|
|
Post by Roger on Dec 27, 2020 22:38:09 GMT
A bit of progress with the injector. I've had it picking up at high pressure having reduced the Regulation Gap, but I still think it needs to be slightly less. I've also been experimenting with another valve idea for the gap, this time using a floating O-ring that moves axially. The valve O-ring is the sectioned Violet one in the middle of the picture. As shown, it's in the RH position so that it's wedged by the chamfered middle portion, causing it to seal against the outside diameter.  O ring valve section by Roger Froud, on Flickr The idea is that it moves left when Steam and Water come out of the gap, allowing that to pass under it and through the slots in the retaining collar. I've deliberately used the same number of slots as there are holes next to it, on the left, so the water doesn't flow over them. Those are the holes for the Delivery overflow Hopefully that will leave enough cross sectional area for the necessary volume. Again, hopefully the reverse flow will pull the O-ring back into position. Anyway, I won't know until I try it.  One piece cone with axial valve by Roger Froud, on Flickr I still want to experiment with the Regulation Gap more before I do any more changes to this. |
|
jma1009
Elder Statesman
Posts: 5,922
|
Post by jma1009 on Dec 27, 2020 23:07:33 GMT
Hi Roger,
The annular gap between end of the Steam cone and the start of the combining cone ought to be easy to establish on your computer as an area, and cross referenced using your computer (or CAD) with the table in DAG Brown's book for your designed output. (I tend towards a slightly more generous figure than he does, but that might not apply to your restricted overflow design).
If it had been me, I would first have tested your cones in a traditional injector body, then taken out the classic check valve ball above the 2 halves of the combining cone once the above had been proved, and then adding the sprung overflow check valve instead.
Cheers,
Julian
|
|
jma1009
Elder Statesman
Posts: 5,922
|
Post by jma1009 on Dec 27, 2020 23:19:18 GMT
(I've never had a problem establishing the annular gap between steam cone end and start of the combining cone. But then I've got that file of detailed notes you saw in January)
|
|
dscott
Elder Statesman
Posts: 2,440
|
Post by dscott on Dec 28, 2020 0:34:36 GMT
We are all sending you support in kind over the internet Roger.
Much experimenting done in most breakthroughs over the years.
Suddenly you hit it and the rest is easy.
Much of my working life with Students was to help them achieve the impossible.
Claim to fame of making a modern Mobile and a Tablet for a forward concepting Student 10 years before the technology caught up.
He famously got slated by the tutor, and me for making the model in 10 minutes. So sorry about the buttons guys.
I suggested the corners being more rounded!!
David and Lily.
|
|
|
|
Post by Roger on Dec 28, 2020 8:56:06 GMT
Hi Roger, The annular gap between end of the Steam cone and the start of the combining cone ought to be easy to establish on your computer as an area, and cross referenced using your computer (or CAD) with the table in DAG Brown's book for your designed output. (I tend towards a slightly more generous figure than he does, but that might not apply to your restricted overflow design). If it had been me, I would first have tested your cones in a traditional injector body, then taken out the classic check valve ball above the 2 halves of the combining cone once the above had been proved, and then adding the sprung overflow check valve instead. Cheers, Julian Hi Julian, It certainly would have been more sensible not to go straight in with lots of different new features. I worked out what I thought the regulation gap should be using the cross sectional area of the annular gaps in D. A. G. Brown's book as a starting point, but I don't think you get the right answer for End Regulation. The resistance to flow seems to be different with End Regulation geometry. I may be wrong, but that's how it looks at the moment. Currently I've got a square end to the Steam Cone and no lead in to the Condensing Cone. Both of these things might prove to be needed, but I wanted to start with something as simple as possible. The way the water and Steam mix is very different in this arrangement, it's not a nice smooth flow, it's much more turbulent to start with. This may be completely the wrong approach, but it's interesting to just try things and see what happens. If I can get it to work after a fashion when these conditions are less than ideal, adding those changes ought to make it easier. I really want to see what difference each change makes and how important they are individually. |
|
|
|
Post by Roger on Dec 28, 2020 22:56:39 GMT
This is the new One Piece Cone first operation. All of the diameters were machined with the 1.1mm wide parting tool. The through hole was drilled at the same time.  20201228_120552 by Roger Froud, on Flickr I've kept the holes for the Combination Cone the same, but decided to drop the size of the ones at the Delivery end from 1.1mm to 0.95mm because I thought the gap didn't need to be that big there.  20201228_153351 by Roger Froud, on Flickr This is a 1mm cutter making the slots on the O-ring stop piece.  20201228_163559 by Roger Froud, on Flickr  20201228_170452 by Roger Froud, on Flickr You can see there's a sizeable gap underneath the O-ring, but there was too much restriction and it wouldn't draw a vacuum.  20201228_171634 by Roger Froud, on Flickr I went through several stages of widening the gap, but it didn't really help...  20201228_175126 by Roger Froud, on Flickr ... so I broke off every other lug on the O-ring stop to see if that would help.  20201228_210926 by Roger Froud, on Flickr That allowed it to pull a vacuum, but I think there's too much restriction when the water reaches the cone too. So that didn't really resolve anything. I could reduce the inner diameter a bit, and that might make enough difference, but I'll probably have to make another one to try that out. I've had this cone deliver some water to the boiler with the O-ring removed, but it's not stable and won't work at low pressures. I might experiment with the lead in to the Condensing Cone which is currently a sharp corner. I may also round off the nose of the Steam Cone. In all of this I've come to the conclusion that 0.16mm regulation gap is in the right ball park. I've managed to get repeatable results, albeit failures in the same way! I've figured out the best way to remove and re-insert the cones and spacers without damaging the O-rings. It only takes a couple of minutes now I'm practiced at it, so I can get more tests done. There's a little way more to go before I try a different tack, possibly with a chunky experimental body just for test purposes. |
|
|
|
Post by 92220 on Dec 29, 2020 9:18:55 GMT
Hi Roger.
Super bit of machining there!!
If anyone else wants to look closer at the machining, Left Click on the photo and it comes up in a new window full screen.
Bob.
|
|
|
|
Post by Roger on Dec 29, 2020 18:40:41 GMT
Next up is trying to smooth the flow with a fluted entry to the Condensing Cone as shown below.  Cone with fluted entry by Roger Froud, on Flickr This is a piece of 4mm diameter Silver Steel which has a 0.8mm radius on it from the Carbide tipped lathe tool shown. I've aimed at a diameter which gives a full 0.8mm radius on the whole corner.  20201229_130924 by Roger Froud, on Flickr I filed most of the material away then ground the rest...  20201229_115204 by Roger Froud, on Flickr ... before hardening and tempering.  20201229_164031 by Roger Froud, on Flickr  20201229_165407 by Roger Froud, on Flickr This does seem to have helped, and it does pick up a little better. However it's hard to know what's happening inside and why the pressure seems to be just below what's needed to feed the boiler. It makes me think that the proportions aren't quite right. I need to chew a few things over and decide what's the best way to proceed. Ideally I'd like to see where the overflow is coming from. The fact that they're both coming out of the one pipe means you can't tell. Ideally I'd like a separate overflow for the two and a way to adjust the regulation gap while it's running. That would make life a lot easier. I could also use a water valve up close to the injector. The one I've bought doesn't look suitable for fine control. At the moment I'm just folding the feed hose to create a restriction. There are so many ideas to try, but without being able to see their effects more clearly, I'm working in the dark. |
|
|
|
Post by chris vine on Dec 29, 2020 20:40:27 GMT
Hi Roger,
If it is picking up and establishing a flow, albeit through the overflow (mostly?), is this the time to try the experiment of restricting the water supply and see if that helps.
It might not be the proper solution, but it just might give you a clue as to what to adjust next.
Another thought would be to drill some tiny tappings into the body, at two or three positions along the length, for pressure gauges. They, again, might tell you something. Because there is no flow through a gauge, they might not interfere with the operation of the injector??...
Chris.
|
|
|
|
Post by Roger on Dec 29, 2020 21:08:37 GMT
Hi Roger, If it is picking up and establishing a flow, albeit through the overflow (mostly?), is this the time to try the experiment of restricting the water supply and see if that helps. It might not be the proper solution, but it just might give you a clue as to what to adjust next. Another thought would be to drill some tiny tappings into the body, at two or three positions along the length, for pressure gauges. They, again, might tell you something. Because there is no flow through a gauge, they might not interfere with the operation of the injector??... Chris. Hi Chris, Those are some interesting ideas and thoughts. I'm not sure it's establishing a flow though, I don't see it filling the boiler. The pressure seems to be just under what's required to do that, so it may be what I'm seeing is overflow that's stalled at the point where it should be entering the delivery cone. To my way of thinking, if you block the outlet, you ought to be able to see the static pressure rise above boiler pressure, but I'm not seeing that. It may be that I've got something wrong with the proportions, so there's not enough kinetic energy in the flow. If I can solve that problem, it might behave much better. I'm not sure how you would tap into the cone without disrupting the flow, but it would certainly be useful to know what the pressure is along the cones. One interesting thing is that sometimes when it picks up to a high pressure, you can see the pressure slowly rise, as if the ball valve is being pulled closed and the steam inside the injector condenses. A lot of the time, there are violent oscillations, threatening to destroy the pressure gauge! Of course, it may be that it's simply impossible to just use a single non-return valve on the outlet, but I'm not convinced yst that this is the case. |
|
|
|
Post by Oily Rag on Dec 29, 2020 21:33:21 GMT
My Wife decided that we should do a family 'Secret Santa' for the first time this year, so I did all mine online a few weeks ago. No visits to the shop for me, and far less presents to buy. That's my kind of Christmas! There's normally far too much waste and buying of things for the sake of it, so this is a welcome change. Long may it continue! I've also not sent a single Christmas Card this year. I honestly don't think people care if they get a card or not, and writing them is a real chore. These days we're in touch so easily by email or instant messaging, and you can say so much more with much less effort. I Maybe the Christmas Card has had its day? |
|
|
|
Post by steamer5 on Dec 29, 2020 21:37:00 GMT
Hi Roger,
If you google laboratory hose clamp you will find something to help adjust your water flow. You want the square shaped ones, I see now that there are also C shaped, didn’t have those in my day.
My dad has my DAG’ Browns injector book from memory, I sure Julian would be able to tell us, but doesn’t the steam cone also have a bell mouth on it as well?
Keep at it you’ll beat it!
Cheers Kerrin
|
|
|
|
Post by Oily Rag on Dec 29, 2020 21:41:09 GMT
Crikey. What a model engineering thread this is. I am struggling to follow and comprehend but it is marvelous. Bravo Roger, lots to learn here.
|
|
|
|
Post by Roger on Dec 29, 2020 22:23:10 GMT
Hi Roger, If you google laboratory hose clamp you will find something to help adjust your water flow. You want the square shaped ones, I see now that there are also C shaped, didn’t have those in my day. My dad has my DAG’ Browns injector book from memory, I sure Julian would be able to tell us, but doesn’t the steam cone also have a bell mouth on it as well? Keep at it you’ll beat it! Cheers Kerrin Hi Kerrin, Thanks for that suggestion, that's a simple fix for the regulator problem. I suppose I could use one of my miniature Toolmaker's clamps like that too. You're quite right that most people do indeed have a taper and bell mouth for the Steam Cone. However, Bob Bramson say's that's unnecessary. However, I may follow suit just to see if it makes a difference. |
|
|
|
Post by Roger on Dec 29, 2020 22:25:08 GMT
Crikey. What a model engineering thread this is. I am struggling to follow and comprehend but it is marvelous. Bravo Roger, lots to learn here. Thanks Dazza, It's not just you struggling to keep up, count me in too! |
|
|
|
Post by Roger on Dec 30, 2020 20:25:19 GMT
This is as far as I've got with a special Test Injector design which is intended to explore the One Piece Cone design so I can understand what does and doesn't work in a more controlled way. It looks a bit complicated, but that's because I have some very specific requirements as follows... 1) Be able to adjust the Regulation Gap while it's running 2) Allow for disassembly without changing the Regulation Gap 3) Be able to measure the Regulation Gap once it's been set. 4) Leave the pipework assembled when changing the cones 5) Have separate overflows for the Combining Cone and Delivery zones 6) Be able to combine the two overflows into one Although it would be nice to see inside, I'm not sure I would be able to get the surface finish of the body good enough to see. It's going to get scratched and stained, so I'm inclined to just make it from Brass. The Steam Cone is unlikely to change very much, so I've made that as one piece with the adjusting arrangement. The Red Steam Cone have two O-rings arranged so that they don't foul the threads and can move in and out as the Cone is turned in the sleeve. The Bronze coloured sleeve is threaded in and again uses O-rings and a thread which pulls the head hexagon against the body. The RH end of the sleeve is is the stop for the One Piece Cone which is pressed against that from the other end. No spacers are required, the end cap will have O-rings that have enough travel to accommodate any changed in length of the cone. I've still got a lot of work to do on the overflow arrangement. The One Piece Cone will have an extra O-ring in between the two rows of overflow holes to separate the overflows. There isn't much room there, so it's not easy to bring them out.  Development injector assembly by Roger Froud, on Flickr As modelled, the OD of the body is 16mm and that might end up bigger. We'll see. There's still a lot of work to do, but I thought I'd share the process as it develops. Anyway, I've not been idle! |
|
|
|
Post by chris vine on Dec 31, 2020 11:49:30 GMT
Hi Roger,
Following your comment about blocking the output and (hoping to) see the pressure rise: I think this is wrong, if I have understood you correctly.
If you block the output, all flow through the injector will stop and the flow which makes it work will fall down.
Maybe you were suggesting partially blocking the outlet and seeing the pressure beginning to rise?
Certainly, an adjustable injector should tell you lots in a short space of time - apart from making it of course!!
Chris.
|
|
|
|
Post by John Baguley on Dec 31, 2020 14:46:55 GMT
Hi Chris,
My understanding is that if you block the output of the injector it will still function but all the water just comes out of the overflow. I definitely read that somewhere but can't remember where !
John
|
|
|
|
Post by Roger on Dec 31, 2020 14:51:26 GMT
Hi Chris,
My understanding is that if you block the output of the injector it will still function but all the water just comes out of the overflow. I definitely read that somewhere but can't remember where !
John
Hi John, That's my undestanding and what appears to happen. I don' think the flow coming out of the Mixing Cone knows what happens to it next unless there is too much restriction in the overflow, causing the flow to stall. I don't think that happens. |
|
|
|
Post by Roger on Dec 31, 2020 14:53:59 GMT
Hi Roger, Following your comment about blocking the output and (hoping to) see the pressure rise: I think this is wrong, if I have understood you correctly. If you block the output, all flow through the injector will stop and the flow which makes it work will fall down. Maybe you were suggesting partially blocking the outlet and seeing the pressure beginning to rise? Certainly, an adjustable injector should tell you lots in a short space of time - apart from making it of course!! Chris. Hi Chris, I think John is right on this point, I think you should get the maximum pressure with zero flow through the Delivery Cone. As long as all the water can escape through the overflow, I don't see how it can stall. Making the adjustable injector is far easier than designing it! |
|
