Trains.com

Why does AC traction provide increased adhesion?

11181 views
62 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    June 2002
  • 20,096 posts
Posted by daveklepper on Thursday, July 8, 2021 9:41 PM

Also, the wheel slip problem may be over-rated as the limitation on DC motor technoloogy.   Most of the time, the limit is the current rating, the ammeter staying in the Yellow Zone too long with the fear of motor blow-out or even going into the Red Zone.  At least that was my experience.

Admittadly, on the routes I rode, the tracks were kept polished by frequent traffic, ten or more trains a day, two or three freights each way, possibly as many passenger trains, and partial coverage by numerable commuter trains.

Engineers who ran the DC diesels can comment whether wheel-slip or current limitations were the main headache on the trains they operated.   Or what else was the problem the AC-traction seems to have solved?

But again, modern computer technology can also mitigate wheel-slip with DC-traction, but, again, why bother?

  • Member since
    June 2002
  • 20,096 posts
Posted by daveklepper on Thursday, July 8, 2021 10:30 PM

Overmod (and others) :  Do any shortline and/or tourist railroads actually prefer power that is not dependent on computer technology?   Not brcause of historic preservation or low purchase price, but because of local on-site repair ability?

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Thursday, July 8, 2021 10:45 PM

daveklepper
Do any shortline and/or tourist railroads actually prefer power that is not dependent on computer technology?

Plenty!  And this is not a new issue; a major stated reason for scarcity of '80s-era power in collections is the absence of electronic support for either the 'boards' or the knowledge how to service or repair them.  I expect this to be worse for the early generation of AC inverter units with 'orphan' components or approaches.  And all this before we get to manufacturers intentionally withholding critical information or support... or owners like Amtrak with a resolve to ban even the least restoration to operation of equipment for which they might be held deep-pockets liable...

Remember that a major reason given for the extinction of the Republic Locomotive 'starship' FL9 conversions was that even if the 'salable' components could have been extracted without severe carbody damage, the resulting components were impossible to convert to be used with 'typical' earlier-generation diesel-electric knowledge.  

I'll admit that 'hope springs eternal' that some version of an 'open-source' hacking protocol, like MegaSquirt for automobile fuel injection, comes to be developed for modern units.  It's almost all a labor of love, though, and as Preston Cook notes, the 'edge of history' has already passed for much of the flat-pack era and even the '80s approaches to systems integration.

  • Member since
    November 2008
  • 1,881 posts
Posted by Leo_Ames on Friday, July 9, 2021 7:34 AM

SD70Dude
NRE was trying to market something like that, I think they called it the SD40-4 with chopper controls or something.  I don't recall if they sold any.  I recall reading that they claimed it gave AC-like performance with DC motors (but probably not the same durability).  

I believe CSX came close to implementing this with their SD40-3's. CSX planned a tractive effort increase from around 86,000 to a maximum of 130,000 pounds when developing the project, but backtracked with the rebuilds instead outshopped with a rating of around ~82,000 pounds.

Unsure if they had early intentions of trying to do anything similar with the GP38-2's and GP40-2's that were planned to follow the SD40-3 program, but doubt it. Those rebuilds when they did show up, carried similar ratings to what they had when first released by EMD in the 1970's. 

While the limits of DC motors would prevent them taking full advantage of it in heavy haul service like coal drags in the mountains, it might've been real useful for the types of work that their SD40-2's typically do nowadays. They sometimes have pretty heavy hauls to do over relatively short distances at low speeds.

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Friday, July 9, 2021 8:44 AM

I'd be highly interested to read exactly what CSX intended to do to get 130K from six motors, and what the physical and electrical assumptions were.

Presumably some of the same approaches used with the AEM-7 motors could be used... if the form factor permitted the necessary field-winding complexity net of the active cooling of the windings (some of which might be done with copper heat-pipes in the winding structure).  To do this for more than momentary rating even with low loss in the "chopper" gating would still involve heroic measures for traction-motor cooling, with the armature current heat dissipation still being difficult without coolant sprayed into the traction-motor-blower ducting...

Don Oltnann may be familiar with what Conrail and then NS might have been doing to increase effective TE with DC motors -- NS was wedded to them for a considerable time past the 'modern AC revolution'.

  • Member since
    July 2004
  • 2,741 posts
Posted by Paul Milenkovic on Friday, July 9, 2021 8:50 AM

Overmod

 

I'll admit that 'hope springs eternal' that some version of an 'open-source' hacking protocol, like MegaSquirt for automobile fuel injection, comes to be developed for modern units.  It's almost all a labor of love, though, and as Preston Cook notes, the 'edge of history' has already passed for much of the flat-pack era and even the '80s approaches to systems integration.

 

 

(Dana Carvey's Grumpy Old Man talks building custom circuits)

When I was young, weeee didn't have any of those fancy-pantsy circuit CAD software programs and supply houses of circuit boards with the flatpack ICs already soldered on.  Weeee had to wire-wrap our electric connections by hand until our fingers bled . . . and we loiked it!

If GM "killed the electric car", what am I doing standing next to an EV-1, a half a block from the WSOR tracks?

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Friday, July 9, 2021 9:51 AM

Paul Milenkovic
Weeee had to wire-wrap our electric connections by hand until our fingers bled . . . and we loiked it!

I am actually old enough to remember using wire-wrap to make connections for high-heat conditions that would melt solder...

... uphill both ways, through the rain, at Mach 27... ah yes, doze were de daiz...

But what's this about wire-wrapping that makes fingers bleed?  We had fancy tools with Pentagon-hammer price that both did the wrapping to spec and loosened the connections for removal.  You couldn't do those by hand, blood or no blood! Wink

(Or at least so I thought.  It would not surprise me greatly to learn there were people who could...)

  • Member since
    December 2006
  • 1,879 posts
Posted by YoHo1975 on Friday, July 9, 2021 12:14 PM
I remember those "chopper" solutions that were put out. I thought at least a handful were sold, but I have no idea what the success rate was.
  • Member since
    December 2001
  • From: NW Wisconsin
  • 3,857 posts
Posted by beaulieu on Friday, July 9, 2021 11:50 PM

daveklepper

Regarding high speed:  All the currently constructed truly high-speed trains Worldwide use AC non-synchronous induction motors.  All.

 Its late night here but I seem to remember that Alstom switched to permanent magnet traction motors used in their AGV series trainsets (Nuovo Transporto Viaggiatori's Italo trainsets used in Italy). 

 

  • Member since
    January 2019
  • 1,673 posts
Posted by Erik_Mag on Saturday, July 10, 2021 11:52 AM

Overmod

I'd be highly interested to read exactly what CSX intended to do to get 130K from six motors, and what the physical and electrical assumptions were.

The GP50's speed versus tractive effort graph that appears in the 1980 Car & Locomotive Cyclopedia shows a max TE of 95,000 lbf and the graph for the GP40 tops out around 85,000 lbf. The latter figure would imply that 130k from six motors would be achievable. What isn't said is how long that the 130k could be produced without overheating.

Using separate chopper control of the field windings and armatures of DC motors provides two benefits with respect to adhesion. First is that the constant field strength would give a faster torque drop off with the start of slip. Second is that taking the field inductance (but not the commutating pole inductance) lets the motor respond faster to changes in the applied armature voltage.

  • Member since
    June 2002
  • 20,096 posts
Posted by daveklepper on Saturday, July 10, 2021 4:28 PM

If the rotating copper or aluminum bars are replaced by permanent magnets, and the motor becomes a sort of iside-out permanent -magnet motor, eith the field coils still requiring stepped computer control. which controls speed, efficiency is (even) greater, but at the price of both greater cost and greater un=sprung weight and track and rail wear.  Are these trains high-speed or commuter and/or regional trains?

Alstom also (in application) pioneered the inside-out non-synchronous motor, a form of wheel motor, where the rotating bars are attached to the hub of the wheel, rotating with the wheel, and the field coils on the fixed, non-rotating axle. This was azpplied in some electric buses and many airport low-floor terminal-to-plane buses. Germany's Magnet Motor, and Derby, England's Stored Energy Systems have rotating magnets on the wheel hub and field coils on the non-rotating axle. 

As far as I know, none of this is i9n really high-speed rail.

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Saturday, July 10, 2021 5:47 PM

One great advantage of a PMM traction motor is that instead of the required 'slip' to induce magnetism in the bars, with the associated heating, the armature can run in perfect sync with the 'field rotation'.

I believe these motors use very strong magnetic material like NIB, so the weight of the armature, and probably its required diameter to produce a desired torque, can be much less.

http://www.imaya.biz/MAYA/Alstom%20Permanent%20Magnet%20Motors%20PMM%20-%20MAYA%20transmission%20Co.,%20LTD.PDF

(I have an information link to Alstom engineering for anyone who actually wants hard data for a particular configuration, if anyone has the actual interest.)

http://repozytorium.p.lodz.pl/xmlui/bitstream/handle/11652/366/Analysis_of_the_traction_Wawrzyniak_2009.pdf?sequence=1&isAllowed=y

  • Member since
    December 2017
  • From: I've been everywhere, man
  • 4,269 posts
Posted by SD70Dude on Saturday, July 10, 2021 6:49 PM

130 klbs is between what notch 4 and notch 5 will produce at very slow speeds on an ES44AC or ET44AC.  This is around the limit of what even they can sustain at 0 to 1 mph without slipping excessively.  

A 216 ton ET44AC can sustain 190 to 195 klbs at 5-7 mph without bucking around too much, with a bit of independent and lots of sand to help it along.  I never have seen the gauge hit the full 200 klbs that they are rated at.

For what it's worth, our Locomotive Engineer operating manual (CN file 8960) contains a list of DC locomotives that short time ratings need not be observed on, basically everything newer than a SD40-2.  From an operator's viewpoint slipping is by far the greater of the two problems.  

DC units are capable of producing tremendous amounts of power, if you go to high throttle on a Dash-8 or Dash-9 with the independent applied (a common strategy for starting a heavy stretched out train on an ascending grade) you can get the ammeter to peg, I think 1800 amps is the highest number on the gauge.  But they can't use it without slipping.

Greetings from Alberta

-an Articulate Malcontent

  • Member since
    May 2003
  • From: US
  • 25,276 posts
Posted by BaltACD on Saturday, July 10, 2021 6:59 PM

I have always felt that tonnage ratings for a territory should be how much tonnage a locomotive CAN GET STARTED FROM A STAND STILL on the ruling grade.  In most cases that is not the standard.

Never too old to have a happy childhood!

              

  • Member since
    December 2017
  • From: I've been everywhere, man
  • 4,269 posts
Posted by SD70Dude on Saturday, July 10, 2021 7:09 PM

BaltACD

I have always felt that tonnage ratings for a territory should be how much tonnage a locomotive CAN GET STARTED FROM A STAND STILL on the ruling grade.  In most cases that is not the standard.

We don't even get published tonnage ratings anymore, I know they exist but they are no longer found in the employee operating manual.  

CN rates a six axle AC unit to pull 10,000 tons westward from Edmonton to either Prince George or Kamloops, where the ruling grade is 0.4%.  I don't know the territory west of those locations but the rating may be higher if the lines follow river grade to the Pacific coast.  

With such a train there are several locations with short, steeper grades where it is a bad idea to stop, and still more where it is a bad idea to stop in the rain.  We've also had AC powered trains stall on ruling grades after the lead unit ran out of sand, or the shop forgot to add any in the first place (they've done the same with water, oil and even fuel on occasion).  

Greetings from Alberta

-an Articulate Malcontent

  • Member since
    July 2004
  • 2,741 posts
Posted by Paul Milenkovic on Saturday, July 10, 2021 8:04 PM

SD70Dude

 

 
BaltACD

I have always felt that tonnage ratings for a territory should be how much tonnage a locomotive CAN GET STARTED FROM A STAND STILL on the ruling grade.  In most cases that is not the standard.

 

 

We don't even get published tonnage ratings anymore, I know they exist but they are no longer found in the employee operating manual.  

CN rates a six axle AC unit to pull 10,000 tons westward from Edmonton to either Prince George or Kamloops, where the ruling grade is 0.4%.  I don't know the territory west of those locations but the rating may be higher if the lines follow river grade to the Pacific coast.  

With such a train there are several locations with short, steeper grades where it is a bad idea to stop, and still more where it is a bad idea to stop in the rain.  We've also had AC powered trains stall on ruling grades after the lead unit ran out of sand, or the shop forgot to add any in the first place (they've done the same with water, oil and even fuel on occasion).  

 

Any foolish locomotive crew can operate their power with fuel.  It is time crews learned how to run locomotives without fuel to meet the Operating Plan?Mischief

If GM "killed the electric car", what am I doing standing next to an EV-1, a half a block from the WSOR tracks?

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Saturday, July 10, 2021 8:12 PM

SD70Dude
... if you go to high throttle on a Dash-8 or Dash-9 with the independent applied (a common strategy for starting a heavy stretched out train on an ascending grade) you can get the ammeter to peg, I think 1800 amps is the highest number on the gauge.  But they can't use it without slipping.

R.J.Russell, who was a NEC passenger engineer in the early Seventies, told me that the normal way to start a commuter train with a couple of passenger Sharks on the Long Branch/Bay Head trains was to open the throttle until the ammeter pegged at 2000.  He said it would sag down off the pin after about 30 seconds (!) and this performance would be repeated at many of the stops.  (These only had four TMs per locomotive, in A-1-A trucks)

These were Westinghouse hexapole motors, so this is likely not a tall tale, but it's interesting that the engines were often kept in pairs even on short trains for what has been described as 'reliability' reasons.

  • Member since
    May 2003
  • From: US
  • 25,276 posts
Posted by BaltACD on Saturday, July 10, 2021 9:38 PM

Paul Milenkovic
Any foolish locomotive crew can operate their power with fuel.  It is time crews learned how to run locomotives without fuel to meet the Operating Plan?Mischief

Before I retired, crews were supposed to report to the Train Dispatcher if their fuel gauge indicated less than 1000 gallons.  Fuel condition of a locomotive was also displayed in the locomotive management computer application which was accessable to everyone in transportation.

If necessary, locomotive management would arrange in route fueling if there was too little fuel on board to make the scheduled fueling location.

Sometimes fuel gauges were inoperative and the computer record for a engine was not current and a engine would run out of fuel.

Never too old to have a happy childhood!

              

  • Member since
    July 2010
  • From: Louisiana
  • 2,310 posts
Posted by Paul of Covington on Saturday, July 10, 2021 11:39 PM

SD70Dude
DC units are capable of producing tremendous amounts of power, if you go to high throttle on a Dash-8 or Dash-9 with the independent applied (a common strategy for starting a heavy stretched out train on an ascending grade)

   How does using the independent help?  I'm guessing that it has something to do with limiting slipping, but I don't see how.

_____________ 

  "A stranger's just a friend you ain't met yet." --- Dave Gardner

  • Member since
    December 2017
  • From: I've been everywhere, man
  • 4,269 posts
Posted by SD70Dude on Sunday, July 11, 2021 12:34 AM

In the case of starting a train, increasing the throttle notch by notch makes very abrupt increases in pulling power, by throttling up first and then slowly releasing the independent you have a lot more control over how much pulling power goes into the train and you can start the train more smoothly.

If you release the independent and pick the right notch to throttle up to it might work out well on some trains.  But if you don't throttle up enough or have slow loading power like older DC GEs the brakes will release before the units load up and you will start rolling backward, or if the units roll into the train and then load up the slack will be jerked out hard, potentially causing a train separation.  And if you throttle up too much the units can slip excessively, jerking the slack around and potentially dropping their load and then rolling back into the train.....

Once you get moving keeping some independent applied at speeds below about 15 mph seems to help reduce wheelslip.  I don't know why, it just does.  If I had to guess I would say the pressure from the brake rigging might keep everything tight and reduce the tendency of trucks to rise up at the front, and when loss of adhesion and wheelsip does occur the brake friction reduces the rate of acceleration of the slipping wheelset, preventing the control system from detecting a "uncontrolled wheelslip" and drastically reducing load to all axles.  

The above is just speculation.  Someone like Dave Goding might be able to provide a more accurate technical answer.

Greetings from Alberta

-an Articulate Malcontent

  • Member since
    June 2002
  • 20,096 posts
Posted by daveklepper on Sunday, July 11, 2021 1:02 AM

Overmod, you are correct about perfect synchonization, anfd the magnet bars are not slanted, but parallel to the axles, and these are, indeed, the most efficient variable-speed motors available.

  • Member since
    July 2010
  • From: Louisiana
  • 2,310 posts
Posted by Paul of Covington on Sunday, July 11, 2021 3:25 AM

   Thanks, SD70.  There are a lot of nuances in train handling that a lot of non-railroaders like me never think about.

_____________ 

  "A stranger's just a friend you ain't met yet." --- Dave Gardner

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Sunday, July 11, 2021 7:21 AM

Paul of Covington
How does using the independent help?

As a partial analogy, from a different perspective: I observed many years ago that the only way to provide truly smooth starts on IRT subway trains involved coordinating the brake release and engaging the controller notches.  That observation was then confirmed by engineers of the NYC commuter MUs operating into GCT, and a bit later by the crew of what were then MP54s on the PJ&B (which used a highly different traction system but with restricted 'notches').  Interestingly enough a similar technique was used on large Baldwins, which had stepless air throttles but relatively high power at idle.

One concern more directly relevant is the eight 'notch' system of MU control, which regulates the diesel prime mover rpm and then controls the electrical generator/alternator loading to match.  That does not give smooth millimetric control of the traction-motor power -- it effectively 'quantizes' the torque output.  Using the brake controls this at the wheels, where it matters; it also provides control of some kinds of motor overspeed that might occur if an axle or two breaks away momentarily.

  • Member since
    April 2007
  • From: Bridgman, MI
  • 283 posts
Posted by bogie_engineer on Sunday, July 11, 2021 6:35 PM

SD70Dude

In the case of starting a train, increasing the throttle notch by notch makes very abrupt increases in pulling power, by throttling up first and then slowly releasing the independent you have a lot more control over how much pulling power goes into the train and you can start the train more smoothly.

If you release the independent and pick the right notch to throttle up to it might work out well on some trains.  But if you don't throttle up enough or have slow loading power like older DC GEs the brakes will release before the units load up and you will start rolling backward, or if the units roll into the train and then load up the slack will be jerked out hard, potentially causing a train separation.  And if you throttle up too much the units can slip excessively, jerking the slack around and potentially dropping their load and then rolling back into the train.....

Once you get moving keeping some independent applied at speeds below about 15 mph seems to help reduce wheelslip.  I don't know why, it just does.  If I had to guess I would say the pressure from the brake rigging might keep everything tight and reduce the tendency of trucks to rise up at the front, and when loss of adhesion and wheelsip does occur the brake friction reduces the rate of acceleration of the slipping wheelset, preventing the control system from detecting a "uncontrolled wheelslip" and drastically reducing load to all axles.  

The above is just speculation.  Someone like Dave Goding might be able to provide a more accurate technical answer.

 

I can only speculate; I can't recall any adhesion testing where this was tried as we wouldn't expect the engineer to have to do this. You may be right about it stabilizing the suspension, it would add some vertical friction damping on a truck with a pedestal suspension. If the wheels are rotating, the brake shoes will clean the wheel surface enhancing adhesion.

Dave

  • Member since
    January 2001
  • From: Atlanta
  • 11,971 posts
Posted by oltmannd on Sunday, July 11, 2021 8:02 PM

Overmod
Don Oltnann may be familiar with what Conrail and then NS might have been doing to increase effective TE with DC motors -- NS was wedded to them for a considerable time past the 'modern AC revolution'.

NS was highly "siloed".  Mechanical ruled the roost for locomotive purchases but didn't really know what was going on on the RR.  In 1998, I asked a Mech Dept exec "why no AC locos?" as part of an informal interview.  (I was shopping myself around prior to the organized "transition" interview process)

They told me, "they cost more and we don't really run trains with more than two units, so AC gets doesn't get us much".  Reality was they were running most merchandise trains with more than two units.  He really was out of touch with what the transportation plan was.    NS was really behind compared to Conrail at many things - mostly because they had nothing pushing them.

-Don (Random stuff, mostly about trains - what else? http://blerfblog.blogspot.com/

  • Member since
    July 2004
  • 2,741 posts
Posted by Paul Milenkovic on Sunday, July 11, 2021 8:35 PM

oltmannd

 

 
Overmod
Don Oltnann may be familiar with what Conrail and then NS might have been doing to increase effective TE with DC motors -- NS was wedded to them for a considerable time past the 'modern AC revolution'.

 

NS was highly "siloed".  Mechanical ruled the roost for locomotive purchases but didn't really know what was going on on the RR.  In 1998, I asked a Mech Dept exec "why no AC locos?" as part of an informal interview.  (I was shopping myself around prior to the organized "transition" interview process)

They told me, "they cost more and we don't really run trains with more than two units, so AC gets doesn't get us much".  Reality was they were running most merchandise trains with more than two units.  He really was out of touch with what the transportation plan was.    NS was really behind compared to Conrail at many things - mostly because they had nothing pushing them.

 

Yes, but . . .

If NS was running merchandise trains with more than two units, they were doing it for the horsepower-at-speed, not for the lugging ability?  

The 6000 HP AC locomotive unit, at least in US railroading, would have had the lugging tractive effort for drag freight and the HP in fewer units for merchandise.  But because reasons the 6000 HP size didn't work out?

 

If GM "killed the electric car", what am I doing standing next to an EV-1, a half a block from the WSOR tracks?

  • Member since
    June 2002
  • 20,096 posts
Posted by daveklepper on Monday, July 12, 2021 8:15 AM

On the B&M, 1952 and 1953, I obviously never had experience with distributed power.  Techniques like mild independent brake application at starting and the reverse, power train-braking to reduce sliding possibilities as well as jerks for passengers, are widely applied..

But how in the World is anything accomplished with the independent when using distributed power?  Would not some real problems occur?

Again, I nerver had the experience of a head-end freight ride with distrbuted power, so I wonder if the independent is useful under that condition.

  • Member since
    May 2003
  • From: US
  • 25,276 posts
Posted by BaltACD on Monday, July 12, 2021 9:49 AM

daveklepper
On the B&M, 1952 and 1953, I obviously never had experience with distributed power.  Techniques like mild independent brake application at starting and the reverse, power train-braking to reduce sliding possibilities as well as jerks for passengers, are widely applied..

But how in the World is anything accomplished with the independent when using distributed power?  Would not some real problems occur?

Again, I nerver had the experience of a head-end freight ride with distrbuted power, so I wonder if the independent is useful under that condition.

Not being a engineer - I suspect Engineers handling a DPU trains become adept at 'Fencing'.  Raising and lowering the Fence that allow the DPU units to be operated at different settings than the lead locomotive consist.

Never too old to have a happy childhood!

              

  • Member since
    December 2017
  • From: I've been everywhere, man
  • 4,269 posts
Posted by SD70Dude on Monday, July 12, 2021 10:27 AM

In the current DP system the remote's independent brake always does the same thing the lead consist's is doing.  Putting the fence up only affects throttle and DB commands.  

On the older Locotrol I and II systems the remote's independent brake was controlled separately.  This might make things a bit more challenging at times (yet another button to operate) but it also would have given more flexibility.  

I only have experience with the modern DP system, but I've heard from a couple former BC Rail engineers that it was possible to move a Locotrol remote around by itself using the lead unit's controls, just like a beltpack.

Greetings from Alberta

-an Articulate Malcontent

  • Member since
    January 2019
  • 1,673 posts
Posted by Erik_Mag on Monday, July 12, 2021 10:33 PM

bogie_engineer

I can only speculate; I can't recall any adhesion testing where this was tried as we wouldn't expect the engineer to have to do this. You may be right about it stabilizing the suspension, it would add some vertical friction damping on a truck with a pedestal suspension. If the wheels are rotating, the brake shoes will clean the wheel surface enhancing adhesion.

That's been one of my thoughts as to why having some independent brake applied helps with traction. Another thought is that traction motors running with enough current to cause wheelslip will have a mostly saturated field, so the rise in back EMF due to slipping will quickly cause a reduction in armature current and thus torque, which then limits the slipping.

FWIW, GN's original Cascade Tunnel electrification was a three phase electrification using induction motors and the locomotives were noted for high factors of adhesion.

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