How do RR measure grade?

[Boyd]

I'm wondering if the RR's use any small portable devises to measure the grade at any given location of the track?

[zugmann]

The measure the angle the sleeping conductor's head is at...

[wgnrr]

The way they recently did it at the Cass Scenic RR was taking a hand held GPS device, going up track 100 feet, and calculating the grade. The previously marked 11% grades were now found to be no greater than 9%...

Phil

[tree68]

Having used my GPS to verify grades on our railroad, I question the accuracy.  I've passed the same point several times with the GPS and have gotten different elevations each time.  For that matter, I've placed the GPS on a solid surface, with excellent satellite coverage, and watched the elevation change as it sat there.

As for the original question - very easy to do.  Set up a surveyers transit, make sure it's level, then shoot a point some distance away (100' makes for easy calculations).  Using the standard surveyors stick, it's easy to determine the rise or fall (ie, grade).

You could do the same thing yourself using a common carpenter's level (2' or more), your camera tripod or other suitable stand, and an 8' stick or board.

Set up the level so it's level and pointing at the second point you want to measure.  You'll need to find out how far it is from the ground to the top of the level, and mark that distance from the bottom of your 8' stick/board.

Measure out 100' and stand the stick/board there.  With one person at the level and the other at the stick, sight across the top of the level and have the person at the stick mark that point on the stick.  There will be some calling of "up," "down," etc, but it wont take too long.

Measure the distance between the two marks.  One foot will equal 1%, six inches will be 1/2%, two feet will be 2%, etc.

If you want to do the calculations and have a transit that measures vertical angles, you could simply measure the angle between points and figure it out that way, too.

Given a 100 foot chain and a transit which also measures horizontal angles, a crew can determine grade and curvature quite quickly for a pretty good distance.

As for the Cass - I think I'd trust the transit more than I'd trust the GPS.  Given the margin of error I usually see displayed on my GPS, it's not unthinkable that they could go 100' upgrade and discover they're going downhill....

[ndbprr]

Since most tracks have been down since before GPS I would assume that the surveyers who laid out the route also figured the grade with a transit.  Having helped layout a steel mill line over 3/4 of a mile long the accuracy astounded me.  You can hold an 1/8 of an inch over that distance.  In the case of elevation simple trigonometry will allow you to calculate the rise by knowing the distance and the angle of it.  I would also assume that some grades have varying degrees of pitch and are not consistent over the entire grade.

[Modelcar]

....I would imagine most grades were intended to be a constant grade for a given distance and not weaving up and down on it's way.  That is if it was not built on the cheap.  Of course the grades are compensated on curves {lessened}, to allow constant drag on certain up grades.

I too have always been amazed {and then some}, the accuracy of surveying over long distances and also in extreme circumstances such as building tunnels {and digging from both ends, etc...}, and meeting in the process, perhaps near the center with accuracy enough to be just inches off of exact meeting.

 The Chunnel in England's English Channel crossing comes to mind.

But to go beyond that, much of the infrastructure {RR}, was done under less than good conditions over a hundred years ago.  Which adds to it as well....!

I agree with Larry re: GPS signals can wonder.....Have seen it right here in my drive way as the instrument was stationary, figures moved up and down.  But {in a built in one}, it sure did show the vehicle was pulled right into my driveway...!

[Railway Man]

I'm wondering if the RR's use any small portable devises to measure the grade at any given location of the track?

Other than standard surveying instruments?  No.  It would be pointless.  If you want to know the grade you look at the track chart.  If you need greater accuracy you call a surveyor.

RWM

[dehusman]

Unless you have a surveying grade quality device (several thousands of dollars) using a GPS device will not give you accurate enough measurements to calculate grade.  Its only good to plus or minus several feet.  So your 9% measured grade is somewhere between 12% and 6%. And that assumes that the same offset of error at both ends.  If the down end measured high and the up end measured low the difference could be even more.  Since the grade was probably measured by traditional surveying methods to begin with, I would go with them.  they would be accurate to inches, whereas handheld GPS is only accurate to feet.

Over long distances (miles) the couple feet error isn't as big a factor.  Railroads use traditional surveying methods, high accuracy GPS surveying or low accuracy GPS over longer distances to calculate grade.

Dave H.

[Modelcar]

.....GPS is a very important instrument in position location in emergencys.  A few years ago over in our home location of Pennsylvania the location where to drill a rescue bore to get to trapped coal miners....{down 240'}, was very much assisted by GPS as time was of essense.  The bore broke thru the mine cavity right where it needed to and a rescue was accomplished.  One might remember the famous line from that resue...."9 for 9"....meaning all envolved were rescued alive.  That was the Quecreek, Pa. location.

[Railway Man]

.....GPS is a very important instrument in position location in emergencys.  A few years ago over in our home location of Pennsylvania the location where to drill a rescue bore to get to trapped coal miners....{down 240'}, was very much assisted by GPS as time was of essense.  The bore broke thru the mine cavity right where it needed to and a rescue was accomplished.  One might remember the famous line from that resue...."9 for 9"....meaning all envolved were rescued alive.  That was the Quecreek, Pa. location.

Quentin, GPS is useless in mine surveying because there is no reference point inside the mine because there is no radio signal inside the mine.  Mine surveying uses traditional techniques. 

RWM

[anb740]

The measure the angle the sleeping conductor's head is at...

      ....Good one!

[Ted Marshall]

.....GPS is a very important instrument in position location in emergencys.  A few years ago over in our home location of Pennsylvania the location where to drill a rescue bore to get to trapped coal miners....{down 240'}, was very much assisted by GPS as time was of essense.  The bore broke thru the mine cavity right where it needed to and a rescue was accomplished.  One might remember the famous line from that resue...."9 for 9"....meaning all envolved were rescued alive.  That was the Quecreek, Pa. location.

Quentin, GPS is useless in mine surveying because there is no reference point inside the mine because there is no radio signal inside the mine.  Mine surveying uses traditional techniques. 

RWM

He meant using GPS to know where to drill the rescue hole from the surface down, not for mine surveying. Try reading the entire post next time.

[Modelcar]

.....RWM:

Reread my comments.....GPS was used to position the drill to do the 30" bore down to the mine cavity.  I didn't say it was used in the mine.  It was used to find the location the miners were thought to be, and.....they did correctly hit that cavity the men were trapped in.

[diningcar]

Railroad grades were (are) established by locating engineers with criteria given them by the RR's Chief Engineer who is also given criteria (keep construction cost down - ruling grade - etc.). In the 19th century keeping initial construction costs low was a dominate theme and thus many changes to the original construction have occurred over the years. 

When Santa Fe elected to relocate 44 miles of the TRANSCON between Williams and Crookton Arizona in 1959-1960 the gradient criteria chosen was ' no grade at any point will be greater than 1% (compensated) '. And at the location where a minus 1% gradient changed to a plus 0.88% a 10,000 foot vertical curve was created for the purpose of minimizing slack action. This kind of 20th century engineering did not have a ' low cost ' theme and was consistant with the continual upgrading of the TRANSCON. It is one illustraion why the TRANSCON is so heavily utilized today. 

Zeiss level instruments and a level rod were the tools used at Williams, not a transit which was used solely for alinement.

 

[Railway Man]

.....RWM:

Reread my comments.....GPS was used to position the drill to do the 30" bore down to the mine cavity.  I didn't say it was used in the mine.  It was used to find the location the miners were thought to be, and.....they did correctly hit that cavity the men were trapped in.

I read your post both times, Quentin.  I think we're talking around each other.  While GPS is useful to find an underground location that is already mapped, it's of no good otherwise in underground mine surveying.  Having a mine map in hand, you can use GPS to reference yourself to a surface location, but I'd back that up with traditional surveying if it was my reputation and money on the line.

(So far this year I've spent more than $250,000 buying surveying services ... and to quote my friend Mudchicken there's darn few surveyors that have a clue around railroads.  I'd rather send a railroad-knowledgable surveyor 1,000 miles to the project site than hire a local who "might" know what they're doing.)

RWM

[mudchicken]

The way they recently did it at the Cass Scenic RR was taking a hand held GPS device, going up track 100 feet, and calculating the grade. The previously marked 11% grades were now found to be no greater than 9%...

Phil

Well, now we know where the idiots reside. 

[Ted Marshall]

The way they recently did it at the Cass Scenic RR was taking a hand held GPS device, going up track 100 feet, and calculating the grade. The previously marked 11% grades were now found to be no greater than 9%...

Phil

Well, now we know where the idiots reside. 

West By God Virginia

[Modelcar]

....RWM:

Fair enough.  With maps in hand and the GPS to get the surface position located {quickly}, it wasn't money that was on the line....it was human lives....There wasn't time to do all the land surveying you mention.  So GPS was employed and as I stated earlier, it hit the pocket cavity the men were trapped in dead on.  They started two rescue bores and one had trouble with breaking equipment {drilling bits, etc...}, and the other was as I've been saying....successful.  All 9 men came out alive in a rescue cylinder....!  Of course, one at a time.

Edit:  A smaller bore was sunk first to validate the location.  Then the serious bore size needed was put down....No easy task...around the clock by super human efforts.

[mudchicken]

Since most tracks have been down since before GPS I would assume that the surveyers who laid out the route also figured the grade with a transit.  Having helped layout a steel mill line over 3/4 of a mile long the accuracy astounded me.  You can hold an 1/8 of an inch over that distance.  In the case of elevation simple trigonometry will allow you to calculate the rise by knowing the distance and the angle of it.  I would also assume that some grades have varying degrees of pitch and are not consistent over the entire grade.

Level (wye and-or dumpy/tilting dumpy) and babbit faced steel chain - no transit. Most 1840's-1880's engineer's transits did not even have a vertical circle. A Parkhurst theodolite would have been impossible for railroad crews to move around with any economy. The 1873 Gurley Engineer's Transit in my living room does not have one. Even with today's electronic instruments, trig-levelling normally is not used (and we often use conventional levelling techniques with second order levels to find errors in local GPS networks or ignore GPS vertical readout entirely). Optics and collimation issues can eat your lunch.  Those old railroad mudchickens did their job by repeating measurements and level loops  over and over again over the same piece of ground.

With GPS, your timing signal can be interfered with in so many ways (multi-path, etc.), especially in the vertical. The comparison of vectors can throw you way off. "P" codes can be interfered with. If you have no concept of how the equipment you are using works and what your limitations are, you are what most surveyors refer to as a "button-pusher" with zero technical credibility..

[wgnrr]

The way they recently did it at the Cass Scenic RR was taking a hand held GPS device, going up track 100 feet, and calculating the grade. The previously marked 11% grades were now found to be no greater than 9%...

Phil

Well, now we know where the idiots reside. 

 

Well, that's what the loggers thought the grade was, and who was going to challenge the loggers' reliability?? Turned out, the local historian, who back in the 60's told everyone their grades were wrong...

Phil

[erikem]

With GPS, your timing signal can be interfered with in so many ways (multi-path, etc.), especially in the vertical. The comparison of vectors can throw you way off. "P" codes can be interfered with. If you have no concept of how the equipment you are using works and what your limitations are, you are what most surveyors refer to as a "button-pusher" with zero technical credibility..

I would expect the worst sources of error are due variations in the ionosphere (which causes variation in the speed of the radio waves) and terrain blocking some of the satellites (the more birds you can see, the better the confidence in the measurement). As for measuring differences in elevation, results using a surveyor grade differential GPS should be pretty good as the elevation measurement should be least affected by the ionosphere (the key is that the measurements are made simultaneously, so the positional errors for both points will cancel out - assuming that both receivers are seeing the same birds).

As others have pointed out, the reported position of a GPS receiver does wander over time, although accuracy does improve with averaging. 

[tree68]

Disclaimer - I used the term transit because that's about all I know in sighting instruments.  I'll defer to the experts.

[timz]

Of course the grades are compensated on curves {lessened}, to allow constant drag on certain up grades.

Except when they're not.

Somebody said new construction nowadays doesn't bother with curve compensation, which sounds unlikely. He even claimed the UP had removed the compensation from the WP line up the Feather River, which as everyone knows was built to not exceed 1.0% compensated. That sounds unlikelier.

[timz]

You've all seen pics of old-time survey crews

 http://www.co.yamhill.or.us/pics/Places%20of%20Interest/RRsurveycrew.jpg

showing off their level, on our left, and the transit on our right. Nowadays transits can measure grade to 0.01% with no problem-- but they used the level to set the grade and the transit to stake out the curves and tangents.

[diningcar]

The accepted compensation is .04 per degree of curviture. So a five degree curve would be compensated by .2 %, making a 1% grade 0.8%

Perhaps new construction would not have curves of five degrees, but one or two at the most. Then compensation may not be considered necessary. Also, with rail lubrication and modern wheel design perhaps the .04 is out of date. 

[Ted Marshall]

While were sharing photos of surveyors, here are some I took one day at work.

The first is of another survey crew of a different company, doing what survey crews do best:

http://i200.photobucket.com/albums/aa317/TedMarshall/IMG_0565.jpg

The next two I took through the lens of my total station (transit) of my colleagues working at a different control point:

http://i200.photobucket.com/albums/aa317/TedMarshall/IMG_0571.jpg

http://i200.photobucket.com/albums/aa317/TedMarshall/IMG_0578.jpg

This one is of my instrument man taking reference to a backsight target.

http://i200.photobucket.com/albums/aa317/TedMarshall/IMG_0575.jpg

We were doing cross-section of the roadway at 300' intervals along Florida's Turnpike using our Topcon GPT-8000 total station to determine the N,E,Z position of each shot taken at every grade break perpendicular to the roadway which included soundings in canals on each side . The elevations of the control points occupied were established by differential leveling using a Leica (Wild) NA2 32x auto level. We started from a F.D.O.T. benchmark and ran through each control point by balancing our turns (150' each direction) between them until we closed to another F.D.O.T. benchmark, one mile away. IIRC, we closed our bench run to within 0.015' which complied with MTS (Minimum Technical Standards) for second order vertical work. All elevations were based on NGVD 29 vertical datum.

 

 

 

[ChuckCobleigh]

Since most tracks have been down since before GPS I would assume that the surveyers who laid out the route also figured the grade with a transit.  Having helped layout a steel mill line over 3/4 of a mile long the accuracy astounded me.  You can hold an 1/8 of an inch over that distance.  In the case of elevation simple trigonometry will allow you to calculate the rise by knowing the distance and the angle of it.  I would also assume that some grades have varying degrees of pitch and are not consistent over the entire grade.

Level (wye and-or dumpy/tilting dumpy) and babbit faced steel chain - no transit. Most 1840's-1880's engineer's transits did not even have a vertical circle. A Parkhurst theodolite would have been impossible for railroad crews to move around with any economy. The 1873 Gurley Engineer's Transit in my living room does not have one. Even with today's electronic instruments, trig-levelling normally is not used (and we often use conventional levelling techniques with second order levels to find errors in local GPS networks or ignore GPS vertical readout entirely). Optics and collimation issues can eat your lunch.  Those old railroad mudchickens did their job by repeating measurements and level loops  over and over again over the same piece of ground.

With GPS, your timing signal can be interfered with in so many ways (multi-path, etc.), especially in the vertical. The comparison of vectors can throw you way off. "P" codes can be interfered with. If you have no concept of how the equipment you are using works and what your limitations are, you are what most surveyors refer to as a "button-pusher" with zero technical credibility..

GPS is not nearly so accurate in elevation as it can be in the horizontal plane.  This is generally because of the low slant angle from most of the satellites to the receiver.  When the constellation has several high-angle payloads, your chances of better accuracy improve, but still, it's going to have an uncertainty of many feet.

It would seem to me that the most reliable measure would still be with a good surveyor's level, a stick and a fit rodman.  Sort of going back to the KISS approach to engineering, I guess. 

[ChuckCobleigh]

While were sharing photos of surveyors, here are some I took one day at work.

The first is of another survey crew of a different company, doing what survey crews do best:

http://i200.photobucket.com/albums/aa317/TedMarshall/IMG_0565.jpg

The measure the angle the sleeping conductor's head is at...

Did I see a candidate for conductor class there?

[Modelcar]

....Timz:

That's quite a photo.  What an environment to be working in....and look at the accommodations for the winter night.

As I've mentioned before, these fellows worked under less than good conditions to seek and plot the routes.  How on earth did they get it done.   And can one think of wearing vests and suit coats in such a setting....Don't see any parkas and good fur hats and all the rest of the good stuff people would have available to use today.  Not too many radios in sight either.....??

[selector]

The accepted compensation is .04 per degree of curviture. So a five degree curve would be compensated by .2 %, making a 1% grade 0.8%

Perhaps new construction would not have curves of five degrees, but one or two at the most. Then compensation may not be considered necessary. Also, with rail lubrication and modern wheel design perhaps the .04 is out of date. 

I don't follow you here.   My understanding is that a grade compensation factor is additive, thus making the grade effectively steeper than it would have been had it been linear.  As you have described the method above, the 1% grade is reduced in severity by having a curve in it...not so.  In your example above, the resultant grade would be 1% plus the 0.2%, making it effectively 1.2%, and thus requiring a greater effort to raise a train of given tonnage and rolling resistance.

Perhaps I am not seeing something correctly?  I'm out of my area here.

-Crandell