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

[jma1009]

Hi Roger,

I don't know why you are still sticking with end regulation with a very stubby end to the exit end of the steam cone.

You making the actual steam cone to be used out of bronze.

So it won't suffer from dezincification anyway, compared to a brass steam cone, and even in brass you can make the material width at the end 5 thou thick or 10 thou thick in brass.

You won't get dezincification anyway as you have your ultrasound tank thingy!

Thin brass steam cone ends don't dezinctify due to
use, but to over zealous cleaning in acid, and due to rubbish advocated by the ignorant "You have to take the injector off after each steam up and leave it for hours in citric or acetic acid"!

[Roger]

Jan 31, 2021 22:39:28 GMT jma1009 said:

Hi Roger,

I don't know why you are still sticking with end regulation with a very stubby end to the exit end of the steam cone.

You making the actual steam cone to be used out of bronze.

So it won't suffer from dezincification anyway, compared to a brass steam cone, and even in brass you can make the material width at the end 5 thou thick or 10 thou thick in brass.

You won't get dezincification anyway as you have your ultrasound tank thingy!

Thin brass steam cone ends don't dezinctify due to
use, but to over zealous cleaning in acid, and due to rubbish advocated by the ignorant "You have to take the injector off after each steam up and leave it for hours in citric or acetic acid"!

Hi Julian,
I just think it's easier to make and it's more robust with a stubby end. You also don't have to worry about getting it perfectly concentric with the Condensing Cone because the gap isn't affected by that. It seems a more elegant and simple solution to me.
The only benefit I can see for Annular Regulation is that you have to move the Steam Cone more to change the regulation gap, so it might be easier to get the gap right. However, now I'm going to make the Steam Cone adjustable, it's going to be simple to adjust on the locomotive.

[Roger]

I haven't had much time to spend on this today, but I did finish the detail design of the adjuster arrangement of the Steam Cone.

I ended up changing the inner Steam Cone taper from 13 to 11 degrees so that it gave a little more material under the 0.5mm section  O-ring.

So here's the Tool & Cutter Grinder getting set up parallel using a Brass test bar to check it's as close as possible. 

20210131_214857 by Georgia Montgomery, on Flickr

I'm making the back reamer from a 2mm HSS Drill blank. It needs to be 1.8mm diameter and long enough to reach deep inside the cone from the end, so I ground that first. I've used Black marker pen on the blank to make it clear where the taper ends.


20210131_221800 by Georgia Montgomery, on Flickr

The table was then set back parallel for grinding in half. There's a locking key that can be used to stop the chuck from turning.

I ground it almost to the middle and then dressed the wheel for the final cut.

20210131_224841 by Georgia Montgomery, on Flickr

I've ordered the O-rings and the 3.5mm reamer needed for the bore, so hopefully those won't be too long coming. I can't really make the parts until the reamer arrives since I want to do the body in one operation. I'm screw cutting the thread on the Steam Cone, so I want to have the body finished to check that I'm happy with the fit of the thread.

Anyway, it won't take long to make when the reamer arrives because I've already created the dimensioned drawings for the Lathe work and the G-code for the Hexagonal ends and the steam and water cutouts.

[Roger]

While I'm waiting for parts to arrive, I thought I'd look more closely at the effects increasing the Steam Cone throat has on the other geometry. It's interesting that D.A.G Brown advocates opening up the Steam Cone throat to lower the pressure range, but without any accompanying changes anywhere else. At least, that's my undestanding, and that's what I've done in increasing the throat from 1mm to 1.05mm, and in fairness, it did seem to work.

However, if you go back to basics and look at that happens on my model when you change the Steam Cone throat, it makes a huge difference.

This is how it looks with a 1mm throat...

Cone21 by Georgia Montgomery, on Flickr

... and this is how it looks with a 1.05mm throat! Now, I've cheated a bit, because I need to trial this in the same cone bodies I've already made, so the length of the Delivery Cone has had to be shortened.

Look at how much the Delivery Cone throat has moved right in the cone below.

Although it's not clear from the model, the throat diameter at the LH side of the Combining Cone as increased in size quite a bit.

The issue here is all about creating the Ejector part of the Injector. Forget about anything to the right of the LH row of vent holes. What we're trying to achieve is a ratio of throat diameters ie Steam Cone throat to Condensing Cone throat of 1.38 to 1.45
If you open up the Steam Cone throat without moving the vent holes upstream to increase the diameter, it may not lift at all!

The problem here is that if you do move the vent holes upstream, you now have less time for the Steam to be fully condensed, so the vent holes are in the wrong place. That means that you have to increase the length of the cone upstream of the vent holes, keeping the vent holes roughly half way along the whole Combining Cone.

So, to but a long story short, you can't just change the Steam Cone throat diameter and expect it all to work just fine. You might get it to work, but then again, you might not. Everything is tied to everything else, and even small changes in diameters have huge consequences because the taper angles are small.

Cone23 by Georgia Montgomery, on Flickr

Anyway, this is a bit of a bodge, I'd like to try it with a longer Delivery Cone, but for the moment I don't think it will matter because it's not the critical part of the design.

So I'll make one to this new design and see if it's any better.

[jma1009]

Hi Roger,

I make this post in a constructive manner, but you have not explored at all a traditional steam cone and what advantages it may have. You don't have to have a wafer thin end of the steam cone.

We know that George Chiverton made injectors of various sizes with a high lifting characteristic.

I don't consider this characteristic as important as Bob Bramson makes out, and Gordon achieved this outside of Bob's parameters as Gordon's steam cone entered the first part of the combining cone by a bit...

Gordon's injectors work extremely well! I think myself the lifting characteristic is too great on Gordon's injectors - it caused a lot of problems with pipe work and water valves for those who did not appreciate the consequences.

I just don't see why you continue to adhere to a method of Bob's that goes against the very best of UK design and completely omits any reference to Gordon Chiverton and his products for the trade.

I've never tried a Bob Bramson made injector, but I've driven locos with and tested the injectors of Gordon Chiverton many many times!

[steamer5]

Hi Julian,
Gordon, as you mentioned, was in the business of selling injectors, & I guess as such he didn’t divulge or publish his designs?
As you posted above.... “Gordon achieved this outside of Bob's parameters as Gordon's steam cone entered the first part of the combining cone by a bit......” ...it’s the “little” bit that becomes the issue for the majority of us I guess Gordon figured it out by trial & error making lots of injectors to do so, which would require some very careful machining work.......from Rodgers work end end regulation would seem to be a little less fraught with disaster!

“Gordon's injectors work extremely well! I think myself the lifting characteristic is too great on Gordon's injectors - it caused a lot of problems with pipe work and water valves for those who did not appreciate the consequences.” As I’ve said before we had a guy in my club who was a dab hand at making injectors & made them lift very well too....once again he didn’t pass on the “how” so that info has been lost

I think we would all be interested in how you go about making injectors to compare against Rodgers efforts.

On your recommendation I brought DAG Browns book, but my Dad nicked it before I finished reading & digesting it!

Cheers Kerrin

[Roger]

Feb 2, 2021 0:53:37 GMT jma1009 said:

Hi Roger,

I make this post in a constructive manner, but you have not explored at all a traditional steam cone and what advantages it may have. You don't have to have a wafer thin end of the steam cone.

We know that George Chiverton made injectors of various sizes with a high lifting characteristic.

I don't consider this characteristic as important as Bob Bramson makes out, and Gordon achieved this outside of Bob's parameters as Gordon's steam cone entered the first part of the combining cone by a bit...

Gordon's injectors work extremely well! I think myself the lifting characteristic is too great on Gordon's injectors - it caused a lot of problems with pipe work and water valves for those who did not appreciate the consequences.

I just don't see why you continue to adhere to a method of Bob's that goes against the very best of UK design and completely omits any reference to Gordon Chiverton and his products for the trade.

I've never tried a Bob Bramson made injector, but I've driven locos with and tested the injectors of Gordon Chiverton many many times!

Hi Julian,

Following what everyone else does isn't 'exploring' anything, nothing is learned and there will never be any progress. Clearly traditional designs work really well, so there's no point in going any further with those. However, traditional designs are unnecessarily difficult to make because they have too many parts, each of which needs to be made extremetly accurately and set to precise depths. Once made, they can't really be disassembled and serviced. They also suffer from the Steam Cone being fragile. The other issue for scale designs is the requirement for the non-return valve which makes it very difficult to incorporate in a scale body.  As far as I'm aware, nobody has ever made an 8X scale injector that works? That's what I'm making.

I'd dispute that Bob Bramson goes against the very best of UK design. If you take the time to read and digest what he says, he's actually coming up with almost exactly the same proportions as the examples in D.A.G Brown's book. He says is that he always makes his injectors with end regulation. He explains both end and annular regulation though, and how to design using both methods. It's my choice to choose one over the other. Bob's anaylitcal approach enabled me to create a 3D model which follows those rules and generates designs that closely match those in D.A.G Brown's book. Excellent as that book is, it doesn't tell you how any of the dimensions have been arrived at, it's just served up as finished solutions. I wanted to dive deeper and really get to grips with how they work and why.


I think you're missing the point about lifting in the injector. Lifting from a lower source isn't important, but the starting characteristics depend on the ejector part of the design. All of D.A.G Brown's designs have this characteristic because it's necessary. You may not see it manifested by the ability to pull water up into the injector from below, but it's there inside the injector else it won't work. The only difference between what you might consider a lifting and a non-lifting injector is the amount of lift, not whether it has lift or not.


I know how traditional injectors are made, and it simply doesn't interest me to make them like that. I'm just going back to first principles and starting from a blank sheet of paper with just the throat diameters, angles and ratios. You can't change the physics of how it works, but you can change the mechanical arrangement to make it easier to manufacture.
Above all of that is the sheer joy and excitement of doing something that's never been done before!

[flyingfox]

Roger, very interesting project.
I know of most of the work about injectors, but wonder where the Bob Bramson material is published?
Thank you
Brian

[andyhigham]

Remember what the first E stands for in club names xxMEE

[Roger]

Feb 2, 2021 12:35:30 GMT flyingfox said:

Roger, very interesting project.
I know of most of the work about injectors, but wonder where the Bob Bramson material is published?
Thank you
Brian

Hi Brian,
Bob Bramson's booklet is purchased through the Taunton ME club if I remember correctly. It's called "All you need to know about miniature injectors and ejectors". I'd take issue with the title though, there are some key things that aren't explained well enough, if at all.
However, it's a lot more analytical than D.A.G Brown's book which focuses on making them rather than designing them. Both books are essential reading in my opinion.

[Roger]

I've just gone to work out the new offsets for reaming the new cone tapers, and something didn't seem quite right. I knew the new cones wouldn't need the Delivery Cone to be reamed as deeply as the previous ones, yet the figures came out deeper!

Rather than calculate these again, I thought I'd  model the collars onto the reamers as if they were set in the 1.6mm setting piece and also depict them as they would be when setting up on the Lathe.

So below you're looking at the reamers at their full depth, which can be visually checked to see if they appear to match the required taper. The two reamers at the back have the collar face 25mm from the cone face. To find the the depth, I just ask the model the difference between the two faces.

That worked out fine for the Condensing Cone end but not the Delivery end.

Delving into this deeper, I could see that I'd somehow got the model wrong where the reference dimension was taken from. Although it looked ok at a distance, when zoomed in I could see it as lower than it should be. Because the angle is shallow, this makes a huge difference to the distance the reamer goes in. In fact I've reamed them all to  9.094mm when they should have been to 10.576mm! Doh! So working backwards, the 0.66mm throat might have come out at 0.479mm, and it wouldn't have worked at all. However, the #74 pilot drill is 0.572mm, so that's what I ended up with. That means the diameter ended up with a parallel section 0.088mm too small instead of two cones meeting.

Reamer settings by Georgia Montgomery, on Flickr

Annoying as this might seem, I'm delighted to have spotted this, albeit a bit late! It might explain why I've struggled to get it to work even as well as it has. Fortunately, the Delivery tapers are all too shallow, so it's a simple job to maching them to the right depth and try them again.

Such is life, mistakes happen and it's not something I fret about. Actually I'm excited to go back and see what difference this makes, and also to machine the new cones that are modelled on the larger Steam Cone throat size.

[delaplume]

Feb 2, 2021 12:46:26 GMT andyhigham said:

Remember what the first E stands for in club names xxMEE

Expensive ??

[delaplume]

For a light-hearted alternative view on Injector Workings have a look at my "Formula for Locomotive Injectors" in the Humour section...

[Oily Rag]

Snipped for clarity, Roger said    

"I know how traditional injectors are made, and it simply doesn't interest me to make them like that."

It is respectful to recognise, digest and accept that this is Roger's motivation and his chosen path of expression and a very interesting journey it is for us interested onlookers.  I hope when my time comes I will have a path to follow for such scale size working injectors.  Also I am amazed at Roger's productivity. I struggle to get an hour or two most evenings and I am very slow.

[Roger]

Ok, that didn't end well. That's what happens when you drill 12 holes to the centre when you meant to only drill 10. 

20210202_164648 by Georgia Montgomery, on Flickr

I've ground up a couple of gauges to check the throat diamter when I ream it from both ends, and there's something amiss. I'm having to stop about 1mm short of the calculated position on the second taper, else it goes oversize.

I think it might be that the reamers are not quite right. They are very slender, so there's a tendency for them to be pushed away from the grinding wheel more at the end. That would make them cut oversize at the throat.

I'd better set them back up on the grinder and have another go at them. I guess I'll just have to leave them for ages to spark out, even though it seems that they're not touching.

In the meantime, I've ordered a couple of 100mm long  x 2mm diameter Carbide rods to make some new ones. Carbide is so much more rigid than HSS, so it ought to be easier to get them really accurate.

[Gary L]

Feb 2, 2021 23:51:22 GMT Roger said:

Ok, that didn't end well. That's what happens when you drill 12 holes to the centre when you meant to only drill 10. 

20210202_164648 by Georgia Montgomery, on Flickr

I've ground up a couple of gauges to check the throat diamter when I ream it from both ends, and there's something amiss. I'm having to stop about 1mm short of the calculated position on the second taper, else it goes oversize.

I think it might be that the reamers are not quite right. They are very slender, so there's a tendency for them to be pushed away from the grinding wheel more at the end. That would make them cut oversize at the throat.

I'd better set them back up on the grinder and have another go at them. I guess I'll just have to leave them for ages to spark out, even though it seems that they're not touching.

In the meantime, I've ordered a couple of 100mm long  x 2mm diameter Carbide rods to make some new ones. Carbide is so much more rigid than HSS, so it ought to be easier to get them really accurate.

Yeah, I get days that as well Roger!

Gary

[chris vine]

Roger,

Carbide taper reamers!! Now you will be able to make the cones out of PB102, Inconel, whatever you choose!!

Chris.

[Roger]

Feb 3, 2021 17:17:55 GMT chris vine said:

Roger,

Carbide taper reamers!! Now you will be able to make the cones out of PB102, Inconel, whatever you choose!!

Chris.

Hi Chris,
I don't see any problems with HSS reamers for any of those materials, they just need to be sharp. I've just used one of mine on the new adjustable Steam Cone for the 8X scale injector and it was fine. Carbide certainly helps with keeping a sharp edge if you have a lot of them to do though.

[jma1009]

Hi Kerrin,

The "bit" and "how" is no secret.

For Gordon Chiverton's 16 oz per minute as marketed and stamped (but are really 18 oz per minute) delivery injectors, the combining cone starts 0.375" into the body, and the snout of the steam cone is 0.062" diameter (OD), and goes into the start of the combining cone by 10 to 15 thou.

The 10 to 15 thou above makes hardly any difference as Gordon's annular gap is more generous than Derek Brown's, and I've also found the OD of the steam cone snout to have 1 thou either way on the examples I've examined.

When I've been asked to repair Gordon's examples I add a few thou to the steam cone snout OD and length.

You will gather from the above there is some degree of latitude in what will work and what won't in respect of the above, but the annular gap can be easily established in each case, and is not the problem Roger appears to consider as a disadvantage.

Gordon's throat sizes for this size quoted are steam cone 0.041", combining cone 0.034", and delivery cone 0.027"

Gordon's 2 halves of the combining cone follow pretty much the proportions in length of those of Bob Bramson which are pretty much identical to those published by C M Keiller in ME in 1962!

Gordon's have a smaller diameter OD to the steam cone snout than 'standard' and consequently a shorter diverging taper length internally though the taper is the 9 degrees standard.

Gordon's also have a slightly larger combining cone throat ID to 'standard'.

Cheers,

Julian

[Roger]

I've made new 7 and 9 degree reamers today, letting them spark out over a 20 minute period on each operation, moving it back and forth occasionally. That will hopefully get the geometry closer than the first ones. I didn't exactly rush the others, but clearly it needed more time than I gave it. You live and learn.

20210203_110013 by Georgia Montgomery, on Flickr

I've made the 2mm bore sleeve for the Tailstock so I can more accurately set the stop collars. I can push all of the parts together with one hand and tighten the grub screw with the other without having to worry about the alignment.

The old method wasn't exactly rough and ready, but this is definitely better.

20210203_215218 by Georgia Montgomery, on Flickr

So here they are, collars set and colour coded. The Go No-Go gauges for the 0.66mm throat are at the back.

To be honest, I don't like the idea of setting the throat diameter by where two tapers meet, it seems too prone to error. I think it might be better to drill or ream the throat and stop just short of that with the tapers. I may change it to be like that.... we'll see.

20210203_215720 by Georgia Montgomery, on Flickr