Greene Automatic Cutoff Steam Engine

[Craig DeShong]

I’m not arguing with you here Kirk, just adding some information.  I’ve even heard of people making cylinder wall repairs in IC engines with JB weld.  I would never contemplate such a thing, but I’ve heard it has been done with success

[Jasonb]

I think the issue would be that the repair block has a 3/8" hole through it to replicate the core, could not do that easily with JBW unless you poured it around a tube. Plus the CI will give a stronger thread than the JBW even tough it can be tapped.

I'd have stuck the repair piece with JBW myself rather than loctite.

[kvom]

I considered JBW, but decided that if the loctite didn't work I could use it later.  As is, the patches seemed to be quite sturdy.  After a couple of night curing, I milled both sides leaving them proud a couple of thou, and then tapped the air supply hole 1/8-27 NPT.  I then fettled the valves removing a few thou from each side at a time until I couldn't see light between the valves and their plates.  If I can get the steam chest covers made a leak test could be done.



As an aside, I recently acquired this tool as a set from Travers, on sale.  It came with 3 bits of different sizes.  Couple of turns around a freshly drilled hole and I have no burrs and a nice tiny chamfer.

[crueby]

I've seen those deburring tools in catalogs but never tried one. Sounds like you like it.

[Jasonb]

I use my one all the time though it is a smaller one with fixed size - upto 6mm holes, so much better than twisting your wrist on the straight shank type.

[Dennis]

Well done Kirk,  the Green cylinder block is a complicated casting with some challenging machining.  I'm looking forward to seeing the final assembly of the cylinder block.
Dennis

[kvom]

To start with some preliminary fitment tests, I installed the valve plates.  With the valve sitting on it at full travel, it appears that the orientation must be as shown here.  If reversed, the inlet is still open at full travel but one slit is only partially unblocked.  Notice that one slit is closed to the edge by .01".



Valve position when fully open.



Here we see the valves and valve rods assembled.  This is quite a tricky job.  I introduce the rod through the outside hole and then through the inside hole far enough to thread the inner nut onto the rod about half way.  Then pull the rod into the center space so that the valve can be place onto the plate; the valve is then slid onto the rod.  Turn the nut all the way to the inside so that the rod extends through the valve so that the outside nut can be threaded on.  This latter operation is tricky as there is little room for fingers.  I succeeded by pinning the nut against the outer wall and starting it by turning the rod to get it started.  Then tightened with a 3/16 wrench (5-40 threads).  The SW model shows two nuts on either side of the valve.  Both inner nuts can't be tightened too much as the hole the valve allows some vertical motion of the valve itself; thus two nuts to position the valve and two jam nuts.



The two nuts on the inside cavity are jam nuts for the "slide stem".  However, as made currently, there's only 2+ threads after the nut for this attachment.  So I plan to extend the threads once this is disassembled.  I do not believe that tuning the input requires adjusting the valve rod as its travel is controlled by angles of other adjustable parts.

I decided to make the initial valve covers from lucite so as to make the motion visible.  I had a full 12" square of 1/8" clear left over from previous projects, so I cut a 4.5x3.5" piece on the vertical bandsaw. This is a bit small and 1-off to warrant breaking out the vacuum table, so I decided to try a fixture method I saw on a youtube video from CNC NYC.  Basically I covered a fixture plate with painters tape as well as the bottom of the lucite, which is covered with a plastic skin.  The tape has Loctite 4581 applied, and the taped surfaces are clamped for several minutes.  Then off to the CNC mill (this would work for manual milling as well).



Afterwards I removed the skin from the top and separated the bottom with a Exacto knife edge.



I found that the left valve is a few thou proud of the block surface, I'll need to take that side apart and take a bit off the top.

[Johnmcc69]

  very nice!
 I imagine you used an end mill for "drilling" the holes in the lucite?

 John

[crueby]

Quite a mechanism, love the plan for the clear covers to start, should make tuning a lot easier.   

[kvom]

Back in the shop for a short session.  After disassembly of the left valve rod, I extended the threading with a die for 3 extra threads and then reassembled.  Also reduced height of the valve by about .025".  Then  I found that the valve on the right side was rubbing against the lucite cover, so I did the same work there, including reducing the height by .010".

I then put on both covers, plugged the outside valve rod holes with 8-32 screws, set the valves closed, and applied 20 psi air.  No appreciable leakage could be heard or seen, so that was a relief.   

Next up were the valve stem assemblies, show here:



The valve rod screws into the valve stem, while its die block and clevis are joined with a 5-40 screw and nut.  I made these parts a good while back, and now with assembly due ran into a few problems.  The first was that I drilled and tapped the set screws on the top and bottom of the clevis, rather than the sides.  Clearly the bottom would be inaccessible.     So I drilled and tapped one side so that top and side screws will secure the clevis to its rod.  The next problem is that there is not enough room for the model-size 5-40 screw head between the clevis and the back wall of the central pocket.  So my plan is to trim the heads of the screws to fit.  Once I'm able to attach the two assemblies, the rest of the inlet mechanism should be ready to attach,

[kvom]

Quite a mechanism, love the plan for the clear covers to start, should make tuning a lot easier.   

I deleted previous post after thinking session in the shower.     Here's how I see things going.

First, adjust the eccentric rod so that the slide has equal motion on both sides.  Remembering that the dashpots should keep the valves closed, so that when the slide is in the center both are closed.  As the slide moves away from center and the trip on that side contacts the catch, the valve will move open as long as slide/trip are in contact.  Assuming no cutoff, I assume also that the furthest movement of the slide means full opening of the inlet.

Where should the piston be at full opening?  The elevation drawing from Dennis shows that in the center slide position both catches contact their trips;  but photo of the actual engine in RI shows the catches closer together than the trips.  The video of John's running engine shows them further apart.

If the piston and slide move in the same direction then full open can be at TDC.  Otherwise if in opposite directions then full open is at the center of travel as with D slide valves.

In any case, the tuning should be the same as with Corliss engines, so perhaps someone who has built a Corliss can chime in.

[kvom]

I spent the best part of the morning assembling all the inlet valve-related parts, with a but of fettling thrown in.



The good news is that everything goes together more or less easily.  The bad news if that the motion is locked up.  Since the catch housings and dashpot clevises are just attached to the cross rods by grub screws, the problem is clearly with either the valve stems or bearing alignment, or both.  Next step will be to remove all but these parts and see how we can loosen up the action, perhaops with a little oil to help along.

I do need to add a second set screw to the dashpot clevises, and I see that I assembled the valve stem clevis with the horizontal screw on the inside where access is problematic, so I'll reverse those as well.

[kvom]

Some more exploration revealed that the rear bearing carrier is too short so that when screwed down the cross shafts lock up.  If I loosen the nut and allow the bearing to float I can move the valve stem by turning the shaft.  I will use a feeler gauge next time in the shop to try to determine the gap, and then try to use shim stock to level it.

The two bearing carriers are fastened only by a nut on a threaded rod.  What keeps them parallel to each other are the two cross rods.

One observation today that was not obvious earlier is that the base of the valve stem is intended to slide on the bottom of the pocket, this preventing it from cocking and putting pressure on the valve rod.

[Craig DeShong]

It looks GREAT Kirk, it’s really coming together.  Lots of motion to see when you get it running.  You might even be tempted  to leave the clear valve chest tops in place so you can see the d-valves in motion.

[Art K]

Kirk,
I know absolutely nothing about steam and less about the Green Automatic cutoff. So I can't offer you any help, and if I did it would be suspect but you are making good progress and I have been enjoying hollowing along.
Art