|
Post by Roger on Aug 26, 2020 18:37:42 GMT
Auto paint restorer/cutting compounds will do the truck Roger. You need a test piece to try on first to check if the paint has cured enough, this can take weeks or even months. I sprayed a sheet of Alloy today straight after finishing the topcoat on the running boards. That will be my test peice. Pete Thanks Pete, I'll wait for this to properly cure, I don't need to fit it for some time.
|
|
|
Post by Roger on Aug 26, 2020 21:16:02 GMT
I didn't like the result I got from spraying this all in one pass, so this time I'm going to do it in stages. These parts can be sprayed more easily with the AirBrush, so I'll do them first. 20200826_214407 by Roger Froud, on Flickr 20200826_214418 by Roger Froud, on Flickr One of the problems with spraying this in one pass is that you get a shadow from the handle. That means you have to spray it slightly from either side to reach everywhere. Unfortunately, when you do that, it's easy to end up will too much paint on the handle. This time I'll paint the handle separately, and then mask that off so I can focus on what the frame plate needs without having to worry about the handle. 20200826_220153 by Roger Froud, on Flickr I think this is going to be the way forward with the Pannier Tanks and Bunker too. I can't see how they can all be sprayed successfully in one pass.
|
|
|
Post by David on Aug 27, 2020 1:33:58 GMT
Looks pretty good from here. I'd be very happy with that.
|
|
|
Post by Roger on Aug 27, 2020 9:01:21 GMT
I thought I'd just share this next incremental step in the design of the scale injectors. Ignore the RH side, I'm still sorting out the steam inlet arrangement. So this is quite a radical and speculative departure from the usual arrangement, because you can see that there's no ball valve in the usual place. Usually the valve is between the gap in the Mixing/Combining cone zone and the delivery cone zone. On this version of the cones, those two gaps as indicated by the three rows of tiny holes in the LH cone which has the Mixing, Combining and Delivey cones in one piece. So what I've done is to effectively leave that usual ball out altogether. However, you still need to prevent air from being sucked into the gap in the Mixing/Combining cone. To achieve that, I've added the ball valve to the overflow like it is in the full sized injector. The argument is that if air can't get into the whole area, then that achieves the same thing. Space is pretty limited in the overflow valve, So I've cheated and made the seat part of the injector body so it can be further up. That makes it easier to create a clearance round the ball. There will be a spring under the plunger by the way. Injector with overflow valve by Roger Froud, on Flickr The big question is, will this work? I've tried to imagine how the injector starts and restarts in various conditions, and I really can't see why this won't work. We know that injectors will still work even if the ball is removed, they just draw in unwanted air and therefore stall earlier than they otherwise would. We also know that this... Hot water injector by Roger Froud, on Flickr ... design from Bob Bramson's book also works. That particular design has the usual ball and an additional one for the overflow. This is the same principle used in the Chiverton injectors where there are two disc valves performing the same functions. The second valve prevents air being drawn into the boiler. This is desirable for Steel boilers, but it also has the effect of allowing the injector to work with warmer water. Presumably this is because air drawn into the delivery cone is compressible thus causing the injector to stall prematurely. You want a solid column of water in the delivery cone. I guess the only way to find out is to make it and see if I'm right. It simplifies things no end if it can be done this way.
|
|
|
Post by Roger on Aug 27, 2020 16:21:44 GMT
Here's a little more detail finished on the injector assembly. As always, everything's a compromise because I'm not starting with the shape I'd like for the body. However, I can't do anything about that, scale is scale. I'm using 'O' rings thoughout because it makes the fits less critical and assembly/disassembly easier. There won't be a huge amount of compression on the rings so they can pass the pipe diameters. Since this is a pain to take off the locomotive, I thought it would be handy to be able to slide out the innards while it's still in situ. The one piece Combining/mixing/delivery cone makes the alignment of those simpler and it won't be affected by the position on the body. The same goes for the Steam cone which is butted up against a thin walled spacer to set the regulation gap. The axial relationship between the Steam Cone and the Condensing Cone isn't quite as critical since I'm using End Regulation and the Steam Cone doesn't need to create an annular gap inside the Condensing Cone. Sectioned injector V1.0 by Roger Froud, on Flickr Anyway, in case it's not clear what the individual cones look like, here's the Steam Cone. I've tried to allow enough room for the water to pass along the outside and get past the spider that keeps it central. Steam cone by Roger Froud, on Flickr Here's the Steam Cone sectioned too. Sectioned steam cone by Roger Froud, on Flickr The one piece Condensing/Mixing/Delivery Cone looks like this. Although the vent holes break into each other in the middle, there's enough meat left on the outside for them to provide enough support so it isn't fragile. One piece cone by Roger Froud, on Flickr I understand that drilled holes have been tried before without success, but I doubt if it's been tried like this with three rows. I have no idea whether this will work or not. I understand that the flow necks in when it exits an edge, so it may not mind what I'd done here. I won't know until I've tried it. My instincts tell me this has a good chance of working, else I wouldn't be doing it! The holes aren't something I want to index by hand, it's too tedious. So this is likely to be the part that makes me finally make that 4-th axis I've been meaning to make for years but never had a compelling reason to do so. With that, it won't be any more work than drilling a matrix of holes in a flat plate. So although this looks like it's very complicated, it's actually pretty simple to make. The outside is a simple turning job, the only critical thing being the overall length so that I can insert the reamers the correct distance from either end. If the holes don't work, I can put a big slot across there, leaving just enough of the body so it isn't too flimsy to insert into the housing. Sectioned one piece cone by Roger Froud, on Flickr
|
|
|
Post by Roger on Aug 27, 2020 19:17:21 GMT
Ok, I decided to bite the bullet and buy this decent sized (6") Universal Dividing Head with a view to converting it to a 4th axis and replacing the rather light tilting rotary table I've been using. I've seen several different designs, but I like this one the best because it looks really robust. It also looks like it ought to be straightforward to add a toothed belt in place of the graduated ring on the front and a substantial mount for the AC Servo that I bought for this project way back in 2008 when I converted the machine to CNC. I won't flesh out the design until it's here and I can measure it up. I want to have the Servo motor sticking out the front where it won't foul anything on the machine. Ideally I'd like to be able to mount it either way round, so the Servo could be fitted on either side. We'll see.
|
|
|
Post by Roger on Aug 27, 2020 22:02:03 GMT
The quick release fittings arrived today, so here's the first one fitted to the big spray gun. The one on the coily hose was aleady there, so this is much more convenient for cleaning up. 20200827_221606 by Roger Froud, on Flickr I've also replaced the 8mm hose fitting on the filter arrangement to a female quick release type. Here it's being modelled with the 6mm hose extension used for the AirBrush. 20200827_222740 by Roger Froud, on Flickr And finally, this is the 8mm extension used in the Garden Office for the big spray gun. This is all a great improvement, it's much simpler to switch things round now. 20200827_222846 by Roger Froud, on Flickr
|
|
|
Post by terrier060 on Aug 27, 2020 23:14:56 GMT
Where did you get the nice quick release fittings from, and what size are they?
|
|
|
Post by David on Aug 27, 2020 23:15:20 GMT
It will be great fun watching your injector development in the metal :) If anyone's going to show it can be done differently these days...
Rather than a ball sitting on the plunger and forming the seal, why not an o-ring on or around the head of the plunger?
And where you're drilling the holes in the cones maybe you could mill out between the 3 holes axially and form slot. If you're using a tiny carbide PCB drill to make the holes would it stand the lateral forces required for that?
|
|
|
Post by terrier060 on Aug 27, 2020 23:31:48 GMT
Very novel injectors Roger. I really hope they work. We could do with some new designs. I have given in and bought a Len Steel injector which is basically a Chiverton.
|
|
|
Post by Roger on Aug 28, 2020 6:59:17 GMT
Where did you get the nice quick release fittings from, and what size are they? Hi Ed, If you search eBay for 'Rectus type 21' you'll find loads of them. Mine came from Context Pneumatics, but you can also get them from any mainstream supplier. They're designed for water systems, but they work just as well for pneumatics.
|
|
|
Post by Roger on Aug 28, 2020 7:03:57 GMT
It will be great fun watching your injector development in the metal If anyone's going to show it can be done differently these days... Rather than a ball sitting on the plunger and forming the seal, why not an o-ring on or around the head of the plunger? And where you're drilling the holes in the cones maybe you could mill out between the 3 holes axially and form slot. If you're using a tiny carbide PCB drill to make the holes would it stand the lateral forces required for that? Hi David, The ball is being held up to the seat on the upper part of the injector by the plunger, there's no seal against the plunger. You certainly could mill out between the holes, but I was trying to make a consistent edge all the way around. My thought was that it would create a smoother flow? Who knows, there's plenty of room for experimentation. You'd need to use a PCB routing bit for that, but it would certainly be possible.
|
|
|
Post by Roger on Aug 28, 2020 7:04:58 GMT
Very novel injectors Roger. I really hope they work. We could do with some new designs. I have given in and bought a Len Steel injector which is basically a Chiverton. If that's a Chiverton design then it ought to be good. The ones I've seen have two valves, so it ought to work with warmer water too.
|
|
|
Post by 92220 on Aug 28, 2020 7:36:07 GMT
Ok, I decided to bite the bullet and buy this decent sized (6") Universal Dividing Head with a view to converting it to a 4th axis and replacing the rather light tilting rotary table I've been using. I've seen several different designs, but I like this one the best because it looks really robust. It also looks like it ought to be straightforward to add a toothed belt in place of the graduated ring on the front and a substantial mount for the AC Servo that I bought for this project way back in 2008 when I converted the machine to CNC. I won't flesh out the design until it's here and I can measure it up. I want to have the Servo motor sticking out the front where it won't foul anything on the machine. Ideally I'd like to be able to mount it either way round, so the Servo could be fitted on either side. We'll see. Hi Roger. I have one of these (the 4.1/2" version), and it has just started falling apart!! The lever that operates the pin that locates in holes in the dividing ring, just pulls straight out now because something has fallen off the inner end. I also want to try and adjust the considerable backlash in the internal gearing and it only has a large slotted screw, and nothing I do will turn this screw! It should be a hex head bolt, like in all the photos in the various listings I've seen, but it is just a slotted screw head about 15mm in diameter. I am now watching out for an old Tom Senior dividing head, on Ebay, or any of the other decent industrial dividing heads. Bob. Hi Roger. I've just noticed in the listing you point to, you can see the backlash adjusting screw in the side opposite the operating handle....the large screw near the bottom of the rotary section. Bob.
|
|
|
Post by Roger on Aug 28, 2020 8:17:13 GMT
Ok, I decided to bite the bullet and buy this decent sized (6") Universal Dividing Head with a view to converting it to a 4th axis and replacing the rather light tilting rotary table I've been using. I've seen several different designs, but I like this one the best because it looks really robust. It also looks like it ought to be straightforward to add a toothed belt in place of the graduated ring on the front and a substantial mount for the AC Servo that I bought for this project way back in 2008 when I converted the machine to CNC. I won't flesh out the design until it's here and I can measure it up. I want to have the Servo motor sticking out the front where it won't foul anything on the machine. Ideally I'd like to be able to mount it either way round, so the Servo could be fitted on either side. We'll see. Hi Roger. I have one of these (the 4.1/2" version), and it has just started falling apart!! The lever that operates the pin that locates in holes in the dividing ring, just pulls straight out now because something has fallen off the inner end. I also want to try and adjust the considerable backlash in the internal gearing and it only has a large slotted screw, and nothing I do will turn this screw! It should be a hex head bolt, like in all the photos in the various listings I've seen, but it is just a slotted screw head about 15mm in diameter. I am now watching out for an old Tom Senior dividing head, on Ebay, or any of the other decent industrial dividing heads. Bob. Hi Roger. I've just noticed in the listing you point to, you can see the backlash adjusting screw in the side opposite the operating handle....the large screw near the bottom of the rotary section. Bob. Hi Bob, I'm scrapping the internals, I just want it for the casting and bearing arrangement. I want to replace the graduated ring with a toothed belt. I'm going to attach an AC Servo with a small toothed pulley to the side with a bracket. That will have a tensioner and provide the indexing electronically. The idea is to have the motor stand out well to the front. Hopefully the bearing arrangement is good enough. I might have to re-engineer it if not. So the purchase is really just to get the basic structure of a 4th axis without having to design and make the whole thing. It looks pretty robust, and it's larger than most of the ones listed on eBay at 160mm. The problem I have with my existing one is it isn't that rigid. I think it was intended for use in a metrology lab. It doesn't have a fine adjustment either, which isn't ideal. Hopefully this one will replace that altogether.
|
|
|
Post by 92220 on Aug 28, 2020 8:36:34 GMT
Hi Roger.
Please ignore everything I wrote in my post above. I did a bit more digging on the 'net, and found a reference to backlash adjustment. It IS there, but not where the manual says it is. If you take off the dividing handle and plate, the adjustment is hidden inside the cast boss that has the oiling point for the spindle. Backlash can be adjusted out but it is very easy to go too far and bind it all up. There are a couple of holes that look as if they may be for a pin type 'spanner'. I adjusted the backlash until it was too tight and then with a rod in one of the spannewr holes, I tapped it looser with a little pin hammer. It only needed a few thou movement to release it and allow almost backlash-free rotation of the handle. What that screw is that I tried to adjust backlash with I have no idea. The manual I downloaded said that is what it was but it definitely isn't, and there is an adjustment behind the dividing plate, so I was quite wrong in what I wrote in the post above.
Bob.
Edit: Hi Roger. If you search on Ebay for just 'dividing head', there are a lot of 4th axis heads for CNC which might be worth looking at as they are around the same price as the Vertex 6" head. Maybe one of them would do.
Bob.
|
|
|
Post by Roger on Aug 28, 2020 8:52:13 GMT
Hi Roger. Please ignore everything I wrote in my post above. I did a bit more digging on the 'net, and found a reference to backlash adjustment. It IS there, but not where the manual says it is. If you take off the dividing handle and plate, the adjustment is hidden inside the cast boss that has the oiling point for the spindle. Backlash can be adjusted out but it is very easy to go too far and bind it all up. There are a couple of holes that look as if they may be for a pin type 'spanner'. I adjusted the backlash until it was too tight and then with a rod in one of the spannewr holes, I tapped it looser with a little pin hammer. It only needed a few thou movement to release it and allow almost backlash-free rotation of the handle. What that screw is that I tried to adjust backlash with I have no idea. The manual I downloaded said that is what it was but it definitely isn't, and there is an adjustment behind the dividing plate, so I was quite wrong in what I wrote in the post above. Bob. Edit: Hi Roger. If you search on Ebay for just 'dividing head', there are a lot of 4th axis heads for CNC which might be worth looking at as they are around the same price as the Vertex 6" head. Maybe one of them would do. Bob. Hi Bob, Thanks for that. I'll be removing all of the internal gears though. I don't think it's feasible to create a zero backlash drive using that sort of mechanism. I'll drive it directly using a belt over the outside, behind the chuck. That way there's no backlash. I've looked at lots of 4th axis units on eBay and I don't think any of them are suitable. There are ones with a fixed axis and they look very light weight. Others are full CNC versions that cost many thousands. I haven't seen any that do what I want. Hopefully all will become clear when I show the model of what I'm proposing.
|
|
uuu
Elder Statesman
your message here...
Posts: 2,849
|
Post by uuu on Aug 28, 2020 14:28:16 GMT
I've seen a small 4th Axis in use - like a tiny rotary table with a stepper motor turning the worm. I doubt it had much in the way of backlash adjustment, but for the small jobs it did it was entirely satisfactory. Things like drilling holes in the heads of small bolts for wiring them to prevent undoing.
So really very similar to your project. But no doubt, once you've got yours, you'll be refining your part-on-a-stick procedures to flip things over and do both sides in one run.
Wilf
|
|
|
Post by Roger on Aug 28, 2020 15:46:22 GMT
I've seen a small 4th Axis in use - like a tiny rotary table with a stepper motor turning the worm. I doubt it had much in the way of backlash adjustment, but for the small jobs it did it was entirely satisfactory. Things like drilling holes in the heads of small bolts for wiring them to prevent undoing. So really very similar to your project. But no doubt, once you've got yours, you'll be refining your part-on-a-stick procedures to flip things over and do both sides in one run. Wilf Hi Wilf, I think Doug has done this too. I seem to recall he used grinding paste to work the worm into mesh and improve the backlash. I'm sure it can be done, but I think a direct toothed belt operating on the back of the chuck is a more satisfactory solution. I don't want this to just be of the odd light job where I need a 4th axis. I want this to be my main rotary table too, rigid enough to take decent cuts. With about a 1:6 reduction from the powerful AC Servo, that ought to be pretty rigid and with zero backlash for all practical purposes. I can't really see me using for true 4 axis work unless I use Fusion360 for the tool paths. Alibre CAM won't do it unless you have the Expert version, for which they want a King's ransom. I can see it being useful for gear cutting though, not that I do much of that. I'm sure I can write a little utility to create a G-Code program with any number of equally spaced divisions.
|
|
|
Post by Doug on Aug 28, 2020 22:34:35 GMT
Hi yeah I used the worm/wheel setup I ended up getting a bigger servo so I could tighten up the worm into the gear. I have noticed that closed loop stepper servo systems are getting much cheaper so that would be an excellent way to do it if you were using steppers. Definitely think you have the right idea Roger going with direct drive, much better if you have the torque in your drive system.
|
|