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

[Roger]

I thought I'd explore the overflow arrangements, starting with a new design of valve that uses a 0.5mm thick Silicone sheet.

This is the little punch I made for the purpose...

20210121_130711 by Georgia Montgomery, on Flickr


... the washer is stretched over the head of the valve shaft.

20210121_131344 by Georgia Montgomery, on Flickr

The 4th axis was already set up, so I decided to use it in that orientation for the hex on the blanking cap.

Here's a video of that.


20210121_151023 by Georgia Montgomery, on Flickr

New Cone 6 is on the left, that's the one with the Delivery overflow holes moved slightly left.



20210121_155115 by Georgia Montgomery, on Flickr

Here I'm trying it with just the Delivery overflow valve, and that has too much restriction, so it won't start the syphon. I know it doesn't strictly need to do that, but I think it's desirable.


20210121_160826 by Georgia Montgomery, on Flickr


This is the other way around, and that didn't work either, again, not starting the syphon.


20210121_165248 by Georgia Montgomery, on Flickr

I'd already tried this Cone 6 which worked exactly the same as Cone 5, so I decided to repurpose it by removing the centre restriction to see if that made any difference.

20210121_174513 by Georgia Montgomery, on Flickr

It still isn't good enough, so that needs a little more investigation.

20210121_193812 by Georgia Montgomery, on Flickr

I thought I'd also give this design a go. It has the same total area of overflow holes, but spread along the length of the combining cone.
The idea is that it allows fewer holes in each row, so the gaps between them are greater. It also allows more room for the flow to stabilise between rows. I also wondered if it would help with restarting if any excess could escape wherever it needed to. It probably won't work, but if you don't try, you won't know!

Cone7 by Georgia Montgomery, on Flickr


Edit:- Ok, I'm absolutely stunned. I've just tried this new Cone 7 above, and it certainly works although I haven't tried it over a wide range of conditions. However, I've just run a whole 6 Litre bottle of water through into the catch tank, and decided to see if it would work with the hot water. To my surprise, it picked up straight away at 70psi, and when I checked the temperature of the feed water, it was 46C! I've just tried it at 90psi, and although it's not happy, it still feeds, although it's cooled down to a mere 44C!

Note that this is with one overflow valve with the new design of flat Silicone Valve.

[simplyloco]

Jan 22, 2021 10:40:46 GMT Roger said:

I thought I'd explore the overflow arrangements, starting with a new design of valve that uses a 0.5mm thick Silicone sheet.

This is the little punch I made for the purpose...

SNIP

I have to say Roger, that you have more time on your hands than I have...
John

[Roger]

Flushed with the surprise success of Cone 7, this is a further iteration to see if the larger vent holes at the upstream end are really necessary. You could argue that the pressure downstream is going to be greater than it is upstream, so the holes may not need to be larger. I have no idea.

Anyway, this is a simplified version with all of the holes being the same size ie 0.4mm

Cone8 by Georgia Montgomery, on Flickr

And here's the first use of my new aquisition, an ER32 collet on a hardened and ground square mount. You can also get these with a hexagonal mount.

I thought this would be a quicker way to do repeat jobs, and I can adjust it too if the runout isn't quite as good as I would like.

20210122_173311 by Georgia Montgomery, on Flickr

This is Cone 8 which has five staggered rows of 4 x 0.4mm vent holes...

20210122_205249 by Georgia Montgomery, on Flickr

... and this is a video of how long it took to swap it over.

20210122_205656 by Georgia Montgomery, on Flickr

[terrier060]

This is brilliant stuff Roger. R and D at its best. You may have to patent this before it is stolen by the full-size boys! It seems such a simple idea - but then thinking back aren't often the simplest and most elegant designs the one that last in history! I can feel the excitement that this develops!
Ed

[Oily Rag]

Jan 22, 2021 12:15:24 GMT simplyloco said:

Jan 22, 2021 10:40:46 GMT Roger said:

I thought I'd explore the overflow arrangements, starting with a new design of valve that uses a 0.5mm thick Silicone sheet.

This is the little punch I made for the purpose...

SNIP

I have to say Roger, that you have more time on your hands than I have...
John

He is not of this world, he has two hearts and a time machine

[Roger]

Ok, so far I've just explained what I've found, but you haven't seen any Steam and water flowing, so let's put that right.

Here's a video of Cone 8 being put through its paces. It's feeding the boiler as you can see from the handle on the 3-way valve. I've made that in a 'T' shape to mimic the port holes inside the valve. So in this position, the bottom of the 'T' is on the left ie the inlet, and the top of the 'T' is vertical so that the flow is blocked at the top, but it's connected to the boiler via the bottom outlet.

In this test, I've got both overflows connected. I tried this test again later with just one overflow valve, and the result was the same, so that's a blessing. There's a lot less dribbling from the overflow when there's just one valve for some reason.


20210122_213738 by Georgia Montgomery, on Flickr

This is just a short test with slightly hotter water. I'm really encouraged by these tests, I think it's very promising.

Ok, the 0.4mm holes might give some people nightmares, but they aren't that difficult to machine, especially in Brass.

20210122_214345 by Georgia Montgomery, on Flickr

[andyhigham]

I know this may be irrelevant but I do a bit of work with adiabatic coolers, these use a fine mist spray on the heat exchanger to allow cooling below ambient temperature (latent heat of evaporation)
I wonder if similar principles could be applied to our models?

[Roger]

Jan 22, 2021 23:01:39 GMT andyhigham said:

I know this may be irrelevant but I do a bit of work with adiabatic coolers, these use a fine mist spray on the heat exchanger to allow cooling below ambient temperature (latent heat of evaporation)
I wonder if similar principles could be applied to our models?

I suppose you could just water cool the body, either directly by passing feed water through it, or sprayed over it. Alternatively, I suppose you could submerge the injector in the feed water. I guess it's a matter of whether there's enough benefit in performance for the added complexity.
It's probably easier to design the injector to cope with higher temperature water. From my experiments, it would appear that some designs will work with feed water of 45C, and I doubt if that temperature would be reached unless you had a saddle tank with no insulation between it and the boiler, and you didn't draw much water from it. I'd be interested to know what the maximum feed water temperature is that people see on Tank Locomitives.

[Roger]

I've been looking on YouTube for Steam Injectors and this came up. Ok, this is for some kind of cleaning or spraying system by the look of it, with perhaps a feed of detergent fed into the centre of the flow with a lance. Looking at their web site, it doesn't look like it's intended for boiler feed water.

However, the overflows are done exactly how I've done them, and a single overflow valve is there too. So clearly I'm not the first to have trodden this path! 

[andyhigham]

I ran my sweet pea with pannier tanks for many years. 10mm space between the boiler cleading and the tank at the narrowest point. I never had an injector fail to pick up, but I always played safe and left the water valves open filling up at each stop

[Roger]

Jan 22, 2021 23:58:03 GMT andyhigham said:

I ran my sweet pea with pannier tanks for many years. 10mm space between the boiler cleading and the tank at the narrowest point. I never had an injector fail to pick up, but I always played safe and left the water valves open filling up at each stop

That's pretty much how I imagined it to be. The only fly in the ointment for my injectors is that the overflow valve would mostly stop any water flow, bar a little leakage since the spring isn't very strong. That's a double edged sword though, because it's that valve that's responsible for being able to use water at higher temperatures.
I honestly don't think it's going to be an issue on my Locomotive. I'm adding a generous layer of insulation behind the tanks, and all of the feed water has to come from the tanks if I'm using the injectors because they take their water straight off the bottom of the tanks. If I'm drawing water from the riding car, it still ends up in the Pannier Tanks, so unless I ran entirely using the Axle Pump, the water is constantly being used.

[steamer5]

Hi Rodger,
Great progress! Looks like the injector how to books are going to have a new chapter added!
On the water temp in the tank front, the boys from Aussie are the ones to ask on hi temp! I’ve seen articles were they add ice to the tanks to keep the temp down so the injectors will work! Pretty sure they will be keen on one that will work up in the mid 40’s!

Cheers Kerrin

[David]

Yes, we sometimes have to add ice to the feedwater, and hose the injector down with water to cool it. I would have been interested to see how the right injector worked on the red loco this summer because it does not have a water tap, just a straight through connection to the tender, so it has a constant stream of cooler water through it.

[racinjason]

David
I run my blowfly with the water on to both injectors all the time and just turned the steam on fast and they always picked up. I don't have any pumps of any kind.
Cheers Jason.

[Roger]

Thanks for all those comments, I can see how hot water would be an issue in hotter climes. A very light non-return valve on the overflow ought to help in those situations, although that does stop water freely flowing through the injector unless you pull on the valve to open it. That probably wouldn't happen if you used a gravity type of setup like you have on the other valve. Bob Bramson's book shows such an injector with a double valve on the top.

I'm convinced that it's necessary to have a valve on the water to get a clean delivrery at lower pressures. Interestingly though, whereas conventional Injectors seem to drop out and pick up very sharply, my variant doesn't do that. It tends to flutter rather than just instantly drop out, and picking up can be quite gentle and slow. I suppose the action of the usual ball valve is responsible for that. Not having anything to dampen the flow of water and steam in either direction is surely bound to alter the dynamics. Also, a conventional Injector lets any water coming out of the ball valve to drain straight away. I imagine that my setup leaves the inside of the Injector flooded with water, so when it starts to pick up, there's water at the overflows, not air.


Anyway, here's another iteration of the same idea, but this time using 3 x 0.5mm holes at each of the five positions instead of 4 x 0.4mm holes. That actually increases the overflow area from 2.01mm2 to 2.36mm2

Cone9 by Georgia Montgomery, on Flickr

I've also discovered that the clear food grade Silicone O-rings are too soft. Both of the ones I tried on the latest Injector cone had broken up, causing some very confusing test results.


I also found that much of the excessive overflow on the last two videos was due to the regulation gap being bigger than it needed to be.


There's still a long way to go before I'm happy that I've got something reliable. 

This cone has only three holes in each row, so I can't look right through from one side to the other to see if they are clear. However, my little LED inspection light makes that easy to see.

20210123_131513 by Georgia Montgomery, on Flickr

Cone 9 complete. Now I'm all set up and I know how to go about them, these only take half an hour to make from the pre-machined blanks. I've only got three more left after this one, so I guess I'll be making another batch soon!

I dread to think how long it would take to have made nine conventional sets of three cones ie 27 pieces! I can see why anyone would be reluctant to experiment that much when you consider the amount of work involved.

20210123_131749 by Georgia Montgomery, on Flickr

[ettingtonliam]

I freely admit I don't understand a word of the injector theory, or the programming and processes that produce the cones, but I am hugely impressed by Roger's persistence in developing these injectors.

Respect!

[Roger]

Jan 23, 2021 13:12:23 GMT ettingtonliam said:

I freely admit I don't understand a word of the injector theory, or the programming and processes that produce the cones, but I am hugely impressed by Roger's persistence in developing these injectors.

Respect!

I'm not smart, but I am very stubborn, and that goes a long way! Fortunately for all of us, the really hard work has been done by the likes of D. A. G Brown and Bob Bramson to name but a few, who have provided us with the absolutely fundamental necessities for a working Injector, namely the throat sizes and angles.
Although my designs look very different in many ways to what we're used to seeing, they still have to follow these figures else they won't work. However, outside of those fixed points, there's a surprising amount of wriggle room when it comes to the overflows. In reality, that's all I'm playing with, in the hope of making them easier to make and to better fit inside scale bodies.

[Roger]

Cone 9 didn't seem to perform as well as Cone 8. It seemed to struggle with the lower pressures, not wanting to keep feeding, regardless of what I tried with the water valve and the regulation gap.

I thought I'd give this a try which is in the other direction. This time the holes have gone down in size to 0.3mm 

Cone10 by Georgia Montgomery, on Flickr

This proved to have too much back pressure, which isn't that surprising. I either need to increase the number of holes or perhaps have larger ones in the first two rows. It might pay to make the final set bigger too.
Looking back at the position of the gap in conventional designs, I can see that I've got an extra row of holes upstream of that on these designs, and that's not going to help the Ejector portion of the design. It's probably worth bunching the holes up a bit and moving them further downstream.

So many options!

Anyway, a decision needs to be made, and here it is. This is back to four holes per row, but they have been squeezed downstream so that the leading edge of the first upstream hole is on what would have been the end of the Condensing Cone. With hindsight, this is what I should have maintained on the first version of this, and I think it's why the ejector action isn't satisfactory. 

You can also see that I've boosted the size of the Delivery Cone holes. I don't think there's any benefit from adding more restriction than necessary at this final stage where the fully formed jet is about to do its job.

Cone11 by Georgia Montgomery, on Flickr

[keith1500]

What is the lower pressure you are aiming hoping for?

Personally I wouldn’t worry about an injector that won’t run below 40psi. In some respects it’s an advantage; we have all done it. Come to a halt with a great fire going put on the injector to make use of that heat and then got distracted or into a natter about something. Next thing you know you have flooded the boiler! However, Speedy boiler is a good size and I often find the pressure drops before you over fill. Seeing the injector shooting water can be a reminder that you have just done something daft! Now with 40 psi and decent fire you can turn that around and no one will know.

[David]

One of my injectors still has a tap so I can restrict the water on that one. The one without the tap is also good for my youngest son. Now I have them working better, and a steam valve further away from the hot boiler and manifold, that side is easier for him to operate. Just one control and less chance of touching something hot.

If my pressure gets down to 40psi, I have to pull over into a siding to get i back together anyway, I'm not a good enough driver to do it on the run most times, esp with passengers. That means if I have any reasonable amount of water I can bring pressure back and then use the injector. If I was filling the boiler at 40psi, so lowering the pressure, I might not have enough steam for the blower to be effective so would be finished anyway with a dead fire.