Hello Asteamhead,
I think that you know much more about building these miniature air compressors than I do. I am good at finding or looking up information but Richard is the one who can start with a blank sheet of paper and come up with something completely new which has never been done before.
Yes, I agree we should be in touch and try and learn what we can from each other both in terms of the construction and the operation of these miniature compressors.
Mission Impossible - Designing and Constructing a 3/4" Scale Westinghouse Cross Compound Air Compressor
Re: Mission Impossible - Designing and Constructing a 3/4" Scale Westinghouse Cross Compound Air Compressor
Hi,
Here we go again.
Steve, Big Boy is still a work in progress, but I work on it as well as these other things. If you keep checking our Big Boy thread, you will see additions every now and then. The boiler was the big headache, but using Locktite, I was able to get a "proof" pressure of 200 psi. which I think is good enough. As I said on the thread, its all down hill from there, but its still a long hill.
All these other "side trips" like power reverse, working air compressors, automatic pressure control, etc. are going to be a part of Big Boy, so that's why they are in the picture too. Because of the way the boiler is built, it should steam well, but at present, that is still one of the "unknowns." The plan is to fire it with propane.
The picture in the previous post shows the original pilot control rod. That fits inside the hollow piston rod which is also shown in the picture. The little pilot slide valve just fits between the two bronze bits on the one end. The other end was threaded on the rod with a #0-80 thread. The original rod was made from 1/16th in. brazing rod. I played around with this for a while trying to get it to work but it was too flimsy so I had to redo it. I used 1/16th inch dia. stainless welding rod and silver soldered the whole thing together to make it stronger. This shows up in subsequent pictures (picture after the next one in this post). To be able to assemble things, I also had to make a new round disk with a "keyhole" in it so I could put it on the shaft. When assembled, it works just fine.
This picture shows the center section which fits between the steam cylinder section and the air compressor section. Again, it is easy to see how impressive the casting and machining is. I planned on using Viton "O" rings as rod seals in these glands, so that's what is inside. The "O" rings make a good seal with low friction and should last for a long time. I used the little CNC mill to cut the vertical slots on the nuts. The mill makes a tedious job easy and the result (other than being a shiny bronze colour) looks prototypical.
As the caption says, this picture shows the top cover with the shuttle sleeve in place. There is a hole in the sleeve that matches the threaded lifting eye you can see in the picture and that is what lines up the sleeve. The sleeve has external grooves which channel the steam to the proper ports in the spool. In a model, this is the way to manage all the cast in passages that were in the full size pump. It takes quite a while to figure out where each passage goes and what they all do, both for supply and exhaust.
As the caption says, this is the revised pilot valve control rod. I made this version because the original wasn't strong enough. Everything is silver soldered together for strength. It took quite a while to get the dimensions right. As an aside, I used a taper pin reamer to make a taper in the casting for the lower guide then filed the guide to fit. This made for an easy assembly that was repeatable and removable. You can also see the keyhole in the round disk. I did this so I could assemble the parts and remove the disk if I needed to (and in the many fittings, I did need to!). Once the rod is assembled in the piston, everything works the way it should because the big hole is off to the side.
Hopefully, this should give everyone enough for a bit. Just to remove some of the suspense, the pump is at the stage where it runs very nicely on air (at about 15 psi.), but I still have to figure out the check valves on the air side before it will compress air. I also have to make a steam flow control to prevent the pump from running too fast as the discharge pressure varies. Got it in my head, just haven't drawn it up yet.
Richard Trounce.
Here we go again.
Steve, Big Boy is still a work in progress, but I work on it as well as these other things. If you keep checking our Big Boy thread, you will see additions every now and then. The boiler was the big headache, but using Locktite, I was able to get a "proof" pressure of 200 psi. which I think is good enough. As I said on the thread, its all down hill from there, but its still a long hill.
All these other "side trips" like power reverse, working air compressors, automatic pressure control, etc. are going to be a part of Big Boy, so that's why they are in the picture too. Because of the way the boiler is built, it should steam well, but at present, that is still one of the "unknowns." The plan is to fire it with propane.
The picture in the previous post shows the original pilot control rod. That fits inside the hollow piston rod which is also shown in the picture. The little pilot slide valve just fits between the two bronze bits on the one end. The other end was threaded on the rod with a #0-80 thread. The original rod was made from 1/16th in. brazing rod. I played around with this for a while trying to get it to work but it was too flimsy so I had to redo it. I used 1/16th inch dia. stainless welding rod and silver soldered the whole thing together to make it stronger. This shows up in subsequent pictures (picture after the next one in this post). To be able to assemble things, I also had to make a new round disk with a "keyhole" in it so I could put it on the shaft. When assembled, it works just fine.
- Center section with "O" ring sealing glands added.
- Top cover showing shuttle sleeve in place but with the holding boss still attached.
- This is the revised pilot rod unit.
Hopefully, this should give everyone enough for a bit. Just to remove some of the suspense, the pump is at the stage where it runs very nicely on air (at about 15 psi.), but I still have to figure out the check valves on the air side before it will compress air. I also have to make a steam flow control to prevent the pump from running too fast as the discharge pressure varies. Got it in my head, just haven't drawn it up yet.
Richard Trounce.
Re: Mission Impossible - Designing and Constructing a 3/4" Scale Westinghouse Cross Compound Air Compressor
Hi,
Here comes some more.
This view shows the top of the control piston with the hole for the control rod and the tapped screw holes (#2-56) for the rod locating cxover.
A better view of the intermediate casting showing the dummy valve covers and the rod packing glands.
Again, a lot of thought went into the design of this model with bosses to fit into corresponding recesses in the steam side casting so that things would be easily aligned on assembly.
This shows the top cover complete with the lifting eye where it should be and the shuttle valve covers in place. The hex hd. screw head maintains the alignment of the shuttle spool and it also controls the orientation and travel of the spool itself. Although you can't see it, the shuttle valve is also in place.
This and the view below are taken at the same time. They show what the low pressure piston looks like when it is fitted inside the cylinder
In this view, you can also see the recess that the boss on the top cover (#A0426A) fits into.
These pictures give a better idea of the bits and pieces and how they fit together. More later.
Richard Trounce.
Here comes some more.
- Control piston showing screw holes for the plate.
- A different view of the intermediate casting.
- This shows the boss on the underside of the top cover.
- Top cover complete.
- A view of the underside of the steam cylinders.
- A view of the top side.
These pictures give a better idea of the bits and pieces and how they fit together. More later.
Richard Trounce.
Re: Mission Impossible - Designing and Constructing a 3/4" Scale Westinghouse Cross Compound Air Compressor
Hi,
This is the first time I had the steam side running. It wasn't perfect, but it went.
This is a closer view of the top cover. You can see the lifting eye is where it belongs and you can see the top of the hex head bolt that keeps both the sleeve and the shuttle spool aligned and in their proper places. You can also see the copper plug that seals the end of the pilot valve hole and the screw on the right side. There is a spring inside the screw (a piece of rubber band) that applies pressure through a stud to the round slide valve to hold the slide valve in place against the ports.
Just a short addition this time, at least partly because I couldn't post the pictures last night. While they check marked green in the "Add Attachment" status section, they may have been too big.
Richard Trounce.
- First time running.
- A closer view.
Just a short addition this time, at least partly because I couldn't post the pictures last night. While they check marked green in the "Add Attachment" status section, they may have been too big.
Richard Trounce.
Re: Mission Impossible - Designing and Constructing a 3/4" Scale Westinghouse Cross Compound Air Compressor
Hi,
Yes, the file size was too big, because this time it worked after I used Photopaint to reduce the file size even more.
This has nothing to do with this particular thread, but for those of you who use Photopaint or Photoshop to crop and otherwise manipulate pictures, to get picture file sizes to an acceptable level I have found that a file size of 1.2 meg. works. 1.6 meg. or higher is too big and will cause problems. In the "Add Attachment" section, even if it checks green in the "status" column, the "size" column must be at least less than 1. Too high and just the text appears in the post together with the file names, but no pictures.
This picture shows the bits and pieces that were listed in the previous post. You can see where they go in the previous close up picture of the top cover.
This picture shows the complete assembly of the high pressure piston including the pilot valve rod, the rod drive plate mounted on the piston and the lower guide which is a tapered fit in the top cover. The round slide valve just fits between the two discs at the end of the rod. If you look very closely, you can even see the overlapping ends of the piston ring including the little "hooks" I mentioned before.
Here is the slide valve that forms the pilot for the main shuttle valve. It fits in the space between the two discs at the end of the rod.
These pictures show the details of the pilot valve assembly so they form a "unit." more to come.
Richard Trounce.
Yes, the file size was too big, because this time it worked after I used Photopaint to reduce the file size even more.
This has nothing to do with this particular thread, but for those of you who use Photopaint or Photoshop to crop and otherwise manipulate pictures, to get picture file sizes to an acceptable level I have found that a file size of 1.2 meg. works. 1.6 meg. or higher is too big and will cause problems. In the "Add Attachment" section, even if it checks green in the "status" column, the "size" column must be at least less than 1. Too high and just the text appears in the post together with the file names, but no pictures.
- Copper plug, locating screw and slide valve pressure screw.
- This shows the assembly of the pilot valve rod in the piston.
- This is the little slide valve plate.
These pictures show the details of the pilot valve assembly so they form a "unit." more to come.
Richard Trounce.
Re: Mission Impossible - Designing and Constructing a 3/4" Scale Westinghouse Cross Compound Air Compressor
Hi,
Some more.
Another view of the round slide valve. Remember, this thing fits in a 3/16" dia. hole, so it isn't very big. Also because its so small, it wouldn't be that easy to make.
This shows both the shuttle valve and the sleeve it slides in. You can see the passage grooves on the outside of the sleeve and the drilled ports in the shuttle valve. You can also see the milled slot in the spool which keeps the spool from rotating and also controls its travel. In some ways, the lower view shows things better. The hex bolt you can see in the top cover controls the position of both the spool and the sleeve.
This gives a better view of the shuttle valve showing the drilled ports and the locating slot.
A good place to stop for now.
Richard Trounce.
Some more.
- Another view of the slide valve.
- Shuttle sleeve and shuttle valve.
- Shuttle valve.
A good place to stop for now.
Richard Trounce.
Re: Mission Impossible - Designing and Constructing a 3/4" Scale Westinghouse Cross Compound Air Compressor
Hi,
I stopped where I did because now we start on a different part of this narrative.
When I got it running, I found that if I let the pressure slowly drop to zero, when I started up again, the pump wouldn't always restart. After taking things apart a few times, I found that when it didn't start, the shuttle spool was somewhere in the middle of its travel so the pistons couldn't move.
After more playing around, I found that if I epoxied a little magnet on each end cover (the covers are bronze, so the field would penetrate the cover), the magnets would reliably hold the spool at one end or the other of its travel. I also found that if the magnet was in the middle of the cover, it was too strong, so that's why in the picture you see the magnet offset to the side. As a side note, as the air pressure increases from zero, the pump starts at about 16 psi. and on the way down, it stops working at about 10 psi.
In this project, we knew we had to do something about the missing air side cylinder casting. I saw on Anthony Duarte's website (Eccentric Engineer) that he was going to be offering unmachined bronze castings for a 1.6" scale Xcompound air pump in 2020. We sent him an e-mail asking if we could buy just the air side casting (we had everything else). He e-mailed back saying that he had two prototype castings and he was willing to sell us one of them for our project. He sent a drawing of the casting and I thought it would work, but the larger bore would have to be sleeved down (too big). We purchased the casting and you can see it temporarily positioned just to see what it would look like.
In the picture, you can also see the little neodymium magnet on the shuttle cover and the small hammer is also holding the bottom cover of the pump in position, again to see how things would look. As you can see, Anthony's casting is very detailed and looks just right when you compare it with the parts we already have. We were lucky that he was willing to let us have it.
The next posts will cover the required machining of the casting.
Richard Trounce.
I stopped where I did because now we start on a different part of this narrative.
When I got it running, I found that if I let the pressure slowly drop to zero, when I started up again, the pump wouldn't always restart. After taking things apart a few times, I found that when it didn't start, the shuttle spool was somewhere in the middle of its travel so the pistons couldn't move.
After more playing around, I found that if I epoxied a little magnet on each end cover (the covers are bronze, so the field would penetrate the cover), the magnets would reliably hold the spool at one end or the other of its travel. I also found that if the magnet was in the middle of the cover, it was too strong, so that's why in the picture you see the magnet offset to the side. As a side note, as the air pressure increases from zero, the pump starts at about 16 psi. and on the way down, it stops working at about 10 psi.
- Shows air cylinder temporarily in place.
In the picture, you can also see the little neodymium magnet on the shuttle cover and the small hammer is also holding the bottom cover of the pump in position, again to see how things would look. As you can see, Anthony's casting is very detailed and looks just right when you compare it with the parts we already have. We were lucky that he was willing to let us have it.
The next posts will cover the required machining of the casting.
Richard Trounce.
Re: Mission Impossible - Designing and Constructing a 3/4" Scale Westinghouse Cross Compound Air Compressor
Hi,
Next installment.
When working on Anthony's casting, I had to proceed slowly and plan everything out carefully step by step because I couldn't make any mistakes. The first thing I had to do was to machine the casting to the proper length. I checked the length shown on the drawing and compared that to what was required because of the stroke on our pump. That turned out to match Anthony's drawing. I decided that facing the casting on the lathe was the best way of doing this. To make sure everything was true, I even took a light "skin" off the faceplate to make sure that was flat and true.
This view shows how the casting was mounted on the faceplate. If you look closely, you can see that the cut hasn't covered the complete surface of the casting yet (down at the bottom of the casting surface). More cuts are required and then the casting has to be turned over to do the other face. For some reason, I don't have any pictures showing the boring operation on the mill, but that had to be done to clean up both bores and to make sure that their separation was right. When I measured the piston rods, I found that they weren't completely parallel, so I had to file the glands on the intermediate casting to make them right. That also affected the bore spacing on the air side casting. By the way, the green you see on the part is a reflection of the green paint on the lathe.
As I said before, the larger bore had to be sleeved and this picture shows the sleeve being bored. I made the boring bar holder and the bars (copied from a commercial one). It is relatively simple to make and it results in a very rigid assembly and is very useful.
This picture shows the final finish cut on the sleeve. When doing this kind of thing, to fit the sleeve in the bore, get very close to the right O.D. on the sleeve, then step down a thou. or two on the end of the sleeve and keep on doing this until the casting will just slide on the stepped down part. This way you have a controlled interference for the press. I did this because I couldn't afford to crack or split the casting.
A good place to stop for now.
Richard Trounce.
Next installment.
When working on Anthony's casting, I had to proceed slowly and plan everything out carefully step by step because I couldn't make any mistakes. The first thing I had to do was to machine the casting to the proper length. I checked the length shown on the drawing and compared that to what was required because of the stroke on our pump. That turned out to match Anthony's drawing. I decided that facing the casting on the lathe was the best way of doing this. To make sure everything was true, I even took a light "skin" off the faceplate to make sure that was flat and true.
- Facing the casting to length.
- Boring the sleeve.
- Finishing the bore.
A good place to stop for now.
Richard Trounce.
Re: Mission Impossible - Designing and Constructing a 3/4" Scale Westinghouse Cross Compound Air Compressor
Hi,
When taking measurements, I found that the bottom cover center hole was out of position for the high pressure cylinder.
This picture shows the start of the slotting of the hole in the cover so the high pressure piston will fit in the hole without striking the side of it. If you look closely you can see the partly finished high pressure piston on the lower left side of the face plate.
This shows the piston in use to check for location. The piston threads on the piston rod and then a locking nut holds the assembly together.
Finally the slot is long enough to fit. Clearance is all that is required. In this case I decided it was better to plunge cut the slot because the cutting forces would be lighter and I wasn't really sure just how firmly the part was held in the setup. Again, I didn't have any extra bottom covers, so I couldn't afford to make a mistake.
This is the stage I'm at now (June 18, 2020). The unit is assembled and runs, but it doesn't pump air yet. I still have to make and install all the check valves on the air side so it will actually work as a compressor. I also need to draw up and make the flow control valve on the steam side so the compressor won't run away as the air pressure varies on the air side.
Also, based on what I have learned, I have to change the pilot valve design so it works with "O" rings, not the present set up.
More to come.
I still have a way to go on the 1/8th scale version before I can even seriously look at a half size version. I know that the pilot valve is likely going to stay the same size, but the shuttle valve can be made smaller and still work. Physics limits what is doable, but I think it can be done. We'll have to see. It looks like I'll never run out of something to do.
Richard Trounce.
When taking measurements, I found that the bottom cover center hole was out of position for the high pressure cylinder.
- milling the bottom cover.
- Checking alignment.
- Finish bore location.
This is the stage I'm at now (June 18, 2020). The unit is assembled and runs, but it doesn't pump air yet. I still have to make and install all the check valves on the air side so it will actually work as a compressor. I also need to draw up and make the flow control valve on the steam side so the compressor won't run away as the air pressure varies on the air side.
Also, based on what I have learned, I have to change the pilot valve design so it works with "O" rings, not the present set up.
More to come.
I still have a way to go on the 1/8th scale version before I can even seriously look at a half size version. I know that the pilot valve is likely going to stay the same size, but the shuttle valve can be made smaller and still work. Physics limits what is doable, but I think it can be done. We'll have to see. It looks like I'll never run out of something to do.
Richard Trounce.
Re: Mission Impossible - Designing and Constructing a 3/4" Scale Westinghouse Cross Compound Air Compressor
Hi,
Here we go again. Now to cover the machining of the discharge valve pockets in Anthony's bronze casting. This part is complex and has to be done precisely. It is very easy to have a drill wander and come out the side.
This shows the casting with the external bosses that have to be drilled and tapped and passages drilled in the casting to align with the outlet. The first step is to drill and tap for 7/16"-32 Model Engineer thread in the top face of the casting (the bottom pocket is threaded 3/8-32 ME because the boss on the bottom is smaller).
As a side note, there are two series of ME threads. The first is the fractional sizes from 3/16" to 1/2," all at 32 threads per inch and the second series is 1/16" to 1/2," all at 40 threads per inch. The fist series is used for adequate strength without taking up a lot of room and the 40 TPI. series is used mainly for threading thin wall tubing. They are stepped in 1/16th in. increments with one or two in 1/32."
Remember, both these series were meant to be used for 2 1/2" gauge models and also for 3 1/2" gauge, which back in the 20s (and later) was known as "Millionaires gauge," because building in that size cost more than most modellers could afford. In those days, it wasn't uncommon to use a treadle lathe (the good old days in England where all this "nonsense" started).
This picture shows the head angled to drill the passage from the bottom of the valve chamber to the vertical passage to the pump compressed air discharge. Most of the time, you see a Bridgeport with the head vertical. However the head can be tilted forward or back or left and right. This makes the Bridgeport a very versatile machine. Even although the Bridgeport is considered to be a "light" machine, it can still take fairly substantial cuts under these conditions. Sorry for the messy shop, but that's the way it is most of the time.
This picture is a closeup allowing you to see the threads, ports, etc. You can see why there are so many steps and why you can't make any mistakes (castings are expensive and there is already a lot of time invested in this one).
This picture shows the head tipped back to allow the connecting passage to be drilled into the upper part of the valve chamber.
This final picture shows the finished valve chamber on the casting. The picture is clear enough that you can see all the detail, the 7/16-32 threads for the cover plug and also the passages that are required for the air flow through the check valves.
Unfortunately, this is just the top side of the casting. Now I have to turn it over and do the same thing on the bottom face. After that, I have to make the cover plugs and the actual check valves that fit inside. Even then, I'm still not finished because I have to make the inlet check valves and the intermediate check valves that go between the low pressure and high pressure cylinders.
As you are beginning to see, making one of these even in 1/8th scale isn't a "walk in the park." However, as you can see, it can be done as others have done before.
Richard Trounce.
Here we go again. Now to cover the machining of the discharge valve pockets in Anthony's bronze casting. This part is complex and has to be done precisely. It is very easy to have a drill wander and come out the side.
- Casting before pocket machining.
As a side note, there are two series of ME threads. The first is the fractional sizes from 3/16" to 1/2," all at 32 threads per inch and the second series is 1/16" to 1/2," all at 40 threads per inch. The fist series is used for adequate strength without taking up a lot of room and the 40 TPI. series is used mainly for threading thin wall tubing. They are stepped in 1/16th in. increments with one or two in 1/32."
Remember, both these series were meant to be used for 2 1/2" gauge models and also for 3 1/2" gauge, which back in the 20s (and later) was known as "Millionaires gauge," because building in that size cost more than most modellers could afford. In those days, it wasn't uncommon to use a treadle lathe (the good old days in England where all this "nonsense" started).
- Showing Bridgeport head angled to drill passages.
- A closer view showing the drill in position to drill the joining passage.
- Showing the head tipped back to allow drilling the hole into the upper part.
- Showing the finished valve chamber and connecting passages.
Unfortunately, this is just the top side of the casting. Now I have to turn it over and do the same thing on the bottom face. After that, I have to make the cover plugs and the actual check valves that fit inside. Even then, I'm still not finished because I have to make the inlet check valves and the intermediate check valves that go between the low pressure and high pressure cylinders.
As you are beginning to see, making one of these even in 1/8th scale isn't a "walk in the park." However, as you can see, it can be done as others have done before.
Richard Trounce.