The Electrification discussion

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.
I fell down a rabbit hole earlier this week on Indian suburban rail networks and hadn't realize how early some had been electrified (Mumbai, Chennai and maybe Kolkata [can't remember exactly] ) as early as the mid-20's.
Mumbai (Bombay) was (Broad Gauge) 1925 1.5kV DC 2016 25kV AC conversion completed
Chennai (Madras) was (Meter Gauge converted to Broad Gauge in the last few decades) 1931 1.5kV DC, 1967-71 25kV AC
Kolkata (Calcutta) was (Broad Gauge) 1957 3kV, 1961-64 25kV AC

Incidentally, both Madras (Chennai) and Mumbai DC system used equipment from Metro-Cammel. Initial AC equipment came from the Hitachi-Mitsubishi-Toshiba Japanese consortium and the 50Hz Group in France/Europe. Progressively Chittaranjan Locomotive Works, BEML and Jessops started manufacturing equipment in India. At present a mix of local manufacture and imports are used to meet the insane size of the demand for new equipment. Foreign suppliers include Siemens, Alstom and CRRC.
 
Last edited:
Those would make a great re-use project.

Funnily enough, if you drive on Route 12 in Indiana, you'll also see electrified railway infrastructure - coinkydink?
Route 12, the insiders guide to electrified railways? Maybe.
One other thing I did not realize was that after the Milwaukee Road closed their substation, the rural electric decided they needed one there so they built a new, more modern one. Not surprisingly, I noticed the old one, not the newer one.
 

Attachments

  • Image 7-19-24 at 11.32.jpeg
    Image 7-19-24 at 11.32.jpeg
    383.8 KB
  • Image 7-19-24 at 11.33.jpeg
    Image 7-19-24 at 11.33.jpeg
    482 KB
Mumbai (Bombay) was (Broad Gauge) 1925 1.5kV DC 2916 25kV AC conversion completed
Chennai (Madras) was (Meter Gauge converted to Broad Gauge in the last few decades) 1931 1.5kV DC, 1967-71 25kV AC
Kolkata (Calcutta) was (Broad Gauge) 1957 3kV, 1961-64 25kV AC

Incidentally, both Madras (Chennai) and Mumbai DC system used equipment from Metro-Cammel. Initial AC equipment came from the Hitachi-Mitsubishi-Toshiba Japanese consortium and the 50Hz Group in France/Europe. Progressively Chittaranjan Locomotive Works, BEML and Jessops started manufacturing equipment in India. At present a mix of local manufacture and imports are used to meet the insane size of the demand for new equipment. Foreign suppliers include Siemens, Alstom and CRRC.
With India (and SE Asia as a whole) growing by most metrics and China, Africa, South America and "the West" experiencing some serious headwinds, I am curious to see how Asia, and the world, develop over the next 20 years.

"Slowly, then all of a sudden."?
 
Last edited:
With India growing by most metrics and China, Africa, South America and "the West" experiencing some serious headwinds, I am curious to see how Asia, and the world, develop over the next 20 years.

"Slowly, then all of a sudden."?
Assuming we are talking about railway electrification, since the broader subject area would be way out of scope in this thread....

Generally, except for North America, so far everyone else is steadily electrifying their railroad or at least are talking about it in the context of meeting Net Zero goals. I suspect it is Canada that will take the lead in electrification, but US will get additional electrification at least in the garb of HSR.
 
Assuming we are talking about railway electrification, since the broader subject area would be way out of scope in this thread....

Generally, except for North America, so far everyone else is steadily electrifying their railroad or at least are talking about it in the context of meeting Net Zero goals. I suspect it is Canada that will take the lead in electrification, but US will get additional electrification at least in the garb of HSR.
Mea culpa.
Sorry.
 
Route 12, the insiders guide to electrified railways? Maybe.
One other thing I did not realize was that after the Milwaukee Road closed their substation, the rural electric decided they needed one there so they built a new, more modern one. Not surprisingly, I noticed the old one, not the newer one.
Hahaha, yes!

That's a great picture actually - probably different electric loads and needs (plus not having owned or wanting to own/maintain a brick building). That equipment doesn't - and won't last long if not used - last forever.
 
I drive along US-12, US-89 and MT-294 from Townsend to Harlowton occasionally but I remember the first time I did and noticed a mid-sized electrical power station (abandoned) in the middle of nowhere. It is just west of the Martinsdale Colony if memory serves which is just 40 to 50 miles from Harlow but since there were only a few ranches in the area the Milwaukee Road had to build a power station in the middle of the empty quarter. That could not have been cheap.
I do not think it was a power generation plant, though I could very easily be wrong. I think it was a sub-station but it still probably cost a lot to build and maintain. I could not find the sub-station I drive by but it was similar to this photo of the Primrose sub-station.

https://historicmt.org/items/show/2216

On edit:
Ah, here it is. The Loweth Substation. 47 miles west of Harlow. The land there has a stark beauty to it. Reminds me of some of Ivan Doig's books.
The 3kv DC system required closer substation spacing than modern AC systems. That's one reason that the Soviet railways were able to rely on the DC system -- substation labor was readily available.
 
The 3kv DC system required closer substation spacing than modern AC systems. That's one reason that the Soviet railways were able to rely on the DC system -- substation labor was readily available.
On a system as vast as the (former) Soviet one, you have a very large scale of standardisation. The Soviets didn't do custom solutions or one offs. They went in with their cookie cutter designs and just repeated them over and over. Just take a look at their urban architecture for example. This extreme standardisation of designs and parts in substations must have enabled low costs of maintenance and repairs.

Furthermore, 3kV requires more feeder and sub-stations than higher-voltage AC. On the other hand the designs and requirements are simpler, so there may be some trade-offs here and the overall penalty not as huge as it may seem.

One place where 3kV is more questionable than elsewhere is on very long lines across essentially empty territory such as the Trans-Siberian, the BAM etc. Higher voltage AC with fewer feed-in points would have been more economic here. But then I guess the principle of a single system and economies of scale outweighed the possible temptation to do something different.
 
On a system as vast as the (former) Soviet one, you have a very large scale of standardisation. The Soviets didn't do custom solutions or one offs. They went in with their cookie cutter designs and just repeated them over and over. Just take a look at their urban architecture for example. This extreme standardisation of designs and parts in substations must have enabled low costs of maintenance and repairs.

Furthermore, 3kV requires more feeder and sub-stations than higher-voltage AC. On the other hand the designs and requirements are simpler, so there may be some trade-offs here and the overall penalty not as huge as it may seem.

One place where 3kV is more questionable than elsewhere is on very long lines across essentially empty territory such as the Trans-Siberian, the BAM etc. Higher voltage AC with fewer feed-in points would have been more economic here. But then I guess the principle of a single system and economies of scale outweighed the possible temptation to do something different.
The Transsiberian is mostly 25 KV 50 Hz now. Looks like most of the 3 KV is in pockets around Moscow and other cities such as Ekaterinberg and Novosibirsk.
 
The Transsiberian is mostly 25 KV 50 Hz now. Looks like most of the 3 KV is in pockets around Moscow and other cities such as Ekaterinberg and Novosibirsk.
The Russians did not convert all legacy suburban DC systems to standard 25kV AC like the Indians did, probably because they simply chose not to invest in such, and they had more of them than the Indians did..

In India Electrification design is as standard as is track designs and layouts at wayside stations. Stations are categorized into about half a dozen categories. The huge stations are of course one of a kind, but the wayside enroute little stations are very very standardized with standardized track layout, interlocking, staffing and such. And electrification is almost cookie cutter or as close to it as you can get.
 
On a system as vast as the (former) Soviet one, you have a very large scale of standardisation. The Soviets didn't do custom solutions or one offs. They went in with their cookie cutter designs and just repeated them over and over. Just take a look at their urban architecture for example. This extreme standardisation of designs and parts in substations must have enabled low costs of maintenance and repairs.

Furthermore, 3kV requires more feeder and sub-stations than higher-voltage AC. On the other hand the designs and requirements are simpler, so there may be some trade-offs here and the overall penalty not as huge as it may seem.

One place where 3kV is more questionable than elsewhere is on very long lines across essentially empty territory such as the Trans-Siberian, the BAM etc. Higher voltage AC with fewer feed-in points would have been more economic here. But then I guess the principle of a single system and economies of scale outweighed the possible temptation to do something different.
There were times on my 2010 trip into Siberia that I thought a lot about the Milwaukee Road and its influence on what the Russians in the late 1920's called "Amerikanomania."

There was at least one non-standard transport feature in the Soviet Union -- bus shelters. It turns out that they were used as design student projects.
 
Last edited:
About conversion to 60 hZ. We keep forgetting that the 25 hZ NEC system is going to become more and more isolated from new US 60 hZ systems every year some are entering service. It is also about last mile service. No matter how slowly HSR becomes being built it will need the last miles into WASH, BAL, PHL, & NYP. Why would hundreds to thousands of HSR units want 25 / 60 transformers? Could not get Siemens to provide a value of weight of 25 / 60 and just 60 Hz transformers for ACS-64s. But dragging that extra weight around 95% of time that is not needed is energy wasteful.

As well extra weight on the rail decreases the useful life of rail and roadbed. A speculation -- if a 2 transformer HSR unit weights 6 tons extra and 10 trains a day on a track section that means 60 tons a day extra or 21,900 tons a year extra on every inch of that rail. This does not even count for the freight RRs who would never equip with 25 / 60 hZ transformers. Even CSX and NS would not want to be saddled with what they call unnecessary weight, capital, and operating costs for a captive fleet.
 
About conversion to 60 hZ. We keep forgetting that the 25 hZ NEC system is going to become more and more isolated from new US 60 hZ systems every year some are entering service. It is also about last mile service. No matter how slowly HSR becomes being built it will need the last miles into WASH, BAL, PHL, & NYP. Why would hundreds to thousands of HSR units want 25 / 60 transformers? Could not get Siemens to provide a value of weight of 25 / 60 and just 60 Hz transformers for ACS-64s. But dragging that extra weight around 95% of time that is not needed is energy wasteful.

As well extra weight on the rail decreases the useful life of rail and roadbed. A speculation -- if a 2 transformer HSR unit weights 6 tons extra and 10 trains a day on a track section that means 60 tons a day extra or 21,900 tons a year extra on every inch of that rail. This does not even count for the freight RRs who would never equip with 25 / 60 hZ transformers. Even CSX and NS would not want to be saddled with what they call unnecessary weight, capital, and operating costs for a captive fleet.
I think you are beating up a Red Herring into pulp. Considering the fact that a significant portion of the European HSR fleet is either 16.67 Hz or 16.67Hz and 50Hz dual system, the issue of train weight due to lower frequency is clearly not any hindrance to HSR operation. CSX and NS being saddled with the slightly higher weight of a transformer is completely laughable considering the axle loads that they run in their freight trains.

Why are we even talking of two transformer etc. The same transformer works just fine in both 16.67/25Hz and 50/60Hz. It just has a slightly heavier core. Do you really believe that the ACS64s and ALP46s carry two separate transformers?

As for dragging around extra weight, it appears that dragging around an entire diesel prime mover over the electrified network is not a problem, but that is exactly what the entire Regional fleet of Airos is going to be doing, and that weight difference is way greater than the difference in weight of the two transformer cores.
 
I think you are beating up a Red Herring into pulp. Considering the fact that a significant portion of the European HSR fleet is either 16.67 Hz or 16.67Hz and 50Hz dual system, the issue of train weight due to lower frequency is clearly not any hindrance to HSR operation. CSX and NS being saddled with the slightly higher weight of a transformer is completely laughable considering the axle loads that they run in their freight trains.
Furthermore the transition from locomotives or power heads to EMUs means that large monolithic transformers are being replaced by smaller distributed units, which lessens the weight per axle issue. Germany is transitioning to an all EMU HS fleet, France is still sticking with power heads but showing latent interest in EMUs. Spain is doing a bit of both. The Japanese and Chinese have gone with EMUs from day one.
 
JIS: -I was speaking of 2 dual frequency transformers. One at each end such as Acelas.

Why has the infrastructure costs for not be considered? What did the new solid-state converter in North NJ cost. and what is the max power output? What about the additional real estate for the converters and transmission lines to the tracks? How many more converters will it take for the higher power draws coming with the increase of number and speed ups of regionals and more Acelas-2s?

Then additional costs for the power transformers for converting to 12 kV 25 hZ vs normal 60 hZ transformers. Suspect adding a new stepdown transformer at Bryn Mar is much more costly for 25 hZ one. More static inverters are going to be needed as the rotary converters die.
 
JIS: -I was speaking of 2 dual frequency transformers. One at each end such as Acelas.
Transformers are limited as to the lowest frequency they can carry. To push this limit down lower you need to add more iron to the core.

There is nothing in physics that prevents you using a transformer at a higher frequency than the one for which it was designed.

There is thus not actually such a thing as a specifically designed dual frequency transformer, because every transformer can accept any frequency down to its lower limit.

Possibly when it comes to protection and monitoring you may need extra equipment. But that's a different story.
 
JIS: -I was speaking of 2 dual frequency transformers. One at each end such as Acelas.

Why has the infrastructure costs for not be considered? What did the new solid-state converter in North NJ cost. and what is the max power output? What about the additional real estate for the converters and transmission lines to the tracks? How many more converters will it take for the higher power draws coming with the increase of number and speed ups of regionals and more Acelas-2s?

Then additional costs for the power transformers for converting to 12 kV 25 hZ vs normal 60 hZ transformers. Suspect adding a new stepdown transformer at Bryn Mar is much more costly for 25 hZ one. More static inverters are going to be needed as the rotary converters die.
You have to balance the incremental costs of adding more converters vs. the huge cost of converting the system to 60 HZ. You could no longer feed it from the 25 HZ generators at the Susquehanna River dams, so you have to find new power sources from grids that are already at capacity. This is complicated by the fact that the railroad is a single phase load and the power feeds are three phase so the load has to be balanced out. By the way, I believe all of the rotary converters have been replaced now, if any still exist they are strictly there as standby units.
 
JIS: -I was speaking of 2 dual frequency transformers. One at each end such as Acelas.

Why has the infrastructure costs for not be considered? What did the new solid-state converter in North NJ cost. and what is the max power output? What about the additional real estate for the converters and transmission lines to the tracks? How many more converters will it take for the higher power draws coming with the increase of number and speed ups of regionals and more Acelas-2s?

Then additional costs for the power transformers for converting to 12 kV 25 hZ vs normal 60 hZ transformers. Suspect adding a new stepdown transformer at Bryn Mar is much more costly for 25 hZ one. More static inverters are going to be needed as the rotary converters die.
25Hz transformers are basically off the shelf equipment from Europe which produces them for their extensive 16.67Hz rail electrification network. So the cost differential may not be as huge as you imagine.

I would recommend that anyone interested in the NEC South electrification, its history, current state and future plans pay attention to the Wikipedia article on Amtrak's 25 Hz traction power system, I would draw your attention to the section titled "Zoo to Paoli transmission line" which is the specific segment that started this discussion over in the Keystone thread in relation to the new Bryn Mawr Substation.

This article also answers many of the questions being thrown around here.

BTW, the article is a bit out of date as it still lists the NJHSR projects as yet to be completed. Those have been substantially completed except for about 20 miles of constant tension catenary as envisaged in the original plan.

There are two ways of converting the entire 25Hz network to 60Hz:

1. Replace all the 60Hz to 25Hz converters to 60Hz 3Phase to 60Hz single phase converter and continue using the single phase transmission network as is. This is the lesser expensive of the two choices and is also pretty pointless make work project IMHO.

2. Rearchitect the entire transmission network decommissioning most of it and feed all the Substations directly from the commercial network. This is an enormous project involving major network architecture work, replacement of all SCADA and much of the transmission network with an alternate one. Extremely non-trivial. But at the end of it one gets a reliable essentially completely new electrification system. This is equivalent to what MNRR did on the New Haven Line. While it would be a nice to do thing, it is really not necessary to continue providing reliable service on the NEC. There are much less expensive ways of doing this using the 25Hz network.

At present there is absolutely no plan to convert the 25Hz network to 60Hz at least in the next decade or two. My guess is that the funding situation will only keep getting worse as time passes for reasons that are beyond the scope of this thread or even AU as a whole, so I don't foresee any conversion taking place for a long long time.
 
Back
Top