Corridor Electric Power Generation and Distribution

[David Benton]

I presume the Navy follow the color coding of the rest of the world , bar america . black is neutral , not ground . ground would be green , green / yellow , or bare conductor . Neutral may be tied to earth , but the key difference is the neutral is current carrying , the earth only carries current under fault conditions .
the 50 kv / 25 kv system auto transformer system does carry current at 50 kv , and does have less losses consenquently . you posted a link to an article stating the savings at 75 % , yet you say you still dont believe it works ??? . i have explained it to you at least 3 times .
the key disadvantage of the auto transformer is the secondary circuit is not isolated from the primary circuit , under fault conditions (or even open circuit ??? ) all parts of the cirucit can reach the voltage of the highest (primary ) side . you asked why not have a 75kv / 25 kv sysytem , the answer is your potential fault voltage is 100kv in all parts of the cirucit . i dont think many railroad personnel would be keen to have 100 kv floating around their locomotives .
i think the second reason against a bigger voltage differential is that all parts of the circuit carry the same current , so you lose efficency bove a 50 :50 ratio. but im not too sure on that one .
Autotransformers are usually used where a small step up or down in voltage is required . i.e you may have one that is 100:110 volts , to compensate for voltage drop in a domestic circuit .

[farecard]

I presume the Navy follow the color coding of the rest of the world , bar america . black is neutral , not ground . ground would be green , green / yellow , or bare conductor . Neutral may be tied to earth , but the key difference is the neutral is current carrying , the earth only carries current under fault conditions .

I don't recall the whole USN color code, but it was 180 out from NEC in many aspects. A friend bought a house wired by a retired CPO, and found out the hard way. White was hot in every outlet... It took him weeks to fix it.

Historic USN wire gauge was also backwards vs NEC. There, large gauges had larger numbers. (I can't readily find an old Navy wire gauge chart on line; I assume it went away sometime after WWII, but can't be sure.)

(BTW, green is not always ground even in the US. The local utility PEPCO uses it as a phase color in some of their 3-phase cable. "Green is ground the world around.... except at PEPCO" the lineman told me when I asked him about it...I was freaked...)

the 50 kv / 25 kv system auto transformer system does carry current at 50 kv , and does have less losses consenquently . you posted a link to an article stating the savings at 75 % , yet you say you still dont believe it works ??? . i have explained it to you at least 3 times .

What I have questioned and continue to is not that autoformer scheme has less drop, but that it has 75% less drop than a simple parallel feeder scheme of the same gauge.

the key disadvantage of the auto transformer is the secondary circuit is not isolated from the primary circuit , under fault conditions (or even open circuit ??? ) all parts of the cirucit can reach the voltage of the highest (primary ) side . you asked why not have a 75kv / 25 kv sysytem , the answer is your potential fault voltage is 100kv in all parts of the cirucit . i dont think many railroad personnel would be keen to have 100 kv floating around their locomotives .

That is true; you'd need to have some system to limit voltage excursions in case of isolated segments, and that would drive the costs up. But with an ever-increasing concern for the price of power, it may be worth that.

i think the second reason against a bigger voltage differential is that all parts of the circuit carry the same current , so you lose efficency bove a 50 :50 ratio. but im not too sure on that one .

You do need windings suitable for the needed load, which can be split. I've seen transformers where the gauge changed at the tap.

[farecard]

Along with the 25Hz, there's 100Hz signal power. What ran and didn't run on that?

Did it include grade crossings [I assume at some point in the past the NEC/PRR had same..], switch power, and other trackside gear?

[farecard]

New questions on an old thread...

The 91.667 Hz "signal" power that I believe is carried at 7KV+ along with the 25Hz.
Where is it generated now? How?

Second question:
How many MW are the 55-odd periodic 25Hz subs rated for?

[Jersey_Mike]

New questions on an old thread...

The 91.667 Hz "signal" power that I believe is carried at 7KV+ along with the 25Hz.
Where is it generated now? How?

Second question:
How many MW are the 55-odd periodic 25Hz subs rated for?

Signal power is generated at each Substation via a motor-generator set or a solid state inverter (ha!). Any grounds affecting the signal power feed can be isolated between substations without causing wider scale disruptions.

Substation ratings are based on the number of installed transformers with most containing two and others in high density areas four. For capacity the number 5MVA rings a bell, but I am not sure if that is per transformer or for a two transformer substation. The original Reading electrification transformers are rated for 1.9MVA and I can see the PRR units being double that for its freight and locomotive hauled passenger services. Almost all NEC substations have pads and infrastructure available for 4 transformers, even if only two are present.

[farecard]

Signal power is generated at each Substation via a motor-generator set or a solid state inverter (ha!). Any grounds affecting the signal power feed can be isolated between substations without causing wider scale disruptions.

Substation ratings are based on the number of installed transformers with most containing two and others in high density areas four. For capacity the number 5MA rings a bell, but I am not sure if that is per transformer or for a two transformer substation. Almost all NEC substations have pads and infrastructure available for 4 transformers, even if only two are present.

OK, so the mentioned 7KV/91 hz seen overhead is local to each sub, no further?
And it's powered by the 25Hz? Or locally acquired 60Hz?
Still curious what such does and does not power - switches, block signals, etc. {Does the 25 directly power anything such as switch motors?}

I'd SWAG that's 5MW/xfmr, even 10MW strikes me as low. Doesn't an Acela traction system alone draw 5MW/car? And then there's the HEP/hotel load. Granted the oil-filled transformers can be overloaded far beyond that rating for low duty cycle use, but.....

[David Benton]

I would think its Mkva . as in a million thousand watts . although va and watts have slighlty different meanings theyre round about the same .

[David Benton]

on review , 5mva appears correct .that would be 5000 Kva . 12 000 hp of an Aclea would be around 20 kva draw , assuming 50 % efficency .

[farecard]

on review , 5mva appears correct .that would be 5000 Kva . 12 000 hp of an Aclea would be around 20 kva draw , assuming 50 % efficency .

Per transformer??

[David Benton]

sorry ,my head is all over the place today .
1 h.p is .750 k.w . 50% efficency would mean 1.5 k.w power use . 1.5 k.w* 12000hp = 18000 k.w per acela set . which would be 18Mva.
i wont even attempt the transformer side at the moment . remember there are 3 phases in , 1 phase out , but i would think the rating is for all phases anyway .

[PRRTechFan]

Concerning signal power, the 91.667Hz frequency was specifically chosen so that the frequency and its' harmonics ( 2x, 3x, 4x, etc; of the original frequency) would be as far removed from the 25Hz traction power and 60Hz commercial power and those harmonics. The signal department track relays that were vital to detecting track occupancy were manufactured to be extremely frequency selective, so picking the correct frequency for track signalling purposes was important for them to be most sensitive to signalling frequency power while ignoring stray currents from 25Hz traction and 60Hz commercial power.

The signal power is transmitted at 6600 volts and the circuit is essentially continuous up and down the right of way. The original frequency converters for signal power were motor-generator sets. Some M-G sets had 25Hz motors and were powered from 25Hz traction power. Other M-G sets had 60Hz motors and were powered from local commercial power. But all of the 91.667Hz generators ran in parallel on the 6600 volt line. So signal power was really sourced from two different sources of power, giving some additional redundancy.

I said the signal power circuit was essentially continuous. There were isolating switches that would allow a damaged line segment to be disconnected. Under these circumstances, the circuit could remain in operation as separate systems fed by converters above and below the break.

I believe the signal power was generated at 480 volts and was stepped up to 6600 volts for distribution. The 6600 volt distribution power was stepped down to 120, 240 or 480 volts as required at each signal equipment installation. I know the track relays operated at the 91.667Hz signal frequency. I believe that the 91.667Hz power was used for all other signal relays and signal lamps as well. Some of that may actually have been DC (direct current), but it was the 91.667Hz signal power that was rectified to DC. The DC circuits could also have included batteries to ride through power outages. Railroad crossing equipment was DC and had backup batteries.

As for switch machines, a lot of the switch machines on the PRR were originally pneumatically operated via compressed air. I believe I remember seeing a schematic somewhere that showed air compressors running off of signal power; or transferring to signal power in the event of a commercial power failure.

As for substation power transformers, I have never been able to find a definitive answer as to their exact MVA ratings. I somehow recall hearing something about them being 4MVA each. This is right in the ballpark of the 5MVA figure mentioned earlier.

The Acela was referred to as having a maximum draw of 18MVA, which would seem to be in conflict with having only a 5MVA source of power. The answer to this lies in the fact that transformers for traction power have much greater "short-time" ratings and they are never continuously loaded (...we hope!) at their "nameplate" or long-time rating.

Transformer life depends primarily on how hot the internal insulation gets and for how long. Oil filled transformers use the oil both for electrical insulating capabilities and for thermal cooling. The oil represents a large thermal "sink" for the heat given off when the transformer is operated at it's much higher short term rating for a limited period of time.

An accelerating Acela may only draw maximum power from that one transformer for a very short period of time... maybe only a minute or so; then the power consumed will drop off. But even if this reduced power is greater than the transformer rating, it will only continue for several minutes as the train accelerates away from one substation and closer to the transformer at the next substation. The result is that the transformer is extremely loaded for a rather short time. This is how a transformer with a lower "long time" rating can successfully supply significantly more power for short periods of time.

[farecard]

Thanks for the signal power data. I knew about the 25Hz vs 91.667; the harmonics from 25Hz must be quite rich....
So do you concur there were rotary converters at every substation, not just at the power plants? Any idea

4MVA sounds like a reasonable transformer size; given there are 2 or more in each substation. Oil-filled power transformers do have max outputs many times their continuous rating; here WMATA's traction power stations transformers creep up into the red zone during rush hour, I'm told. Some utility substations use external oil circulation pumps to extend that time constant.

[farecard]

Separate question:

Traction systems use low pass filters called Wee-Z bonds to pass traction voltage return from the rails back to the power station. This so the rails can also be used for block signaling.

How does the NEC block signaling system work? (The track occupancy detection.) Where can I read up on it?