|
Post by terrier060 on Jan 10, 2021 21:59:54 GMT
I should add that the measurements above are approximate to the nearest 1/64th. The final drawing will be to nearest half a thou. Sorry for all those of you working in mm, but all my Terrier drawings are in Imperial as were the originals. So is my machinery, though I have to work in both Imperial and Metric on the Tormach!
|
|
jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Jan 10, 2021 23:03:00 GMT
Hi Ed,
There is a thread on here somewhere for a 7.25"g loco injector with a sliding first half of the combining cone.
It was for 'Charles' narrow gauge Penryn loco same as 'Linda' and 'Blanche' now of the Ff R.
But it was quite a bit bigger than your drawing as you are going down the 16 oz per minute delivery route same as Roger.
I wish you well with this experiment.
|
|
|
Post by nick952 on Jan 10, 2021 23:06:40 GMT
|
|
|
Post by terrier060 on Jan 10, 2021 23:43:28 GMT
Brilliant both of you - thanks Julian and Nick - yes I understand what I am seeing now. Probably the one I have drawn has a parallel rather than tapered cone. Moving cone injector detail by ed cloutman, on Flickr
|
|
|
Post by David on Jan 11, 2021 5:33:42 GMT
How does water flow over the vanes to move the cone? It either goes through the overflow gap or into the delivery cone doesn't it? Or does the pressure of the water going out somehow push the cone back?
The other thing that struck me about the one in Dave's photo and the one in the diagram you posted is the severe path the water has to take out of the delivery cone. Do you think that has any effect, or given it's later slamming into a check valve held closed by boiler pressure it's of no consequence?
I'm looking forward to seeing what you come up with and how you do it!
|
|
|
Post by terrier060 on Jan 11, 2021 9:56:11 GMT
Hi David I am sure there are others can help you with this as I am no expert only having A level physics, but I assume as you say, once the injector gets going and passes water into the boiler, there is not enough room between the vanes for water to pass so it slams the valve shut. Stroudley used this twisting action on his crosshead pumps on the first Terriers so that it had a lapping action on the valve and I suspect also threw off any debri that might have stuck to a non-rotating valve. Very ingenious!
|
|
|
Post by Roger on Jan 11, 2021 17:18:15 GMT
How does water flow over the vanes to move the cone? It either goes through the overflow gap or into the delivery cone doesn't it? Or does the pressure of the water going out somehow push the cone back? The other thing that struck me about the one in Dave's photo and the one in the diagram you posted is the severe path the water has to take out of the delivery cone. Do you think that has any effect, or given it's later slamming into a check valve held closed by boiler pressure it's of no consequence? I'm looking forward to seeing what you come up with and how you do it! When the injector picks up, you've got water above boiler pressure at the delivery end of the moving cone, and below atmospheric pressure at the inlet end. We know this is true because we normally have a valve there to stop air being sucked into the gap. With that sort of pressure differential, the cone is going to move away from the delivery end. Any excess water escaping around the moving cone as it moves will tend to impart a twisting force which will turn it slightly as it's floating on a film of water and there's next to no resistance to its motion. Once it's running cleanly, there's no flow part the vanes. Even if there was, the cone wouldn't turn because by then it would be held firmly against the valve seat in the inlet end. I think the inside of the outlet cover is curved to reduce the hydrodymamic losses. Personally I don't think it's that severe because the flow rate isn't huge.
|
|
|
Post by terrier060 on Jan 11, 2021 17:18:20 GMT
This is a rough drawing which I need to refine. Open for comments. Vanes not shown. I thought it might be possible to machine the vanes parallel and then twist the body before final machining? Moving mixing cone injector by ed cloutman, on Flickr
|
|
|
Post by Roger on Jan 11, 2021 17:32:55 GMT
This is a rough drawing which I need to refine. Open for comments. Vanes not shown. I thought it might be possible to machine the vanes parallel and then twist the body before final machining? Moving mixing cone injector by ed cloutman, on Flickr That's a neat arrangement. I presume there will be straight vanes on the moving cone to keep it central. Personally, I think I'd go with the dimensions and geometry you've already got for your injector, with a single taper through both Combining Cone parts to start with, and a flat nosed end regulated setup. That would make the moving cone shorter and everything simpler to make. There are so many differences in this design, I think you'd find the same issue I'm having in figuring out what's causing any problems. Another thought about alignment... you could have a 45 degree chamfer on the valve seats so that they self centre when pressed home. A fixture could hold the two together for drilling and reaming the tapers. I don't think the moving cone would need to be an especially close fit so long as it centres itself when it's running.
|
|
jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Jan 11, 2021 19:39:22 GMT
W H Pullen in his Injector book 2nd edition 1900 p. 81 has this to say (amongst a lot else):-
(Paraphrasing somewhat by me) "the loose cone is shown in the position it occupies when the injector is exhausting the water pipe, the steam being able to exhaust into the overflow channel, through the space between the combining cone and the loose cone... directly the water arrives and condenses the steam, the partial vacuum formed in the combining cone allows the loose cone to fall until its flange rests upon the combining cone, it being assisted and held there by the pressure of the atmosphere without. In this closed position the two cones form one long continuous cone."
|
|
jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Jan 11, 2021 19:50:30 GMT
The vanes or wings to the loose section of the combining cone are to guide it and allow any overflow from the gap with the delivery cone to escape.
They need not be swirly.
One would have to consider how a sliding part of the combining cone could be accessed easily if it seized up. In full size, the delivery cone is screwed in and removable after removing the screwed on end cap.
|
|
jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Jan 11, 2021 20:14:58 GMT
I dislike vertical miniature injectors of the types that have been designed/described/sold in the UK because of the difficulty of accessing the gap between the 2 parts of the combining cone to clean them of obstructions.
However, in the USA, Andrew Duarte of Eccentric Engineering markets a design that avoids all this problem. His range is something quite else - all vertical injectors - and sort of 'wow'! The price tag is also 'wow'! And the smallest is really for a large 7.25"g loco by UK standards.
But well worth looking at the website
|
|
|
Post by terrier060 on Jan 12, 2021 10:08:50 GMT
Thanks Roger and Julian. One advantage is that it is only the condensing cone that needs to be a push-fit. The mixing cone is an individual part and the steam and delivery (with it's chamber) cones need only be a light register fit so can easily be moved for cleaning etc. I agree a tapered valve seat may be an advantage.
|
|
|
Post by terrier060 on Jan 15, 2021 10:59:55 GMT
Here are the drawings. I have to wait for some new belting for the Quorn. The old belt keeps breaking with the large, heavy grinding wheel on. I think maybe the outside of the belt is contaminated with oil or aluminium dust and will not stick when I try to heat the ends together. Terrier Moving cone injector by ed cloutman, on Flickr
|
|
mbrown
Elder Statesman
Posts: 1,724
|
Post by mbrown on Jan 15, 2021 11:03:54 GMT
If your use of a sliding cone can produce a reliable small injector without a bulbous ball valve chamber on top, the possibilities for a much improved scale appearance are massive. This will be a tremendous boon to us all.
Good luck!
Malcolm
|
|
mbrown
Elder Statesman
Posts: 1,724
|
Post by mbrown on Jan 15, 2021 11:07:18 GMT
I meant to add that a small team on the Talyllyn are working on producing new injectors for all the locos there. The ones currently in use are Gresham & Craven 4mm injectors, with a sliding cone (with swirly vanes!) but they are quite old and are getting worn. The team is makings some modifications to the design so that it can be produced from stock material rather than complex castings. It is being written up in the Talyllyn News.
It is very similar to your project, just a bit bigger....
Malcolm
|
|
|
Post by steamer5 on Jan 15, 2021 12:21:49 GMT
Hi Ed, Having just read thru the link Nick posted, the moving cone has a helix machined on the outside, this would be required to keep the moving cone centered, your drawing doesn't show it.... but they are a work in progress! Are you planning on a helix or just straight vanes? Its going to be fun watching you make one of these!
Cheers Kerrin
|
|
|
Post by Roger on Jan 15, 2021 12:45:09 GMT
Here are the drawings. I have to wait for some new belting for the Quorn. The old belt keeps breaking with the large, heavy grinding wheel on. I think maybe the outside of the belt is contaminated with oil or aluminium dust and will not stick when I try to heat the ends together. Terrier Moving cone injector by ed cloutman, on Flickr Hi Ed, Are you following the throat sizes you've worked out for the conventional version? I don't think it will work unless you abide by those. It looks like there's something amiss with the clearances on the valve. As drawn, there appears to be a very small gap at the LH end and the moving cone is already almost as far right as it will go. Personally I'd also add a chamfer to the mating valve face on the Condensing Cone to make sure it centres properly.
|
|
|
Post by terrier060 on Jan 16, 2021 20:06:03 GMT
Thanks Malcomb. Is it possible to get a copy of the Talyllyn News? It would be very interesting to follow what they are doing. Roger - although you cannot see it in the drawing, the tapered front of the moving cone does engage in the condensing cone rather than the flat seat. Also the reason the moving cone appears to hit the delivery is because the way it is drawn you cannot see the stops on the front end of the cone. There needs to be a gap at the front end of the splines to allow flow through the overflow. If you remember the cone only seats at the front end so I have shown the valve open at the rear end of it's travel. You are correct about the throat sizes. I will have a look at those. I just wanted to get the bulk of the drawing done, so that there was something on paper to discuss. The belt has arrived for the Quorn so I can now make the cutting tools. Being an awkward sod I am going to try longer tapers (as per full size) rather than the traditional model short cones. Today had been making bits while waiting for the belt. I have made a box spanner to open the blowdown quickly, as it has been a tedious job with a spanner, getting fingers burnt! The Tormach makes making small tools and spanners to pleasing to do. I would never have bothered if I had to drill and file by hand (Julian will be grimacing as he is a traditionalist). I measured the size of the square and added a thou all round. It was a perfect fit. I drilled out the corners as the first operation, then pocketed with a 4mm endmill, followed by contouring with a 2mm endmill. Job done. The other job was to fit a mini quick-release connector to the blowdown. This makes filling the boiler a simple job without water spraying everywhere. The other end of the hose has a standard garden hose connector on it. Blowdown Box spanner by ed cloutman, on Flickr
|
|
|
Post by Roger on Jan 16, 2021 20:16:33 GMT
Hi Ed, That's a neat box spanner, I too drill out the corners, particularly on the very small sizes. I like the quick release hose fitting too, I use something similar on my paint spray guns.
On the injector drawing, the other point I was making as that the moving cone can't go much further right because it's almost hitting the Delivery Cone.
I'm not sure why the conical seat on the fixed part doesn't show in the section, normally you would see it but there appears to be a sharp corner there. Was the section taken before the chamfer was added?
|
|