injector design - a few musings

[jma1009]

just a further tale re Bill Carter. Bob Youlden has related how Bill was nothing but a perfectionist! i hope that Bob will contribute a few more reminiscenes!

on the first occasion that i was introduced to Bill by his old friend Bert Brock, Bill was being shown a revolving tailstock centre that Tom Bartlett had made. old Tom was very proud of his new bit of workshop tooling he had spent hours/days making, but Bill was very dismissive! the conversation went something along the lines of a dead centre being far more accurate said Bill because of no play in the revolving centre bearings after a coursery glance at Ted's pride and joy!

i do not doubt for one moment that Laurie Lawrence's injector articles showed Bill's influence in the design details, though had Bill written the articles himself instead of Laurie they would have been a much better read.

it is rather interesting to note that Ted Linden started making his own injectors commercially in 1939 (at the huge price of 30 shillings, so they remain incredibly rare), and form the basis virtually unaltered in every detail to the nearest thou by many subsequently because they perform faultlessly and are totally reliable. Ted's other claim to fame was that his superb 3.5"g LMS 4-6-2 loco was driven by the Duke of Edinburgh at the 1951 ME exhibition!

cheers,
julian

[jma1009]

hi Roger,

the Bob Bramson steam cone nozzle end is very interesting. i made one for a damaged Linden original injector, and it works perfectly. there are however sizing difficulties which are not adequately explained by Bob. the annular gap between steam cone nozzle and entrance to the combining cone is crucial for a self regulating 'automatic' miniature injector.

when you receive Derek Brown's book all this will be made clear if you read the whole text.

one very interesting feature of the Chiverton injectors is that the steam cone nozzle OD is less than that stated by Derek for his designs, and Bob Bramson's tables, and Laurie Lawrence's 'standard' design, and Arthur Grimmett's very successful 'medium' injector and the excellent Linden injector of the same size as Arthur's. Gordon's combining cone size is greater than those shown by Derek and the above. the annular gap must be greater as a result, unless Gordon uses a lesser/smaller taper which i have never been able to ascertain. they are self regulating and automatic and self re-starting and have a strong lifting characteristic - much stronger than 'standards'. the combination of a slightly smaller steam cone nozzle OD and larger combining cone throat must be what gives the Chiverton injectors their characteristic strong lifting characteristics. but i cant prove this!

incidentally the 'chirping' of an injector indicates the combining cone 'throat' (smallest internal diameter) may be too small in relation to the delivery cone throat size. it is not anything ive ever attempted to correct. as stated above it is worth noting that the Chiverton injectors avoid this characteristic due to their proportionately larger combining cone throat when compared to the delivery cone throat.

the John Cashmore type deserves a mention. John used a 3/16" bore in the injector body for medium sized injectors whereas Derek and Arthur Grimmett and Ted Linden and Laurie Lawrence used a 7/32" bore. as a result there is quite an explosive effect when starting the injector before 'pick up' as the overflow isnt big enough. it only lasts a few seconds and isnt really worth bothering about, but does show how trimming down the internal sizes can have an effect that might be surprising!

cheers,
julian

[uuu]

Is Arthur's first floor place the location of the famous "window test"? Where a visitor would proudly show off their work and, if it didn't pass inspection, it was thrown out of the window into the nettle patch below.

I didn't visit this workshop, but the sprit (and much of the equipment) lives on at The Pump House. Although the window test is now less severe as the bee hives have gone.

Wilf

[houstonceng]

Bob Bramson gave a lecture to the SM&EE and an article in 5 parts was published in " The Journal" of that Society, which is, probably, the basis of his book, previously mentioned.

I was interested in his reference to end regulation in Part 3 of the article. Essentially, the designs from the other aforementioned MEs in this thread use angular regulation, whereby, the water quantity and it's ratio to the steam cone size is regulated by the angular space between the nose of the steam conne and the first part of the combining cone into which it is partially inserted.

The form of the steam cone in the end regulated design appears to have a significant advantage over the usual design in that it doesn't have the sharp edge that wears over time. The regulation is effected by the distance between the end of the steam cone and the entry to the combining cone into which it is not inserted.

Unfortunately, Bob only gave a couple of tables of dimensions, for 80 and 100 psi, in a 24 oz/min injector for the end regulated version. Does he give more in his book ?
Obviously, I don't want to buy the book if it is only the same as the SM&EE articles I have.

Has anyone else investigated or made end regulated injectors ?

[Roger]

I'll take a look in the book when I get back, I'm away for the weekend.

[jma1009]

hi andy,

i am pretty sure the SMEE articles by Bob Bramson are exactly the same as the booklet, as the booklet is the same as his articles that appeared in EIM December 2011 to February 2012.

only one set of sizes is given for his steam cone end which as you state are for a 24 oz per min injector. i have copied these details with complete success, but despite a lot of experimentation i have been unable to successfully arrive at correct sizes for other sizes of injector. i should add that Bob's 24 oz per min nozzle drawing as published appears to be incomplete and misses off many vital dimensions, and the diverging taper is shown completely out of proportion and far too short.

in my opinion there is no point Bob Bramson advocating a new theory about steam cone nozzle shape if the relevant formula to calculate the sizes is missing, and the only drawing is incomplete!

cheers,
julian

[jma1009]

hi wilf,

the 'window test' was indeed at Apse Heath at the 'Aptule' works. i used to get there early on the famous saturday afternoons on my bike and retrieve from my bike saddle bag my week's work to pass inspection with Arthur Grimmett before anyone else arrived so i avoided the 'window test' at tea time. i wasnt daft!

Arthur looked liked like Jimmy Durante and had the same sort of charisma and fun and humour (but in Arthur's case very dry humour). he had a black and yellow capri called 'Banana', always sat at the head of the clubroom table in 'King Arthur's Chair', everything was 'fun' and if you made something quite ok 'Cant be bad' or 'That'll do' was his response in a deep sonerous voice.

my second daughter was born a few days after Arthur died - had the baby been a boy instead he would have been called Arthur!

Arthur was even more conservative than me re injectors. in 1986 i wanted a high pressure injector. i'd worked out what was required and i suggested we make up some high pressure steam cones. this was a bit out of Arthur's comfort zone but he knew what i was 'onto' so i was 'humoured' and so was allowed to do what i wanted on the old worn out Atlas. he was pleasantly surprised when the injectors worked!

happy days!

cheers,
julian

[Roger]

Oct 16, 2015 21:45:49 GMT jma1009 said:

hi andy,

i am pretty sure the SMEE articles by Bob Bramson are exactly the same as the booklet, as the booklet is the same as his articles that appeared in EIM December 2011 to February 2012.

only one set of sizes is given for his steam cone end which as you state are for a 24 oz per min injector. i have copied these details with complete success, but despite a lot of experimentation i have been unable to successfully arrive at correct sizes for other sizes of injector. i should add that Bob's 24 oz per min nozzle drawing as published appears to be incomplete and misses off many vital dimensions, and the diverging taper is shown completely out of proportion and far too short.

in my opinion there is no point Bob Bramson advocating a new theory about steam cone nozzle shape if the relevant formula to calculate the sizes is missing, and the only drawing is incomplete!

cheers,
julian

I can confirm that the tables in the booklet for the dimensions of Annular and End regulation refer only to 24fl Oz injectors.
I see that Bob suggests that two different sized injectors are fitted, the logic being that you can accommodate either rapid or slowly changing conditions more readily. I quite like that idea.

[oldnorton]

Hello Julian

Very interesting comment about the dimensional construction data available in Bob Bramson's booklet. Are we saying that this document entitled "All you need to know about miniature injectors and ejectors" does not in fact tell one how to make anything other than a 24 oz injector, and then incompletely? My cursory reading through had previously led me to assume that I could extrapolate from the various tables (e.g. table 2 listing cone sizes from 12 -70 fl ozs/min) to arrive at a design for a given size of injector. But I am a complete novice in this subject and bow to any greater authority.

Norm

[jma1009]

hi Norm,

if you are thinking of making some injectors yourself then i would suggest Derek Brown's book. if you have access to old MEs then it is also worth reading

1. Don Young's description of the County Carlow injector in 1970

2. the Eric Rowbottom injectors May 1976 particularly if you are thinking of a small injector

3. the Laurie Lawrence articles 1975 and reprinted 1986

Derek's 12 oz per min injector is exactly the same as the 'weeny' injector by Bill Carter described by Laurie (with one minor difference), and Derek's 26 oz per min injector is exactly the same as Laurie's 'Standard 1' and 'Standard 1A'.

Derek's 18 oz per min injector is a very good design, as are all the above.

cheers,
julian

[oldnorton]

Thank you very much for the suggestions from past ME copies. I am at the club tomorrow so will be able to go through our library of old copies. Your 'precise' reply tells me all about the question I raised! and I much respect your politeness on the matter. Derek Brown's book is arriving from Santa - I often send Santa's helper (my dear lady) an email pointing to a specific book and I think I saw the packet arrive the other day!

Norm

[jma1009]

hi Norm,

the reason for suggesting the above ME reading material is so you can digest same and get an idea of different construction/manufacturing methods. best to read at the same time as Derek's excellent book when your Xmas presents are delivered by Santa.

i dont think that making the bodies presents any difficulties.

however you will notice i hope considerable differences how say Laurie Lawrence makes the cones when compared to Derek Brown. Laurie's method of pushing in the combining cones is frankly bizarre, and the Derek Brown method and Don Young method how to achieve accuracy with this very vital part is to be recommended IMHO. i know how they were done commercially, and it wasnt Laurie's method! i would suggest making the combining cone parts undersize and re-reaming once pressed into the bodies.

note also that Laurie reams his steam cone nozzle before turning down the outside diameter of the nozzle. this rather suggests that his reamers were a bit iffy in sharpness and cutting properly. the Derek Brown method is what ive always done which is the reverse of what Laurie suggests.

Derek's book covers a range of detailed designs of well proven and tried automatic and self re-starting lifting injectors that are 100% reliable. there is no need to experiment, unless you get sent damaged and problems injectors like i sometimes get sent!

it is somewhat telling that Ted Linden perfected the standard medium size injector in 1939, and that Derek and Laurie's standard designs of this size in effect copy same, many years later. although Roger and I have discussed this subject via email, i do not think that the Linden proportions can be improved upon. if you get an extra fine finish on the cones the operating range is extended particularly with smaller injectors. a very old late friend of mine Ron Sheppard made blank tapered reamers to same size as the reamers then made into 'D' bit reamers, but all had a high finish, and the 'blanks' were used to burnish the cones and polish same. they were all quite small injectors to the Eric Rowbottom designs, and all worked perfectly. obviously very good filters were required as well for small injectors!

cheers,
julian

[robmort]

Oct 12, 2015 7:24:26 GMT flyingfox said:

....re the combining cone halves, I have seen a design somewhere that does not separate the cones completely, but cuts a small slit part way on each side, allowing a web joining the two halves to maintain concentricity. Wonder if it worked?
Regards
Brian

Oct 12, 2015 9:25:38 GMT jma1009 said:

......the type of combining cone you refer to has never been used in miniature successfully, and it's operation would be regarded these days as far from satisfactory for passenger hauling on a continuous track if you wanted the injectors to pick up ok 'on the run'. the lifting and automatic and self re-starting characteristics are lost, and you cannot incorporate the 'secret' Linden chamfer on the second half of the combining cone.

unless the webs are made quite big there is also a danger of the webs being compressed when pressing in the combining cone.

LBSC described the above type of slotted combining cone as the 'Sellers' type, and our more familiar 2 part combining cone as the 'Holden and Brooke' type.

cheers,
julian

Julian, thanks for this interesting thread.
But regarding the above on combining cone setting, do you not consider Geoff King's webbed cones solution in ME July 1997 as a success?
This seems to me to be very useful and simple, and I was going to try it soon.

Another simple way of inserting, setting and aligning the combining cones is given by Peter Cawley which involves using a taper press tool fitted inside the two cones, and with a shoulder if necessary as a stop. This ensures correct alignment and separation of the cones in one go. A detailed build using the latter method is given on an interesting thread at www.modelenginemaker.com/index.php/topic,5054.0.html.

Both of these offer potential improvements over fitting the cones separately and then reaming.

Rob

[ejparrott]

I would be worried about a taper tool marking the passages in the cone and getting stuck in it

[robmort]

Oct 20, 2015 10:17:03 GMT ejparrott said:

I would be worried about a taper tool marking the passages in the cone and getting stuck in it

that does not seem to be a problem if the the tool is smooth and clean: the builder says he's made 5 and they all work!

[jma1009]

hi Rob,

the Peter Cawley method is exactly as used and described by Laurie Lawrence in 1975 and 1986. it is a method that is fraught with problems. Laurie uses the same OD for the 2 halves of the combining cone when pressed in. this can easily lead to the second cone being loose. he also uses a light interference fit, far too light a fit for me. i also do not believe that the first cone being pressed in will stay in perfect allignment as it goes over the internal holes, or can be pressed in accurately by just a 0.149" length of 9 degree taper.

re-reaming after pressing in takes a couple of seconds and is so easy to do! there must be a good reason why the commercial makers use this method!

re Geoff King, i am not prepared to compromise the excellent characteristics of an automatic and self restarting lifting injector, for the sake of a webbed one piece combining cone.

cheers,
julian

[jma1009]

(and with reference to Rob's link, Stu Hart says he made 5 that worked but has admitted to me only one of them worked! he states the annular gap between steam cone nozzle and combining cone is 2 thou, but for the 26 oz per min injector he was making it should be 7 thou)

[Roger]

I agree with Julian that the most satisfactory way to ensure the best alignment of the two tapers in the combining cone is to finish ream them in situ. I've not seen one of these apart, but looking at the sectioned pictures I've found on Google, it would appear that there is only a very short length of the outside diameter that's available to locate the cone. I'd make pushers to press them home, and those would keep them reasonably true, but I don't think I'd rely on that unless the location length was longer.

I can certainly see why people have proposed making the cones in one piece, it would solve several manufacturing issues. The added complexity in making a one piece combining cone is enough to put me off that idea. If those solutions compromise the self starting then that's not much good anyway. Making a simple slug of Brass is pretty trivial, so there would need to be a very good reason to move away from that.

Has anyone made injector bodies which have multiple small holes in them instead of one large hole for the water feed and ball valve? It's obviously more work to do them that way, but it resolves the issue of cones tipping as they pass over them.

[oldnorton]

Hello Julian

I think I wish to remove any impression of criticism I have given of Bob Bramson's booklet, which is a compilation of his EIM articles. In his introduction he says "it is not… intended to describe how to make injectors" so it was my error entirely not to see and appreciate this. Having now read it, it is an interesting discussion of the many variables in the design of injectors and a report of his many measurements made on test apparatus.

I have compiled your postings on this thread into a document that now runs to ten pages of small 10 point print! You have told a most interesting story of the development of the small injector through the names of those involved. It must have been an exciting time for model engineering in the 1970s when new engine designs were coming out and there was technical innovation. The Model Engineer magazine of today can sometimes seem a very stale publication when compared with the older copies. I have found those early articles you suggested and I am most grateful. In 1970 Don Young says that he wanted a vertical injector for his 3 1/2" County Carlow but his attempts to make the LBSC design failed, and he subsequently doubted it could ever work. His friend Arthur Grimmett then 'allowed' enough 'know how' to be released through Don's article to enable the reader a "reasonable chance of success" if he made the injector. It does make me wonder what information Arthur held back to ensure that any manufacture was only 'reasonably' successful?! I note your comment here that the Don Young County Carlow injector 'will not work' so 'reasonable' turned out to be 'pretty slim' or 'hopeIess' I suppose. I do like the stories repeated here about Arthur, especially the work judging sessions involving a trip through the nettles and bee hives for the failures :-)

I am interested to hear whether you think that the vertical injectors were solely the result of Don Young and Gordon Chiverton not wanting to upset Arthur Grimmett, or whether there was also a design/appearance wish to replicate prototypical designs? Also, why can a vertical injector not disassemble in exactly the same manner as a horizontal one - you mention pressing in the combining cones. Surely one can copy a horizontal design exactly, but swivel the air valve seat 90 degrees, and you have a vertical injector - or have I missed something?

Norm.

[jma1009]

hi Norm,

LBSC described a couple of vertical injectors - one was for his 3.5"g Virginia, so when Don Young and Gordon Chiverton went down the vertical injector route it wasnt anything new as such.

i am trying to remember exactly what was the 'understanding' between Don and Arthur Grimmett in 1970 and what transpired with Don's own attempt. the first thing was that Don's County Carlow design was a small injector plus it was vertical. Arthur made 3 horizontal sizes, small, large, and medium. the medium (standard 26 oz per min) was by far the most popular and very few of the other 2 sizes were sold and hence made. under no circumstances would Don have been allowed to see the drawings Arthur worked to (Don was always refered to by Arthur as 'The Industrial Spy'). it went something along the lines of Don being given one of the small injectors, on the understanding that if the innards were described in ME it would only be in vertical format so as not to affect Arthur's business of horizontal injectors for Reeves and Kennions. Arthur hated making the small injectors anyway. they were never as fool proof or reliable as the medium injectors. (with the benefit of hindsight this was due to their length being compressed into the standard LBSC body length of 7/8").

it was always understood that until Arthur retired Don would not sell (and Gordon would not make) horizontal injectors. by the time Don started out on his own in 1975 Arthur was 64 and winding down. he sold his boiler fittings business plus all the jigs and drawings to 'Fyne Fort Fittings' (Frank Lockwood and family, forever after refered to by Arthur as 'The Goons'), but carried on with his medium injectors for a few more years. he made very little money out of the boiler fittings and injector business.

Don would never have done anything to affect Arthur's livelihood. this was a 'gentleman's agreement' based on Arthur's extreme kindness and long friendship with Don, plus those whom Don needed to do business with.

going back to your questions,
Don always specified vertical injectors on his designs so as to create a market for Gordon's products!
it is significant that when Gordon did start making his horizontal injectors, production of the vertical injectors tailed off. the vertical injectors of Gordon's design are very clever and require a lot of extra work in making the bodies.

there are 2 types of vertical injector. essentially there is the Derek Brown type in a chapter devoted to same in his book. this is in effect (like the County Carlow vertical injector) a horizontal injector turned 90 degrees. both the steam cone and delivery cone are pushed/slid into the ends of the body and held in place with the union nuts and cones. as such, there is hardly any advantage with pipe runs. there is the added disadvantage that any muck between the 2 halves of the combining cone cannot be seen.

Gordon's vertical injectors have a side connection on the bottom for the delivery with a cap on the end, and the delivery cone is pressed into the body and cannot be removed. you have the disadvantage of the delivery cone not being able to be removed to clean the injector (plus, as above, not being able to see if there is muck between the 2 halves of the combining cone), but the advantage of much better pipe runs.

as to Don's attempt in 1970, i believe that as described in 1970, Arthur had to work his magic on it before it worked! (this probably involved Arthur fitting a new set of cones but to different dimensions and length as Don's version had a body length of 1 1/8"). it was not lifting or self re-starting or automatic. it was not amongst Don's workshop stuff when he died, and the only loco fitted with an injector when he died was his Railmotor which had a Chiverton vertical injector fitted. his K1/1 had the injectors missing. i do not know what injectors he fitted to his 5"g Fishbourne as the pics i have of it dont show the injectors - according to Don this loco was sold to someone in Canada shortly after the design appereared in ME in 1967.

when reading some of Don's stuff a certain amount of salt pinching is required with some of the stories. however, he was incredibly kind and generous with me especially when i was an impoverished 'nipper' still at school and starting out on this lark many years ago.

cheers,
julian