jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Nov 19, 2014 0:41:07 GMT
i have the opportunity to discuss with Jos Koopmans recently this fascinating topic. i was able to do him a favour and he is for loco draughting what don ashton is for valve gears!
anyway, i think ive read everything that has been written on locomotive draughting for ordinary single orifice types, including Goss and Young. in 1904 Henry Greenly wrote a book on miniature locos and set out the Greenly rules of 1:3 and 1:6 tapers, reproduced many many years later in martin evans' manual of model locomotive construction. what is apparent is that the miniature loco people realised the fundamental principles well before most fullsize designers did, probably because there is less margin for error in miniature locos and they were easier to test.
steam exits from an orifice with an angle of about 5 degrees (the 1:6 taper). this can be observed and measured by various means. if this cone of exhaust steam doesnt fill the top of the chimney just before the top there are problems - with air actually being sucked down into the chimney as a result.
the 1:3 taper is more complex but results due to sides of the exhaust steam mixing with the combustion gases, and this has also been measured to be equivalent near enough to a 1:3 taper or cone.
incidentally you will never read of the Greenly formula mentioned by LBSC due to their big bust up in 1924!
in fullsize in the UK Sam Ell of the ex-GWR carried out extensive testing of draughting of a number of fullsize locos and formed the basis of numerous official BR test reports. these included small adjustments to locos such as the LNER V2s and lots of other locos to transform their draughting and steam output and efficiency. most of these results were summarised in E.S.(Stuart) Cox's 'Locomotive Panorama' volumes published in the early 1970s, Stuart Cox being chair of the Rugby testing plant committee.
with hindsight many CMEs and drawing ofice staff had forgotten or not applied properly what really amounts to nothing more than the Greenly rules.
why does any of this matter? a loco both in miniature and in fullsize needs to produce sufficient draught in the smokebox/expell a given quantity of the gases of combustion to get the fire to burn at an optimum rate to produce enough steam. if this can be optimised and produce less back pressure at the same time then you have a well behaved free steaming loco...
anyway enough for tonight - ive done a few drawings as part of a query re the Boxhill thread, and will post these tomorrow.
cheers, julian
|
|
|
Post by Deleted on Nov 19, 2014 2:24:34 GMT
In my post in the thread re}-- Keith Wilson's passing I referred to a demonstration that he gave about fluids and gases leaving an orifice...It was to show this 1:6 taper.....In order for you to set/check your blastpipe and chimney I have heard of people using a card cut-out or a turned wooden or Aluminium cone.......Vertical alignment and concentricity are equally important to get the desired results......Finally, the smokebox must be air-tight in order for the Vac. created to be maintained........Favourite culprit for this a poorly fitting door ( Badly designed OR}-- not cleaned down properly prior to closing !!)........Any steam or water leaks within the smokebox will also have a detrimental effect upon the locos performance....
|
|
|
Post by keith6233 on Nov 19, 2014 7:11:08 GMT
Of course all this theory is lost when you have a narrow gauge loco with a tall chimney or a traction engine.
|
|
steam4ian
Elder Statesman
One good turn deserves another
Posts: 2,069
|
Post by steam4ian on Nov 19, 2014 7:37:41 GMT
Kieth.
The theory is not lost, just the drafting efficacy of the tall chimneyed engine.
My little 0-4-0 (virtually a scale up Tich to 5") had a piece of 1" pipe as the chimney stack and I always felt it was some how choking the blast and draft just from the appearance at the top of the stack. To prove that there was a better way I made a new stack with a taper much as for a Lempore system. The throat and petticoat were about right
I can say the improvement was significant and in the words recorded by Joe Koopmans "the fire burns much hotter now". I was a new driver at the time and driving the loco became much easier, maintaining a fire was no longer a problem.
I am sure NG locos and traction engines could benefit from tapering the stack because there will be less friction on the way up because the gas velocity at the walls will be lower.
Regards Ian
|
|
smallbrother
Elder Statesman
Errors aplenty, progress slow, but progress nonetheless!
Posts: 2,269
|
Post by smallbrother on Nov 19, 2014 9:57:18 GMT
From my fluid mechanics days (ok, studying civil engineering not mechanical), I recall the edge of an orifice has a large influence on the way the fluid exits. Sharp, square, rounded edges would influence things quite dramatically.
Does this come into consideration with the construction of the blastpipe?
Pete.
|
|
jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Nov 19, 2014 10:02:33 GMT
hi keith, i agree entirely with ian. none of the theory is lost when you have a tall chimney. in fact a tall chimney will always be more efficient. the only reason lots of locos have 'short' chimneys in miniature is to follow a prototype fullsize loco which would foul the loading gauge if fitted with a tall chimney. ideally the length of chimney (including petticoat pipe from the choke upwards) should be 5 times the internal diameter of the chimney, and not less than 2 times this figure. a slight divergent taper on the inside of the chimney of 2 degrees is also of benefit , though tests have shown that a parallel bore is as effective but needs to be slightly larger in diameter as a result. there is a point at which a loco can have too tall a chimney. cheers, julian
|
|
|
Post by ejparrott on Nov 19, 2014 10:03:48 GMT
There are various ideas on tapered orifices in blast nozzles
|
|
jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Nov 19, 2014 10:25:51 GMT
hi ed, when you start considering multiple jets and lempor diffusers etc then the blast nozzle can become rather complicated. however on a straightforward well designed ordinary single chimney with ordinary single blast nozzle the results should be perfectly effective. as will be seen later when i post a few pics/drawings i have always used a blastpipe the same diameter of the exhaust passages from the cylinder, then a taper inside to meet with a parallel blast nozzle. as i make the blast nozzle and blast pipe in one piece this arrangement of a parallel nozzle has certain advantages when lining everything up with the petticoat pipe and chimney. i should add that a divergent taper on very top of the nozzle serves no purpose on an ordinary chimney arrangement, though it is something one often sees on drawings (martin evans' Boxhill being one example). alan's point about concentricity is of vital importance - many is the time ive driven a badly steaming loco to then peer down the chimney to find that the blast pipe is not central! cheers, julian
|
|
|
Post by ejparrott on Nov 19, 2014 12:01:35 GMT
Yes, I'm not going to bother with convergent/divergent blast nozzles on the Manning Wardle, not to start with at least. My Lilla is perfectly sized and that's a plain hole.
|
|
|
Post by keith6233 on Nov 19, 2014 12:31:23 GMT
hi keith, i agree entirely with ian. none of the theory is lost when you have a tall chimney. in fact a tall chimney will always be more efficient. the only reason lots of locos have 'short' chimneys in miniature is to follow a prototype fullsize loco which would foul the loading gauge if fitted with a tall chimney. ideally the length of chimney (including petticoat pipe from the choke upwards) should be 5 times the internal diameter of the chimney, and not less than 2 times this figure. a slight divergent taper on the inside of the chimney of 2 degrees is also of benefit , though tests have shown that a parallel bore is as effective but needs to be slightly larger in diameter as a result. there is a point at which a loco can have too tall a chimney. cheers, julian Thanks for that Julian if your figures are correct i should have the correct proportions with the chimney on my new engine.The chimney is 3.625 ID and 18 " long the loco is finished but i will leave it to next spring for it's first steaming ,have you a link to this information?. By the way if you Google Sweet creek locomotive you will find some information on it (not mine) Keith
|
|
smallbrother
Elder Statesman
Errors aplenty, progress slow, but progress nonetheless!
Posts: 2,269
|
Post by smallbrother on Nov 19, 2014 12:43:45 GMT
I am not sure my question on orifice conditions was understood. A sharp orifice would have a chamfer to give a clean edge at the point the fluid leaves the constriction. It would not entail any convergence/divergence. I just wondered if this was ever taken into account.
I know for sure it has quite an impact on liquid measuring devices.
|
|
jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Nov 19, 2014 13:24:11 GMT
here's a few drawings/sketches the first one is my smokebox draughting arrangement as made for Stepney. you can see the 2 cones of 1:3 and 1:6 the 1:3 cone intersects with the 3/4" dia bore of the petticoat pipe just above the choke so that if the blast nozzle is opened out later on i dont have to make a new blast nozzle (applying the 1:3 rule for say every 1/32" the blast nozzle is opened out it lowers the top of the 1:3 cone by 3/32"). there is slight divergent above the choke to top of the chimney. (the above arrangement differs in a number of important respects from the martin evans Boxhill drawings). the second sketch is a very old sketch that i made when making my Railmotor. don young specified only a 9/16" dia choke in the petticoat pipe which i thought far too small (and which many years later in LLAS don admitted as such). anyway i used a 11/16" dia choke, which incidentally is what don used on his 'George' design. close attention is paid to grate area, and i always do a calculation for free gas flow through the tubes as a percentage of grate area as this affects resistance to the gases of combustion plus the volume of these gases to be expelled. my Railmotor ended up with a blast nozzle of 7/32" dia and steamed extremely well. the third sketch is another old sketch of the smokebox arrangement on my 3.5"g GWR KING. this is where 'scale' causes problems! the single chimney is dead on scale size and of large diameter. it will be noted that the 1:6 cone doesnt fill the chimney, and the 1:3 cone is too high. i was never very happy with how the loco steamed and with a 17" square of grate area and quite a restricted nest of tubes it isnt a particularly free steaming boiler, so draughting is perhaps more crucial than ever. anyway i never did anything about it because i do like big chimneys. this will be an ideal candidate to try out a multiple nozzle blastpipe when i get round to finishing the loco's overhaul when Stepney is finished. anyway i hope the above is of some interest and might get you all making up templates and checking your smokeboxes or checking the commercial drawings! cheers, julian
|
|
jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Nov 19, 2014 13:38:16 GMT
hi keith, apart from reading all i could on the subject over many years (and dismissing most of it as twaddle), i also spent a very long time drawing out the smokebox arrangements of all of martin evans' and don youngs' designs in my possession. very rarely do the 1:3 cone and 1:6 cone dimensions appear on any drawings so these all had to be drawn out and double checked. virtually all of don young's designs after 1970 show a marked superiority and great care taken over the draughting. i knew that in my 5"g Railmotor and 5"g GWR 0-6-0 loco i had by chance more than anything else achieved optimum results plus low back pressure with the ordinary arrangement, and these empirical results confirmed the examination of all the weighty texts and designs.
hi peter, extensive tests have shown that having a 'sharp' exit on top of the blast nozzle makes not the slightest difference to the way the exhaust steam behaves in the ordinary arrangement.
cheers, julian
|
|
|
Post by Roger on Nov 19, 2014 14:36:58 GMT
Hi Julian, Does compressed air behave in a way close enough to steam to use that for experiments? It would be easy to replace the smokebox door with perspex and perhaps do something similar to the rear and then measure the pressure with a manometer while using compressed are at different pressures. Allowing smoke to be drawn through holes from the back would allow you to see what's going on too. That might be an interesting experiment to verify the design and see what happens if you change things.
|
|
smallbrother
Elder Statesman
Errors aplenty, progress slow, but progress nonetheless!
Posts: 2,269
|
Post by smallbrother on Nov 19, 2014 15:03:59 GMT
Thanks Julian for clarification.
You owe me a PM by the way!
Pete
|
|
jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Nov 19, 2014 15:06:32 GMT
hi roger,
you apparently have to use steam through the blast nozzle, though apparently it doesnt matter if air is used instead of 'smoke' from a fire on the grate but part of any such testing will involve replicating the resistance to air flow through a coal fire on the grate etc.
however any driver who knows a loco well will know if the steaming is improved or has deteriorated by any changes made.
you have to be a bit careful about level of vacuum. you could create a high level of vacuum simply by restricting the blast nozzle more than necessary with a badly designed petticoat pipe and chimney etc. the excessive back pressure would mean that the loco uses more steam to do the same amount of work (when compared to a well designed arrangement) and therefore needs a higher vacuum. this is where many of the amateur tests with manometers can be dismissed. awhile ago john (baggo) quoted some articles in ME by a guy who went completely 'down the wrong path' so to speak using a manometer. off the top of my head i seem to remember this guy fitted a 1/4" (or was it smaller?) dia blast nozzle to a Speedy and said he got much better vacuum and steam production from the boiler. of course he would with such a small blast nozzle! but at what expense for the free running of the loco and back pressure and wasting most of this extra steam plus forcing the fire too much and making loads of clinker?!
cheers, julian
|
|
uuu
Elder Statesman
your message here...
Posts: 2,808
|
Post by uuu on Nov 19, 2014 17:53:24 GMT
I've no empirical evidence with steam to support it - but a sharp edge to an orifice "feels" right. In the same way that a sharp-edged spout on a teapot or kettle will pour cleanly.
Wilf
|
|
|
Post by Roger on Nov 19, 2014 17:53:51 GMT
Thanks for that explanation Julian, I can see that it's a balancing act involving the system as a whole. That's an interesting thing that the guy did with the blast nozzle for SPEEDY. A little knowledge and all that...
|
|
jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Nov 19, 2014 19:47:16 GMT
hi wilf, i suppose it depends what you mean by 'sharp edge'. if you have a few hours to spare and fancy a long read have a look at the following... www.ideals.illinois.edu/handle/2142/4435this is a very exhaustive study of loco draughting carried out by Ernst Young at Illinious University in 1933. i read this many years ago before it was online. Young carried out extensive tests on all shapes of blast nozzle as well as every possible arrangement of the ordinary arrangement of chimney and blast nozzle. he did what roger suggests plus very much more besides. apart from the 'pepperpot' blast nozzle (which takes us into the realms of multiple blast nozzles) there was very little to choose between the usual type of blast nozzle as in all cases the steam exits from a hole at an angle of 5 degrees approx. what is important is to get the petticoat pipe and chimney to match a size of nozzle to produce efficiently a certain expulsion of combustion gases with as low a back pressure as possible. cheers, julian
|
|
|
Post by ejparrott on Nov 19, 2014 20:06:06 GMT
Question.
At what point do you start to construct the 1:3 and 1:6 tapers? Looking at your sketch, you seem to have started with the 1:3 from wherever it intersects the blast nozzle, and started the 1:6 from the same 0 point of the 1:3 taper.
|
|