|
|
Post by Roger on Jan 17, 2021 23:43:39 GMT
Agreed, they simply don't make sense in my opinion. They also assume that your control will be able to make use of them, and mine certainly can't. There's a nice function in Peck Drilling where you can select an initial peck distance and have it reduce by a set amount, clipped to a minimum. You can also add a dwell. Neither of these things work in Mach4, and I suspect most other controllers. What they really ought to do is to remove functions based on what post processor you select, but that's too user centric. They just take the easy option and throw in everything the most complex machine could do. In the end, you just find what works and ignore the controls that seem pointless or just don't do what you might reasonably expect them to do. Frankly I don't think some the the height controls work as intended, it's pretty buggy in my opinion. I've sometimes thought the same, that the height settings are buggy, but I'm not sure. It's so complicated in the end I figure it was probably my mistake when something goes wrong. You do get some weird Z heights in the output though, especially when you use the stay down feature. While doing my mods to the post-processor I would put easily recognised values into the fields so I could calculate which combination of settings the final value came from. It was still confusing and didn't always make sense to me! I'm not convinced they should limit the exposed settings based upon the target machine. The post-processor would have to be changed to support a call back where it can report what g-codes and options it supports for one. That would be tedious but doable. Then you'd have the problem of a user choosing one target, setting the parameters, then choosing a different target that supported more or less parameters before generating the g-code. That's unlikely, but it is possible so would have to be handled. That would really worry me. I know that the top of the job is Z0, so anything less than that at the start is a red flag. I always know the clearance height I've set, say 1mm, so if the output doesn't do an initial move to Z1 then I won't run it. You need to be confident your output is what you expect or crashes soon follow. An alternative for the CAM output options would be to have a basic setup and then extended ones that aren't normally shown. I'm sure they haven't debugged the height settings properly in the drilling cycles. If that's not the case, they need to better explain why they work the way they do, because it certainly doesn't make sense to me. |
|
|
|
Post by Roger on Jan 18, 2021 12:19:53 GMT
Here are the first three One Piece Cone designs I'm going to try. They all share the same 8 x 0.5mm holes for the delivery overflow so this is just a series of experiments to see what different overflow arrangements for the Combining cone will work. One interesting aspect of using drilled holes for the overflows is the way they interact when they overlap. You can see from the delivery overflow, that adding ever more holes cleans up the remaining edge and also creates a kind of undercut which increases how much water can escape. These first three cones explore the situation where the holes don't overlap. The idea is to stagger the hole patterns so that the flow only has to jump one gap while the rest of the flow remains unaffected. Whether this is any better than a single large gap is anyone's guess. It's not been tried to my knowledge, so let's find out. The first cone has 6 x 0.5mm holes on both rows. The rows are as far apart as I can achieve in the available gap.  Cone1 by Georgia Montgomery, on Flickr The second cone has 3 x 0.75mm holes in two rows...  Cone2 by Georgia Montgomery, on Flickr ... while the third has just 2 x 0.95mm holes in two rows.  Cone3 by Georgia Montgomery, on Flickr Place your bets as to what any of this will do! |
|
jem
Elder Statesman
Posts: 1,075
|
Post by jem on Jan 18, 2021 17:23:38 GMT
Would these holes be better at an angle to the main axis, rather than at 90 degrees, or am I talking complete rubbish?
Jem
|
|
|
|
Post by Roger on Jan 18, 2021 18:03:34 GMT
Would these holes be better at an angle to the main axis, rather than at 90 degrees, or am I talking complete rubbish? Jem Hi Jem, That may be beneficial, but I don't think there's any way of achieving that in the space available. |
|
|
|
Post by Roger on Jan 18, 2021 23:42:50 GMT
This is the first of the three test cones I showed yesterday...  20210118_210641 by Georgia Montgomery, on Flickr ... and here's a video of the second one getting the two rows of 0.75mm Combining Cone. I've now drilled all of the holes in the first three cones. They just need finishing off so I can give them a try.  20210118_213355 by Georgia Montgomery, on Flickr When I tried the output from the post processor, post-post processed by my utility there were a few issues. The Drill output on the post processor did output the Feedrate, but for some reason, they didn't do that on the Peck Drill. That left it with an F1000 rapid feed which it would have used to drill the hole if I didn't do a dry run as a sanity check! I've added that to the Post Processor now. That was after I deleted the L0 command it also output on the G83 Peck Drill command, because for some reason Alibre CAM seems to think that's the Dwell time whereas in fact it's the number of repeats. I'm never going to use that, so I've just deleted it. I don't know why anyone would want to drill exactly the same hole L times, but I suppose you could skip the hole if you just set it to zero. Who knows. Another problem I thought I might have to deal with was the M3 turning on of the spindle. Mach4 isn't smart enough to realise that the spindle is already running, and so it puts in the startup delay every time the sub-program is repeated. The solution to this involved making bookmarks in the calling program where they would go if they were found. On the second pass I could then find any M3 lines and move them to the appropriate places. Because I'm making this to work with multiple repeated sections, the following bookmarks needed moving down each time I added a line, else it ended up in the wrong place. It took a minute for the penny to drop as to why this wasn't working. Anyway, the raw output now looks like this from Alibre CAM. You can see the F5 on the end of the G83 Peck Drill command now, and there's no L0 You can also see that there's S1500M03 command in the first of the three repeated sections. (Drill size 1) (Combining vent row 1 ANGLEREP6) (Tool Dia. 0.5 Corner radius 0.25) T2 S1500M03 G01 X2.750 Y0.000 Z0.498 F1000.0 G83 X2.750 Y0.000 Z-2.950 R-1.000 Q0.5 F5.0 G80 G01 X2.750 Y0.000 Z0.498 F5.0 (Combining vent row 2 ANGLEREP6 START30) G01 X3.900 Y0.000 Z0.498 F1000.0 G83 X3.900 Y0.000 Z-2.950 R-1.000 Q0.5 F5.0 G80 G01 X3.900 Y0.000 Z0.498 F5.0 (Delivery vent hole ANGLEREP8) G01 X7.487 Y0.000 Z0.498 F1000.0 G83 X7.487 Y0.000 Z-3.000 R-1.000 Q0.5 F5.0 G80 G01 X7.487 Y0.000 Z0.498 F5.0 M5M30 So now, this is the output I get from my CNC Utility.exe program. You can see that the S1500M03 line has been moved before the M90, A0F10000 which moves the 4th axis to 0 degrees. I've put it in front of that so it gives the spindle more time to get up to speed before it starts cutting. I've also added the F10000 on the A0 4th axis move because it was being left very slow from the previous move. (4th axis commands done) S1500M03 M90 A0F10000 M98 P1001 L6 M90 A30F10000 M98 P1002 L6 M90 A0F10000 M98 P1003 L8 M5M30 (Drill size 1) (Combining vent row 1 ANGLEREP6) O1001 (Tool Dia. 0.5 Corner radius 0.25) T2 //The S1500M03 command before this line has now gone.G01 X2.750 Y0.000 Z0.498 F1000.0 G83 X2.750 Y0.000 Z-2.950 R-1.000 Q0.5 F5.0 G80 G01 X2.750 Y0.000 Z0.498 F5.0 G91 A60 F10000 G90 F1000 M99 (Combining vent row 2 ANGLEREP6 START30) O1002 G01 X3.900 Y0.000 Z0.498 F1000.0 G83 X3.900 Y0.000 Z-2.950 R-1.000 Q0.5 F5.0 G80 G01 X3.900 Y0.000 Z0.498 F5.0 G91 A60 F10000 G90 F1000 M99 (Delivery vent hole ANGLEREP8) O1003 G01 X7.487 Y0.000 Z0.498 F1000.0 G83 X7.487 Y0.000 Z-3.000 R-1.000 Q0.5 F5.0 G80 G01 X7.487 Y0.000 Z0.498 F5.0 // note that the M5M30 command has been removed too. Spindle stop program rewind.G91 A45F10000 G90 F1000 M99 So, not that interesting if you're not into CNC, but it's been an essential journey for making the 4th axis more useful |
|
|
|
Post by Roger on Jan 19, 2021 12:44:37 GMT
|
|
|
|
Post by richp on Jan 19, 2021 16:55:43 GMT
Hi Roger
I'm seriously considering building SPEEDY, but currently in the planning stages and studying drawings. Before I start purchasing Laser plates and castings I'd like to obtain more details about the Don Ashton mods.
I understand Engineering in Miniature Jan 2009 is the main source but I am unable to get a copy.
Could you point in the direction of another source of the drawings, or pass anything on.
Any assistance would be greatly appreciated.
Kind regards
Rich
ps. I am familiar with fusion 360 for other projects
|
|
|
|
Post by Roger on Jan 19, 2021 17:30:58 GMT
And the surprising winner of this first test is... drum roll... This one, and by a large margin. This picks up nicely from 90psi down to 40psi with 18C water temperature, the water needing to be throttled of course at the lower pressure. I've tried this with an initial Regulation Gap of 0.314mm (as near as I can measure it) and just played with the gap at 90psi to find a happy middle spot where it feeds and also has the driest overflow. I'm waiting for it all to cool down so I can drain some water from the boiler and then retest the 40psi performance. Now, maybe 40psi isn't that ambitious for low end performance, but it's a reasonable cut off point while I'm still finding out the general parameters of what what works best. I can see how low it will go when I'm further down this road, but for the moment, if it won't pick up at 40psi, it's a reject. What's very interesting is that my initial test of this cone seems to show a cleaner overflow when they are combined. That may be because any overflow from the Delivery Cone end can fill the cavity and end up being drawn back into the stream through the Combining Cone overflow. I need to try this a bit more in different conditions and see what the limits of performance are. In any case, the other designs are rejected, I need to start with this one and see where to go next. Cone1 by Georgia Montgomery, on Flickr I'm not very happy with the outlet pressure gauge, it reads up to 150psi while the other one reads to 100psi. I've just ordered a couple of new ones that go to 100psi which will make reading them more ovbious as I can glance from one to the other and see the difference. I'm not convinced that my valve is working as well as it should, that's leaking a bit. I'm also not sure the balance valve is quite right either which is why I've ended up filling the boiler a bit too much. In the end, I need to be certain it will fill the boiler! Anyway, this is very promising indeed, it's a good start, especially with the combined overflows. |
|
mbrown
Elder Statesman
Posts: 1,790
|
Post by mbrown on Jan 19, 2021 17:53:41 GMT
Very well done Roger!
By way of comparison, the Talyllyn Railway's project to design "home made" 4mm injectors for our locos is looking to establish a working pressure range of 40 - 180 psi. 180 is the highest pressure used in service and 40 psi seemed a reasonable bottom figure. I once kept one of our old G&C injectors feeding down to less than 15 psi.... but the amount it fed at that pressure was negligible, so 40 psi seems reasonable.
Onward and upwards!
Malcolm
|
|
|
|
Post by Roger on Jan 19, 2021 17:53:46 GMT
Hi Roger I'm seriously considering building SPEEDY, but currently in the planning stages and studying drawings. Before I start purchasing Laser plates and castings I'd like to obtain more details about the Don Ashton mods. I understand Engineering in Miniature Jan 2009 is the main source but I am unable to get a copy. Could you point in the direction of another source of the drawings, or pass anything on. Any assistance would be greatly appreciated. Kind regards Rich ps. I am familiar with fusion 360 for other projects Hi Rich, I'm probably not the right person to ask this, because I abandoned the plans almost immediately when I realised just how far they differ from 1501. If you go to the start of this thread, you'll see how that came about. I started a Wiki which you can find here which explains some of the larger errors in appearance and also addresses the valve gear question. Many have successfully been built to the plans, but there are issues with the full gear cutoff not being quite long enough. I've shown Don's gear there, and why it's better. However, don't imagine that it's trivial to use Don's dimensions because they cause issues with clearance on the Gear Frame and other valve gear parts. If you model it in Fusion360, you'll see where the issues are. I moved the Gear Frame and the Weighshaft slightly and also modified the Expansion link so a more scale looking one would fit. I also cheated an added 1/8" to the front of the locomotive frames to give enough clearance for the front brake hangers which don't fit as drawn. So in the end it depends on whether you want to build SPEEDY or 1501. If you want something closer to scale, put the plans in the bin because they are hopeless on every detail. I'd already gone too far down the road to start again, but that's what I would do if I was starting today. Don't get me wrong, there's absolutely nothing wrong with SPEEDY, it's simple and it works as drawn by LBSC. You will of course have to change the boiler design to use Phosphor Bronze bushes rather than tapping straight into the Copper which isn't allowed any more. Standing back 15 feet, you won't know it's not a dead ringer for 1501, so it all depends on what you want at the end of the build. I changed my mind completely and it's been a lot of fun. However, there's nothing left of the LBSC design with the exception of the frames and overall boiler dimensions. Pretty much everything else is an attempt to get the dimensions of the real thing to fit on a platform that isn't quite right. Whatever you decide, it's a fun project and a very powerful locomotive for its size and a good looking one too. |
|
|
|
Post by Roger on Jan 19, 2021 18:25:10 GMT
Ok, here are the next two candidates for the One Piece Cone, this one has two staggered rows of five 0.5mm holes instead of the six on the best cone so far...  Cone4 by Georgia Montgomery, on Flickr ... and this one has two staggered rows of four 0.5mm holes  Cone5 by Georgia Montgomery, on Flickr Of course, each time I reduce the number of holes, it reduces the disruption to the flow, allowing more room for the water to pass by the holes. However, the flip side of this is the reduction of the total area available to let the water and steam escape. There's no point in having more area for the overflow than necessary, so hopefully this will help me find out where we cross that boundary and don't provide enough I could make another with just three holes, but this may already be too far and there's no point in wasting the cones I've made. This will already be five out of the original set of twelve. |
|
|
|
Post by Roger on Jan 19, 2021 22:42:28 GMT
Ok, I machined Cone4 which has two staggered rows with 5 x 0.5mm holes instead of the six before, and that won't pick up at 90psi because the overflow is too restricted. So it's a waste of time makine the other one I'd modelled because that would be even more restricted. Instead, I'll pursue this a little further by using the same reduced arrangement of 5 holes, but make the first ones 0.6mm instead to increase the volume they will pass. Obviously that now widens the disturbance they create, but something's got to give! In reality, it's not too bad because that first row breaks out on a larger diameter, being further up the taper. Everything else stays the same, so this will prove whether I've got enough area now. If not, I could potentially go up to 0.6mm on the second row.  Cone5 by Georgia Montgomery, on Flickr I've just tried this, and it's a different animal to Cone4. It happily starts the syphon at 90PSI although it's not an aggressive suction. It seems to work down to 40psi, and it's similar to Cone1 in that area. ie it can be encouraged to pick up, but it's not keen. This will produce over 100psi into a blocked outlet when the boiler pressure is 90psi, so this looks very promising. I think I need to chew this all over and maybe do some more tests with higher water temperatures to see how they perform. |
|
|
|
Post by Roger on Jan 20, 2021 19:38:31 GMT
This is a new valve element to replace the leaking one that I made too loose. This one is probably a bit too tight, but as long as I can turn it that's fine.  20210120_121717 by Georgia Montgomery, on Flickr I've also been making a second adjustable Steam Cone with a 1.05mm throat compared to the 1mm throat of the existing one. Reading D.A.G Brown and Bob Bramson's books, it's clear that the size of the Steam Cone throat controls the performance at the top and bottom of the pressure range. The smaller the nozzle, the higher the pressure it will work at, but the lower end pressure increases too. So there's no point in having an injector that works at any higher pressure than the working pressure of the boiler at the expense of low pressure performance. At the moment, the cones that work pick up very well at 90psi, so it may be that the Steam Cone throat could usefully be increased. Obviously if you go too far, the Injector is overwhelmed with too much steam and it won't work. I need to find out where this point is. The charts in the books give a good idea, but it's worth trying this to see what will work.  20210120_145851 by Georgia Montgomery, on Flickr Fortunately, I managed to use the main spindle with the 4th axis set to the front and still reach to machine the hex with an 8mm cutter without moving anything.  20210120_151950 by Georgia Montgomery, on Flickr  20210120_152454 by Georgia Montgomery, on Flickr  20210120_163245 by Georgia Montgomery, on Flickr  20210120_163251 by Georgia Montgomery, on Flickr  20210120_192331 by Georgia Montgomery, on Flickr |
|
don9f
Statesman
Les Warnett 9F, Martin Evans “Jinty”, a part built “Austin 7” and now a part built Springbok B1.
Posts: 961 
|
Post by don9f on Jan 20, 2021 19:49:10 GMT
Hi Roger, I’m loving this pioneering stuff....can’t wait for the next Instalment!
Don
|
|
|
|
Post by Roger on Jan 20, 2021 21:27:09 GMT
I'm glad you're enjoying it Don, it's interesting and a lot less frustrating now I can quickly change things around. I've taken the overflow valves apart and polished the stems so they move a bit more smoothly. Of course, combining the two overflows means I've got two valves to seal. I need to make a blanking plug and take one of them out when running like that so I've only got the one to seal. That will more faithfully simulate the way it is on the scale body. I'm not convinced that I've got the best arrangement for the valve yet. I've got some Silicone O-rings coming for them instead of the Nitrile ones I happened to have. Those will be softer and that ought to seal more readily. Another eBay find are these Silicone LED Covers which are available in several sizes. They look really thin, so I was wondering if they could be used as the valve element? Anyway, I've ordered some to have a look at them, and those might be a discovery. Anyway, here's a quick summary of the test with the new Steam cone, checking to see if it will still pick up at the higher pressure. Tests repeated with combined and separate overflows. Regulation gap - 0.29mm 1.05mm Steam Cone Water temperature – 36C Picks up at 90psi and runs down to 35psi, feeding water into the turned off Boiler to see how it performs. However, I think the extra volume of steam could benefit from slightly more room to escape through. So, the water temperature was really warm, up on the limit of what's possible. It's a promising result. |
|
|
|
Post by Roger on Jan 20, 2021 22:11:08 GMT
Just one interesting observation when testing with the overflows combined and then tested when separated.
When it's running at medium to low pressures, there's a little stream from the Delivery Overflow which is only to be expected. What's interesting though is that with separate overflows you can see that the Combining Cone is sucking because any leakage causes any water in the overflow pipe to rise slightly.
Now take the identical situation with a single overflow. Because the valves aren't perfectly identical, one opens and, in these conditions, the other stays closed. So you have the interesting situation internally where the Combining overflow is sucking, but there's an excess of water that needs to escape so the valve still has to open just enough to let that out.
It would be interesting to see if the whole of the chamber is completely full with water when this is happening. My guess is that it probably is.
Even if it isn't, the non return valve ought to ensure that whatever is in the body of the injector is Saturated Steam, which presumably can be condensed if it's drawn into the Combining Cone.
Anyway, all of the tests done to date seem to indicate that you don't need a valve on the Combining Cone so long as you've got one on the Delivery Cone. This is what I've been hoping for because it makes the design simpler to make, and will probably help with dealing with hot feed water.
|
|
|
|
Post by Roger on Jan 21, 2021 0:09:05 GMT
While chewing over what to try next, I spotted that I'd changed the Delivery Overflow hole size without keeping the upstream edge where it would be if it was a separate cone. If you look at this cone, which is the one I've been experimenting with today... Cone5 by Georgia Montgomery, on Flickr ... compared to this one, you can see that the RH set of holes is slightly upstream and hence creates a different throat diameter. That will have the effect of slightly increasing the area available for the overflow. To be honest, I don't think this will make any difference at all, but I ought to just correct this to make sure. I suppose you could argue that the flow as a little more time to settle down after jumping the gap before finally reaching the Delivery Cone throat. Anyway, I'll probably make this and give it a try tomorrow.  Cone6 by Georgia Montgomery, on Flickr Ideally I'd like to get this to pick up at ever lower pressures while maintaining the same characteristics at high pressure. I don't want to change more than one thing at a time else it's hard to know what's made any difference, if any. |
|
uuu
Elder Statesman
your message here...
Posts: 2,858
|
Post by uuu on Jan 21, 2021 8:51:45 GMT
... compared to this one, you can see that the RH set of holes is slightly upstream and hence creates a different throat diameter... Your eyes must be better than mine.  Wilf |
|
|
|
Post by Roger on Jan 21, 2021 8:56:47 GMT
... compared to this one, you can see that the RH set of holes is slightly upstream and hence creates a different throat diameter... Your eyes must be better than mine. Wilf Hi Wilf, I probably didn't make it clear enough where to look. I've moved the single row of holes on the RHS left so they are closer to the flange. |
|
uuu
Elder Statesman
your message here...
Posts: 2,858
|
Post by uuu on Jan 21, 2021 8:59:53 GMT
Thank you. I was looking at the right hand of the other rings, i.e. the middle row.
Wilf
|
|
