AC vs DC?

[Tenspeed]

What are the merits of AC traction vs DC? Why do some railroads order both? Why do some railroads prefer DC locomotives over AC?

I just received my copy of "Locomotive" and have read the articles about GE's Evolution Series and EMD's new 70-series. There is a chart comparing the GE locomotives and they are both the same size, weight and horsepower, but the AC units have substantially higher tractive effort and slightly higher top speed. Wouldn't any railroad prefer the AC units?

[chad thomas]

With AC units you don't have a minimum continuous speed rateing. In other words DC traction motors burn up at low speeds. 12mph is the typical minimum speed for a freight unit. AC traction motors can give full power down to a crawl (or even stopped). That makes AC units ideal for low speed lugging bulk comodities. At higher speeds there is no real advantage over DC. It boils down to do you want to pay the extra 1/2 million cost for AC on new units. Some railroads don't see the need.

[tree68]

Believe it or not, there is actually a comparison here that can be made to model railroading - pulse throttles vs straight DC.  Obviously loading is not a problem - we don't often stall (or even slip) our model locomotives, but control is much finer, and slower speeds are possible  with pulse throttles than with DC.

As Chad said, minimum speed is essentially zero for an AC locomotive.  I saw one account where a train was running at about 1/2 MPH - in notch 8. 

It's all about return on investment.

[Tenspeed]

With AC units you don't have a minimum continuous speed rateing. In other words DC traction motors burn up at low speeds. 12mph is the typical minimum speed for a freight unit. AC traction motors can give full power down to a crawl (or even stopped). That makes AC units ideal for low speed lugging bulk comodities. At higher speeds there is no real advantage over DC. It boils down to do you want to pay the extra 1/2 million cost for AC on new units. Some railroads don't see the need.

They don't give costs in the table so I did not realize that AC costs more.

[chad thomas]

You can figure a DC unit cost roughly 1.3 million and a AC unit costs roughly 1.7 million (standard freight locos). More capitol costs but in the long run AC maintainance may be significantly cheaper as there are no comutators (brushes) in an AC traction motors. Comutator maintainance is a factor in DC TMs. AC units require very expensive simiconductors and other components initially but require little maintainance on an on going basis.

[Anonymous]

Is that because the field has the windings, and not the armature?

[oltmannd]

The other guys have it right, but I'll toss one more word into the mix:  adhesion

Besides being thermally limited, a modern DC locomotive's wheel slip/creep system is only capable of 28% or so on a day in day out, all weather basis.  An AC locomotive can do 35%.

All that wonderful power handling ability of those AC motors would mean nothing if you couldn't depend on getting it down to the rail reliably.

For a DC locomotive, you want the reliable adhesion limit to occur at the roughly the thermal limit for the traction motors.  Too little weight and you'll have a "slippery" locomotive.  Too much balast and your just adding dead weight that'll cost you fuel and wear and tear that only gives you a little more TE for very short periods of time.

Part of the reason that AC units have more capable wheel slip/creep systems is the basic physics of how the motor speed is controlled.  On a DC unit,  motor speed varies with voltage.  To vary the voltage, you vary the excitation on the main generator.  The main generator field is proportional to the current flowing through it.  The main generator windings are inductive and, therefore, they resist having the current flow changed.  So there is a limit to how fast you can react to detected slip/creep. 

On AC units, the motor speed is controlled by frequency.  You can change the output frequency of the inverters almost instantaneously, so can react faster to slip/creep and gain much finer control over power delivery to the motors.

 

[oltmannd]

Is that because the field has the windings, and not the armature?

The AC motors are squirrel cage motors.  You are correct that the windings are in the stator.  The rotor is just a bunch of copper bars welded together - very rugged!

[PBenham]

Is that because the field has the windings, and not the armature?

The AC motors are squirrel cage motors.  You are correct that the windings are in the stator.  The rotor is just a bunch of copper bars welded together - very rugged!

 AC motors are expensive to buy, and repair. The rationale there is that the AC motors will last longer, and absorb more abuse than a like rated DC motor.

[Tenspeed]

Part of my question has been answered, but why do some railroads order both AC and DC units? In another  "Locomotive" chart, for example, BNSF has ordered 415 AC units and 195 DC units from GE. I presume it depends on the use they will be assigned to.

[PBenham]

Careful, tenspeed you might be giving BNSF too much credit, they don't always put the right unit on the right job! CSX, for example, will always assign their units on the FIFO principle, regardless of the type of train the next train out of a terminal will be! Hence, a 40+ year old SD40 from a lease fleet,an equally old rebuilt SD40 of their own, another leaser, and an SD50 can go out on a hot freight, and then they'll put a pair of AC4400s on a string of garbage gons! BNSF is a bit smarter than that...

[zapp]

Part of my question has been answered, but why do some railroads order both AC and DC units? In another  "Locomotive" chart, for example, BNSF has ordered 415 AC units and 195 DC units from GE. I presume it depends on the use they will be assigned to.

It really depends on the type of train your trying to run. If you'll notice alot of unit trains (ie, coal, grain, potash, etc...) will have AC units. That's due to the weight of those types of trains. I routinely run 133 car coal trains that weight 19,000 tons (summer weight). Where as a mixed freight or "Auto Rack" will only weigh about 2500-9000 tons.

Sitting in the engineer's seat you can really feel that horsepower being applied to the rails in AC. Those things will vibrate your brains out!!!!

[tree68]

Part of my question has been answered, but why do some railroads order both AC and DC units? In another  "Locomotive" chart, for example, BNSF has ordered 415 AC units and 195 DC units from GE. I presume it depends on the use they will be assigned to.

As has been alluded - different locomotives for different jobs.  Back in the day when there were a plethora of locomotives available in the catalogs, the railroads bought the locomotive best suited for the job at hand.  GP40's for fast freight, SD's for drag work, etc.  Today, there are basically what, two D/E locomotives available?  EMD and GE.  4400HP, more or less, take it or leave it.  And some special purpose stuff, as in switchers.

 

[GP40-2]

You can figure a DC unit cost roughly 1.3 million and a AC unit costs roughly 1.7 million (standard freight locos). More capitol costs but in the long run AC maintainance may be significantly cheaper as there are no comutators (brushes) in an AC traction motors. Comutator maintainance is a factor in DC TMs. AC units require very expensive simiconductors and other components initially but require little maintainance on an on going basis.

A lot of the comments on this thread and another thread on CSX's use of AC units on all types of trains and "not getting a return on the capital expense" is based on many of the posters ASSUMING we actually paid that much more for the CW44ACs.

If you have the ability to negotiate the costs of the AC units close to what the builders are willing to let a DC unit go out the door for, does it really matter what type of train you assign it to?

The notion that AC units are only for "drag freight" is a misconception amoung railfans. AC units have better performance characteristics across the enitre speed range.

[beaulieu]

<snipped>

A lot of the comments on this thread and another thread on CSX's use of AC units on all types of trains and "not getting a return on the capital expense" is based on many of the posters ASSUMING we actually paid that much more for the CW44ACs.

If you have the ability to negotiate the costs of the AC units close to what the builders are willing to let a DC unit go out the door for, does it really matter what type of train you assign it to?

The notion that AC units are only for "drag freight" is a misconception amoung railfans. AC units have better performance characteristics across the enitre speed range.

Which suggests that the builders are selling the DC locomotives with less markup, on total dollar basis that might be true. CSX's bargaining power should be just as good on DC locomotives. The question on return depends on utilization, if having more all AC units improves fleet tm/hph ratio enough to offset the higher intial cost then you are correct. How would you explain the recent large CSX order for ES44DC locomotives. The vvvf Invertors are an additional piece of equipment not needed on a DC locomotive, the $400,000 price difference seems about right for them.

[MichaelSol]

<snipped>

A lot of the comments on this thread and another thread on CSX's use of AC units on all types of trains and "not getting a return on the capital expense" is based on many of the posters ASSUMING we actually paid that much more for the CW44ACs.

If you have the ability to negotiate the costs of the AC units close to what the builders are willing to let a DC unit go out the door for, does it really matter what type of train you assign it to?

The notion that AC units are only for "drag freight" is a misconception amoung railfans. AC units have better performance characteristics across the enitre speed range.

Which suggests that the builders are selling the DC locomotives with less markup, on total dollar basis that might be true. CSX's bargaining power should be just as good on DC locomotives. The question on return depends on utilization, if having more all AC units improves fleet tm/hph ratio enough to offset the higher intial cost then you are correct. How would you explain the recent large CSX order for ES44DC locomotives. The vvvf Invertors are an additional piece of equipment not needed on a DC locomotive, the $400,000 price difference seems about right for them.

If financed over 10 years, that extra cost with interest totals $720,000 at 8%. A company could buy three DC hp for every two AC hp. In justifying that huge extra cost, how much do commutator brushes actually cost, and how much attention do DC motors really require?

[rrandb]

The cost is probably much greater than the replacement parts alone. Besides the labor required to replace said parts the more expensive cost may be the lack of availabilty while the units are off line for both repairs and inspections. The actual parts will be the least of your expenses.
 

[eolafan]

Anybody who wants a really good explanation of the benefits of AC versus DC units should read the article in TRAINS magazine this month about the issues CSX has historically faced on Cranberry Grade and how their AC units with significant added weight solved those problems.  Enjoy the article, I did.

[GP40-2]

... How would you explain the recent large CSX order for ES44DC locomotives...

Two reasons: (1) GE has made a number of  improvements to the DC version of the GEVO over the DC Dash 9 that were very attractive performance wise to us, and more importantly (2) The GEVO is a HOT seller right now. GE has more orders than they know what to do with, and the were not willing at this time to give us the same discounts they gave on the CW44ACs.

[Tenspeed]

Anybody who wants a really good explanation of the benefits of AC versus DC units should read the article in TRAINS magazine this month about the issues CSX has historically faced on Cranberry Grade and how their AC units with significant added weight solved those problems.  Enjoy the article, I did.

I just flipped through both the September and October editions and did not see the article. Do you already have the November edition?

[eolafan]

Anybody who wants a really good explanation of the benefits of AC versus DC units should read the article in TRAINS magazine this month about the issues CSX has historically faced on Cranberry Grade and how their AC units with significant added weight solved those problems.  Enjoy the article, I did.

I just flipped through both the September and October editions and did not see the article. Do you already have the November edition?

YES, my November TRAINS came in Friday's mail and I read the subject article while trackside yesterday at my beloved Eola Yard in the afternoon.

[MichaelSol]

The cost is probably much greater than the replacement parts alone. Besides the labor required to replace said parts the more expensive cost may be the lack of availabilty while the units are off line for both repairs and inspections. The actual parts will be the least of your expenses.
 

For the most recent available year BN has 5,790 locoomotives, and the annual locomotive repair and maintenance charges are $538,253,000. [2004 ANNUAL REPORT FORM R-1, SCHEDULE 210]

Say, for the purposes of discussion, traction motor maintenance and repair is 10% of that number, $53,000,000. That strikes me as high, nearly $93,000 per unit, but this is a hypothetical to determine ballpark feasibility.

If the fleet was all AC, the extra costs of electrical equipment and carrying charges at 8% over 8 years adds $498,000,000 per year to the "cost" of the fleet over DC, in order to save up to $53,000,000 per year in repair costs. The extra cost of the AC locomotives is approximately ten times the possible savings on traction motor maintenance. You can adjust this accordingly if you think the traction motor maintenance cost is higher or lower than 10% of overall maintenance and repair costs.

Over the 8 years, an all-AC fleet would cost BN nearly $4 billion more in extra capital requirements over an all-DC fleet. The opportunity cost of the AC investment is $2.56 billion.

Insofar as the partly hypothetical parameters here are concerned, traction motor maintenance savings are not a justification for the expenditure. A performance metric that reads to the effect of "goes real well at slow speeds," is difficult to quantify and justify in a network where slower speed reduces capacity and increases other operating costs and capital needs.

This change alone would drop the Operating Ratio from 86% to nearly 91%.

Clearly, any use of an AC locomotive would necessarily be limited by economic feasibility, rather than considerations of technical superiority unless you could specifically quantify that technical superiority in some fashion to justify the cost.

 

[JSGreen]

Well, since by your own admission, "but this is a hypothetical...", everything else is just a boilerplate.  Find the real numbers, plug them in, and then the diccussion might be meaningful.  Until then, its a great big game of What IF.  

Here's an Idea....What If we talk in general terms until we actually find some reasonable numbers?

Just my

[MichaelSol]

Well, since by your own admission, "but this is a hypothetical...", everything else is just a boilerplate.  Find the real numbers, plug them in, and then the diccussion might be meaningful.  Until then, its a great big game of What IF.  

Here's an Idea....What If we talk in general terms until we actually find some reasonable numbers?

Just my

Well, I just love "general terms" because anybody can say just about anything and it is meaningless. The hypothetical is based on something slightly better than just "thin air."

The last motive power study that involved maintenance specifics that I participated in was 32 years ago, but my recollection is that the diesel engine constituted approximately 70% of the maintenance and repair requirements of an SD-40-2 locomotive, which is what we were looking at. Brakes and rigging were about 10% and "other electrical" about 10%. I don't specifically recall that the DC traction motors were a big item at all, but what's left is around 10%.

If the economic advantage is solely in traction motor repair costs, the "break-even" on AC would require that the DC traction motors cost approximately 93% of total locomotive maintenance and repair costs.

That's not reasonable.

[GP40-2]

Well, since by your own admission, "but this is a hypothetical...", everything else is just a boilerplate.  Find the real numbers, plug them in, and then the diccussion might be meaningful.  Until then, its a great big game of What IF.  

Here's an Idea....What If we talk in general terms until we actually find some reasonable numbers?

Just my

That's something you will never find. Placing a locomotive order worth hundreds of million dollars is a complicated drawn out negotiation. The only people who know what it really costs GE to produce a DC vs. AC locomotive is GE, and they aren't going to release their internal accounting to the public. CSX isn't going to release internal costs either, so you guys have no choice but to play "What if" as to why we assign our locomotive fleet the way we do.

[beaulieu]

I think that there are other factors involved, lost time while tractions motors are exchanged, extra inventory of spare traction motors, the fact that either the train must be lightened or extra locomotives must be assigned. The whole run should be run through a simulator with things like cycle times, locomotive utilization, expected fuel burn, crew costs, etc. taken into account. I know that BN reported that 3 SD70MACs would save on transit time from the PRB to Houston via the Denver route versus a 5 unit set of SD40-2s inspite of having 3000 less horsepower, also the SD40-2s needed a 2-unit helper to get over the Palmer Lake grade. While the SD70MACs would just lug it out. the time adding and removing the helper would almost offset the higher speed.

[nbrodar]

CSX's large investment in DC GEVOs is all about Tony Ingram's view on motive power. 

A.) He intensely dislikes EMD for some reason.

B.) He doesn't believe in AC.

C.) He can't see buying 2 locomotives, when he could buy 3 for the same price.  (When you need to replace 500 locomotives and you have the credit rating of toliet paper, this is a big concern).

Nick

[GP40-2]

CSX's large investment in DC GEVOs is all about Tony Ingram's view on motive power. 

A.) He intensely dislikes EMD for some reason.

That's not true. He's just smart enough to realize GE now builds a better, more profitable to run product.

B.) He doesn't believe in AC.

That's not true. He didn't believe in AC for NS operating conditions. Our conditions are much more severe in places. We will always have AC in our fleet.

C.) He can't see buying 2 locomotives, when he could buy 3 for the same price.  (When you need to replace 500 locomotives and you have the credit rating of toliet paper, this is a big concern).

Nick

That's nothing new. We all feel the same way. Like I said, don't assume we paid the "2 for 3" price for our large fleet of CW44ACs.

 

Why do so many of you on this site pass your opinions off as fact?

[rrandb]

Without some sort of economic benift of the AC traction technoligy the odds of selling the additional expense of the engines to the naturally consevative railroad managment would have been an uphill battle.
 They would have never gotten beyond the demonstrator phase. Railroads are not known for buying new technoligy just because it's trendy.

[MP173]

I really dont know much about locomotives, let alone electrical engineering.  I can only report what I see.

The AC units on the heavy coal trains really dig down deep and pull.  I personally have walked with a 18000 ton coal train at less than 1mph til it reached the summit (or in this case the top of the hill here in Valparaiso).  Incredible pulling power at low speeds.

I also have watched stalled trains get moving on the same hill and watched the locomotives vibrate like crazy when taking off.  So much that I moved back...way back.

ed