|
Post by Roger on Dec 19, 2020 15:55:58 GMT
Here's the revised design to allow for internal adjustments. The Red spacer has been added to the delivery end, and the O-ring moved back so I can make that almost 1mm thick. The purpose of the spacer is to allow both the inlet and outlet covers to be fully screwed home. That's important because there are cutouts in the sides of those for the Steam and Water to get past them. The spacer will need to be made to the exact length so it compresses the cones together. It might be that a Delrin spacer is sufficient. That might be compliant enough to allow for a bit of deformation. Section injector assembly with spacer by Roger Froud, on Flickr
|
|
|
Post by 92220 on Dec 19, 2020 17:52:13 GMT
Hi Roger.
To keep the hose in the container, why can't you tie a piece of 3mm stainless welding rod, or similar, to the outside of the hose, to hold it straight? The stainless rod could also just be inside the hose, if it is big enough that the wire doesn't restrict flow. Even a piece of thin wall copper tube would do the same job.
Bob.
|
|
|
Post by Roger on Dec 19, 2020 18:33:12 GMT
Hi Roger. To keep the hose in the container, why can't you tie a piece of 3mm stainless welding rod, or similar, to the outside of the hose, to hold it straight? The stainless rod could also just be inside the hose, if it is big enough that the wire doesn't restrict flow. Even a piece of thin wall copper tube would do the same job. Bob. Hi Bob, Fortunately now I've straightened them with a hot air gun, they're happy to stay where they're put. I've added a third one for the overflow so condence the Steam back in the reservoir rather than filling the room with moisture.
|
|
|
Post by Roger on Dec 19, 2020 20:56:59 GMT
I've just made a new spacer which increased the Regulation gap from 0.17mm to 0.32mm and that's had a dramatic effect. It now picks up but the pressure it forcing a fine spray of water from the outlet flange so that's being stripped and examined to see why that is.
Anyway, this is a lot more promising as I can see the pressure gauge going up to somewhere near boiler pressure.
|
|
|
Post by andyhigham on Dec 19, 2020 22:19:37 GMT
High pressure at the outlet can only be a good sign
|
|
|
Post by Roger on Dec 19, 2020 22:58:33 GMT
Ok, that's sorted out the leak, but I've managed to break the handle of the 3-port valve because it's too stiff. I'll sort that out and try again. I also need to return the overflow to a separate container because it's making the feedwater hot of course, and that's clearly not very sensible. On the bright side, the boiler is working really well, I can set the control knob and forget about it. I've also struggled with the M1.4 bolts on the flanges because my Box Spanners are too chunky. To solve that problem I've designed a modified skinny version of the 2AF Box Spanner to be machined from a piece of 3mm Silver Steel. Skinny 2AF box spanner by Roger Froud, on Flickr It's too small to hold in the 3-jaw chuck, so I'm using the drill chuck which was fitted with a parallel shank for this purpose. 20201219_214353 by Roger Froud, on Flickr 20201219_214402 by Roger Froud, on Flickr The corners are peck drilled with a 0.3mm diameter drill to start with, then the middle is drilled out to 1.8mm 20201219_220655 by Roger Froud, on Flickr I guess I was pushing the Carbide PCB drill a bit hard because I saw a flash of Red out of the corner of my eye and the top went blue. Oh well, that doesn't matter. Here I'm using a 0.5mm Carbide PCB Burr to machine the hex 0.05mm deep at a time and about 4mm/min. I could probably go faster, but I might break the cutter, so what's the point. 20201219_224206 by Roger Froud, on Flickr
|
|
tony9f
Seasoned Member
Posts: 104
|
Post by tony9f on Dec 20, 2020 11:04:56 GMT
Hi Roger,
This is all looking very encouraging. I would suggest that an experiment into the injectors ability to pick up with warm feedwater might be a worthwhile exercise as well especially as you've got a tank engine.
Tony
|
|
|
Post by Roger on Dec 20, 2020 12:41:21 GMT
Hi Roger, This is all looking very encouraging. I would suggest that an experiment into the injectors ability to pick up with warm feedwater might be a worthwhile exercise as well especially as you've got a tank engine. Tony Hi Tony, Absolutely, that's well worth doing. However, I need to walk before I run and get it to work at room temperature conditions. This design is novel in several areas, so it might not be viable as a design as currently made.
|
|
|
Post by Roger on Dec 20, 2020 13:37:39 GMT
Ok, this is a lot more promising. The Regulation Gap is now 0.32mm and there's still a razor sharp edge on the entry of the Condensing Cone. The inlet water temperature is around 17C straight from the tap, so it's not going to be indicative of what's happening on the locomotive. Still it's early days with a lot more investigation to come. I've just swapped the water from the overflow with the feed water containers, and unsurprisingly it won't pick up since I'm measuring 50C water inlet temperature. Whether that would work if you had a longer Condensing cone to give it more time to condense, is open to debate. Here it's been running for over a minute at 80PSI... 20201220_125153 by Roger Froud, on Flickr ... and a short video so you can see what's happening. As you can hear, it's not silky smooth, but it does continue to run. 20201220_125246 by Roger Froud, on Flickr It picks up nicely as long as water is available at the inlet so it doesn't have to draw it up. That's understandable really, because Steam heats the whole assembly when there's no water there, and then it becomes more difficult to condense the Steam. It's interesting that it works with the sharp edge that the water has to pass over. However, this isn't the same as the situation with Annular Regulation where a smooth flow of water through the annular gap contracts onto the Steam in the centre, gradually condensing it. Here, the water flow is at 90 degrees to the direction of the Steam flow, ie it's coming in from all sides and being abruptly carried by the Steam forwards into the Condensing Cone. Maybe the edge has no effect because it's not a smooth flow, or alternatively perhaps it assists in causing a chaotic flow which condenses the Steam more effectively? Who knows. So there are many more questions than answers, and a very long way to go before I know whether this is going to be good enough. At the moment the boiler is too full, so I need to blow some of that down and try filling it using the injector instead of running the Balanced Clack valve to waste. Sorry it's more writing than pictures this time but there are a lot of theoretical considerations to air. One thing that I was concerned about was the Injector body filling with water that would somehow affect the delivery if the area around the Delivery Cone was drowned. However, that doesn't appear to be an issue. I guess the as soon as you get too much Steam, Water or both in the body, it just cracks the valve open to let it out. Oh, and one last thing to mention. I tried the injector with the outlet pipe completely blocked, and the delivery pressure went up noticeably.
|
|
tony9f
Seasoned Member
Posts: 104
|
Post by tony9f on Dec 20, 2020 15:18:01 GMT
I'm impressed, you've got it working pretty much from the off. Perhaps if you had a foot valve on the suction it might pick up and lift as there would be water available near the cones. Likewise I think water at 50 degrees is a step too far but who knows if you have a positive head which you would have on the loco.
|
|
|
Post by Roger on Dec 20, 2020 15:29:23 GMT
I'm impressed, you've got it working pretty much from the off. Perhaps if you had a foot valve on the suction it might pick up and lift as there would be water available near the cones. Likewise I think water at 50 degrees is a step too far but who knows if you have a positive head which you would have on the loco. Thanks Tony, there's still a long way to go though. I may make a simple header tank with a valve along the lines of what will actually be on the Locomotive. I think you're right about getting that water near the cones at all times. Steam heating the injector isn't a good recipe for it picking up. I think Bob Bramson says in his book that injectors with a suction valve on the overflow can work up to 40C. I think that's probably the limit unless you redesign the cones to give it more time to condense hotter water. In principle it seems to me that you ought to be able to go much higher. After all, it's only necessary to completely condense the Steam. I don't know what the temperature of the Steam is on the exit of the Steam Cone. At 90PSI, it would theoretically be 166C on entry, but of course it will be less than that after it has expanded. Maybe there's a thermodynamicist who's better at this sort of thing than I am.
|
|
tony9f
Seasoned Member
Posts: 104
|
Post by tony9f on Dec 20, 2020 15:41:40 GMT
It's temperature may not drop that much initially but will go into superheat as it is throttled at the cone. The other issue with picking up warm water is that it is likely to flash off under vacuum conditions and that miniature injectors just don't have the room to allow for the additional condensing to occur.
|
|
|
Post by Roger on Dec 20, 2020 16:31:45 GMT
It's temperature may not drop that much initially but will go into superheat as it is throttled at the cone. The other issue with picking up warm water is that it is likely to flash off under vacuum conditions and that miniature injectors just don't have the room to allow for the additional condensing to occur. Ok, that makes sense. I guess you've got approximately atmospheric pressure, or just below, in the body. I suppose that means you could work out how much water flow would be required to have enough capacity to condense the Steam. It amazes me that there's time to condense all of it under any conditions to be honest. There's certainly not much spare room in the 8X scale body. The most I could gain would be a millimeter or two.
|
|
|
Post by delaplume on Dec 20, 2020 20:17:49 GMT
Just a thought Roger but the injector on a locomotive is invariably fitted below the water supply, thus giving it a positive head....Wasn't that the original idea for this boiler ie}--- to test injectors in a working environment ??.......My apologies if I missed an explanation earlier on.
|
|
|
Post by delaplume on Dec 20, 2020 20:24:46 GMT
The Injector is basically a heat difference device so warm water ( and a warm or hot injector body )..will always be a no-no....
|
|
|
Post by Roger on Dec 20, 2020 20:42:22 GMT
Just a thought Roger but the injector on a locomotive is invariably fitted below the water supply, thus giving it a positive head....Wasn't that the original idea for this boiler ie}--- to test injectors in a working environment ??.......My apologies if I missed an explanation earlier on. Hi Alan, The original test was just to check that it would lift and prove that the Ejector section worked. If it would work like that it would be a bonus, but it's not needed as you rightly point out. I ought to put a valve on a tank of some kind and rig that up. To that end I've just sourced the necessary parts from eBay and also a 10mm to 4mm reducer to suit the little 12V pump that I used to fill the boiler through the clack. There's an absolute monstrosity of fittings cobbled together at the moment to achieve that!
|
|
jma1009
Elder Statesman
![*](//storage.proboards.com/forum/images/stars/star_red.png) ![*](//storage.proboards.com/forum/images/stars/star_red.png)
Posts: 5,903
|
Post by jma1009 on Dec 20, 2020 21:17:59 GMT
Quite wonderful, Roger!
Very well done and congratulations!
In my own sphere of interests, extremely significant!
Cheers, Julian
|
|
|
Post by Roger on Dec 20, 2020 21:51:51 GMT
Quite wonderful, Roger! Very well done and congratulations! In my own sphere of interests, extremely significant! Cheers, Julian Thanks Julian, but it's a bit too early to claim success. It still feels like I'm in the foothils, figuring out the symptoms each change produces. Still, it's showing some promise, so maybe it will work well enough in the end with enough development. There are so many things I can experiment with, but I need to try to home in on the sweet spot of what I've got, one parameter at a time. All very interesting. Part of the development process is in experimenting with ways to make the changes as easily and quickly as possible. Here's the Injector being stripped in situ, with the end covers removed and the Delrin pusher having just been used to slide out the cones and spacer. That's two parts, soon to become three so that I can accommodate different regulation gaps. 20201220_211857 by Roger Froud, on Flickr
|
|
|
Post by Roger on Dec 20, 2020 22:27:03 GMT
It's proving difficult to make the Steam Cone Spacer to an exact length, so I've designed this simple fixture with a shouldered pocket at a known depth. This will make it possible to just part them off over length, insert them and set the length relative to the end of the fixture. That can be done on the lathe or the grinder. Steam cone spacer fixture by Roger Froud, on Flickr I tried a third setup with a slightly greater Regulation Gap, and that made things worse, so it would appear that I needed to adjust it back the other way slightly. To that end, I've made a new spacer that's 50 microns shorter (2 thou) but I'll now need to increase the shoulder on the One Piece Cone and made a thin spacer for the other end. Up to this point, I've just been reducing the length of the end cap, but now I need to go back the other way.
|
|
milky
Seasoned Member
Posts: 120
|
Post by milky on Dec 21, 2020 9:23:08 GMT
I haven't been around for a while, but I thought I'd look in to find out what the “Froude” has been up to. I wasn't disappointed, the thread as always is fascinating but now we are on to injectors ! So once you have reached ‘Base Camp’ (to use your climbing analogy) can you play around with pressures and the range of the injector. I have some that will pick up at around 40 psi and others that won’t go below 60. Each injector has around 40 psi of working range. It makes me wonder if the ‘Range’ could be extended ?? Hope this will lead to the knowledge to enable them to work in the 100-150 range
|
|