastrous998 I outlined above. Another compounding factor for Wednesday's catastrophe.

I love that we're unsure about which catenary failures we're talking about.


I forgot to mention I heard a train complain just east of Edison that they lost (and restored) power at 5:03PM.
 
https://web.archive.org/web/20110723013215/http://www.amtrakengineer.net/AMT2111505.pdf
Page 21 from AMT-2

It says that signal power is 2.3kV, 91 2/3 Hz between Hack (Hackensack Substation [42] which is located at Bergen, the tunnel portals) and Gate and 6.9kV, 100 Hz for an unspecified section, likely the entire 25Hz system from Hack to Washington excluding Zoo-Paoli.
I stand corrected. Thanks.
Signal wires (at least in PRR design) should be slightly above the catenary but on the side. Check page 26 for a diagram. 138kV lines are the highest. It's actually split-phase so each conductor is 69kV to ground.

Communication lines are probably buried.
Communication lines these days are also optical fiber which would trend not to go "boom" when power lines come in touch. But yeah, in many sections they are now underground.
But the meltdown appeared to be unrelated to a problem on Tuesday morning with wires in a tunnel under the Hudson that led to delays of up to 60 minutes.

There was apparently a separate incident on Wednesday regarding catenary wires in the tunnel(s) as well, at least according to the live-update Reddit post from OneDisastrous998 I outlined above. Another compounding factor for Wednesday's catastrophe.

I love that we're unsure about which catenary failures we're talking about.


I forgot to mention I heard a train complain just east of Edison that they lost (and restored) power at 5:03PM.
I heard later that the tunnel catenary issue was another NJT pantograph performance, but I am not quite sure about which of the several(apparently) catenary damages was being talked about.
 
I heard later that the tunnel catenary issue was another NJT pantograph performance, but I am not quite sure about which of the several(apparently) catenary damages was being talked about.
There was also an Arrow III set that morning that got stuck at Metropark on track 4. I'm missing radio recordings but I saw some people on the roof around the pantograph and the catenary was out.
 
I understand that there were numerous delays in Chicago yesterday, June 18, caused by a warehouse fire which caused tracks to be closed. Track was reopened at 8:30 p.m. I understand. Hopefully someone has more details. I understand that 8 and other trains were delayed for hours.
 
To further explain, the fire was on the north side of the tracks only three blocks east of the busy junction of the old Milwaukee Road lines with the Union Pacific West line. The Amtrak services mentioned use the former, in particular Milwaukee-North.

I don't know how far east of the junction a train intended for the Milwaukee lines has to be switched to the right to not end up on the UP, but I suspect too far east to do the otherwise-obvious: run on the far south/left to avoid the fire and then cut north/right once past the fire.
 
Back
Top