Western Railroad Discussion > Saw-By Meet? Double Saw-By? What Are They?

I've been curious about this for some time. Perhaps Roustabout Al, who had the great "Kicks and Tricks" thread a few days ago, could comment. Or any of the other immensely knowledgable railroaders who so kindly contribute to this board.

I have always thought that a saw-by was a meet where two trains are both too long for a given siding. However, I have seen a couple of references (here's one: http://www.trainorders.com/discussion/read.php?11,860466,860466#msg-860466 ) where it seems that perhaps a saw-by only involves *one* train that is too long for the siding. Is this correct?

So then, two too-long trains meeting at too-short siding would be either a saw-by or a double saw-by. But it would be known as a problem, nonetheless. So can somebody please describe the process for passing these two trains? I think I read about something like this happening recently at Silt, Colorado (UP Glenwood Springs Sub), but they had a house track to work with (cheaters!)... How would you do it with no extra tracks? Let's say, for example, both trains are 300-500 feet too long for the siding.

Thanks guys!
-Wes

drgw Wrote:
-------------------------------------------------------
> ... a saw-by only involves *one* train that is
> too long for the siding. Is this correct?

Correct.

> So then, two too-long trains meeting at too-short
> siding would be either a saw-by or a double
> saw-by.

A double sawby.

> So can somebody please describe the
> process for passing these two trains?

If trains that are both too long for a siding have to meet, it is a very time consuming process, and something that the dispatcher will avoid at all costs:

Let's call them Train A and Train B, running in opposite directions. For argument, Train A is 500 feet longer than the siding, and train B is 1000 feet longer than the siding.

Train A pulls up on the main at a siding and stops short of the turnout at the far end. At least the rear 500 feet of train A will be sitting on the main beyond the clearance point of the siding, due to being over siding length.

Train B starts to enter the siding, but stops at a point where the rear 1000 feet of the train are still fully on the main track. Normally this will also be at least 500 feet from the turnout (the excess length of Train A) The excess length of Train B is then uncoupled from the train, tied down, and left on the main track. The head end of Train B then pulls ahead in the clear between the switches of the siding.

Train A then normals the turnout for the main, and proceeds ahead to couple onto the end of Train B, which is still sitting on the main. The hand brakes are released, and the air connected for a brake test.

Once the tail end of train A clears the switch at the opposite end of the siding, and while the brakes on the standing cut are being properly tested, Train B can pull out onto the main track and proceed far enough ahead so that its tail end is at least the length of train A plus 1000 feet (the excess length of Train B) away from the turnout.

When Train A has completed the brake tests, and Train B has moved far enough away from the siding, train A will back up until its locomotives are nearing the siding switch at the far end of the siding. At this point, the excess length of train B, still coupled to the front of the locomotives on Train A, would be on the main track, between the switches of the siding. Train A would stop, and uncouple the excess length of Train B, tie it down, and leave it on the main between switches.

Train A would then continue to back up until it could reverse direction and enter the siding. At this point, the tail end of Train A would be nearing train B, which would have stopped on the main track.

Train A would then proceed through the siding and depart.

Train B would back up onto its tail end, release handbrakes and upon a satisfactory brake test, would depart.

End of story, two hours after the beginning.

The choice of which train would drop its tail end cars would be based on gradient, train tonnage, and locomotive consists. The exact locations for leaving the cuts of cars would vary according to the difficulty in walking on the ballast, and whether there was a caboose or not.

Read an article by J.D. Santucci at www.railroad.net. He gives a detailed description of how one occurred recently.

RailNet802, out

http://www.railroad.net/articles/columns/hottimes/hottimes_20050230.php
Here's the link to the article.

Josie Moore Crum's book ``Rio Grande Southern'' details an 11-move saw. This was done at the Millwood siding between Mancos and Dolores, CO. It involved a passenger train, (pre Galloping Goose) two frieght trains and a cut of cars left on the siding that the dispatcher forgot about.
Moore's husband John Harvey Crum, was the conductor in charge of the saw. It was done at night on a curved piece of track with the traiman communicating only with lanterns.
Veteran D&RGW engineer Andy Payne said old Arboles was the meeting point for freight trains on the Durango-Chama district and saw bys occurred often. While the railroad lengthened the Toltec siding to avoid that on the Cumbres Pass section no sidings were lengthened west of Chama. In fact Carracas siding, the most favorable point for siding extention about halfway between Chama and Durango, was taken out when the San Juans quit running. This was fuel for veteran trainmen and enginemen to take jabs at management, especially when sawing at old Arboles.

The foregoing description of a double saw-by is very good and describes the most common form. But there others. For example, the two trains can exchange some cars (I saw this on the Milw once, two empty log trains).

Doug Harrop had a good description of one occurring on the ex-WP sometime within the last 12 months of CTC Board.

Best regards,

Rob L.

Thanks everyone for your responses! Thank you, run8, for your excellent description of what I might call a "full" double saw-by (where everyone ends up with all the same cars, in the same order).
Interesting to hear that they sometimes swap cars from one train to the other, or perhaps even both ways. Just makes it that much harder for the next guy, though.
Take care,
-Wes

run8 Wrote:
-------------------------------------------------------
<snip>
> If trains that are both too long for a siding have
> to meet, it is a very time consuming process, and
> something that the dispatcher will avoid at all
> costs:
<snip>

A lot of the pipe trains meets at Arboles were doubleheaders. That added a demension that one of both of the helpers could be used as switch engines in the move.