7.25 inch gauge Milner Hunslet

[terrier060]

I made the two cones today so now comes the test. I shall steam up tomorrow and see if it works!

Injector cones by ed cloutman, on Flickr

[Roger]

This is a clever adaptation of the One Piece Cone. Ed is using that RH 'O' ring as a spring to allow the Regulation Gap to be altered while it's being operated. After some consideration, I think that 'O' ring would be better on the Steam Cone end because there are no other 'O' rings there to restrict the free movement of the cone.

The idea is to adjust the Regulation Gap by tightening the Union nut, and then measure the distance over the union nuts so that the required spacer can be made in place of the 'O' ring. You don't need much movement to adjust End Regulation.

Being able to adjust the Regulation Gap when it's being used is a game changer. It removes the guesswork of whether the gap is too big or too small, and endless messing about with shims and repeated dismantling of the injector.

i really like this design, it's simpler than the one I proposed, and I think there's a good chance it will work. Obviously just scaling up an injector might not work if the change in scale is huge. However, this isn't so different to the one I made, so it may well be successful.

[terrier060]

I made another injector today. I took a bit more time over it and making the first one helped me to improve my technique. It is smaller with a 3/8in body rather than 1/2in. I hope to try them out tomorrow if I have time, but we are putting an extra watering station in at the club as we have added to our steam collection and have got several 5in and 7.25in locos now. It can get quite busy for water! When I joined the club three years ago there were mostly electric or diesel-electric locos there. Steam is now catching up, though the reliable non-steam ones do most of the passenger hauling and can tackle our steep gradient.

[terrier060]

I have just resurrected the Terrier thread which got lost on page 7. I was passenger hauling today and thought some feedback on my vacuum ejector system might be interesting to some of you - so see the Terrier thread. The Froud injector design really got me excited and I am determined to try and get it working. Roger is a lot more positive than me that it will work, so I thought I would take you along with me as I progress. It basically is trial and error until I can get the right proportions. The basic proportions are for a 26oz injector so I started by exactly scaling up the successful one that Roger has built, but of conventional design as far as pipework goes, so that it fits the Hunslet existing pipework.

The most important dimensions are the throat sizes and the regulation gap. For a 26oz injector both Brown and Bramson use the following:
Steam cone No.56 drill
Delivery No.66 drill (this relates to the throat of the single cone design of Roger Froud).

The General arrangement shows the silicone valve on the overflow outlet and the important moving single cone on the delivery side. The compression of the Viton O-ring allows the regulation gap to be adjusted by loosening and tightening the delivery union nut, the single cone being free to slide in the body of the injector and sealed by the two Nitrile O-rings.

The YouTube video shows the failure of the injector, but Roger has suggested that the eight holes may be too large, even though they were scaled up correctly from his 16oz design. So I am making two new cones, the first to exactly the same design but with eight No.63 drill holes rather than No.56.

The second cone is based on Brown's 26oz injector measurements, and you can see that the placing of the eight holes is further along the 9 degree taper as is the throat, from the mouth of the cone. It also uses eight No.63 drill holes. This is equivalent to the distance between the condensing and mixing cones.

GA of injector by ed cloutman, on Flickr

Hunslet cone design by ed cloutman, on Flickr

Video of Injector trying to pick up as tghe regulation gap is very slowly closed.

youtu.be/6hSRLYWpsKA

[jma1009]

Hi Ed,

I watched your YouTube clip of the injector a number of times.

It is extremely difficult to provide any opinion as we can't see whether you are adjusting the injector rear cone, and obviously this is also to Roger's unique design.

I would comment that the annular gap for this size of injector is specified by Bob Bramson with his type of steam cone.

More generally, the injector in the YouTube clip struggles to pick up which if we consider all other things being ok suggests a deficiency of steam supply/too much water inlet.

If the YouTube clip then shows you making an adjustment then potentially you go too far too quick.

Perhaps your water valve is too generous? Has it been regulated?

Very difficult to judge, except the annular gap for this size is known (I've done it myself for a Bramson type steam cone nozzle) so you shouldn't have to mess about with this.

Whether your version of Roger's special valve is operating as it should in Roger's own YouTube clips I would not be able to provide an opinion.

My initial thought was not enough steam of high enough pressure to the steam cone.

Did you check the temperature of the overflow water? And the boiler pressure?

Cheers,
Julian

[Roger]

Apr 24, 2022 20:15:06 GMT jma1009 said:

Hi Ed,

I watched your YouTube clip of the injector a number of times.

It is extremely difficult to provide any opinion as we can't see whether you are adjusting the injector rear cone, and obviously this is also to Roger's unique design.

I would comment that the annular gap for this size of injector is specified by Bob Bramson with his type of steam cone.

More generally, the injector in the YouTube clip struggles to pick up which if we consider all other things being ok suggests a deficiency of steam supply/too much water inlet.

If the YouTube clip then shows you making an adjustment then potentially you go too far too quick.

Perhaps your water valve is too generous? Has it been regulated?

Very difficult to judge, except the annular gap for this size is known (I've done it myself for a Bramson type steam cone nozzle) so you shouldn't have to mess about with this.

Whether your version of Roger's special valve is operating as it should in Roger's own YouTube clips I would not be able to provide an opinion.

My initial thought was not enough steam of high enough pressure to the steam cone.

Did you check the temperature of the overflow water? And the boiler pressure?

Cheers,
Julian

Hi Julian,
You can't use the annular gap from other texts because this design uses End Regulation. The reason for this is because concentricity is less critical and the Steam Cone is more robust. It does mean that you need to be able to adjust the gap to find the optimum setting.

My opinion is that the 8 holes that replace the overflow gap between the Condensing and Mixing cone are too large. Ed has scaled these up in proportion, but that's not how that's done on conventional Injectors. There's no difference in the gap when going from 16 to 26fl oz/min sizes, and the same should be the case here. So the holes are currently 1.2mm and they should be 0.9mm
That makes a big difference to how the Water jumps the gap.

[terrier060]

I made the two new cones today. Top picture shows drilling through the cone with a No. 66 drill. I just use the handwheel on the Myford tailstock, and with frequent withdrawal to clear the swarf, the drill goes through without wandering. There is plenty of 'feel' through the handwheel. I used a fairly high speed on the Myford, but not its highest, which takes a heavy load to wind up to speed with the heavy 4-iaw chuck! I was careful to use my dial gauge which measures to 10ths of thou to centralise the workpiece.

The middle picture shows the setup on the Myford mill. I used the Myford dividing head and centralised the work in a Burnerd Grip-Tru chuck which I was lucky to get while I was apprenticed there in the 1960s.

The bottom picture shows the cone I used in the test failure on the left, with the same cone with eight smaller holes in the middle and the cone to Brown's dimensions on the right. I will test them next time I steam up and take a video.

Machining the cones by ed cloutman, on Flickr

[terrier060]

Thanks for the comments above - I posted this before reading them. As you can see I now have the same cone with the smaller hioles as you suggested to me during our phone conversation Roger. I also thought I would try making a cone using Brown's dimensions. We did not chat about this Roger so I would be interested to hear your views. They both fit into the same test body. I will try and take a video showing me adjusting the cone next time.

Regarding the adjustment of the regulation gap Julian, this can be done very gradually, thousanths of an inch at a time, by screwing the delivery nut tighter against the Viton O-ring as I stated above. Roger has told me that this is a critical adjustment and can only be done under working conditions. Regarding boiler pressure, the safety valves lift at 100psi, but I run at about 80 to 90psi. Roger assures me that when the injector proportions are correct it should pick up from about 50 to 100psi when either the steam or water are turned on first. The water in the tank was cold as it was straight from the mains, and this time of year that is pretty cold! It had not had time in the saddle tank to heat up. So it is just trial and error and making small adjustments until I get it right or give up trying!

[terrier060]

I should add Julian that I did not try adjusting the water. I just opened the steam valve fully and the water fully and only adjusted the regulation gap. I sort of got mesmerized on the gap adjustment and forgot to try and adjust the water as we do on conventional injectors! It may have given a clue as to what was wrong with it, if it had picked up. Trouble is that Roger has given us so much confidence with his design that I do not think he needs to adjust the water - is that correct Roger?

[jma1009]

Bob gives the gap for end regulation for this size of injector! I've used it and checked it! No need to mess about with Bob's end regulation gap for a 26 oz per min injector!

Cheers,
Julian

[jma1009]

"I should add Julian that I did not try adjusting the water."

Well, there we are. Slight hint on a 7.25" Milner Hunslet try throttling the water valve. Been there, done it, got the injectors working when no one else did.

[Roger]

Apr 24, 2022 22:42:02 GMT jma1009 said:

Bob gives the gap for end regulation for this size of injector! I've used it and checked it! No need to mess about with Bob's end regulation gap for a 26 oz per min injector!

Cheers,
Julian

Hi Julian,
I couldn't make head nor tail out of the way Bob tried to explain the gap. In my experience, it's a critical measurement. If you try to just set it to a value, you'll be there forever trying to get this to work. It makes life so much easier to be able to adjust it while it's running.

[Roger]

Hi Ed,
I'm curious to know why you didn't use a PCB drill for the cross holes, it saves having to use a Centre Drill and is much less prone to wandering.

You can't really translate the Condensing Cone length from an Injector designed for Annular Regulation to one for End Regulation. That's because the Steam Cone enters inside the Condensing Cone using Annular Regulation, and that effectively shortens the cone.
It may be that this part of the Cone needs to be made shorter. During all of my many experiments, I didn't change the downstream parts very much, ie the thoat of the Delivery Cone area and the overflow there. None of that seems to be super critical. It's definitely the size, spacing and position of the 8 holes that's important. I didn't change the overall length that I started with because it seemed to be almost working from the start.

[terrier060]

Point taken Roger, but seeing as I have made it as well as the modified one to your measurements, there is no harm in trying it. Regarding the PCB drill, I did use my smallest one to start the holes, but I simply have not got one small enough. Also your reamer is just showing signs of losing it's edge, but it has done wonderfully well and I should not have been able to carry out these experiments without it. The small drills are remarkably good if you take your time with frequent withdrawals. The cones are nearly an imch long and the drill comes through with no discernable runout. So I think the very short length through the eight holes should not be a problem for the No.63 drill which is larger than the one through the cone. I did as you recommended and drilled opposites as I went round, rather like I remember tightening a car engine head down! Gosh! that was a long time ago - my son's mini I think it was when he drove it without any water in the radiator!!

[Roger]

With regard to partially closing the water valve, the Injector regulation is set up to work with a dry overflow at the maximum working pressure and ambient water temperature. Just to be clear, Regulation is just setting a gap or any other internal feature to limit the maximum amount of water that can flow. When the Steam pressure is lower than the design value, there is less water delivered and also less drawn in. However, those things don't exactly balance each other out, and the result is that you end up with more water being drawn in than can be delivered. Water starts to overflow from the Combining Cone gap, and this situation eventually results in the injector stalling altogether. Restricting the water flow externally by closing the valve rectifies this situation which often is all that's required for the injector to pick up again.
So in a nutshell, if there's water coming out of the overflow or you have to regulate the water to get it to pick up, the internal regulation gap is too large.
Sometimes you might want to set it up this way though, because this is how it needs to be when the feed water is hot. You can set up one injector with a larger regulation gap which will be happy to work with hot water, but at the expense of needing to throttle the water using the valve when the pressure drops if working with cool water. Injectors are optimised for one set of conditions only.

[Gary L]

Apr 25, 2022 15:32:29 GMT terrier060 said:

Point taken Roger, but seeing as I have made it as well as the modified one to your measurements, there is no harm in trying it. Regarding the PCB drill, I did use my smallest one to start the holes, but I simply have not got one small enough. Also your reamer is just showing signs of losing it's edge, but it has done wonderfully well and I should not have been able to carry out these experiments without it. The small drills are remarkably good if you take your time with frequent withdrawals. The cones are nearly an imch long and the drill comes through with no discernable runout. So I think the very short length through the eight holes should not be a problem for the No.63 drill which is larger than the one through the cone. I did as you recommended and drilled opposites as I went round, rather like I remember tightening a car engine head down! Gosh! that was a long time ago - my son's mini I think it was when he drove it without any water in the radiator!!

Big advantage of PCB drills is that they don’t wander when starting. For repetition drilling of accurate small holes this can be a big timesaver. But if the workpiece can flex, however little; or for ‘freehand’ drilling (by which I mean with the workpiece not firmly bolted down - we shouldn’t, but I bet we all do it occasionally!) all you will get is a broken drill.

This self-starting ability means that they are particularly good for drilling radial holes on cylindrical workpieces- probably the reason for Roger’s recommendation. In conjunction with a DRO (for centralising) they are beyond excellent for drilling those cross holes in cotter pins. Injector cones too, I shouldn’t wonder

Gary

[jma1009]

My initial observations of the YouTube clip was that the injector was subjected to an inadequate supply of steam for the steam cone jet size (which at No. 56 is standard for this size), and too much water.

You also have to consider whether the water is providing a 'head' which seems to be the case in this example.

We would normally not have this.

I did look again at the Bob Bramson 24 oz per minute injector for end regulation.

Cheers,
Julian

[terrier060]

Will try and make the video clearer when I do the next test on the two new cones. I will try performing the tests in a sequence if possible, first with both water (which has a head) and steam fully on and at about 80psi as near as I can keep it (lot to think about), while I adjust the regulation gap. Then I will try regulating the water. If that fails I will try higher and lower boiler pressures while regulating the water.

[terrier060]

Here is one solution, but not a complete cure to the bouncing (or to use the latest F1 term 'Porpoising' - actually I do not drive it that fast for ground effect to come into play) of my Hunslet! The main problem was the initial springs were too short so I replaced them with springs which were a bit stronger and 1/4in longer. This made quite a difference. The other problem was that the rear axle weighed in at 120kg and the front 100kg. I think being tail-heavy is what may have caused it to derail. It does still bounce, as the combination of a heavy full cab, plus the copper firebox at the rear end, and the cylinders, smokebox and long chimney at the front, both overhanging the wheels, is the worst scenario for stability. Nick's Hunslet has a short chimney and open cab which may explain why his is more stable. So to help cure the problem I have been running with a 12kg ballast weight fixed to the front buffer.

What you see below are the buffers which are going to be bolted to the buffer beam, to represent the old wooden buffers that the quarries used to shunt trucks along. They appear to have been square blocks, probably of oak, bolted to the front beam with coach bolts. Some had a steel or iron plate covering the front. My intention is to fit a 1/8in plate to the front which is about 1/8in wider than the mild steel blocks. The latter will have wood veneer stuck to the sides to represent the wooden blocks. These will be stained a dark-oak colour (probably with a bit of steam oil rubbed in)! This should help to restore even weight distribution on both axles. The only real way of preventing bouncing is to do what Roger has done and fit some sort of shock absorber.

I suspect many of you, like myself, find marking out really hard with our eyes getting more and more long-sighted. How wonderful that we now have CAD and can draw everything large on the computer screen. Those of you that have gone over to CNC have the added advantage of letting the machine do all the accurate milling and drilling. I would not be without it now. Not only did it give me another interest and another technique to learn, it stimulated me into getting back into the workshop and meeting all you fellow engineers. The only marking out I did on the buffers was to scribe a line from corner to corner and centre-pop the crossing. I could have used an edge-finder, but that sort of accuracy was not required. The machine did the rest. The dimples are to clear the rivets on the buffer beam.

Hunslet Buffers by ed cloutman, on Flickr

[jma1009]

Hello Ed,

Going back to page 1 of this thread, we had 'bouncing', then this evening we have "derailing".

Many of us contributed on page 1 of this thread, and we don't seem to have had a proper update. You have fitted stronger longer springs but where are they? Are they still between top of axle box and top of horn cheek? Or as I recall they were on the example of Milner's design that I had experience with?

Cheers,
Julian