It Looks So Easy and Straightforward ON Paper But...

[carrdo]

Hi All,

Moving on.

The partially made valve handles shown are not Superscale but I liked the look so I decided to make up something in that style. They were "eyeballed" from the Superscale photo posted early in the thread.

Construction wise, they were first roughed out on the bandsaw and then contoured on the belt sander. Just careful work but it left a lot of grit to clean up.

 

[carrdo]

Hi All,

Moving on with the valve layout.

Initially, when I did this to the LBSC drawing dimensions, it didn't look anything like what was shown so after some head scratching I decided to make a revision. Some of the dimensions given do not seem to be correct but we shall see.

My rough sketch, with the revised dimensions, are seen in the first photo along with the layout tools used. The odd shaped scrap piece of bronze first had to be machined along two sides at right angles to provide the needed reference straight edges for the layout.

I often use a draughtsman's plastic circle template for small laying out and the toolmakers clamp is seen as one needs three hands to align the circle template to the scribed centreline of the part and I only have two. Also, use good lighting and an eye loupe for this type of fiddly work.

This is another part which is going to be more difficult to machine than what is shown on paper.

 

[carrdo]

Hi All,

The machining of the valve. I knew this part, though it looked so simple on paper, was going to present me with a lot of machining difficulties. It had two internal features, a 7/64" through square hole and a flat bottom circular hole. The flat bottom hole I can deal with as, in the past, I have needed to made up a number of flat bottom drills which, can be a problem to produce in their own right if one does not have the right grinding equipment but fortunately I do so...

To have a chance with the 7/64" square hole, a new single lip square broach needed to be made. This size was smaller than any which had been made before and the start was not good as in the first go around I snapped off the pilot section of the broach while trying to machine in the undercut.

After that I reasoned why not try and machine a two piece broach as it would be much easier and simpler if it was done this way. And it worked! See the first photo. The undercut could now be simply formed through the use of a number 0 or 00 centre drill as seen in the second photo. The centre drill was drilled into the square end of the broach until the 60 degree centre hole just met the edge along the four sides of the square. No back relief was put on the square cutting end. The material was then hardened in oil but not tempered. The pilot piece was left soft as machined.

The broach also uses a head steady as seen in the second photo for reasons which will become clear. The third photo shows the square end test piece which I machined to exactly .1093"(7/64"). I use a spin index in the mill and sharp cutters to produce all of the square sections which I think is quite straightforward to do so no photos.

The setup for the actual broaching operation is seen in the next three photos. The bench arbor press may seem to be standard and nothing unusual but it has a number of features which make it rather unique together with all of the special fixtures made for it over the years so maybe I will describe all of this in a separate thread.

One not often seen feature is the ram has a reamed 5/8" dia. centre hole which is very useful when attempting a broaching operation like this as the broach head steady fits precisely in this hole and as a result one is able to precisely hold, locate and turn the broach to whatever position is needed quickly and accurately. I found out the hard way that small commercial multi tooth broaches are very delicate when pressing them and the slightest misalignment or arbor press ram slippage/wobble will result in a broken tool - hence the head steady. Also, not seen but the ram has been shimmed to eliminate any and all side play.

Another feature is the arbor press base had been drilled (square to the ram and how to set it up this way is another interesting operation) when I got it so special locating/holding fixtures could be used. One of these is being employed here in the broaching operation where there is a slot and centered hole in the fixture (which was designed for another entirely different operation) giving the broach clearance when it passes through the work.

The actual broaching technique needs a few words. It is best if the arbor press lever arm can be horizontal or near horizontal when broaching but things work out that this is rarely the case with these small cheap bench units (not being of the ratchet type) but clever people have figured out a modification to overcome this deficiency which involves a ram modification which I have not yet done.

Do not try and press through at one go. Carefully align the broach first in its required orientation and press slowly into the work ensuring that the pilot end of the broach slides down into the pilot hole in the workpiece without hanging up anywhere on any edge. Press firmly and release, pressing through the workpiece in small incremental stages. After the initial pressing, the pilot can be removed and the broach used without it as it will stay aligned and the open end of the broach allows for the displaced metal to more easily move into. When through gently tap out the broach with a non metallic small hammer as without any back relief the workpiece and broach are now firmly set together. As I have said before this type of broach has its limitations one being the metal thickness which limits it to a maximum of 1/8" thick.

The result is what is seen in the last two photos, a perfect square hole. As an aside, the pilot stem (being so slender in this case) did not survive but it is quite easily remade.

 

[carrdo]

Hi All,

Just for completeness, there are other ways of "broaching" miniature square holes. On Youtube:

1- Clipspring Spare Parts # 13,
2- Joe Pie Mini Milling Machine The Three Ball Crank Arm,
3- Joe Pie Can you Grind Tools on a Milling Machine? YES--Take a Look!,
4- Course You Can How to Make a Square Broach to Cut a Square Hole.

On some of these videos I would use the term "Square Hole Slotting" instead of broaching although a square hole single tooth broach is being used as a slotting tool. I am not saying anything at all negative about these methods as I have used them in exactly the same way to produce true miniature rectangular slots.

 

[Dabbler]

All of the OP's images before post#43 are gone.

 

[carrdo]

Hi Dabbler,

Yes, I know but I didn't remove them. They are gone for me also. Apparently, it is a server update issue now being looked into.

 

[carrdo]

Hi All,

The valve flat bottom drilling and depth check.

Do not try and drill directly with the flat bottom drill as a drill ground this way will jump all over the place. A flat bottom drill is just used to produce a flat surface on the end of a drilled hole.

So, the sequence of operations was centre drill followed by a 5/32" dia. regular jobbers drill followed by a 5/32"dia. end mill and finally the 5/32" dia. flat bottom drill. Depth had to be closely controlled so a small plug gauge in conjunction with the micrometer dial on the quill of the mill were used in combination. Preparation is everything so...

 

[carrdo]

Hi All,

Finishing a valve by contouring the outer (peanut) shape. LBSC says to do this by filing but I couldn't do it this way if my life depended upon it!

Precision filing is a lost art for me and even using a die filer, I couldn't do it this way either as the part, at this stage, is so small and next to impossible to hold. I guess that CNC could make short work of such an operation but since I don't have CNC and the part is already at its final thickness so how would one readily hold it?

Anyway, I had to sweat over this and think about it for a long time.

In the end, the part was first roughed down/cut out from the stock piece by bandsawing and milling but I left two short parallel ends so I could do the milling by holding the rough valve contour in the vise of the mill. Fortunately, the inner curve on the peanut was the radius of a 3/4" dia. end mill so that part of the finished outer contour was formed first as shown in the first photo. Then, by very careful incremental belt sanding all of the other external curves were formed and blended together. It involved a few tense moments as I expected that, at any minute, the abrasive belt would snatch and destroy the valve as there is (needs to be) a slight gap between the belt and the table and I could only hold/locate and move the part very lightly with the end of my fingers. Larger parts are no problem but I have had small parts like this rocket away into destruction but this time...

 

[carrdo]

Hi All,

The valve inner assemblies (consisting of the the valve arm, the valve and the backing spring) are now complete.

The last parts involved some interesting machining.

The valve backing springs are specified to be 5/32" OD and only 5/64" long! There is no chance of being able to commercially locate anything like that locally so I hunted through my spring collection and came across a tiny spring cannibalized from a fountain pen. It would do the job but it had to be cut down extensively lengthwise . See the first photo of how I do this and as well square the ends of tiny springs. If you are going to make any parts like these correctly, every step has to be controlled. The spigot on the end of the mandrel on which the spring fits is made a light push fit on the ID of the spring. Use a light touch when grinding as you don't want to temper the spring.

To make the valves themselves, I used a piece of rectangular scrap bronze set up in the 4 jaw as shown to machine the spigot end. Then, the part was carefully sawn away from the parent material, reversed, and set up again in a collet to finish the other end. You have to realize what you are dealing with here. The valve spigot is only 1/32" long and the entire valve is only 1/16" thick. I didn't know if the collet would hang on to the 1/32" long spigot or if it would run true but it did! (this is the value of having first rate collets and keeping them that way). Here, one needs to take VERY light cuts and have razor sharp cutting tools. Everything proceeded without incident.

As a final operation, the 1/8" dia. by 1/32" deep centre recess in the end of the valve was created using a sharp centre drill followed by a new 1/8" dia. end mill as dimensions here are not critical.

 

[carrdo]

Hi All,

Machining the stainless steel valve spindles. The stainless which I have was not marked but it was not 303 (extremely machinable) or 316 (difficult to machine) and so likely it was 304.

I will only show the machining of the two squares on the valve spindles which is the most interesting and tense operation as the other operations are straight forward lathe work. But everything needs to be dead concentric on these spindles so all of the lathe operations were done using collets.

One spindle square is for the valve operating handle (the smaller one which is 3/32") and the larger square (which mates with the valve arm is 7/64"). Both have to be machined accurately on size.

One of the largest problems here was to gain enough spindle clearance over the spin index division plate as one needs to machine right up to the holding collet to minimize part overhang and to maximize part rigidity as these parts are long and thin and have little inherent rigidity. In this case I chose to use an extra long new carbide four flute end mill to machine these stainless steel parts. Take light milling passes with not more than 0.005" DOC and check the indexing pin is in the correct hole before each pass as it is easy to index the wrong pin hole.

When approaching finished size, reduce the DOC and check, check, check the square width. I had to use my dial vernier calipers on the Inner larger square next to the collet as the square slots produced by the end mill were too narrow for a regular micrometer spindle. Even so, everything came better than 0.001" to the nominal size but it was very exhausting tense work.

Again, the spin index/5C collets used were far from perfect (offshore brands) so so certain upgrades (not seen in the photo) have been added. One side of the cast spin index base has been milled dead parallel to the spin index spindle axis and two hardened, closely fitting, tee slot round head plugs were used to quickly align the spin index in the mill. Also, there is another hold down clamp not seen in the photo.

 

[jcdammeyer]

Again, the spin index/5C collets used were far from perfect (offshore brands) so so certain upgrades (not seen in the photo) have been added. One side of the cast spin index base has been milled dead parallel to the spin index spindle axis and two hardened, closely fitting, tee slot round head plugs were used to quickly align the spin index in the mill. Also, there is another hold down clamp not seen in the photo.

I have found the same issues with my 5C indexer. In the past I've put a support under the extended end to prevent it from deflecting.

 

[carrdo]

Hi All,

As I forgot to show the LBSC article of the valve spindles, here it is. Small and fiddly.

 

[carrdo]

Hi All,

Further work on the boiler blowdown valve handles.

In this regard a new single 3/32" square single tooth broach needed to be made. As usual, 01 was the material used and after forming it was heated to a bright orange and then quenched in oil. I have described this before so no photos.

The attached photos show the latest work performed; basically a slimming down of the handle profile through the milling operation shown, the end drilling of the handles and then finishing them by careful belt sanding. I tried to get closer to the actual curved outline of the original Okadee handles from the photo which I have of them and shown earlier in this thread.

The plug gauge shown in the photos is used for end rounding both handle ends in the rotary table if I go that far.

 

[carrdo]

Hi All,

Proceeding with a tricky layout that of getting the 3/32" square opening in the curved handle for the valve spindle rotated 30 degrees. See the first photo.

I really only needed to determine the orientation of this rotated centreline (photo 280) but for interest I have included the setup photos of how to produce all three centrelines (horizontal, vertical and rotated) as shown in the LBSC article.

The first centreline was the longitudinal one joining the two end holes (photos 276, 277) in the handle which corresponds to the horizontal centre line on the straight handle shown on the LBSC drawing. The hole in one end of the handle is 3/32" and at the other end is the #48 drilled hole. The first and only operation was to make up two 1/8" diameter close fitting stepped plugs for each hole. The part with the plugs inserted was then set in between the jaws of the milling machine vise as seen in the photos for all subsequent layout operations. As I know that the vise jaw top surfaces are dead level/aligned with each other and dead parallel to the top surface of the vise (as they have been surface ground that way), I can use my vise as a type of surface plate for any small part layout work.

The various centrelines were then generated through the setup's shown in the photos. In photo 280, the 30 degree rotation centre line is scribed using a 60 degree plastic draughtsman's square set as shown.

The only thing left to worry about now is there is a LH and a RH valve which means that the second square handle opening may need to be 30 degrees rotated behind and not forward of the valve handle vertical centreline. I really don't know until I form the first rotated square opening, fit this handle to a spindle/valve and assemble a complete valve to see which way everything needs to be operated/oriented. I have always had trouble with LH/RH parts so proceed cautiously.

 

[carrdo]

Hi All,

Proceeding with broaching the 3/32" square in the mild steel handle. In steel, the single tooth broach is really a one trick pony as you will see.

The first two photos (281,282) show the setup used. I have described this type of operation before in this thread but a couple of additional notes. Since the handle square is rotated 30 degrees one needs to have excellent lighting and the support busing needs to be quite stout as well as having as small an opening as possible in the support bushing to give the maximum amount of part metal surround. Because with steel, the broaching force is so high, such a "thin" part part can be swaged, deformed or show "dig in" marks at the circular end where the part is being broached due to the limited amount of part metal in this area. The part must also be supported at both ends and be set dead square and flat as shown.

As stated previously, the setup should be such that broaching is carried out with the arbor press handle as near as possible to horizontal to give ease of leverage when broaching.

And initially, the pilot in the broach should be used to locate/align the part precisely under the broach/ram but when the actual broaching operation is started the pilot is removed or a bent/destroyed pilot will result as seen in photo 283. This is necessary as with the pilot in place, there is no space for the sheared metal to move into as broaching proceeds.

And, in steel, expect to destroy the broach in the process (photo 284) as this definitely is a one trick pony when broaching mild steel.

After broaching, the broach will be stuck fast to the part as I do not have any back relief on the broach. It can be hand wiggled and tapped apart with a non metallic hammer but then the broach is very carefully tapped back in the now square hole many times from both sides to give a bit of additional clearance.

The result was a perfect square but much preparatory work and experience is required.

I can now assemble one complete valve.

 

[carrdo]

Hi All,

One complete valve assembly.

 

[carrdo]

Hi All,

One completely assembled boiler blowdown valve.

But we are not quite out of the woods yet as the operating valve stem has a "packing gland" to make the valve spindle pressure and steam tight. And herein lies a small issue as I would like to use a compressible EPDM O-ring seal here (EPDM is really the only suitable material for this combination of steam, temperature and pressure). However, the dimensions needed (a 1/8" dia. valve stem in a 3/16" diameter recess) are "unobtainum" in imperial O-ring sizes and for the nearest size (1/8" ID by 1/4" OD) the ring wall is far too thick and oversize.

If one goes metric a 3 mm ID by 5 mm OD O- ring (which is also very difficult to find), we are close having only a thin 1mm wide wall thickness (I have ordered some - the shipping was 10x the cost of the O-rings!) but what one will still have to do is make up some simple holding fixtures to hold these O-rings while I "machine" them down to the correct fit.

I have never tried to machine an O-ring before so will see what happens. Always something completely new to contend with.

Another issue which tells me that LBSC never actually made these valves was when trying to assemble and keep together the spring backed valve, when assembling the two body halves, there was NO WAY one could do this and have the valves stay in place or not fall out so finally, in desperation, I glued the spring backed valve in place with some paper glue. My reasoning was, that once in use the steam will melt away the glue.

 

[carrdo]

Hi All,

The two completed boiler blowdown valves with some of the last construction fixtures. Note how the valve handles ended up for a RH and a LH valve.

The valve handles were finished to final thickness by surface grinding and the valve outside contours finished by drum sanding.

 

[Mike R]

Thanks for the write up. I started to build one of these about a year ago from the same LBSC write up and got side tracked somewhere along the line. I need to revisit and I'm sure that your description of the trials and tribulations will come in handy. FYI mine will be for a 5" gauge B1 Springbok that I've picked up half built.

 

[carrdo]

Hi All,

One final operation, to turn the OD of each valve body to remove any and all previously made chucking blemishes.

I made up a chucking piece out of some round aluminum bar threaded 1/8 NPT to hold the entire body by the 1/8" NPT spigot on the rear half of the valve body. Everything was initially turned and threaded in the lathe to ensure concentricity but threading a 1/8" NPT tapered thread with a hand tap even in soft aluminum could only be started in the lathe due to the extreme torque required. I had to finish threading by hand tapping in my large bench vise utilizing soft jaws and a very large hand tap holder with the vise tightened so much I was certain I was going to distort/squash the chucking piece. It worked but what a struggle.

The entire valve body was then threaded into the chucking piece and set up in the lathes' 4 jaw which was adjusted until the valve body half nearest to the chuck jaws ran true. Again I had to ensure that there was NO wobble on the part before any OD machining. This meant that the threaded spigot on the valve body had to be screwed down onto the chucking piece thread pretty much to the backside of the valve body and that the two threads were really locked together. Minor residual wobble could then be corrected by tapping gently with a non metallic hammer.

I call this operation risky as ANY movement of the chucking piece/valve in the 4 jaw would result in the tool digging in and destroying everything as we now have a large eccentric movement of the chucking piece. So sharpen your cutting tool and take only very light shaving cuts. Fortunately, nothing happened.

I actually surprised myself as I finally finished something.