Water well cover is my guess.

Where is the piping located relative to the roundhouse and current tracks? As I no longer live in New Hampshire I can not run over to the library to check my memory, but I remember looking at some old maps showing that there was a watering and coaling facility a hundred or so yards South South West of the roundhouse. So a well head or water pipe in that area would be appropriate. There were several spurs on the West side of the main, but most traces of those have been long covered by Relyco and surrounding businesses.

Sounds like it must be a well head.  It is, indeed, located about 100 to 125 yards SSW of the roundhouse.  

So does that mean there would likely have been a track from the turntable angled off towards the watering facility?  It is way too far the main and the two passing sidings to seem like it would have connected back to the passing sidings. Or could water have been piped from the watering facility closer to the rest of the tracks? 

Thanks for the far more educated assessments of what this might have been than anything I had thought of.

-Ed

The google view below I have annotated with where the mystery item is - now presumed to potentially be a well head.  As you can see, it is far away from the rest of the tracks.

-Ed

of the roundhouse might have been a track at one time ,perhaps with a water tank? .DaveB

The 1941 northeast corner quad USGS Historic Map on the UNH site seems to show that there were more siding in the yard extending all the way over to where the pipes are. See http://docs.unh.edu/nhtopos/Dover.htm. I guess I am going to have to go back to the library the next time I visit NH. 

Dave - the road actually is a pretty steep climb, definitely not suitable to have been a track location (at least 6%, grade, possibly more).

Ken, the link to the USGS maps is fantastic. The 1956 map is exactly what I needed since I am modeling between 1956 and 1959.  The next part of the layout I will be building is Dover, which is why I took the field trip to the roundhouse first.  With the aid of the USGS map, I have altered my Dover track plan to closely reflect the map other than compression and have made room to include where the watering & coaling facility would have been. While they would not have been in use by then, having stopped steam service at that point, it provides a good opportunity for doing some modeling of it overgrown and out of use.  The map also seems to show a couple spurs west of the mainline that I had hypothesized and put in fictitiously for industry but that instead I just had to reverse the direction of to match the map (with further compression of course.)  I do have one extra pair of turnouts not on the map to enable a run-around to work the industry spurs effectively.

The out of service watering and coaling facility should make a good complement to the roundhouse which I got enough outside and inside pictures of yesterday to scratch build it with a good degree of detail and accuracy (The roundhouse will be my main structure modeling project this winter after I get the track laid.)

Here is the revised track plan (the out of use watering and coaling area will be in the open space just above the tracks to the right of the turntable.)

(ignore the spacing variations on the curves - that's my not being that good at XTrackCAD yet.)

Thanks!

-Ed

Easy to fix the spacing variation on your curves. With a single curve in place...

1. Select the curve

2. Click the parallel tracks icon

3. Set the separation (at bottom of screen), same as spacing in your yard if wide enough to work in a curve

4. Click to left or right of curve

5. Use the Join command to connect to you existing track

It helps to place the main body of the curve, place the second parallel curve and third if needed, then use the Join command to connect to existing track. Pay close attention to what the easement setting is doing when joining so that you get the easement sections to also be parallel. A few Undo's are sometimes needed to get it just right. If you can't get the easements parallel with equal spacing try joining the inner and outer curves using different easement settings - sharp on the inside curve, normal on the outer curve, etc.