Help with 'Speedy' valve gear and other issues...

[Neale]

Thanks, Roger - I shall investigate. I do use CNC - a lot! - and my smallest job to date was to cut the 3mm square holes in brake handles for my Black 5 tender. That used a 1mm cutter, although I was able to rough out with larger to remove the bulk. Typical flute length around 3mm, though, so perhaps a bit more robust. I'm glad to say that I did not break any. That time - I'm sure that it will happen. However, those narrow slots can't be roughed out so it's the small cutter for the whole job. And no, I wouldn't even dream of doing something like that with a manual machine. CNC does open up possibilities that are difficult to achieve otherwise, at least in a home workshop.

[Roger]

May 14, 2024 20:30:22 GMT Neale said:

Thanks, Roger - I shall investigate. I do use CNC - a lot! - and my smallest job to date was to cut the 3mm square holes in brake handles for my Black 5 tender. That used a 1mm cutter, although I was able to rough out with larger to remove the bulk. Typical flute length around 3mm, though, so perhaps a bit more robust. I'm glad to say that I did not break any. That time - I'm sure that it will happen. However, those narrow slots can't be roughed out so it's the small cutter for the whole job. And no, I wouldn't even dream of doing something like that with a manual machine. CNC does open up possibilities that are difficult to achieve otherwise, at least in a home workshop.

Ok, that's great. If it's a 1mm PCB burr or less, I use a depth per cut of about 25microns and a feed of 3mm/min. I'll have to measure the RPM though, because it's not something that comes out in sensible numbers using my high speed spindle. I'll try to remember to do that tomorrow.
You can definitely machine almost anything with those cutters, they're extremely sharp and hard. Mild Steel sheet isn't a problem, and I've even machined HSS with a 0.5mm cutter for the gear cutting form tools.

Once you've got the hang of it and some workable feeds and speeds, it opens up a whole new world of possibilities. I use tiny hexagonal pockets with rounded corners for valve handles for example. You can get a bigger AF size with a Hex than a square, and I think it's more robust when the sizes get really small.

Have fun!

[Roger]

I'm trying to get things done that are standing in the way of Steaming the Locomotive, and the Whistles are some of those because they're under the footplate.

I've already made the Servo Valve and experimented with 3D printed whistles to get the frequencies I need, so now it's time to make the Brass ones.

As already mentioned, It's not possible to get the right frequency because the low one would be too long to fit. However, if I swap the High pitch one for the Low pitch one, I can get that one in. So the Brake whistle pitch is higher than it should be by three semitones, as is the Service whistle. Unless you have perfect pitch, you're not going to notice.

This is the thin wall tube being made even thinner. It started as 1mm wall thickness, but I'm making it half that.

PXL_20240514_143741258 by Billy Roberts, on Flickr



PXL_20240514_145510479 by Billy Roberts, on Flickr

I'm using a 1mm Carbide End Mill to machine out the openings rather than a PCB burr because I want to get both of these done today.

PXL_20240515_100819440 by Billy Roberts, on Flickr



PXL_20240515_103040884 by Billy Roberts, on Flickr

I don't have a 4th axis output on my Alibre CAM so I've contrived a way to to that. I'll explain that at the end.

PXL_20240515_123846124 by Billy Roberts, on Flickr

I decided to use some 0.8mm Brass rivets to locate all of the parts for Soldering rather than struggle keeping it all where it should be.

PXL_20240515_130141882 by Billy Roberts, on Flickr

This is one of the end caps...

PXL_20240515_131747338 by Billy Roberts, on Flickr

... and this is the larger of the two whistles.

PXL_20240515_175409583 by Billy Roberts, on Flickr

I made a Steel plug to go inside the body so it wouldn't get crushed by the chuck jaws.


PXL_20240515_175539671 by Billy Roberts, on Flickr

Hopefully this will all fit together. I still need to make the unions to pipe this up.

PXL_20240515_200134521 by Billy Roberts, on Flickr

So, as promised, here's how I created the 4th axis output.

The sketch below is basically a flattened out version of the opening to scale. I used the measuring tool in Alibre to tell me what the length of the aperture was. I had to calculate the short distance to the start of it.

The the cut was treated as if it was being cut out of a flat sheet with the Y-axis being front to back instead of the A-axis wrapping around. The Orange lines are the apertures, and the Blue lines are the 1mm cutter centre lines.

Whistle cutouts using 4th axis by Billy Roberts, on Flickr

My CAM output is directed to a little Utility program that I've slowly added to over the years. Most of the time, I just output the unchanged program as a .nc file

This is what it looks like for the above cutout. You'll notice that there's an additional comment on line 1 that tells the program what the diameter is. This is a line I've typed into the name of the machining operation so that gets included in the output.

If you look to the bottom button on the Utility program, you'll see there's one called 'Scale Y to A'

Cutout program from CAM by Billy Roberts, on Flickr

When you press that button, this is what you get instead.

The Utility program reads the Diameter from the comment, then calculates what angle represents 1mm so that it can define a scale factor. It then creates the line #1000 = 4.6964 which is a parameter that's set in the CNC control. You can have as many of these as you like and then substitute anything in the program with them.

The Utility program also changes every instance where it says Y1234 to be A[#1000*1234] for example. Hopefully you can now see where this is going. When the CNC control runs this program, it makes a note of what parameter #1000 holds, and then calculates a value for A that equates to the angle required to move that distance in mm on that diamter.

Yes, I know that looks complicated on the face of it, but it's just a scale factor in brackets multiplied by what the value was in Y.

Cutout program modified by Billy Roberts, on Flickr

Needless to say, this is a huge time saver, because you can effectively wrap whatever flat pattern you like around a given diameter without using any special software. This was actually created for a commercial program to create a CAM path on a cylinder, but it's also perfect for this.

[uuu]

For some time I had been trying to figure out how to do the Y to A conversion. To do some engraving on a cylindrical surface. Then my CAM program came to the rescue - the latest version of CamBam has a "revolve" post-processor option that wraps the output around.

I just have to remember to turn it off for the next job.

Wilf

[Roger]

May 16, 2024 7:24:15 GMT uuu said:

For some time I had been trying to figure out how to do the Y to A conversion. To do some engraving on a cylindrical surface. Then my CAM program came to the rescue - the latest version of CamBam has a "revolve" post-processor option that wraps the output around.

I just have to remember to turn it off for the next job.

Wilf

That's handy. It's worth looking at the use of Parametric Programming though for other things. I used a local change to Incremental measurements in a Sub Program to allow repeated calls to move the same amount. It would appear that Parametric Programming could make that more elegant and also move the programmable parts to the header of the program where they're more easily modified. The same sort of idea could be used for creating an array of images.
It's not something I've really taken advantage of very much.

[Roger]

These went together nicely, they're just held together with three 0.8mm Brass rivets which were a very snug fit indeed. I don't think I'll bother Soldering the bottom half, other than to attach the inlet unions. The top caps will need Soldering on though, or I suppose I could use Loctite which might look neater.

They sound fine when blown by mouth. I just need to finish all of the pipe fittings and design the brackets.

PXL_20240516_094217223 by Billy Roberts, on Flickr

I've put the Locomotive back on the Weighbridge tonight, along with all the parts that are going to be on the Cab end. I don't think the Pannier Tanks are going to change the balance from front to back very much, they're fairly evenly placed at the middle.
Looking at the figures, I think I'm light on the back by about 4Kg, so I'm looking at making some plates to bolt under the Cab Stretcher. I've got a wide strip of 2.4mm thick Lead Sheet, so I could possibly put three or four thicknesses of that in there. If it's not enough, I suppose the rest will have to go in the bottom of the Bunker. Ideally I'd like to avoid that though because I'd like to fill that with Coal.

[Roger]

Continuing with the Whistle arrangement, here's the Servo Valve that you've seen before, but this time I've added a new valve to the inlet of the Pilot steam that operates the diaphragm.

You can see vent holes around the base of the Hex, and that's where the Condensed water escapes when the Pilot Steam is turned off.

Servo valve with vent banjo by Billy Roberts, on Flickr

Basically, it's a simple Banjo arrangement, but with a tiny ball valve that's normally held open with a spring. The Brown rod is actually just a piece of 0.5mm wire.

The idea is that the Servo valve will end up full of water that's condensed from the Steam coming from the Pilot valve. The body is going to be cold when the whistle is blown, so water is going to accumulate behind the diaphragm pretty quickly. The intention is to let any excess water out of the valve when the Pilot Steam is turned off. As soon as the pressure drops sufficiently, the ball valve will be pushed open by the spring so that there's away to get rid of the water quickly without much restriction.

When the Pilot Steam is first turned on, it pushes the ball against the seat so that the pressure can build up behind the diaphragm and allow the valve to open.

I had used a permanently open small bleed hole, first in the Servo Valve cover plate, and later in the Tee piece in the feed pipe. It didn't work flawlessly as I would have liked, hence the changes outlined here. Hopefully this will do the trick. Something similar is done on Steam Brake cylinders for the same reason.


Banjo drain valve by Billy Roberts, on Flickr

[Roger]

Here are the various parts of the Duplex Servo Valve Banjo being machined from Phosphor Bronze. This is my goto material for fittings because it's so much stronger than Brass, which is useful when the parts are slender.


PXL_20240518_080710140 by Billy Roberts, on Flickr

This is the Banjo Bolt...

PXL_20240518_144412503 by Billy Roberts, on Flickr

... with the hex added and then all of the cross holes. The ones in the hex are the outlets when the ball springs open.

PXL_20240520_130923224 by Billy Roberts, on Flickr

The 0.5mm thick Copper Washers were made from thick wall pipe, annealed to make it soft enough.

This is now just waiting for the 1mm springs and the 1.5mm Silicon Nitride balls which are due next month.

PXL_20240520_200904939 by Billy Roberts, on Flickr

This is the short pipe that connects the small whistle to the Servo Valve...

PXL_20240521_100246467 by Billy Roberts, on Flickr

... this is the other end.

PXL_20240521_101623181 by Billy Roberts, on Flickr

The pipes cross over so that I can get the Servo Valve in a position where it's more easily serviced.

PXL_20240522_171821771 by Billy Roberts, on Flickr

[Roger]

Here's the bracket for the Whistles, this time made from Brass since it's a very damp environment.


PXL_20240523_075621867 by Billy Roberts, on Flickr

The brackets are slightly different lengths so that any condensate will run towards the rear and out of the Whistle mouths.

Yes, I could have made this a folded part, but I wanted to make sure the heights were exactly how they're modelled.

PXL_20240523_092335238 by Billy Roberts, on Flickr

I had used Loctite on these joints, but it wasn't strong enough. I really dislike Soft Solder, it's messy and weak. So I decided to Silver Solder this.
As usual, I've used Nail Varnish as a mask, and put a single turn of 0.5mm Silver Solder wire.

The Brass tubes are very thin, only 0.5mm thick, so I arranged it so that I could heat the fitting with the flame for a long time before finally getting the local tube just warm enough to melt the Silver Solder. It would have been very easy to overheat the Whistle and end up with a molten pool of Brass!

PXL_20240523_152013269 by Billy Roberts, on Flickr



PXL_20240523_154747045 by Billy Roberts, on Flickr

Here's the Small Whistle getting the same treatment.

PXL_20240523_193937269 by Billy Roberts, on Flickr



PXL_20240523_194436467 by Billy Roberts, on Flickr

This has taken a fair bit of figuring out so that it all fits and also is accessible for maintenance.

The bracket has two bolts that go into the bottom of the thick Cab Stretcher. I can get to these with a driver, not a spanner.

PXL_20240523_202643292 by Billy Roberts, on Flickr

The valve will just be supported by the pipes.

PXL_20240523_202655222 by Billy Roberts, on Flickr

[JonL]

Can't wait to hear this.

[Roger]

I started routing the Copper pipes for the Whistles and Steam Brake, but realised that they all have to get past the Dummy Hydrostatic Lubricator which is only partially made. This is what it looks like. I also need to make that extra valve with the Cool Handle, because that's going to get in the way too.

20180110_110924 by Billy Roberts, on Flickr

Anyway, this is how far I'd got with it, and it's missing all of the fiddly little bits.

You can see that there are a couple of holes in the top plate, which is on the lower left of the photo. There are two bosses that go there, but they're impossible to get around with a cutter to machine this in one operation, you can't get around them.

PXL_20230227_214556767 by Billy Roberts, on Flickr

Anyway, before making those, there are 'T' handles to be made and a round one. This is the tapered end of one handle being machined from a piece of Silicon Bronze Brazing wire.

PXL_20240528_104115638 by Billy Roberts, on Flickr

The Compound Slide was kept in the same position, with the lathe run backwards and a tiny Carbide boring Bar used on the back to do the other handle. Having moved it in the collet, it needed clocking up between operations. The collets aren't that good.

PXL_20240528_111557725 by Billy Roberts, on Flickr

That was transferred to the Mill and a 0.7mm hole was drilled 1mm deep across the middle to locate the shaft.

PXL_20240527_203856469 by Billy Roberts, on Flickr

The shaft is 1mm diameter with the 0.7mm location on the end.

PXL_20240528_120353804 by Billy Roberts, on Flickr

Those were held in the 'Helping Hands' for Silver Soldering.

PXL_20240528_193945987 by Billy Roberts, on Flickr

... like this.

PXL_20240528_194249165 by Billy Roberts, on Flickr

That was then parted off and held in a Pin Vice in the Lathe for finishing with a file.

PXL_20240528_211748308 by Billy Roberts, on Flickr

This is the round handle getting the 8 grooves added to the diameter using the 4th axis and an engraving tool. This part is made from Colphos, as are the nuts and other bits and pieces. Brass is too Yellow and doesn't look right.

PXL_20240529_100249057 by Billy Roberts, on Flickr



PXL_20240529_100617736 by Billy Roberts, on Flickr

Here are the two bosses, including the threads to attach other parts to them,

PXL_20240529_195425794 by Billy Roberts, on Flickr

A scrap of 0.5mm Silver Solder wire was put on the back and the whole thing fluxed and heated...

PXL_20240529_200707420 by Billy Roberts, on Flickr

... so it ended up like this.

PXL_20240529_202544644 by Billy Roberts, on Flickr



PXL_20240529_202551692 by Billy Roberts, on Flickr

This is the main tall Stainless Steel flanged bolt that goes on top of the filler.


pxl_20240530_102356401 by Billy Roberts, on Flickr

I made a few extra handles so I can pick the best ones. I need four in all.

pxl_20240530_103020592 by Billy Roberts, on Flickr

I'd mistakenly forgotten to add the boss on the front of the bottom rail, so here I'm adding a 2.2mm diameter location to add that now. It was started with a 2mm Carbide drill to make sure it didn't wander.

pxl_20240530_111519478 by Billy Roberts, on Flickr

I'd also neglected to tap all of the holes in the body, so here's the M1 holes for the threaded rods.

pxl_20240530_214347607 by Billy Roberts, on Flickr

This is the second Silver Soldering operation on the bottom bar. I'd already done the front gland nut, and here the triangular flange is being done. Again, I'm just using scraps of 0.5mm Silver Solder wire. You don't need much and I don't want it to go everywhere.

pxl_20240531_193258210 by Billy Roberts, on Flickr

The top of the body has a few M2 tapped holes in the bosses to attach the various fittings...

pxl_20240531_201621993 by Billy Roberts, on Flickr

... and there are ones on the bottom too.

pxl_20240531_202743601 by Billy Roberts, on Flickr

So here's the state of play tonight with some of the parts added. I've got some 1.5mm Clear Acrylic rod coming from China, but that won't be here for a couple of weeks.

pxl_20240531_214247646 by Billy Roberts, on Flickr

It's an awful lot of work for just a cosmetic item, but it's very prominent on the backhead, so worth putting a bit of effort in.

[Roger]

I was going to fit the various Gland Nuts to the top and bottom bar today, but rather than struggle to cut some M1.6 bolts down to 2.5mm and 6mm lengths, I had a quick look on eBay and found some Stainless Steel Grub Screws that were cheap enough from China instead.

In the meantime, I added the bent handle to the top fitting. That's going to be held in with Screw Lock.


PXL_20240601_214652061 by Billy Roberts, on Flickr

I've modelled the weird looking bottom valve, so that's currently being roughed out. More of that shortly.

[baldric]

Just for information as it w9nt affect your model, the weird looking valve on the bottom is the drain for the condensed water. The valve feeds into a funnel so that the driver can see when all the water is out and oil starts to flow. The warming cock feed also drains in to it, the warming cock is the wire- wound handle.
Baldric.

[Roger]

Jun 2, 2024 20:51:38 GMT baldric said:

Just for information as it w9nt affect your model, the weird looking valve on the bottom is the drain for the condensed water. The valve feeds into a funnel so that the driver can see when all the water is out and oil starts to flow. The warming cock feed also drains in to it, the warming cock is the wire- wound handle.
Baldric.

Thanks Baldric, I wondered about that. I ought to have guessed that really, knowing it's a displacement lubricator!
I notice that there have been two types of the warming cock handles on the photos I've got. One is a wire wound type, the other has 6 spokes with 6 ball ends to them outside of the rim. I'm not sure what type to make. 

[chris vine]

Most people would make the easiest type, but I suspect Roger will make the most difficult/interesting!!…

[Roger]

This is the strange looking Drain Valve and viewing arrangement on the bottom of the Hydrostatic Lubricator. It looks quite a bit different to the crisp outline drawing I have of it on the Blower Valve Works Drawing. However, this is what's on 1501, so it should look like this.


20180110_110321 by Billy Roberts, on Flickr

As usual, I'm making these tiny parts from Phosphor Blonze because they would be very fragile in Brass or Bronze.

The cutouts in the side have been suppressed in the model before outputing the roughing operations. There's no point in it being more fragile than necessary at this stage. I'm roughing it out with a 1mm Carbide End Mill.

PXL_20240601_215541634 by Billy Roberts, on Flickr

That was repeated with the 4th axis at 180 degrees, and then followed up with a finishing pass with a 1mm Ball Nosed cutter and 0.1mm stepover.

PXL_20240602_140330399 by Billy Roberts, on Flickr

I used the same trick as before with the pockets being defined as a flat pattern in X/Y and then converting that with my utility program to be X/A at the appropriate diameter.

I've left 0.4mm of the end still being closed so it's stronger when I need to Silver Solder the two parts together.

The cutouts were repeated at 120 then 240 degrees...

PXL_20240603_181633915 by Billy Roberts, on Flickr

... before swapping to a 1.5mm End Mill to finish the hexagon. The flange will be finished when the thread is added later.

The 1mm hole for the handle was drilled part way through at this stage too.


PXL_20240603_183538492 by Billy Roberts, on Flickr

The 4th Axis was then tilted vertical and the end detail roughed out with first a 1.2mm drill, then a 2mm to the depths taken from the 3D model...

PXL_20240603_184813543 by Billy Roberts, on Flickr

... and finally a 3mm drill. That's got rid of most of the unwanted material, so it just needs finishing with a 1mm Ball Nosed cutter.

PXL_20240603_185521329 by Billy Roberts, on Flickr

The trick as always with this sort of machining job is to do everything while you've still got hold of it in the collet. 

[mbrown]

Gosh Roger - I hope your eyesight is holding up after all this tiny work!

Even if I could make it, I would be sure to drop it and lose it. With arthritic fingers and knees, I not only drop things but can't get down to pick them up!

Very admiring of your work!

Malcolm

[Roger]

Jun 3, 2024 20:18:36 GMT mbrown said:

Gosh Roger - I hope your eyesight is holding up after all this tiny work!

Even if I could make it, I would be sure to drop it and lose it. With arthritic fingers and knees, I not only drop things but can't get down to pick them up!

Very admiring of your work!

Malcolm

Thanks Malcolm. It's fiddly, but as long as you use razor sharp cutters and tiny cuts, it's easy enough. It does take a long time though, around 8 hours machining in that part so far. It's so flexible that you can only take 25-50micron cuts (1-2 thou)

[Roger]

Here's the 3D printed fixture that holds the valve body without risking damaging the fragile end.

PXL_20240604_200559374 by Billy Roberts, on Flickr

The parting off pip was then turned off and the M2 thread added.

PXL_20240604_201615732 by Billy Roberts, on Flickr

This is the funnel part of the valve, which of course is part of the same casting on the real thing.

I turned an over length M3 thread on the stock to enable it to be held firmly in my standard M3 threaded mount with a locking screw up the back.

The first half was roughed and finished with a 2mm Ball Nosed cutter...

PXL_20240604_130719029 by Billy Roberts, on Flickr

... then turned 180 degrees and the other side done too.

PXL_20240604_182057141 by Billy Roberts, on Flickr

The top was drilled with a 1.5mm, 2mm and 3mm drill to remove the bulk of the material.

Then the register was machined with a 1mm cutter, followed by the tapered bore with the same one.

PXL_20240604_203340532 by Billy Roberts, on Flickr

Then it was just a matter of moving the thread further out and cutting it off.

PXL_20240604_203846522 by Billy Roberts, on Flickr

These just need a little attention with a needle file to get them ready for Silver Soldering.

PXL_20240604_204552516 by Billy Roberts, on Flickr

It's pretty small.

PXL_20240604_205819153 by Billy Roberts, on Flickr

[Roger]

I drilled and tapped an M3 hole in a piece of Steel scrap to hold the Condensate Drain Valve. Here it is with a 'U' shaped piece of Silver Solder around one of the cutouts in the hope it will flow everywhere.


PXL_20240605_190916793 by Billy Roberts, on Flickr

Ok, that almost worked, but I don't think it was quite hot enough.

PXL_20240605_191325148 by Billy Roberts, on Flickr

I cleaned it up and repeated the process with another piece of Silver Solder, and now it looks fine. Fortunately, the two parts were a gentle push fit, so they stayed together through the process with gravity assisting.

The Silver Solder has migrated up the top part, I should have masked that really.

PXL_20240605_202927818 by Billy Roberts, on Flickr

Anyway, it's probably good enough. I've cleaned off some of the excess Silver Solder, but I might do a bit more. I'm not sure if it's alloyed into the surface such that it will never all come off. Something to remember for next time.

PXL_20240605_203735934 by Billy Roberts, on Flickr