Trains.com

How much does double-stacking save?

11965 views
15 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    January 2007
  • 2 posts
How much does double-stacking save?
Posted by jgjnestor on Monday, July 30, 2018 8:36 PM

I have heard different opinions as to how much is saved by double-stacking containers v single stacking on say lighter-weight acticulated "5 packs" (five cars each taking a container of up to 48' on six trucks total.  These opinions range from the modest to the ridiculous.  Who is correct?

  • Member since
    January 2002
  • 4,612 posts
Posted by M636C on Monday, July 30, 2018 9:17 PM

The big advantage of double stacking is in train length.

This is important on long single track lines.

In areas of limited clearances there are often double tracks which reduces the need for double stacking.

There are links to some videos of Cajon Pass where whole trains are shown.

http://cs.trains.com/trn/f/111/t/271225.aspx

Many of these are all double stack containers, but others are double stack leading trailer on flat car and single stack containers, just on this route on BNSF, which matches what I've seen.

I recognise your avatar as the steam "Spirit of Progress".

As you no doubt know, that line is to be modified for double stack for the Inland Rail line to Brisbane, which will be single track for most of its length.

That line to Sydney is double track north of Junee and between Albury and Seymour, so there is less need for double stacking. However, the Inland Rail section to Illabo is about half way to Sydney anyway....

In Australia, clearances and axle loads are not as favourable for double stacking as in the USA. Double stack wagons can't use articulated trucks between wells because the axle loads would exceed the allowable limit for high speed operation, but articulation is fine for single stack trains.

So the weight differential for double stack would be higher in Australia since each well has to have two trucks, even if they are in groups of five. There were five articulated well sets built in Australia for high cubic light weight traffic (like breakfast cereal boxes).

In Australia there are very few 53' containers, with 48' being the norm and most articulated sets are to this length. The RRYY and TRAY type flats with small diameter wheels can all carry 53' boxes, and were used with half height boxes double stacked west of Adelaide.

In the USA, clearances encouraged New York Central to go with Flexi-Vans rather than TOFC, and some well vehicles were equipped to carry trailers in the wells for similar service. The Road Railer vans were mainly used in the East, where clearances could be a problem.

CSX is working to improve clearances in Baltimore and Washington DC where tunnels run under the city.

So the answer about weight differential depends on the allowable axle load and the maximum weight of container to be carried. The differential would be less in the USA than in Australia.

Peter

  • Member since
    June 2003
  • From: South Central,Ks
  • 7,170 posts
Posted by samfp1943 on Monday, July 30, 2018 11:02 PM

To jgjnestor;

 Member since 2007! Bow    First post! Wow   Welcome

 And your first response was from Peter [636C], a fellow Ausie...

                      Pretty Cool! Cool

 

 

 


 

  • Member since
    May 2003
  • From: US
  • 2,593 posts
Posted by PNWRMNM on Tuesday, July 31, 2018 1:42 PM

jgjnestor

I have heard different opinions as to how much is saved by double-stacking containers v single stacking on say lighter-weight acticulated "5 packs" (five cars each taking a container of up to 48' on six trucks total.  These opinions range from the modest to the ridiculous.  Who is correct?

Figure it out yourself based on your knowledge and assumptions.

Big double stack advantages are in train length, which comes into play on single track lines AND in terminals. The other is tare weight.

First look at your line and determine what maximum train length is, net of power, in feet. That sets your train length. Determine ruling grade which will determine trailing tonnage per unit, subject to drawbar limitations.

In the US trailer flats are about 93' coupled length and carry a maximum of two 48' trailers. 100 such cars are 9,300 feet, say 10,000 with power and you can carry 200 trailers. In line haul service determine train mile operating cost, or simply pick a number but try to come close. Determine tare weight of cars and trailers and then assume typical average load. That will give your tonnage. Check proposed lengths and tonnage agains real restrictions OR assume you are OK. If I was doing in US this would be my base case train.

Your base train my be different, say 53' flat cars. Just do the same calculations to get a reasonable single stack train with available equipment.

Now do same thing with double stack. We have 5, 3, and one box platforms. I would default to either 5 or 3 platform cars. A real train will be a mix but the figures for a pure train of each will be about the same. Lets use 5 platform cars which are about 280 feet long. 33 such cars are 9,240 feet long and can carry 330, 48 or 53 foot containers.

Do same weight calculation as for singles. Double stack tare per box is usually a bit less than single stack.

The key is that absent weight related constraints your train capacity is 350 with stacks vs. 200 single, or about 1.75 double to 1 single.

The cost to run either train, excluding equipment costs, will be identical, but more trains will cause more congestion, so a double stack system will impose less congestion cost than will single.

Very crudely, for every $100 cost per container for single stack, line haul cost for double stack will be $57.

Terminal costs are a whole different kettle of fish with tradeoffs going both ways that I have insufficient knowledge to attempt to discuss.

Mac

  • Member since
    January 2017
  • 48 posts
Posted by CMQ_9017 on Tuesday, July 31, 2018 6:00 PM

Doublestacking exact cost savings ranges depending on the lane but largely put it is about 40% -- rather significant enough to warrant the clearnace improvements. In some cases, it is the make or break factor.

Also keep in mind the toupet/filet model, so for example on Pan Am Southern you have a single stack train for about 200 miles then gets double stacked the rest of the trip to Chicago. The success of this operation relies on quick and efficient stacking operations so as to not slow down container velocity significantly, but it works well as a hybrid model. 

  • Member since
    January 2007
  • 2 posts
Posted by jgjnestor on Thursday, August 2, 2018 1:27 AM

Thank you to everyone for this info.    Because of the much lower typical axle load limits on Australian main lines, it would figure that here the transportation savings would be considerable lower than 40%; I had guessed at something above 20% after considering all the extraneous factors, so this is probably about right.

In particular, thanks Pete; and let me put to you an idea I have not seen, even after reviewing various ARTC studies:   The Melbourne- Brisbane Inland RR has the "small" disadvantage of avoiding the area containing half the population and much of industrial base and ports of Australia (Syd-New-Wollongong).  Although ARTC considers eventually converting Illabo to Moorbank for double stacking, this presents major problems north of Mittagong ie retunneling and raising bridges on the existing line in this section would be very expensive and still leave a substandard and busy layout; but a new alignment would be grossly expensive.  But double stack ability from Cootamundra (x Parkes or Melbourne) to Moss Vale would present modest expense, from where B double trucking via freeway to Sydney and Port Kembla, or drayage by single stack trains would be possible:  This would bring Syd-New-Wollongong into the Perth/Darwin double stack area.   I assume Parkes-Cootamundra is already or could easily be, double stackable.   What do you think?

  • Member since
    May 2003
  • From: US
  • 2,593 posts
Posted by PNWRMNM on Thursday, August 2, 2018 9:47 AM

If you are starting from scratch in terms of equipment, I would design a container optimized skeleton flat car for single stacks. Concept is the same as log cars in US Pacific Northwest. It is an underframe with cross pieces over the truck centers. You may or may not need to make them adjustible for different lengths.

Coupled length would be about 58' for 48' box. That allows for draft gear and coupler pocket, and you might be able to shorten each end by a couple of feet. A 100 box train would be about 5800 feet long excluding power. You could set them up in 3 or 5 unit configurations to minimize couplers, air hose connections and air leakage. The objective here is to minimize tare weight, and thus fuel consumption.

The advantage to double stack is still train length. If your railroad has short sidings, then the greater revenue per train becomes important. If the line is a busy single track, then minimizing the number of trains becomes important. Remember the rule of thumb that the number of meets per unit of time (typically a day) increases as the SQUARE of the number of trains. If double stack will require a lot of clearance improvements then it is out. Of course you could lengthen some or all sidings and your terminal track capacity will probably constrain train length.

So many choices!

  • Member since
    March 2016
  • 1,568 posts
Posted by CandOforprogress2 on Thursday, August 2, 2018 11:54 AM

My Hobo freinds say that single stacks and Trailer only UPS trains move fastera and have higher accelerration

 

  • Member since
    December 2001
  • 3,139 posts
Posted by chutton01 on Thursday, August 2, 2018 1:36 PM

PNWRMNM
If you are starting from scratch in terms of equipment, I would design a container optimized skeleton flat car for single stacks. Concept is the same as log cars in US Pacific Northwest. It is an underframe with cross pieces over the truck centers. You may or may not need to make them adjustible for different lengths.

Sounds a lot like the standard intermodal wagons in the UK and other countries.  (FEA wagon). Actually sounds like current US practice spine cars, but without the platforms for supporting trailer wheels (since it wouldn't be in TOFC service it seems).
I guess the thing is, you have well car designs which handle both single and double stacks already, why have two type of captial-intensive (that railroad accounting fright word) specialized equipment when one type can do both jobs more or less the same...

  • Member since
    May 2003
  • From: US
  • 2,593 posts
Posted by PNWRMNM on Thursday, August 2, 2018 6:07 PM

Chutton01,

The photo that came up for FEA wagon looks like a battle ship, far too heavy. Think ATSF spine cars set up for containers, but really GN/NP/MILW log flats are the place to start. Underframe 12x12" channel. Two cross bars (log bunks) and hang air brake parts on the underframe, even if in a strung out arrangement probably with truck mounted brake cylinders for simplicity. Tare should not exceed 50,000#. 40' log cars were about 35,000 pounds IIRC.

Can proponent avoid Interchange Rules? US standards impose a weight penatly that cars in dedicated service can avoid.

The design goals are light and cheap.

  • Member since
    January 2002
  • 4,612 posts
Posted by M636C on Sunday, August 5, 2018 12:30 AM

jgjnestor

Thank you to everyone for this info.    Because of the much lower typical axle load limits on Australian main lines, it would figure that here the transportation savings would be considerable lower than 40%; I had guessed at something above 20% after considering all the extraneous factors, so this is probably about right.

In particular, thanks Pete; and let me put to you an idea I have not seen, even after reviewing various ARTC studies:   The Melbourne- Brisbane Inland RR has the "small" disadvantage of avoiding the area containing half the population and much of industrial base and ports of Australia (Syd-New-Wollongong).  Although ARTC considers eventually converting Illabo to Moorbank for double stacking, this presents major problems north of Mittagong ie retunneling and raising bridges on the existing line in this section would be very expensive and still leave a substandard and busy layout; but a new alignment would be grossly expensive.  But double stack ability from Cootamundra (x Parkes or Melbourne) to Moss Vale would present modest expense, from where B double trucking via freeway to Sydney and Port Kembla, or drayage by single stack trains would be possible:  This would bring Syd-New-Wollongong into the Perth/Darwin double stack area.   I assume Parkes-Cootamundra is already or could easily be, double stackable.   What do you think?

 
Parkes to Stockinbingal will be adapted for double stack operation as part of the Inland Railway. New track will be built from Illabo to Stockinbingal to avoid the Bethungra Spiral (a bit like the Tehachapi loop but cut through solid rock not conducive to enlarging tunnels for double stacking).
 
There were no tunnels on the original main line from Picton to Mittagong, apart from one at Picton that has been bypassed by part of the new line north of the junction point. Using the old line would remove all tunnels between Mittagong and Moorebank. Of course it comes with significant problems not the least being the grade of 1 in 30, but it is five miles shorter and has far fewer curves. There is one tunnel now, under the M31 motorway which I understand is double track and suitable for double stacking.
 
It would be possible to build a new line from Douglas Park to Yanderra following the M31 motorway but I have no idea of the gradients involved. This is regarded as shortening the distance significantly, and would remove a number of the tunnels but the railway can't match the gradients on a road. It would cost much more than putting modern track on an existing right of way and might not give an easier gradient.
 
The Inland Railway runs through every National Party Electorate in NSW and that party is the main supporter. I fear that it won't be complete until well after the Pacific Highway upgrade and the intended freight will have already been lost to road. As the Highway upgrade progresses, the short lengths of two lane road become less of a problem to through truck traffic.
 
There is no possibility of double stacking over the Blue Mountains is nil, given the tunnels and the electrification. It might be possible to reconfigure trains in Parkes to single stack for high value cargo, but if the south line can be made to take double stack through to Moorebank that would probably meet the needs.
 
Peter
  • Member since
    January 2017
  • 48 posts
Posted by CMQ_9017 on Sunday, August 5, 2018 8:29 PM

I'm a bit confused on the weight problem here, why do you want lighter cars? Tare weight of a container is between 1.5MT and at heaviest 3.5MT, depending on container design/dimensions. Max payload into ports is typically about 26MT, with certain places going up to 26.5MT. Here in the US, the majority of roadways prohibit higher than 20MT payload, although with special permitting some places will let you push up to 63K LBs (27.5MT) with the vehicle gross weight being restricted to 100,000lbs with tri-axle chassis, tractor & container (or just a trailer in cases).

So when railcars push payloads of 100T (90MT)+ tare weights of 23MT to 28MT, why are you trying to limit weight with a fully loaded container being tare + payload at max possibility 30MT? So with 2 units per car (either doublestacked or circus style), you'd be looking at 60MT and you still have plenty of weight for the tare of the railcar and be well under the max. In fact, that is why you can share weight on a single axle in well cars... the payload will never be as heavy as tradition railcar even doublestacked.

Light & cheap = derailment & fast asset depreciation. I've not heard of anyone looking to design intermodal equipment light (in some instances I believe they add ballast to keep them heavy so they stay on the rails when empty mid consist). In fact when a certain Northeast regional was running spine cars on the head pin on heavy freight trains in the early intermodal days they had quite a few problems including a spectacular derailment in Massachusetts. The practice was ended then and there.

Your note on log flats -- wood is a dense fiber material and will quickly max out a payload, so you need a lighter car to maximize utility. I would not advise on a container train. 

  • Member since
    May 2003
  • From: US
  • 2,593 posts
Posted by PNWRMNM on Sunday, August 5, 2018 9:21 PM

CMQ,

Since I brought up weight I suspect your comments are aimed at me so I will respond.

Weight is the arch enemy of fuel effeciency and tonnage ratings. Fuel effeciency particularly on lines with a lot of redundant grades. Since this is proposed for Austrailia I will make the happy assumption that buff strength requirements are less then here, so is more opportunity for weight reduction since single stack car tare is higher per revenue unit than is case for double stack.

That said, here is a real single well car for North American service, Greenbrier 'All Purpose 53' which is designed for 53' boxes, is 76'9" coupled and 54,000# tare, or 27,000# tare per box. The OP says 48' box which would shorten car by 5 feet and lighten by maybe 2,000#.

Wiki has a NP log flat diagram, which is typical in general look but converted from a box car so probably not optimized in terms of length or tare, so it is a conservative starting point. Car is 42 feet over end sills, 45'6" coupled length, with 32 foot truck centers and bunk centers. 40 ton trucks were 14,200# and body was 19,300#. I suspect 70 ton trucks are 18,000 pounds. Need to add 16 feet between truck centers/bunks at about 400#/foot, say 6,400# added to the body for total body weight of say 26,000# for a total tare weight of 44,000#. Note this tare per box is almost double that of the stack car.

That seems heavy enough to stay on the rail. ATSF fuel foilers would be a design to look at in terms of derailment history.

I presume unit trains and have assumed there may be short siding issues, which are relatively cheap to fix but may have to be lived with. There is no question that double stack is far more effecient in terms of tare weight and box capacity for given train length IF you have the clearances. If you do not, then shed all possible tare off single height equipment since it starts out heavy and you have burn fuel to haul tare weight both ways and up and down. Of course, if unit trains, then can set up as units of five or ten platforms between couplers which would shed the tare of at least some trucks and a bunch of couplers and give a tighter train line.

Mac

  • Member since
    January 2002
  • 4,612 posts
Posted by M636C on Sunday, August 5, 2018 11:34 PM

There is no need to guess or approximate the details of Australian container wagons. All the data is available:

https://www.artc.com.au/library/GI_10_loco_rs_data.pdf

To compare the most common types:

An RRZY five pack well set with 20 axles weighs 102 tonnes empty and is 106.5 m long. It can be loaded to 350, 380 or 420 tonnes depending on whether 21, 22 or 23 tonne axle loads are permitted.

An RRAY five pack single level articulated set with 12 axles weighs 53 tonnes empty, 73.1 m long and can be loaded to 228 tonnes at all times.

The load to tare ratio is 350/102 = 3.43 for the RRZY and 228/53 = 4.3 for the RRAY.

The axle load is the restricting feature.

To carry ten 48' containers you need 102m double stacked or 146.2 m single.

So a single stack train will be 40% longer.

Peter

 

  • Member since
    May 2003
  • From: US
  • 2,593 posts
Posted by PNWRMNM on Monday, August 6, 2018 7:46 AM

Peter,

Thank you for the real info. Your designers have done a very good job in limiting tare weight of skeleton type container cars.

Mac

  • Member since
    January 2002
  • 4,612 posts
Posted by M636C on Tuesday, August 7, 2018 8:13 AM

PNWRMNM

Peter,

Thank you for the real info. Your designers have done a very good job in limiting tare weight of skeleton type container cars.

Mac

 

 

There is one problem with the ARTC listing. In general, all the details of vehicles are correct but the details of the operators are a little strange. As with many official lists, things get added but things aren't often edited out.

The passenger train cars shown as belonging to "QR National" are now in service in Peru to Macchu Pichu, and I don't think they ever belonged to QR National, but to QR, the parent organisation.

One reason the container cars are light is that none need floors, since there is virtually no TOFC operation in Australia. This is a result of the tight clearances in most of the East. Where clearances allowed, there was a rapid adoption of double stacking.  One reason for this is that only circus train loading has been used in Australia, nothing like USA Z vans with direct fork lift loading has been used. This is strange, since TOFC in Well cars would fit many areas in the East and might open new markets. Rail has a very small share of overall interstate freight and carrying vans instead of containers might widen the customer base.

The reason for the axle load limits is the use of rail as light as 94lb/yd in main lines, and old bridges. This is being replaced by 120lb/yd rail and most main lines already have concrete ties.

I note that there is little difference in loaded weight between single and double stack trains with the examples I've given although there are significant length savings. Current train lengths on main lines are between 1500 m to 1800 m, so just under a mile to a bit over a mile in length.

Peter

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