It seems that many of the switches on mainlines & sidings through my area are the powered/DS controlled type. My question is this:
What are the costs involved (generally speaking) with converting or changing a switch from non-powered to powered?
Does the money spent on converting yield savings by not having to have a crewmember walk the length of a train (long term)?
What would make a railroad want to make the change either TO powered switches or FROM powered switches (short of abandonment)? I realize this may have bearing on signalling as well.
I don’t know dollar figures but it is easy to assume that the cost of a power switch is weighed by use, frequency, speed. None power would be weighed the same. Is it used once a week? day? hour? Track speed is 5mph? 10mph? 90mph? Number of trains 1 a week, three a week, 10 a day in each direction? every 10 minutes every day?
Yes it does. For my story to make sense, here is a little briefing. On the DM&IR line to Steelton from Proctor (Steelton Hill), there is a spot called Nopeming Junction. Nopeming Junction is the spot where DW&P trains would reach home rails on their way north from Pokegama Yard in Superior. DW&P trains would go to the left, while DM&IR trains would go to the right.
Nopeming Jct:
Now once the CN took over, they decided to run all NB DWP trains over the Missabe, and all SB trains down the DWP, which is actually the reason they bought the Missabe. So at Proctor Jct, which is off of the Missabe leg (duh) there is a Y. Right leg goes towards Proctor Yard, and the left leg goes onto the Missabe Sub mainline at Adolph.
Here is a shot showing the leg towards Proctor. Ore train is coming on the Missabe Sub, manifest is coming out of Proctor Jct. Missabe Sub mainline go straight past the curving manifest.:
Now the CN has built a manifest yard at Proctor, and assigned a pair of GP9RM’s with a slug to switch it out:
Now with alot of transfers being brought from Pokegama to Proctor, and some road trains straight from the WC to the DM&IR, Proctor Jct. is getting alot of use. The only problem is that the switches were hand thrown. Conductors would have to walk 100 cars back to the head end after lining a switch back normal (normal position is for the mainline). Pretty soon the CN got sick of
Although I’ve had no direct involvement with these, they don’t look too complicated - I’d guess something around $10,000 - $15,000 or so for what you see in these photos. Compare its ‘guts’ to that of any automobile, and you’ll see why I venture that uninformed opinion. . . [swg] Some research should yield a more reality-based figure.
That price range works out to from $3 to $15 per day, depending on what interest rates and time frames you use for the cost recovery calculation, so it’s likely worthwhile in many situations when the time of the crewman walking, the train and track time consumed while waiting, the time and market values of the commodities carried, and the loss of track capacity value, etc. are added up.
More significant are other variables and costs - power ot the site if not solar-powered as in the link, and most especially, the signal installations and changes to control the thing and interface it with everything around it. I’ve seen some shocking numbers cited for installing a CP at the end of a siding or for a universal cross-over, most of which is not the track - so it must be for the switch machines, controls, signals, and interfacing with the rest of the railroad’s signal and dispatching systems. Well into $6-figures is a definite possibility - against that, the cost of the power switch machine itself is minor.
Perhaps someone else with more familiarity can provide better data on this.
If you want a full-blown CTC end-of-siding installation with absolute signals operated by a dispatcher from a central office, figure $500,000 for a one-turnout end-of-siding not including approach signals, track circuits other than within the control point itself, and any significant costs for extending utility power to the location, or building any communications backbone. If you can settle for a DTMF-type switch machine that is crew-operated via the locomotive radio, figure $100,000.
. No, that is not the cost item that matters. Besides, generally you leave the switch open, at least on every railroad I have worked for that had single-track lines without spring switches without CTC without cabooses. The cost item that matters is the loss of train capacity. If you don’t mind only collecting the revenue from 20 trains a day when you have the business opportunity for 30, then don’t install power switches. (Or don’t have train meets.)
What would make them change is the desire to operate more trains than is within the capacity of a unsignalled railroad with hand-throw turnouts. If you are cash-poor and bootstrapping yourself using your own internal free cash flow, you might install DTMF, get your capacity per day up, bank some profits, then put in real CTC.
Phenomenal information RWM, thanks again. Can I presume that for a more complex “plant” the costs ramp up at a similar rate? I’m thinking dangerously close to the “Track Warrant vs CTC” thread that was up a while back…but is the “once installed it’s not coming out” thing what prevents, even the Class 1 I’m thinking of, from improving certain lines? Or is it likely that the line doesn’t see enough traffic to justify the expense because the current method of control offers enough capacity (all other things being equal) to get things over the rail without further improvement? The sub I’ve got in mind sees (to my limited observations):
-coal service (loads & empties) roughly once every week-sometimes twice in ten days,
-8 to 10 local trains (out and back to service an industry),
-4 daily ‘through freights’ (from yard X & Z to Y & back, X & Z being on an adjacent Sub and trains from them usually average 100 cars).
Finally…on a line already equipped with CTC would about the only reason to leave a switch non-powered be to serve an industry (unless it’s relatively huge)?
. No, it’s non-linear. For example, if it costs $500K for a one-turnout CP, a two-turnout will be in the range of $750K.
No, it’s just the sheer cost of CTC, and the many competing demands for scarce rail capital. Rail capital improvement projects must have a very high rate of return to compete for the company’s capital. In the example you give below, if the line is already operating successfully without CTC, and there isn’t any traffic that’s being turned down or highly likely to appear in the near future because the line lacks capacity, the ability of a CTC project to generate a high rate of return is unlikely. The incremental savings in equipment utilization and labor probably are not going to be very impressive.
[quote]
Or is it likely that the line doesn’t see enough traffic to justify the expense because the current method of control offers enough capacity (all other things being equal) to get things over the rail without further improvement? The sub I’ve got in mind sees (to my limited observations):
-coal service (loads & empties) roughly once every week-sometimes twice in ten days,
-8 to 10 local trains (out and back to service an industry),
-4 daily ‘through freights’ (from yard X & Z to Y & back, X & Z being on an adjacent Sub and trains from them usually average 100 cars).
Finally…on a line already equipped with CTC would about the only reason to leave a switch non-powered be to serve an industry (unless it’s relatively huge)?
Thanks once again. Sorry if I’m being a pain but I came up with a couple more questions.
By accelleration/decelleration track, would that be a double ended track, separate from the mainline? Who would pay for maintenance of the a/d track-the RR or the industry & would the RR be free to use the track for meets (if long enough) and it’s not being used for switching?
I see several industries of similar size & usage off of a CTC mainline in my area. Some have single headed signals with three lights (R/Y/G) facing traffic coming out of the industry and some have none. Why would there be signals at some and none at others?
Earlier you mentioned DTMF switches, why don’t those work well with CTC?
Typically yes, but if trains will only arrive/depart from one direction, sometimes they are not.
Maintenance allocation agreements vary. Typically the railroad has the right to use the track when not needed by the industry.
I’d have to look at it in the field to give a definitive answer. There are many possible reasons. Often it’s a historical artifact. An industry spur installed a long time ago when main line traffic was low may have not been required to be signaled because it didn’t create an impact on main track capacity. Later main track volumes grew, and a new industry was required to provide signaling. Another common reason is the desire to have a higher main track or controlled siding maximum authorized speed at that location. Also, it matters if a train will ever clear the main track at that location while switching the industry. If the train never clears, a leaving signal may not be required.
This is hard to explain without going into a lot of explanation of FRA regulation and how signaling technology functions. I’ll see if I
For what seems to me to be a good example and discussion of one of these on a CSX line, see the “Casky Case Study Overview” (4 pages, approx. 25 KB in size) on the public portion of the AREMA website under “Education & Training - Committee 24 - Case Studies” - http://www.arema.org/committees/c24.aspx - at:
And since RWM has now provided more reliable figures for the cost of a DTMF power switch several posts above - in the $100,000 range - I’ll note that the ‘daily cost’ figures I posted above should then be multiplied by a factor of about 10, to get from $30 to $150 per day for each such switch, to be recovered over a period of 3 to 5 years or so. That’s still pretty cheap and worthwhile to me if it saves a hour or more of train occupancy and waiting time - in the aggregate - per day, unless there’s just no other use that the train, locomotives, crew, track, or track time/ capacity could be used for in addition or instead.
(1) and some people can’t see why running through a power switch (usually in hand operation) is a capital Bozo-no-no on a railroad (ie - fireable offense)…$10K+ to repair the stripped or otherwise damaged gearing of those GRS or the other 8 or so common types.
(2 ) …and RWM didn’t explain that the signal bubbas want everything gold-plated and new[swg] and secondhand/used is almost not heard of[:-,]
PDN’s case study brings up the adding of CTC power switches and new sidings on CSX’s A&WP subdivision from Fairburn intermodal to LaGrange, Ga. That route segment approx 40 miles in length until about 3 years ago. CSX first added an appproach operated switch at the south end of Union City so all south bounds took siding and northbound had to restore north end switch to main. Shortly the same setup was installed at Hogansville siding. There was a siding in Newnan that could take a short train. To improve fluidity trains were often fleeted over this segment. A opposite direction train to fleeting could take 3 - 5 hrs+ to cover this segment.
Then CSX got the haulage contract to move 3 or more BNSF intermodals round trips / day. It appears the contract specified these delays would no longer happen in normal operaqtions. So 2 additional 10K ft sidings were added and CTC installed on the segment. Now trains almost always make this segment in !:15 or less. Fluidity greatly improved.
CSX already had code line CTC on the old ACL Lineville sub from LaGrange - Birmingham but that was gradually replaced with RCL the past 3 years along with the longer sidings required for the BNSF intermodals. Dispatcher nightmares to give priority to the BNSF trains for a while until enough sidings lengthened.
All of which begs the question of why, in a power operated switch…which like ANY other switch WILL get run through from time to time…the design of the switch machine causes the roughly $10K to repair rather than a broken $10 switch lug. When you know something is, from time to time, going to be broken, why make that occurrence as expensive as possible, rather than inexpensive as possible.
Would some kind of failsafe breakfree protection be possible with the power switch? Or would the re-engineering of the mechanical linkages/etc cost more or be to cumbersome to be practical in the field?
The simple failsafe in this instance would be to make all switches spring switches. However, the ability to derail is a failsafe in the sense that it may keep the violating train from running into another or in the face of another.
Don’t get me wrong…running through ANY switch is a big No No! Making a power switch expensive or inexpensive to repair has no effect on the fact that rules have been violated in getting the switch run through in the first place, and crews will pay the appropriate disciplinary price for the occurrence. I just find it incongruous that in this day and age of minimizing expenses that the carriers are still permitting power switch machines that cost thousands to repair to be designed and installed.
In the real world, it isn’t just the cost but also the down time in getting the repairs effected. Many times the switch that is controlled by the damaged switch machine will be out of service for several days while the necessary parts are secured and then the Signal Maintainer will have the entire track out of service for the period of time it takes him to install the new parts…all in all a very expensive and time consuming occurrence that in my mind is totally unnecessary and could be eliminated by having the part that fails when the switch is run through, designed to be easily and economically replaced.
On that line of thought…let’s say the switch would be a power AND spring switch. Normally lined through and could be powered to align for diverging but going diverging to through route would be sprung. Or would that not jive with signalling (presuming a work-around would be needed for continuity) and would the increased wear over the spring loaded frog be worse long-tem? Or would such a monstrosity not exist in the first place (as I suspect it doesn’t)?
So you’re saying having parts designed to break away when too much force is applied in the wrong direction? Something that could then have it’s broken parts unbolted or torched out and a new piece bolted or welded back in? Presumably there would be some room in the design for something like that somewhere…but surely some engineer(s) already thought that one up.
Power switches do not derail equipment when they are run through…they will only derail equipment when a reverse movement is made over the switch after it is run through without the points having been clamped in the desired direction of movement.
You also have to take into consideration how many times a power switch will be run through. Generally these are all on the mainline, or at the throat of a yard, and I would guess it is very rarely that power switches are run through. There is a reason most yard switches are floppers nowadays.
