|
Post by Roger on Oct 3, 2019 6:59:48 GMT
That's a neat filter element, that could definitely be used if the other options don't work out. I rather liked another one which had even the threaded part sintered as one with the filter. Unfortunately that one came with a Stainless Steel housing from Germany with a total cost of about £30 when you added the carriage. Pity because it was ideal. Hopefully this one will prove to be acceptable, but I'll need to modify the housing to get to the spanner flats. That's not a problem though, I've already modelled the new end shape.
|
|
|
Post by jon38r80 on Oct 3, 2019 8:17:48 GMT
I had a look on Aliexpress ,the Chinese market, and you can get the sintered element on its own which with your ingenuity and some O rings I'm sure you could make a workable filter. I know it's nice to have the threaded element for assembly but that seems to be the source of the problem. I searched for sintered bronze to find the media but I can't find a way yet to give you a link to a specific product on Aliexpress .
|
|
|
Post by Roger on Oct 3, 2019 8:26:34 GMT
I had a look on Aliexpress ,the Chinese market, and you can get the sintered element on its own which with your ingenuity and some O rings I'm sure you could make a workable filter. I know it's nice to have the threaded element for assembly but that seems to be the source of the problem. I searched for sintered bronze to find the media but I can't find a way yet to give you a link to a specific product on Aliexpress . I hadn't thought to look that far afield, but that's certainly another possiblity. I've bought Carbide cutters from there in the past. I'm hoping that the Brass body is going to be compact enough on the all metal one I've ordered. If not, I suppose I could always turn off the bit that gets in the way. The thread is only to stop it falling out, there's no force on it so I could just srew it in lightly with my fingers. I think I prefer an 1/8" BSPP threaded attachement, it's a convenient size for the application.
|
|
|
Post by Roger on Oct 3, 2019 19:56:05 GMT
Ok, don't take this too seriously, it's just an experiment.... a 3D printed whistle! This is based on a design kindly sent to me by Gary L which looks like this when it's made from separate pieces in Brass... Whistle assembly section by Timothy Froud, on Flickr This is my interpretation of what that looks like to make it printable. It's shown here without the top cap which would have to be separately printed and glued on. It might be possible to bridge across the top, but it's probably more sensible to just print it separately. The first thing to note is that the wall thicknesses are significantly greater. Since this is plastic, condensation may not be such a big issue as it doesn't conduct heat that well. 3D printed whistle by Timothy Froud, on Flickr Here are a few sections taken at various angles to show the internals. The feed hole splits into two oval shaped ones that supply the cavities under the languid. I've kept the bridge piece that splits the large single opening into two for the purpose of giving more strength. The shallow taper from the base is necessary to be able to print this without supports. Section1 by Timothy Froud, on Flickr Supports will be needed for the feed hole, but those can be removed easily enough and then that can be tapped M8 coarse for a union. Section2 by Timothy Froud, on Flickr Here's how the steam splits to get under the languid... Section3 by Timothy Froud, on Flickr ... and another view along one of the oval passages. Ovals are easy to print as long as the tall side is vertical to the direction of the print. Section4 by Timothy Froud, on Flickr Now I've probably made an awful gaffe somewhere in the design, so do correct me if something looks hopelessly wrong! It's just a bit of fun that mught just work, who knows until you've tried it. This can be printed in one of the higher temperature plastics such as ColorFabb HT which ought to cope with the relatively short blasts of steam it has to deal with.
|
|
timo
E-xcellent poster
Completing 3 1/2 Rainhill .Building 5" Railmotor and waiting to start 3 1/2" King
Posts: 234
|
Post by timo on Oct 3, 2019 21:47:29 GMT
Roger, "It's shown here without the top cap which would have to be separately printed and glued on" Perhaps better to make it a sliding fit and then you can tune the whistle, finally gluing it in place when you have the right tone. Why not go the whole hog and make a chime whistle
Tim
|
|
|
Post by Roger on Oct 3, 2019 22:10:28 GMT
Roger, "It's shown here without the top cap which would have to be separately printed and glued on" Perhaps better to make it a sliding fit and then you can tune the whistle, finally gluing it in place when you have the right tone. Why not go the whole hog and make a chime whistle
Tim
Hi Tim, Well, the length should be exactly right for G# which is for the main whistle. 1501 doesn't have a chime whistle but it does have a second whistle which is D# but that's too big to fit. I'll print it soon if there are no glaring errors then see if it works.
|
|
|
Post by chris vine on Oct 4, 2019 7:40:52 GMT
Presumably, Roger, with 3D printing you could make the resonating tube any shape you like in order to get the full scale length correct. Like a Tuba/Trombone etc.
Chris.
|
|
|
Post by Roger on Oct 4, 2019 8:07:50 GMT
Presumably, Roger, with 3D printing you could make the resonating tube any shape you like in order to get the full scale length correct. Like a Tuba/Trombone etc. Chris. Hi Chris, I honestly have no idea, I don't understand these things to any great depth. There are constraints as to what you can 3D print, but there's a lot of flexibility. I'm not sure what your suggestion would entail so I can't really make a sensible comment about whether you could to what you say or not.
|
|
|
Post by Roger on Oct 4, 2019 18:47:50 GMT
One quick update on the transfer valve actuating arm which was going to have a collar to prevent it from sliding away from the frame. In the end I decided to go with the same solution I used on the lubricator, ie using an 'E' clip. Here's the 0.7mm wide groove going in to a diameter of 3.2mm on the 4mm shaft. I couldn't take the levers off, so I had to hold it by the short end. 20191004_153853 by Timothy Froud, on Flickr This is how it looks in situ, you can't really see it from outside of the frames. 20191004_154542 by Timothy Froud, on Flickr
|
|
|
Post by Roger on Oct 4, 2019 22:09:27 GMT
This is my first draft of the Whistle servo valve, based on a sketch of a whistle with this integrated into the base. I don't recall who sent it to me, but thanks to whoever that was. The idea is very simple in that the diaphragm is 0.5mm Silicone Rubber sheet which is moved to the right by steam from the small bore whistle valve. The pipe will be in the region of 1.6mm OD for that. The bottom pipe will be connected to the middle of the three turret valves with at least a 4mm pipe. I've shown this withfittings for 5mm OD pipe, I'm not sure what I'll use as yet. The design I've copied uses a spring to close the valve, presumably to stop it weeping when raising steam. I might add one there too. It won't be doing much when there's full steam pressure on the RH side of the diaphragm. The inlet pipe is Silver Soldered into position so it can all be as compact as possible. Anyway, I'm sure quite a few details will change but it's more or less in the shape it needs to be. The outside is 18mm diameter so you can get an idea of how large it is. Servo valve assembly sectioned by Timothy Froud, on Flickr
|
|
|
Post by Roger on Oct 5, 2019 8:21:54 GMT
Just a quick update on the blower design... I've been experimenting with how to make some shapes easier to 3D print, particularly circular shapes that are on their side. These things can't be printed without some sort of support because the ever increasing overhang evenually results in the filament sagging as it's laid down. I've tried different supports generated by the slicer without much success, the finish is usually terrible if you're using a more difficult plastic. So to that end, I've added thin wafers of material to the 3D model which translate into supporting wafers that allow the filament to bridge between them better. Obviously you need to carefully remove them with a scalpel or similar very sharp blade, but the finish is better overall, and more importantly the dimensions are better when you come to fit the Steel parts. NB:- To slice this with the support wafers, you need to tell the slicer to detect thin walls else they will be too thin to print. This forces the slicer to generate a single nozzle width of material for each of the wafers. Blower body with support wafers by Timothy Froud, on Flickr
|
|
Gary L
Elder Statesman
Posts: 1,208
|
Post by Gary L on Oct 5, 2019 12:25:48 GMT
Ok, don't take this too seriously, it's just an experiment.... a 3D printed whistle! This is based on a design kindly sent to me by Gary L which looks like this when it's made from separate pieces in Brass... Whistle assembly section by Timothy Froud, on Flickr This is my interpretation of what that looks like to make it printable. It's shown here without the top cap which would have to be separately printed and glued on. It might be possible to bridge across the top, but it's probably more sensible to just print it separately. The first thing to note is that the wall thicknesses are significantly greater. Since this is plastic, condensation may not be such a big issue as it doesn't conduct heat that well. [Snip] Ha ha, it had to happen! That's Roger all over! First of all Roger, I can confirm that your first model (as above) is correct. For those without access to a 3D printer it is a very neat principle (not mine) which gets around the difficulty of cutting a very narrow, accurate slot for directing steam past the mouth of the whistle. In this design, the slot width is determined by the material thickness of the tube, and issues with the slot filling with solder as in the traditional construction are a thing of the past- though obviously care is needed. Solder paint can be used for the entire assembly, so there is little risk of blobs ending up in the wrong place. As long as you use thin-wall brass tube (not plumber's pipe!) there is little to go wrong. 0.015" is a suitable tube thickness, but not critical within reason. As far as tuning it goes, one dimension only is critical, and tables exist, so a specific note is not hard to achieve if your work is accurate. For example, to get G# , the correct note for a GWR service whistle, requires the bell (tube) dimension from the top of the mouth to the inside of the top of the bell to be either 9.61" or 4.80", depending on what you have room for. The shorter dimension gives a note that is one octave higher than the longer; it is impossible (sadly) to replicate the octave of the original as the whistle would be longer than the loco! The mouth needs to be rectangular in shape, the height should be half the dia. of the bell, and it should extend for half the circumference around the bell (though for strength, this need not be contiguous). But making it of plastic, that is novel, and we might have guessed Roger would try something different! I can't see any reason why it wouldn't work, in principle. Traditional referees' whistles are available in plastic as well as metal. Purists might discern a difference in timbre, but I doubt if an average person would notice any difference. The sound is generated by the vibration caused by a rapid alternation of gas flow between the inside and the outside of the bell, so I would expect the bevel on the outer top edge of the mouth to be critical, and might need some experimentation. I doubt that unsuperheated steam would have a bad effect on plastic, but if it softens appreciably it would probably not be good. And if the plastic expands a lot, the tuning might suffer. My (limited) experience is that if you can get any kind of note from a whistle by mouth, you will get a much stronger one from steam (though the pitch will be a bit different) so it won't be hard to judge if your plastic steam whistle is worth persevering with. We await results with interest! Gary
|
|
|
Post by Roger on Oct 5, 2019 12:52:12 GMT
Ok, here's the result of the overnight print. Actually it wasn't that long, just under 7 hours using 0.1mm layer height on the first more intricate section and 0.2mm above that. I'm sure it could be printed better with a smaller nozzle and optimising the print to get finer detail. 20191005_103656 by Timothy Froud, on Flickr So this doesn't look that promising... 20191005_132542 by Timothy Froud, on Flickr ... although it cleaned up really easily, taking a bit of care on the bevelled edge. 20191005_134330 by Timothy Froud, on Flickr Anyway, against all expectations, it seems to work surprisingly well! If you blow it very hard with the air line, it goes up an octave, but the natural gently blown tone is as in this very wobbly video. 20191005_133940 by Timothy Froud, on Flickr Out of interest, it takes very little flow to get a nice deep tone, so I was wondering if an air mover venturi arrangement could be used which would further reduce the temperature that the whistle has to deal with? Just a simple arrangement like a vacuum ejector which would draw air in and entrain it with the steam. I feel another experiment coming on...
|
|
|
Post by delaplume on Oct 5, 2019 14:01:28 GMT
Oh dear---------- sounds like an 08 shunter's whistle ???..........keep trying !!!
|
|
|
Post by Roger on Oct 5, 2019 15:18:21 GMT
Oh dear---------- sounds like an 08 shunter's whistle ..........keep trying !!! Sounds fine to me, but what do I know. Maybe it's an octave too low. I'll have to measure the frequency and see.
|
|
Gary L
Elder Statesman
Posts: 1,208
|
Post by Gary L on Oct 5, 2019 15:27:02 GMT
Oh dear---------- sounds like an 08 shunter's whistle ..........keep trying !!! Nah! I disagree. It is not far off the required note to my tin ear. It is 3 or 4 octaves high, but that can't be helped, it is down to the laws of physics- you can't scale sound. Under steam the note will be a bit different. Pretty good for a first attempt I would say, remarkable in fact. I don't think even Roger will opt for an electronic sound system to get a truly accurate sound Roger, if you arrange to drive it with your venturi device (using pure air?) you will need to recalculate the bell length. On the plus side, there has been lots of work done on organ pipes, so the tables can be relied on, whereas the steam calcs use a conversion factor and steam atmospheres can vary. -Gary
|
|
|
Post by delaplume on Oct 5, 2019 15:42:55 GMT
Oh dear---------- sounds like an 08 shunter's whistle ..........keep trying !!! Nah! I disagree. It is not far off the required note to my tin ear. It is 3 or 4 octaves high, but that can't be helped, it is down to the laws of physics- you can't scale sound. Under steam the note will be a bit different. Pretty good for a first attempt I would say, remarkable in fact. I don't think even Roger will opt for an electronic sound system to get a truly accurate sound Roger, if you arrange to drive it with your venturi device (using pure air?) you will need to recalculate the bell length. On the plus side, there has been lots of work done on organ pipes, so the tables can be relied on, whereas the steam calcs use a conversion factor and steam atmospheres can vary. -Gary I only mentioned it because I recently spent a couple of hours surrounded by three Class 08 shunters all tooting their little air-powered whistles off ........ it kinda stays in the old noddle a wee bit !! As Roger says --- it's only a bit of fun after all.........
|
|
|
Post by Roger on Oct 5, 2019 16:08:48 GMT
Oh dear---------- sounds like an 08 shunter's whistle ..........keep trying !!! Nah! I disagree. It is not far off the required note to my tin ear. It is 3 or 4 octaves high, but that can't be helped, it is down to the laws of physics- you can't scale sound. Under steam the note will be a bit different. Pretty good for a first attempt I would say, remarkable in fact. I don't think even Roger will opt for an electronic sound system to get a truly accurate sound Roger, if you arrange to drive it with your venturi device (using pure air?) you will need to recalculate the bell length. On the plus side, there has been lots of work done on organ pipes, so the tables can be relied on, whereas the steam calcs use a conversion factor and steam atmospheres can vary. -Gary If I blow it really hard, it whistles like this... 20191005_170008 by Timothy Froud, on Flickr ... but that doesn't seem to be the natural frequency. That's more like you get when you blow really hard on a recorder.... I'm sure we've all done that! Good point about the venturi though, I hadn't thought of that. I think I'll leave it at that, maybe try another print when the smaller nozzles I've ordered arrive.
|
|
|
Post by Roger on Oct 5, 2019 20:17:51 GMT
Here's a bit more detail of the finished Whistle Servo valve. The main steam fittings now use 4mm pipe which gives a 3.1mm bore. This doesn't have to be especially compact, but I don't want it to be massive either. Silver Soldering the inlet pipe to the side keeps the diameter down to a minimum. I've slightly increased the amount of room for steam to pass the valve, and reduced the volume on the RH side of the diaphragm since that's got to be filled and emptied through the 1.6mm pilot piping. Moving the pilot inlet away from the centre line allows that to clear the nut on the valve shaft, enabling the cover to be thinner. I've added a raised part around the valve to accommodate a spring which isn't modelled here. I might replace the valve part on the shaft with a PTFE ring that slides onto the shaft. That's more likely to be a good seal. Servo valve assembly sectioned1 by Timothy Froud, on Flickr The outlet thread and 'O' ring seating and OD have been machined on the lathe, so here I'm adding the pocket where the inlet union will be Silver Soldered. 20191005_170942 by Timothy Froud, on Flickr I can find the location of the sloping channel from the 3D model relative to the end chamfer, so that's what I'm using here for the setup. 20191005_171755 by Timothy Froud, on Flickr 20191005_172337 by Timothy Froud, on Flickr I hadn't made any of the 4mm unions and nuts, the only 4mm fittings I've made have been the flanged type for scale appearance. I'm going to use traditional fittings for the valve, so here's one of the nuts being finished on the lathe. I didn't have a mandrel for M7 x 0.75 (fine), but now I have. 20191005_205313 by Timothy Froud, on Flickr I need to get a bottom tap for these, I've only got a second. 20191005_205448 by Timothy Froud, on Flickr
|
|
jma1009
Elder Statesman
Posts: 5,913
|
Post by jma1009 on Oct 5, 2019 21:29:42 GMT
Hi Roger,
Sorry, your 'very wobbly video' didn't do it for me.
When I try my own whistles on my big compressor at 60 psi they don't over blow jumping an octave but give a clear loud note without sounding asthmatic like yours did in the above clip. I should add that lest one suspect plastic provides a poor note quality consider any penny whistle.
"2 out of ten, go to the back of the class" I say tongue in cheek, and know you will come up with something far more resembling full size in due course. And of course all GWR locos ought to have 2 whistles... see the first few minutes of the Ealing film Titfield Thunderbolt.
I personally consider the pressure valve unnecessary, but Roy Amesbury was very keen on them on his last 2 steam locos.
Cheers,
Julian
|
|