Home Up Flue Failure Lagging Test Engine Stand 844 Photos Automatic Air Brakes Steam Air Pump

 

Cylinder Sizing

Machining a 1.6" scale Cross Compound Air Pump (Compressor)

 

Updates:

12/04/2004 Page started
12/05/2004 Added second lathe method (here)
12/06/2004 Added milling machine approach (here) and moved the page to a more permanent location
   

 

The project:

This page deals with some discussion we have been having on the LiveSteamers email list about the best way to machine a set of 1.6" scale Cross-Compound pump castings I have.

 

A prototype Westinghouse Cross Compound Air Pump

 

I have been thinking about how I will approach machining the castings I have for a 1.5" scale cross-compound air pump.

The main challenge is to machine the three main castings such that the upper and lower cylinder bores, as well as the corresponding piston rod glands are all concentric and parallel within a high tolerance.

Photo 1: The three main castings (Castings from Ulin Locomotive Works)

 

Photo 2: Close-up of the spacer casting

 

On a simplex pump (only one set of cylinders) this problem is somewhat easier because you can machine close-fitting bosses on the face of the spacer casting to register the cylinder accurately to the piston rod gland.  However, in the cross-compound, not only would the bosses have to be EXACTLY the same distance apart but they would both have to be exactly concentric to their respective piston rod glands.  I don't see a way to accomplish this on opposite faces of the spacer casting to a sub-0.001 tolerance.

There is also not a lot of "meat" on the face of the spacer casting to machine much of a boss and still have sufficient thickness remaining.

 

My current thoughts are to proceed as follows:

1 Mill (fly cut) the bottom (side that mates to the spacer casting) of the steam cylinders (upper casting) and bore the larger (low pressure) steam cylinder in the same milling setup.  This guarantees that the cylinder is 90 degrees to the lower face of the casting
2 Flip over the upper casting and fly cut the upper surface parallel with the lower surface
3 Fly cut the top (side that mates to the spacer casting) of the air cylinders (lower casting) and bore the larger (low pressure) air cylinder in the same milling setup.  This guarantees that the cylinder is 90 degrees to the upper face of the casting
4 Flip over the lower casting and fly cut the lower surface parallel with the upper surface
5 Fly cut the upper and lower surfaces of the spacer block (center casting) flat and parallel
6 Bolt the sections together, adding two locating pins between the spacer block and each set of castings to be absolutely sure that they always go together in the same relationship
7 Mount a suitable piece of stock in the lathe and make an expanding mandrel with a drawbar extending thru headstock.  It must be a tight fit to one of the larger bored cylinders.
8 Without disturbing the mandrel, slide the large cylinder onto the mandrel and tighten.  The centerline of the mandrel is now parallel and concentric with the axis of the large cylinder.
9 Bore the opposite (smaller) cylinder using a very low RPM due to the imbalance of the castings.
10 Without disturbing the mandrel or the cylinder block and spacer mounted on it, unbolt the outer cylinder casting from the spacer block
11 Drill, bore and ream the rod hole thru the spacer section glands (again at low RPM)
12 The axis of the two cylinders and rod glands are now concentric to a high degree of accuracy determined primarily by the fit of the mandrel in the large cylinder and the accuracy of the lathe
13 Repeat steps 7 thru 12 using the second large cylinder (this assumes that the second large cylinder is a different size than the first.  If not, skip step 7)

 

 

(added 12/05/2004)

After some discussion on the LS list I came up with a second method which eliminates the need for the mandrel and is somewhat simpler:

1 Mill (fly cut) both mating surfaces of the three castings flat and parallel
2 Bolt the sections together, adding two locating pins between the spacer block and each set of castings to be absolutely sure that they always go together in the same relationship
3 Mount the assembled castings on the lathe faceplate centering it on one of the cylinder bores.
4 Bore the cylinder on the outer end (all turning operations are done using a very low RPM due to the imbalance of the castings)
5 Remove the outer cylinder
6 Drill, bore and ream the rod hole thru the spacer section glands
7 Remove the spacer casting
8 Bore the cylinder on the inner end
9 Repeat steps 2 thru 8 centering on the second cylinder set

 

(added 12/06/2004)

After some discussion on the LS list I realize that I could accomplish getting the cylinders concentric on the mill.  This had several advantages including not having to rotate an out-of-balance casting.  It also avoids any tendency on the lathe to turn a taper (not a large factor, but there).

1 Mill (fly cut) both mating surfaces of the three castings flat and parallel
2 Rough drill the cylinder bores in the upper and lower castings
3 Drill bolt holes and holes for locating dowels in the upper and lower faces of the castings (see note below)
4 Bolt the sections together, using two dowel pins between the spacer block and each set of castings to be absolutely sure that they always go together in the same relationship
5 Mount the assembled castings vertically on the milling machine, locate the spindle over the centerline of one of the cylinder bores and lock the table.
6 Bore the cylinder in the upper casting (see note below)
7 Remove the upper cylinder casting keeping the table locked
8 Drill, bore and ream the rod hole thru the spacer section glands
9 Remove the spacer casting keeping the table locked
10 Bore the cylinder on the lower casting
11 Reassemble the castings and repeat steps 4 thru 10 centering on the second cylinder set

Notes:

bulletI'm thinking of doing all boring operations from the bottom up.  This way chips collecting in the bottom of the bore don't crowd the cutter.
bulletSince I have a CNC mill I can, in effect, "match drill" the bolt holes in the mating surfaces by mirroring the CAD drawing and G-code and locating the casting by using the dowel pin locations on each face as references. Since this is being done prior to critical alignment operations I just need to get this orientation (drill pattern to casting face) close, not within 0.0001's. All the holes remain in perfect alignment because they originated from the same drawing.
 

 

 

 

 

Hit Counter