Trains.com

AC vs DC-what's the difference?

59015 views
20 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    April 2003
  • 305,205 posts
AC vs DC-what's the difference?
Posted by Anonymous on Sunday, December 16, 2007 5:38 PM
I hope this doesn't sound like too much of a dumb question-with regards to the new EMD'S and GE'S what is the difference between running AC vs DC? I have read articles in Trains about both but I still don't understand.Is there a difference in the way they run and why do some roads prefer one over the other? Also on the designation SD70ACe-what does the small "e" stand for? Thanks for any insight.Bob
  • Member since
    November 2007
  • 2,989 posts
Posted by Railway Man on Sunday, December 16, 2007 5:49 PM

A.C. vs. D.C. refers to the electrical transmission:

D.C.:  The engine drives an A.C. main generator, which generates A.C. current rectified to D.C. by diodes that powers D.C. traction motors.  The advantage is significantly lower initial cost.  Disadvantages are higher maintenance costs for the traction motors, less tractive effort at low speeds (less than ~ 11 mph).

A.C.:  The engine drives the same A.C. main generator, which generates A.C. current rectified to D.C. by diodes.  From that point it differs: the D.C. current is converted to A.C. which powers A.C. traction motors.  The advantages and disadvantages are opposite a D.C. locomotive. (You might be wondering why the A.C. locomotive goes from A.C. to D.C. and back to A.C.; the answer is that the D.C. rectification provides "clean" current that the invertors can use to create the proper frequency needed by the traction motors. The reason the main generator is A.C. and not D.C. is that D.C. generators are too large, too expensive, and too complex once horsepower climbs past 2,000 or so.)

North American railroads such as BNSF, UP, and KCS have determined that A.C. is more cost-effective for heavy-haul trains such as coal, and D.C. more cost-effective for all other trains.  NS has favored D.C. for everything and CSX A.C. for everything.  The decision is based on the physical characteristics of the railroad, the operating plan, and the traffic mix.

The only difference in the way A.C. and D.C. run is that a train powered with A.C. locomotives can have a very low horsepower-per-ton ratio and still move without exceeding the heating limits on the traction motors.  A.C. motors in fact can run to a stall at full current without damaging the motors, whereas D.C. motors will quickly overheat at low speeds and full current and experience fatal damage. 

The "e" in SD70ACe stands for "enhanced'; it is a major redesign of the previous SD70MAC but shares the same basic specifications, horsepower, and prime mover. 

RWM

  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Sunday, December 16, 2007 5:59 PM
Railway Man-Thanks for the reply and explanation. Now I understand it better. Bob
  • Member since
    December 2005
  • From: Cardiff, CA
  • 2,930 posts
Posted by erikem on Monday, December 17, 2007 12:42 AM

To amplify what Railway Man said.

The limiting factor with a DC generator is the commutator, more power requires a larger diameter commutator. A larger diameter commutator requires a larger diameter armature, which then requires a larger diameter stator (frame) and pretty soon the thing is too big to fit in a locomotive. In addition, a larger generator requires a lower speed (to maintain a sane limit on peripheral speed of the commutator and rotor) - optimum power output for a 900 RPM generator is 1200KW (about 1600HP).

Making a large AC generator (actually alternator) is much easier - the slip rings only carry the field excitation current, the power of which is a few percent  of the alternator output. The alternator allows for a higher magnetic flux density which reduces size in comparison to the DC generator. Alternators only became practical for locomotives (and cars) after the development of silicon rectifiers.

It is possible to convert fixed frequency AC to variable frequency AC with a cycloconverter. Inverters, however, are smaller and simpler. The original enabling technology was the GTO thyristor, but the current trend is with IGBT's. I worked on one project using an IGBT H-bridge module that occupied less than 1.5 cu ft and could put out in excess of 100KW - locomotive modules are 'a bit' larger.

AC induction motors have an inherent adhesion advantage over a DC series motor. The AC motor has a steep torque-speed curve when run at a given frequency. A little bit of slip will cause a significant reduction in torque, where the same amount of slip with a DC motor will cause little reduction in torque (and keeps slipping).

FWIW, the first application of 3 phase induction motors for traction on a US railroad was the original Cascade tunnel electrification on the GN. The locomotives were able to maintain a noticavly higher factor of adhesion than an equivalent loco with DC series motors.

  • Member since
    November 2007
  • 2,989 posts
Posted by Railway Man on Monday, December 17, 2007 12:49 AM

That's the clearest explanation of the size/output limitation on a D.C. generator I've ever read.  Same for the clear and concise comparison of the torque/speed characteristics of A.C. and D.C. motors.

RWM 

  • Member since
    October 2006
  • From: Chicago, Ill.
  • 2,843 posts
Posted by al-in-chgo on Monday, December 17, 2007 9:45 PM
 Railway Man wrote:

A.C. vs. D.C. refers to the electrical transmission:

D.C.:  The engine drives an A.C. main generator, which generates A.C. current rectified to D.C. by diodes that powers D.C. traction motors.  The advantage is significantly lower initial cost.  Disadvantages are higher maintenance costs for the traction motors, less tractive effort at low speeds (less than ~ 11 mph).

A.C.:  The engine drives the same A.C. main generator, which generates A.C. current rectified to D.C. by diodes.  From that point it differs: the D.C. current is converted to A.C. which powers A.C. traction motors.  The advantages and disadvantages are opposite a D.C. locomotive. (You might be wondering why the A.C. locomotive goes from A.C. to D.C. and back to A.C.; the answer is that the D.C. rectification provides "clean" current that the invertors can use to create the proper frequency needed by the traction motors. The reason the main generator is A.C. and not D.C. is that D.C. generators are too large, too expensive, and too complex once horsepower climbs past 2,000 or so.)

North American railroads such as BNSF, UP, and KCS have determined that A.C. is more cost-effective for heavy-haul trains such as coal, and D.C. more cost-effective for all other trains.  NS has favored D.C. for everything and CSX A.C. for everything.  The decision is based on the physical characteristics of the railroad, the operating plan, and the traffic mix.

The only difference in the way A.C. and D.C. run is that a train powered with A.C. locomotives can have a very low horsepower-per-ton ratio and still move without exceeding the heating limits on the traction motors.  A.C. motors in fact can run to a stall at full current without damaging the motors, whereas D.C. motors will quickly overheat at low speeds and full current and experience fatal damage. 

The "e" in SD70ACe stands for "enhanced'; it is a major redesign of the previous SD70MAC but shares the same basic specifications, horsepower, and prime mover. 

RWM

  I understand that in D.C., there's always a risk of burning out the motor(s).  But why is that?  Is there something inherent in D.C. that keeps a governor or some sort of energy expenditure low enough to keep burnout happening?

This is a great thread, BTW.  - a. s.

 

 

al-in-chgo
  • Member since
    December 2005
  • From: Cardiff, CA
  • 2,930 posts
Posted by erikem on Monday, December 17, 2007 10:41 PM
 al-in-chgo wrote:

I understand that in D.C., there's always a risk of burning out the motor(s).  But why is that?  Is there something inherent in D.C. that keeps a governor or some sort of energy expenditure low enough to keep burnout happening?

This is a great thread, BTW.  - a. s. 

The main work of a DC motor is done by the windings on the armature (rotor) and that poses a minor problem with cooling. In addition, only a fraction of the windings are active at any given instant - this creates a more severe heating problem (less copper to carry the current) and also imposes a minimum speed requirement (motor needs to turn fast enough that the active winding doesn't overheat). The copper area is also reduced by the 'teeth' on the armature - which strengthens the magetic field induced by the field windings. A final problem is that the electrical insulation also tends to be thermal insulation, interfering with cooling - and it is the electrical insulation that burns out. The commutator is also an issue.

An AC induction motor has a much simpler armature - typically copper bars forming the 'squirrel cage'. The copper cross section can be larger, more effective cooling is possible and there is no need for electrical insulation.  With a variable frequency drive, it is possible to design the induction motor for high efficiency operation (i.e. low slip) - with fixed frequency induction motors there is a severe tradeoff between starting torque (rquiring high slip) and efficiency. An additional benefit from the high efficiency design is that the torque speed curve is steep providing an inherent traction control.

  • Member since
    December 2005
  • From: Cardiff, CA
  • 2,930 posts
Posted by erikem on Tuesday, December 18, 2007 12:47 AM
 Railway Man wrote:

That's the clearest explanation of the size/output limitation on a D.C. generator I've ever read.  Same for the clear and concise comparison of the torque/speed characteristics of A.C. and D.C. motors.

Thanks for the nice compliment.

The information for the DC generators was from the chapter on DC generators in the "Standard Handbook for Electrical Engineers" published every ten years or so by McGraw-Hill. I have several editions of the book and the chapter on DC generators hadn't changed much between the 10th edition published in the 60's and the 13th edition published in the 90's. As the name implies, these books are a nice reference on electrical (as opposed to electronic) engineering.

The info on DC series motors versus AC induction motors was from an electrical machinery course I took in my senior year at UCB - turned out that was one of the most useful courses I took at Cal as part of my job over the last decade and a half has been designing magnets.

  • Member since
    December 2007
  • From: Georgia USA SW of Atlanta
  • 11,919 posts
Posted by blue streak 1 on Tuesday, January 8, 2008 10:59 PM
one question. is the AC output to the traction motors always the same or is it variable in relation to the motor's speed and what is the frequency?
  • Member since
    November 2006
  • From: Southington, CT
  • 1,326 posts
Posted by DMUinCT on Wednesday, January 9, 2008 12:29 PM

 VVVF (Variable Voltage - Variable Frquency) Drive works on the principle that an AC motor (any AC Motor) RPM (Revolutions Per minutes) is a product of the number of Poles (windings) and the Frequency of the applided Voltage.  Example, a 2 pole motor at 60 HTZ turns at 3600 RPM.

  If you wind the motor with more Poles the base RPM is lower (Synchronize Speed).  Apply more Voltage and you get more Torque, not speed.  If the you are to run an AC Motor at a Frequency less than the designed Frequency you WILL develope Heat but it still works, just add Blowers.  If you design a railroad locomotive for 50 mph service, the gearing and motor design RPM should be near that speed.

  This is why AC drive CAN be much better than DC could ever be.   Apply a low Frequency and the motor turn slowly under complete control.  As you Raise the Frequency, the motor accelerates smoothly.   Apply a high Voltage in starting and you have the needed Torque, which can be reduced as the motor takes the load and comes up to speed.  You end up with complete and separate control of Speed and Torque.

  How is it done ?  With a Micro-Processor Computer.  In a locomotive, the diesel turns an AC generator, this AC is Rectified by Diodes to a pure DC power Bus. Integrated Gate Bi-Polar Transistors, under the control of the Micro-Processor, turn the DC on the Power Bus back into AC at the Frequency requested by the Throttle setting.  As the Throttle is advanced, the AC Frequency is raised.  

The Drives (both in an AC or a DC locomotive) are refered to as "Four Quadrant".  Four sets of IGBTs (or SCRs in a DC locomotive) are required.   One set for pulling Forward, another for coasting Forward (Dynamic Breaking) - one set for powering in Reverse and another for coasting in Reverse.  

 

 

Don U. TCA 73-5735

  • Member since
    February 2005
  • 164 posts
Posted by blade on Wednesday, January 9, 2008 1:40 PM
the biggest diiference is cost,the ac traction motors cost less to maintain than do the dc traction motors.
  • Member since
    December 2005
  • From: Cardiff, CA
  • 2,930 posts
Posted by erikem on Wednesday, January 9, 2008 11:47 PM
 DMUinCT wrote:

 Integrated Gate Bi-Polar Transistors, under the control of the Micro-Processor, turn the DC on the Power Bus back into AC at the Frequency requested by the Throttle setting.  As the Throttle is advanced, the AC Frequency is raised.  

The Drives (both in an AC or a DC locomotive) are refered to as "Four Quadrant".  Four sets of IGBTs (or SCRs in a DC locomotive) are required.   One set for pulling Forward, another for coasting Forward (Dynamic Breaking) - one set for powering in Reverse and another for coasting in Reverse.   

A couple of minor corrections:

IGBT stands for Insulated Gate Bipolar Transistor (a more accuate term is conductivity modulated FET as an IGBT is basically a power MOSFET with modified doping profiles).

The four quadrant refers to the voltage and current operating in all four quadrants of output voltage and current - which is an absolute requirement when dealing with a reactive load such as an induction motor.

  • Member since
    November 2006
  • From: Southington, CT
  • 1,326 posts
Posted by DMUinCT on Thursday, January 10, 2008 8:50 AM

 Sorry Erikem, for using the wrong word, Intergrated Gate would be one big short circuit !

 By the way, I'm not into Locomotives, Elevators are designed and operate the same way. Yes we run up with a load or up empty.  We run down with a load or down empty.  BUT up empty or down with a load we regenerate power back into the building.

Don U. TCA 73-5735

  • Member since
    December 2005
  • From: Cardiff, CA
  • 2,930 posts
Posted by erikem on Friday, January 11, 2008 12:03 AM
Your explanation of four quadrant makes sense with a DC motor load (assuming fixed field).
  • Member since
    November 2006
  • From: Southington, CT
  • 1,326 posts
Posted by DMUinCT on Friday, January 11, 2008 9:05 AM

 Talking elevator control with DC motors (much the same as locomotives), "name plate" voltage gives us about 60% to 80% of elevator speed.  At that point we weaken the motor field to reach "contract speed".  At the start of a slowdown we first go to "full field" which puts us into a Dynamic slowdown and then hold the current negitive into the floor level. (+/- 1/4")  Dynamic Brake Resistors (DBR) are used in an Emergency Stop.

  With VVVF drive, we must rectify the AC back to DC and then use the IGBTs to produce exactly 60 Hz to push back into the power companies 440 volt, 3 phase lines.  That's the big added cost of 4 Quadrant VVVF drives, but component cost is falling.

BOY, IS THIS OFF TOPIC ! 

Don U. TCA 73-5735

  • Member since
    December 2005
  • From: Cardiff, CA
  • 2,930 posts
Posted by erikem on Friday, January 11, 2008 10:59 PM
 DMUinCT wrote:

 Talking elevator control with DC motors (much the same as locomotives), "name plate" voltage gives us about 60% to 80% of elevator speed.  At that point we weaken the motor field to reach "contract speed".  At the start of a slowdown we first go to "full field" which puts us into a Dynamic slowdown and then hold the current negitive into the floor level. (+/- 1/4")  Dynamic Brake Resistors (DBR) are used in an Emergency Stop.

  With VVVF drive, we must rectify the AC back to DC and then use the IGBTs to produce exactly 60 Hz to push back into the power companies 440 volt, 3 phase lines.  That's the big added cost of 4 Quadrant VVVF drives, but component cost is falling.

BOY, IS THIS OFF TOPIC ! 

I think it is on topic (motor control is at least as important for locos as elevators) - back in the 70's there was a fair amount of talk (more European than American) about using separately excited fields for traction motors.

Your comment about 4 quadrant converters covers what needs to be done to enable regenerative braking on electric locomotives with AC catenary. 

Thanks for the info.

  • Member since
    December 2001
  • From: NW Wisconsin
  • 3,857 posts
Posted by beaulieu on Saturday, January 12, 2008 10:22 AM
 erikem wrote:

I think it is on topic (motor control is at least as important for locos as elevators) - back in the 70's there was a fair amount of talk (more European than American) about using separately excited fields for traction motors.

 

The Brush built Class 60 Diesel-electric for British Rail (now on EWS) used separately excited fields in the traction motors for better traction control. They use the term SEPEX to describe their system. 

 

Your comment about 4 quadrant converters covers what needs to be done to enable regenerative braking on electric locomotives with AC catenary. 

Thanks for the info.

 

Did you see that Alstom used new generation Synchronous Permanent Magnet motors for the additional power bogies on the recordsetting TGV run. It's interesting that the powercars used Asynchronous 3-phase drives, while the bogies under the coaches used Synchronous Permag drives, that are being designed for the next generation TGV (to be called the AGV). 

  • Member since
    December 2005
  • From: Cardiff, CA
  • 2,930 posts
Posted by erikem on Saturday, January 12, 2008 11:41 PM

beaulieu,

I wasn't aware of separately excited fields on modern diesel-electrics (modern being post WW2), but it does make sense for traction control - wheel starts slipping, torque drops dramatically (although not quite as fast as an AC motor). My understanding (perhaps mistaken) was there was even more of an interest to use this for electric locomotives (SCR choppers made variable voltage DC drives practical in the late 1960's).

GE is working on permag traction motors for locomotives, expectations are to get 25% more torque for a given motor size (or the motor could be made smaller and lighter for a given power output - which may be the reason for uisng them on the coach bogies). Another advantage of a permag motor is more efficient operation at high torque output, especially at low speeds. A disadvantage is that the maximum speed will be a pretty hard limit set by the voltage rating of the inverter, where an induction motor can trade off some torque for extra speed.

Thanks for the tidbits. 

  • Member since
    October 2006
  • From: Chicago, Ill.
  • 2,843 posts
Posted by al-in-chgo on Sunday, January 13, 2008 2:40 AM
 erikem wrote:

beaulieu,

I wasn't aware of separately excited fields on modern diesel-electrics (modern being post WW2), but it does make sense for traction control - wheel starts slipping, torque drops dramatically (although not quite as fast as an AC motor). My understanding (perhaps mistaken) was there was even more of an interest to use this for electric locomotives (SCR choppers made variable voltage DC drives practical in the late 1960's).

GE is working on permag traction motors for locomotives, expectations are to get 25% more torque for a given motor size (or the motor could be made smaller and lighter for a given power output - which may be the reason for uisng them on the coach bogies). Another advantage of a permag motor is more efficient operation at high torque output, especially at low speeds. A disadvantage is that the maximum speed will be a pretty hard limit set by the voltage rating of the inverter, where an induction motor can trade off some torque for extra speed.

Thanks for the tidbits. 

Hope this isn't too OT, but I have heard for years that Chicago's L system cars brake with "reverse induction," instead of applied pads or calipers, compressed air, whatever.  That means the train can be quickly (if necessary, VERY quickly) slowed from the top speed of 55, 50 or so  and gets down to less than human-walking speed -- about 3 mph -- smoothly.  At that time apparently the motorman (or engineer, excuse me) does apply some kind of friction brake and it's only then that the train goes bumpy-bump-bump.  And there is no squealing or shrieking of brakes. 

A.  Does this, as I am guessing, have to do with the properties of DC current?

B.  In 1965 the L system introduced its first (and streamlined) "High Performance" trains.  They could cruise at about 5-7 mph faster than the old "green machine" type, so in some cases they can hit sixty, but mostly on the newer lines that have fewer way stations.  Did that 1965 "High Performance" series introduce an improved type of motor, better torque or acceleration, whatever?  (They were also the first to have air-conditioning.) 

C.  If this seldom-found type of braking is so great, why don't more American RT systems use it?  It's easy to tell when most of the subway systems in this country are deferring maintenance because not only do the trains squeal around corners, they screech as the brakes are applied.  IIRC Washington's Metro (Bechtel) was not supposed to screech like that but guess what:  a few years into its operation it screeched, too.  Apparently the brakes were silent only given tip-top condition, not "good enough to get by" condition.   -- a. s.

 

Help? 

 

 

 

 

al-in-chgo
  • Member since
    December 2005
  • From: Cardiff, CA
  • 2,930 posts
Posted by erikem on Sunday, January 13, 2008 11:59 PM

Al,

I'm not familiar with the control systems used on the Chicago cars, so anything I say is a guess and shouldn't be taken as proven fact. I seem to recall that many of the Chicago cars were built with PCC technology, which used both dynamic braking and magnetic track brakes. The PCC's dynamic brakes were good to a few MPH, which is consistent with your obeservation.

I'm also not familiar with the Washington Metro cars, but am familiar with the BART cars. Those use a combination of regenerative/dynamic braking plus disk brakes (which get used more often when problems crop up with the regeneartive/dynamic braking). The substations used on BART were not set for regenerative braking, so if a train goes into braking and there is no other train nearby to use the power, the control system starts switching in resistors to disapate the regenerated power. This latter system was causing BART problems in the mid-70's which led to a much greater than anticipated use of the disk brakes.

- Erik

 

  • Member since
    December 2001
  • From: NW Wisconsin
  • 3,857 posts
Posted by beaulieu on Tuesday, January 15, 2008 4:23 PM
Al there are two types of braking systems similar to what you describe; The first is magnetic track brakes which is probably what CTA uses, they are common on European trams and some narrow gauge railways. Basically you have a long brake shoe mounted on the frame of the tram that is clamped by magnetic attraction directly to the railhead. This is typically an emergency braking system because it rapidly decelerates the railcar. The other system is an eddy current braking system and is used for emergency braking on highspeed trainsets like the TGV and ICE, this works like a reverse linear induction motor working against the rails, this is used only in an emergency as it rapidly heats the rails and can cause kinks and other disturbances in rail alignment.

Join our Community!

Our community is FREE to join. To participate you must either login or register for an account.

Search the Community

Newsletter Sign-Up

By signing up you may also receive occasional reader surveys and special offers from Trains magazine.Please view our privacy policy