steam4ian
Elder Statesman
One good turn deserves another
Posts: 2,069
|
Post by steam4ian on Jul 3, 2020 12:45:55 GMT
I note a bit of talk and even some despair here regarding cylinder diameter and blast nozzle diameter. Whilst Jos Koopmans mentions this relationship in his book he doesn't endorse it. He ignores the relationship because his discussion is focused on the nozzle to stack relationship, that is in getting an ejector with maximum potential efficiency.
I question so much angst regarding the cylinder Bore to nozzle relationship. It might be true for full size with adequate exhaust passages and steam chest to cylinder passages; look at a modern cylinder casting to see the porting arrangements. Now look at your average model where the steam chest to cylinder passage is at most three parallel drillings; hardly aerodynamic. There will be significant pressure loss between the cylinder and steam chest which would swamp that of the blast nozzle.
The Simplex design is mentioned here. I have a 1.5 times Simplex running 7.25". It pulls very well and has no difficulty steaming. It even steamed well with with a hole in the superheater header blasting steam into the smoke box; the reason it is now out of service for repair.
In summary, don't sacrifice cylinder diameter just to satisfy some arbitrary formula regarding the blast nozzle. Do make every effort to improve the porting so that the blast nozzle does become the place of greatest pressure loss.
Ian
|
|
|
Post by rogsteam1959 on Jul 3, 2020 18:29:55 GMT
Because I can’t change the height of the blastpipe I made a nozzle that has the cone to give me the chance to put the effective diameter deeper into the blastpipe. Had her on the rolling road today. youtu.be/lINB_7Yl3ns
|
|
jma1009
Elder Statesman
Posts: 5,900
|
Post by jma1009 on Jul 3, 2020 21:26:49 GMT
Hello Dave (kipford),
You need to raise your blast nozzle upwards by some 10 to 15mm (I would go for 15mm as that is Jos's 6 times 'x' via his correspondence with me). I would try and keep the blast pipe as big as possible (before it reaches the blast nozzle), perhaps squaring it off internally (which I have done many times), due to the constraints in the cylinder block.
The dremel and a dental burr can work wonders on all of this, and you ought to round off the exhaust passageways where they meet the bottom of the blast pipe - which in your case is well within the cylinder block.
I've always filed or later on dremeled out with dental burrs the drilled passageways from port face to cylinder into slots, as the exhaust steam ought to escape/exit freely, and this will aid the smokebox draughting. Churchward had a good quote on all this that applies as much to miniature as to fullsize.
Cheers,
Julian
|
|
kipford
Statesman
Building a Don Young 5" Gauge Aspinall Class 27
Posts: 566
|
Post by kipford on Jul 3, 2020 21:54:11 GMT
Julian Thanks, already done, I spent 35 years designing air systems where pressure drop was always at a premium. Dave
|
|
Gary L
Elder Statesman
Posts: 1,208
|
Post by Gary L on Jul 3, 2020 23:53:19 GMT
I had a good look through this discussion, referenced in an earlier post: www.chaski.org/homemachinist/viewtopic.php?f=8&t=82682I paid more attention to the contributions of Dr Koopmans, following Julian's thoughts that this man knows what he's talking about. It seems to me that in Jessie's design the choke is too large. At 1 1/2" it's six times the blast nozzle size of 1/4". And the distance between the blast nozzle and the choke is too much - 11.6 times the nozzle diameter - more than Dr Koopmans preferred ratio of 7. In my first sketches, I was raising the blast nozzle, but it might be better instead to extend the petticoat downwards, when I can reduce the choke by incorporating a taper. So here's another sketch: Draughting2 by Wilf, on Flickr This shows three different choke sizes, but in all cases the blast nozzle to choke distance is 2.1", being seven times the nozzle diameter of 0.3" (which is larger than the designer's 0.25"). Also I've brought the base of the petticoat down to just above the superheater flues, so they're still accessible. Jessie is blessed with a decent chimney, so the 1:6 cone intersects 1/2" below the top. - The smallest choke is 1.125". This is 3.75 times the blast nozzle diameter. This results in a petticoat pipe taper of 8°, which is perhaps too much. The 1:3 cone intersects 0.638" above the choke.
- The medium choke is the result of choosing a 6° taper (which is the most Dr Koopmans recommends). The choke ends up at 1.206" and the 1:3 cone intersects 0.800 above.
- The largest choke size is 1.265". This is what results if the chimney taper of 4.34° is continued downwards. The 1:3 cone intersects 1.029" above the choke.
All of this is confined to the smokebox alone - since fluid dynamics is not my specialist subject, I've not extended the analysis into the grate area, or tubes, cylinders, passages etc. Most of these I can't change, whereas I can put a modified petticoat and blast nozzle in easily, and revert to standard if it all goes wrong. For what it's worth, the cylinders are 1 1/2" bore, the grate area is 20.2 sq in, and the total tube area is 3.78 sq in (making no deduction for the area of flues occupied by superheaters). Please tell me if this is all rubbish. Wilf PS - I could make things easier with a bigger blast nozzle. The petticoat would not need to be so long, and the choke could be larger - but I'm already quite a bit bigger than the designer, and I'm not confident enough to stray too far away from conventional wisdom. Hi Wilf Sorry for the late response. I had a lot of trouble with bad steaming on my Bridget when made exact to drawings. To be fair, a friend built a Jessie exact to drawings and never had any bother at all (and for the info of other readers, Jessie and Bridget are almost the same loco, just dressed differently). To digress for a moment, I suspect that it is not only the drafting that caused my problem; I suspect the proportions of the coaxial superheater may be partly to blame, but that is not an easy fix. Certainly even now, when smokebox char blocks the lowest 2 small tubes the steaming is noticeably impaired. But to return to the drafting... The first thing I tried was to reduce the blast aperture. This did not help matters and made it less free running, as you would expect. The next thing I tried was the 'jet pump' theory. This was very interesting. (If you put "jet pump" into the Search on this Board you will find my post on the subject, a couple of years ago.) Julian said it gave a result very close to the Koopmans system, which is quite likely. You are right; you have to work back from the chimney top diameter, because you can't change that. (My Bridget has the stovepipe chimney, and my opinion now is the chimney is simply too big for conventional draughting.) Following the jet pump principle, I ended up with a new blastpipe that was mounted higher than its predecessor, and much bigger in diameter than Ken Swan's. if I remember offhand it is 7/16" dia, which is the same diameter as the manifold, so not really a nozzle at all, just a plain pipe. Nevertheless the difference was striking. The loco was dramatically more free-running; the blast was soft to the point of being inaudible, and (this is the bit that surprises people) the steaming was distinctly better. However it was still not good enough. The big chimney determined the diameter and location of the blast nozzle, and I reckon the cylinders could not supply enough steam to feed such a big nozzle. The ultimate solution was to keep the Jet Pump proportions unchanged, but to modify the nozzle by inserting a central spider so as to divide the blast into four smaller parallel jets. This has proved to be the complete answer. The free running and soft blast are little changed, but (at last!) the boiler can keep up with the cylinders. The Jet Pump theory applied to loco drafting uses a standard layout of all the components; you just apply the template to the original drawings, magnifying or reducing the scale till it fits the chimney top, then make a blastpipe in the position and diameter given. It might be necessary to sleeve (or perhaps bore out) the choke, but I didn't need to do this. The Koopmans formula does effectively the same thing, and as far as I can see, will produce broadly similar dimensions as a result. A particular feature of the Jet Pump principle is that the blast nozzle must be sharp-edged (i.e 90 degrees, the corner not broken in any way) and it must be parallel internally for at least (IIRC) 2 diameters before the opening. This is a fluid dynamics thing, which is applicable to all blast pipes and similar devices regardless of the choke proportions. Hope this helps Gary
|
|
uuu
Elder Statesman
your message here...
Posts: 2,807
|
Post by uuu on Jul 4, 2020 15:31:16 GMT
Thank you - yes this is useful stuff.
You and I have had similar thoughts - that the diameter of the chimney is too big and/or the blast nozzle is too small. Looking at my chimney casting, I could not have made the internals any smaller. In boring out the taper I've had to leave some as-cast surface near the bottom - I may apply some smoothing filler over this.
I'm in agreement with you over the sharp edge of the blast nozzle, although I know that Julian has found documented experiments which suggest otherwise.
I hadn't considered the multi-nozzle solution. I suspect I will start with a single and swap it out if I'm not happy (or if I get the urge to experiment).
Where you and I differ is that you raised your blast nozzle, where I'm proposing to extend the chimney downwards. It will be a long while yet before I get to see any results.
Thanks again
Wilf
|
|
jma1009
Elder Statesman
Posts: 5,900
|
Post by jma1009 on Jul 4, 2020 22:16:08 GMT
Hi Wilf,
Don't worry about adding filler; the smokebox gunk of ash and oil will actually provide it's own filler, and will also smooth out any steps internally, and one example of a miniature loco before I got interested in all this stuff of smokebox draughting would actually have the 'gunk' form the Jos Koopmans dimensions within the petticoat pipe and chimney, though I didn't know this at the time.
Cheers,
Julian
|
|
Gary L
Elder Statesman
Posts: 1,208
|
Post by Gary L on Jul 5, 2020 1:29:03 GMT
Thank you - yes this is useful stuff. You and I have had similar thoughts - that the diameter of the chimney is too big and/or the blast nozzle is too small. Looking at my chimney casting, I could not have made the internals any smaller. In boring out the taper I've had to leave some as-cast surface near the bottom - I may apply some smoothing filler over this. I'm in agreement with you over the sharp edge of the blast nozzle, although I know that Julian has found documented experiments which suggest otherwise. I hadn't considered the multi-nozzle solution. I suspect I will start with a single and swap it out if I'm not happy (or if I get the urge to experiment). Where you and I differ is that you raised your blast nozzle, where I'm proposing to extend the chimney downwards. It will be a long while yet before I get to see any results. Thanks again Wilf Hi Wilf If you do this properly, you won't have a lot of choice about how high to set the blast pipe or how low the chimney can extend downwards. The only dimension that can be varied is the distance from the chimney top to the choke, and that might be controlled by the need to incorporate the 4 degree included angle in the chimney itself. Otherwise, the proportions are fixed in relation to the choke D, but that in turn (in the practical situation of a model locomotive) has to be worked back to the chimney top dimension, as in this diagram. It is important to follow all the factors exactly. The only ones that have been found to be non-critical are those of the petticoat- as long as there is one. Koopmans' formula works in much the same way. It was 1.5 diameters minimum for the parallel portion of the jet below the nozzle top BTW. I didn't swap out for a multi-nozzle, I just silver soldered the spider in the centre. It is important to maintain the outer edges of the blast cone. The Jet Pump principle is supposed to be optimised for all rates of flow, but I think like most things it is marginal at the extremes, and Bridget's blast was one of those. It needed a bit more velocity, and by (in effect) blocking the central part of the cone somewhat, this had the effect needed, without compromising the essential proportions of the cone. HTH Gary
|
|
|
Post by ettingtonliam on Jul 5, 2020 2:39:32 GMT
Isn't this central 'spider' the same as the infamous 'jimmy' used by generations of drivers (illicitly) to sharpen up the blast of their steed, and universally banned by the locomotive departments of virtually all railway companies?
|
|
|
Post by steamer5 on Jul 5, 2020 2:42:50 GMT
Hi Gary, Any chance you could post a picture of your nozzle please? I get the idea but a picture would be great
Cheers Kerrin
|
|
Gary L
Elder Statesman
Posts: 1,208
|
Post by Gary L on Jul 6, 2020 0:46:55 GMT
Hi Gary, Any chance you could post a picture of your nozzle please? I get the idea but a picture would be great Cheers Kerrin Hi both A Jimmy was indeed against the rules, but I've read that shed fitters would make them for drivers for a consideration. In principle they were a bar that clamped across the blastpipe, with a safety chain attached, because dropping it down the blastpipe (or shooting it into the air with the exhaust) were equally disliked. However this is definitely a Loco Dept approved (by me) modification! The original nozzle was 7/16" dia. The spider was made by drilling 4 holes in some stainless stock, then turning the outside to break open the holes at the required 7/16" dia. Part off, making sure the spider is deep enough to maintain the 1.5 diameters depth rule. Silver solder into the nozzle and skim the top face flat to get some nice sharp edges. You will see how this maintains the designed outer edge of the blast cone, at least for about 40% of the circumference. Slightly ironic because I intended to try a multi-nozzle blastpipe before I was introduced to the Jet Pump principle... but I could find nothing whatever in print or on the web about the geometry of a Lemaitre or any other mutiple nozzle setup. A Jet Pump (or a Koopmans for that matter) blast should not need this extra embellishment, but as I explained earlier, the oversize stovepipe chimney on Bridget required an oversize blast nozzle, to the extent that there was effectively no nozzle at all. The fact that this created enough blast to draw the fire at all is testament to the efficiency of the proportions, and it would keep the fire going on a run too, but there was no reserve whatsoever. This mod sharpened the blast enough to fix the problem entirely. Hope this helps Gary
|
|
chrisb
Part of the e-furniture
Posts: 340
|
Post by chrisb on Jul 9, 2020 10:17:52 GMT
I need to have a look at my Wren to see what I can do with it and the limitations of no petticoat pipe and a tall chimney, albeit with a slight taper
|
|
|
Post by ettingtonliam on Jul 9, 2020 14:27:21 GMT
Locomotion has a 2" dia chimney, 12" long, with the exhaust pipes feeding in about 7" from the top. There is a gunmetal casting sleeved into the chimney above the blast nozzle, and this forms petticoat pipe and choke. Because its sleeved into the chimney, it will be easy to adjust the height above the blast nozzle. I did discuss this with Julian some time ago, and we concluded that the whole thing was pretty sub-optimal by modern standards, so I'll let you know how it works, about 5 years time probably.
|
|
|
Post by keith1500 on Jul 9, 2020 20:58:38 GMT
Locomotion has a 2" dia chimney, 12" long, with the exhaust pipes feeding in about 7" from the top. There is a gunmetal casting sleeved into the chimney above the blast nozzle, and this forms petticoat pipe and choke. Because its sleeved into the chimney, it will be easy to adjust the height above the blast nozzle. I did discuss this with Julian some time ago, and we concluded that the whole thing was pretty sub-optimal by modern standards, so I'll let you know how it works, about 5 years time probably. Sounds similar to my OS model of Rocket. The copper exhaust steam pipes simply enter the chimney turn 90 degrees to point upwards. The ends of the pipes are reduced to form a nozzle. There is no choke and the pipes are not in the centre of the chimney but it works extremely well. In fact I never driven a model steam locomotive that produces a near incandescent fire as Rockets does.
|
|
kipford
Statesman
Building a Don Young 5" Gauge Aspinall Class 27
Posts: 566
|
Post by kipford on Jul 9, 2020 21:25:58 GMT
Tweaked the Aspinall this evening. Starting from the top of the chimney I started the 1:6 2 mm from the top. Keeping the nozzle diameter moved it up about 14 mm (ties in with Julians thoughts). If you then plot in the 1:3 it comes about 0.5 mm above the existing throat. So the only mod required is to split the blast nozzle pipe and put a spacer in, I will probably take the opportunity to make it a screw in part so the nozzle diameter can be changed if required. Dave Aspinall Draughting 1 by Dave Smith, on Flickr
|
|
|
Post by rogsteam1959 on Jul 10, 2020 22:37:15 GMT
Hi Gary, Any chance you could post a picture of your nozzle please? I get the idea but a picture would be great Cheers Kerrin Hi both A Jimmy was indeed against the rules, but I've read that shed fitters would make them for drivers for a consideration. In principle they were a bar that clamped across the blastpipe, with a safety chain attached, because dropping it down the blastpipe (or shooting it into the air with the exhaust) were equally disliked. However this is definitely a Loco Dept approved (by me) modification! The original nozzle was 7/16" dia. The spider was made by drilling 4 holes in some stainless stock, then turning the outside to break open the holes at the required 7/16" dia. Part off, making sure the spider is deep enough to maintain the 1.5 diameters depth rule. Silver solder into the nozzle and skim the top face flat to get some nice sharp edges. You will see how this maintains the designed outer edge of the blast cone, at least for about 40% of the circumference. Slightly ironic because I intended to try a multi-nozzle blastpipe before I was introduced to the Jet Pump principle... but I could find nothing whatever in print or on the web about the geometry of a Lemaitre or any other mutiple nozzle setup. A Jet Pump (or a Koopmans for that matter) blast should not need this extra embellishment, but as I explained earlier, the oversize stovepipe chimney on Bridget required an oversize blast nozzle, to the extent that there was effectively no nozzle at all. The fact that this created enough blast to draw the fire at all is testament to the efficiency of the proportions, and it would keep the fire going on a run too, but there was no reserve whatsoever. This mod sharpened the blast enough to fix the problem entirely. Hope this helps Gary Hi Gary You are talking about the 1,5 rule for the height of the spider. How critical is the factor? What if it would be made longer?
|
|
Gary L
Elder Statesman
Posts: 1,208
|
Post by Gary L on Jul 10, 2020 23:46:53 GMT
Hi both A Jimmy was indeed against the rules, but I've read that shed fitters would make them for drivers for a consideration. In principle they were a bar that clamped across the blastpipe, with a safety chain attached, because dropping it down the blastpipe (or shooting it into the air with the exhaust) were equally disliked. However this is definitely a Loco Dept approved (by me) modification! The original nozzle was 7/16" dia. The spider was made by drilling 4 holes in some stainless stock, then turning the outside to break open the holes at the required 7/16" dia. Part off, making sure the spider is deep enough to maintain the 1.5 diameters depth rule. Silver solder into the nozzle and skim the top face flat to get some nice sharp edges. You will see how this maintains the designed outer edge of the blast cone, at least for about 40% of the circumference. Slightly ironic because I intended to try a multi-nozzle blastpipe before I was introduced to the Jet Pump principle... but I could find nothing whatever in print or on the web about the geometry of a Lemaitre or any other mutiple nozzle setup. A Jet Pump (or a Koopmans for that matter) blast should not need this extra embellishment, but as I explained earlier, the oversize stovepipe chimney on Bridget required an oversize blast nozzle, to the extent that there was effectively no nozzle at all. The fact that this created enough blast to draw the fire at all is testament to the efficiency of the proportions, and it would keep the fire going on a run too, but there was no reserve whatsoever. This mod sharpened the blast enough to fix the problem entirely. Hope this helps Gary Hi Gary You are talking about the 1,5 rule for the height of the spider. How critical is the factor? What if it would be made longer? Oh yes, it can be longer, but shouldn't be shorter. The rule is that you don't constrict the jet itself (which is counter-intuitive; most traditional blastpipe designs, full size or miniature do this) because the constriction causes turbulent flow. To smooth out the turbulence, the nozzle should be straight and parallel for not less than 1.5 diameters below the top edge, which must be sharp to induce a clean breakaway. (There is a learned-sounding name for the point where this parallel section joins the ordinary pipework below, but I've forgotten it!). It goes without saying that in the case of my 4-fold nozzle, the parallel portion is 1.5x the dia of an individual nozzle, not the 7/16" dia of the original calculated Jet Pump nozzle. Hope this helps Gary
|
|
|
Post by rogsteam1959 on Jul 11, 2020 11:12:54 GMT
Hi Gary Sorry, but I’m confused right now. You chose the same outer dia for your spider. 7/16” and the 1,5 height of the spider is not the height to 7/16” nozzle ? It’s the height for the bores you chosed for 4 holes to make the spider? Let’s assume you made 4 holes of 6mm diameter, so the height of the Spider needs to be 9 mm minimum in height? Where can I find something about this? Thank you.
|
|
Gary L
Elder Statesman
Posts: 1,208
|
Post by Gary L on Jul 12, 2020 0:34:56 GMT
Hi Gary Sorry, but I’m confused right now. You chose the same outer dia for your spider. 7/16” and the 1,5 height of the spider is not the height to 7/16” nozzle ? It’s the height for the bores you chosed for 4 holes to make the spider? Let’s assume you made 4 holes of 6mm diameter, so the height of the Spider needs to be 9 mm minimum in height? Where can I find something about this? Thank you. Correct in every detail. I can't give you a reference on Jet Pumpery, my good friend Dr Mark Phillips is our club expert, and he doesn't post here. He's a Chemical Engineer who specialised in a different sort of blast (on behalf of HMG I might add!), and he needed a device to pump unstable liquids around without stirring them into unseemely aggravation. It was something of a boom industry at the time. Applying it to steam loco draughting was a flash of inspiration you might say. Certainly one or two IMLEC winners have benefited from his advice. You will have noticed that there is no mention of 1:3 and 1:6 cones as such in the drawing, and from what I remember, not in the Koopmans' formula either (though neither are a long way out; it is just that the science has moved on.) There is also no mention of grate areas, cylinder volume or any other extraneous factors. Mark is adamant about that; the only existing dimension that matters, is the diameter of the top of the chimney. Everything else is developed from that. This is parallel to draughting developments in the last days of steam. You might remember the Geisl ejector. It came too late to make a significant difference, but BR tested a few, and one of the Talyllyn locos still carries one. It came in different sizes; the internal proportions were always the same, you just ordered the size that best fitted your loco. HTH Gary
|
|
mbrown
Elder Statesman
Posts: 1,718
|
Post by mbrown on Jul 12, 2020 7:30:54 GMT
Talyllyn No.4, carried the Giesl Ejector from 1959 to 1969. It worked fine, but despite dramatic claims of fuel savings, the main difference when it reverted to a conventional chimney was that it became much quicker to raise steam from cold. Maybe the smoke found it harder to escape via the narrow opening without the help of the blower...
It was taken off in the end because the thin plates it was made of corroded badly. Conventional cast chimneys sometimes corroded and lost the lip at the top, but never right through as far as I know.
I believe the Giesl fitted Bullied pacific was well thought of - not sure about the 9F.
Malcolm
|
|