|
|
Post by David on Feb 15, 2021 22:32:30 GMT
Being able to adjust it for water temperature is a great feature, although probably not one most of us would use. Even down here we'd just put ice in the water and hose the injector down rather than fiddle with it on the day. But it might be good to have each side set for slightly different water temperature ranges. People more enthusiastic than me might change the setting in spring and autumn :)
If your higher temp setting still works down to 30deg just having to adjust the feed water valve, that's no shame. Plenty of injectors seem to need the feed water valve adjusted at any water temperature and pressure before they'll pick up and run anything like dry.
I couldn't make the reamers. I could possibly make the rest of it to a good enough standard though. I can't turn a parallel shaft or get an accurate taper on anything. I can't (or won't) set my machines up properly so everything comes out some degree of wrong.
|
|
|
|
Post by Roger on Feb 15, 2021 22:42:23 GMT
Now I know that the scale 8X Injector body is viable, I've made the few detail changes I wanted to make, and now it's time to make the two I need for the Locomotive. The first operation is to perform a deep profile clearance cut to the full depth of the injector. This is the flat inside of the Injector half which is Silver Soldered to it's mirrored counterpart. The idea is to make it so that this outer part falls away when it's parted off.  20210214_203618 by Georgia Montgomery, on Flickr The second operation is to rough out the inside cavities that can't be machined when it's complete, and to get into some of the tighter places. This time I'm not roughing out all of the inner spaces because it makes it difficult to get a hole truly down the middle. It's easier to make it as solid as possible so it can be drilled and reamed afterwards. This is not only more accurate, but it saves time at this stage.  20210214_215709 by Georgia Montgomery, on Flickr I know you can reach the openings on the overflow etc, but it's helpful to have something to clock up for a sanity check when using the fixture to machine the flanges etc.  20210215_095747 by Georgia Montgomery, on Flickr The M1.4 holes worked really well last time to hold the part for the inverted operations, so I'm doing that again. The small holes are for 0.8mm Steel rivets, again the same as last time because it worked well holding the parts together.  20210215_104509 by Georgia Montgomery, on Flickr Parting this off is easy enough, but it's a bit brutal when you get the swarf getting trapped in gaps in the part. Here's a video at the start...  20210215_105322 by Georgia Montgomery, on Flickr You need to leave at least 0.5mm over the finished size because the parting tool is prone to bending even though the front is ground straight. Here's a video of it chewing on some of the swarf.  20210215_110043 by Georgia Montgomery, on Flickr This is the second one...  20210215_164204 by Georgia Montgomery, on Flickr ... and them both parted off. The first of the mirrored halves is just finishing off now, so tomorrow I ought to have this first step completed.  20210215_205649 by Georgia Montgomery, on Flickr |
|
|
|
Post by Roger on Feb 15, 2021 22:57:42 GMT
Being able to adjust it for water temperature is a great feature, although probably not one most of us would use. Even down here we'd just put ice in the water and hose the injector down rather than fiddle with it on the day. But it might be good to have each side set for slightly different water temperature ranges. People more enthusiastic than me might change the setting in spring and autumn  If your higher temp setting still works down to 30deg just having to adjust the feed water valve, that's no shame. Plenty of injectors seem to need the feed water valve adjusted at any water temperature and pressure before they'll pick up and run anything like dry. I couldn't make the reamers. I could possibly make the rest of it to a good enough standard though. I can't turn a parallel shaft or get an accurate taper on anything. I can't (or won't) set my machines up properly so everything comes out some degree of wrong. Certainly I'd set the Injectors to have a slightly wider gap if I lived where you do, that makes perfect sense. It's no big deal to restrict the water if you ever get below 50psi. The one thing I've found in all the dozens of variants of cones I've tried is that almost everything works if you're in the general ball park. However, getting it work really well over a wide range of conditions is another matter. I very much doubt if manufacturers can afford the time to make injectors to a level of accuracy required to get the best out of them. The design of them is not conducive to mass production, there are just too many dimensions that you have to tightly control. So it's no surprise to me that they often fall below our expectations. I thing we tend to assume that if they struggle to make them, and it's their business, then it must be next to impossible for us to make them. I think this is really wide of the mark. We can afford the time to get it right because we don't value our time the way they do. Ideally you need someone with a Cutter Grinder to make the reamers from HSS drill blanks. Sure you can make them from Silver Steel and harden and temper them, but they are much longer lasting in HSS. With care they ought to last for 100 or more cones before needing sharpening. |
|
dscott
Elder Statesman
Posts: 2,440
|
Post by dscott on Feb 16, 2021 2:31:51 GMT
There must be a tease in there somewhere.
"AH Burt, wonderful Guy MMM. Yes he only comes to run his Locomotive on the HOTTEST days of the year...
He once told me that he got some half price injectors from Australia!"
Then again if you look at the Plimsoll Line on Ships they have 2 settings. One for Up North. and the other down South.
Yes and painted over in the case of ferry's, in places like the Philippines!!!
David and Lily.
|
|
|
|
Post by Doug on Feb 16, 2021 8:53:36 GMT
Hi regarding tank temperatures (think I have already posted about that but here are the results of the testing I did a few years ago  So 94 deg. C at the safety valve brass fitting  And 33 deg c down between boiler and the tank, this is a butch model not my speedy but a fair comparison I think, I doubt the tank water will not be heated above 30 deg. The boiler is insulated with 3mm ceramic wool and there’s an air gap of 1mm, it would be fair to say the probe was touching the side of the boiler and the tank so the 33deg will be made up of cooler water and hotter boiler. The test was done after an hour of running, the tanks being filled or topped up every 20 mins or so. Hope that information is useful. All the best Doug |
|
|
|
Post by Roger on Feb 16, 2021 8:59:13 GMT
Hi regarding tank temperatures (think I have already posted about that but here are the results of the testing I did a few years ago So 94 deg. C at the safety valve brass fitting And 33 deg c down between boiler and the tank, this is a butch model not my speedy but a fair comparison I think, I doubt the tank water will not be heated above 30 deg. The boiler is insulated with 3mm ceramic wool and there’s an air gap of 1mm, it would be fair to say the probe was touching the side of the boiler and the tank so the 33deg will be made up of cooler water and hotter boiler. The test was done after an hour of running, the tanks being filled or topped up every 20 mins or so. Hope that information is useful. All the best Doug Hi Doug, Thanks for re-posting that, yes I do remember us discussing it. It seems clear to me that insulation is needed if you intend to use that water for Injectors like I am. The boiler surface, less the cladding, is 166C at 90psi, and the back of the tank isn't that far away. It's hardly surprising that it gets hot since the boiler insulation isn't that thick. It does bring the temperature down enough to use other insulating materials on the back of the tanks though. |
|
|
|
Post by Roger on Feb 16, 2021 23:32:22 GMT
The four mounting lugs need shortening before the outside can be machined, so I'm doing them one by one. The part is only held down by M1.4 bolts, so I can't put too much force on it.  20210216_191126 by Georgia Montgomery, on Flickr  20210216_210017 by Georgia Montgomery, on Flickr  20210216_210026 by Georgia Montgomery, on Flickr Here's a video of the roughing cut which uses a 2mm cutter. I thought I could hear it struggling a bit at times, so I stopped it and was about to replace the cutter but it seemed razor sharp.  20210216_211249 by Georgia Montgomery, on Flickr Anyway, I returned to find this... most of it done, but the tool has broken. I swapped that for a new one and went to start it again, only to find that the reason for the problem is that the spindle has seized. Fortunately, I have a couple of spare ones, so I won't have to wait while I repair this one. These spindles are used on the PCB Routing machines that I support, so I always have at least one spare ready to go. I know this looks a mess, but it's just a roughing cut, and Phosphor Bronze isn't the easiest material to machine dry like this.  20210216_230911 by Georgia Montgomery, on Flickr Anyway, so far so good, I ought to get this one finished tomorrow. |
|
|
|
Post by Roger on Feb 17, 2021 21:04:05 GMT
This is how the outside looks after roughing with the 2mm cutter. I just ran it again from the start with the new spindle. I used the wobbler on the fixture to set it up again from scratch. The new spindle ought to locate in the same place, but better safe than sorry. It still looks a mess, but it doesn't matter, that's just getting rid of the unwanted material.  20210217_105123 by Georgia Montgomery, on Flickr My 1mm Ball nosed cutters aren't long enough to reach down the sides of the vertical edges, so here I'm using a diamond wheel to give me some more clearance. I only need a total of 3mm, so there isn't much material to come off.  20210217_115557 by Georgia Montgomery, on Flickr There are three operations with the 1mm Ball nosed cutter. The first one is a Horizontal Finishing operation that works down in 0.25mm slices. I'm doing it that way so that there's more control over the entry and exit. If I used a parallel finishing operation that follows the contours of the job like you see in the video shortly, you'll see that it would plunge and probably break the cutter. Figuring out what strategy to use for getting into the deep narrow spaces is one of the biggest challenges in this type of machining.  20210217_124841 by Georgia Montgomery, on Flickr Here's a video of the second of three operations performed with the 1mm ball nosed cutter.  20210217_124850 by Georgia Montgomery, on Flickr The third operation is a very find Parallel Finishing operation leaving zero stock. The step over is only 0.05mm because with such a small radius Ball nose, you end up with lines that are too visible if you use anything bigger. This is how that came out, not too bad really. It will clean up nicely. I think I need to give it the odd drop of cutting oil while it's finishing so it's a little less ragged. Anyway, it's got to go through the Silver Soldering process next, so there's no point in wasting any time doing any cosmetic work at this stage.  20210217_203404 by Georgia Montgomery, on Flickr The second one of these is just roughing out now, so that one should be get finished tomorrow and that will be the half way point. |
|
|
|
Post by mugbuilder on Feb 18, 2021 0:59:12 GMT
Absolutely mind boggling!
|
|
|
|
Post by Oily Rag on Feb 18, 2021 2:51:27 GMT
Absolutely mind boggling! Can you imagine what would occur if the Rodg started from scratch with the works drawings of a Beyer Garratt AD60 or a QR BG Garratt, besides requiring a longer workshop?
|
|
|
|
Post by Doug on Feb 18, 2021 12:56:59 GMT
Hi Roger,
Can I ask which program your using for the CAM? Is it still geomagic?
Thanks
Doug
|
|
|
|
Post by Roger on Feb 18, 2021 14:19:03 GMT
Hi Roger, Can I ask which program your using for the CAM? Is it still geomagic? Thanks Doug Hi Doug, Yes, it's the one they imbed in Alibre Design, they've changed the name back I think. Mecsoft make the CAM, you have to buy it separately. I wouldn't recommend either to be honest, Fusion360 is a better option. However, I'm not in a position to change. |
|
|
|
Post by Roger on Feb 18, 2021 20:57:27 GMT
Trouble at mill... Hmmm, I have an issue with the spindle output on the mill which has now wrecked my spare spindles.
While I can using the Mill as a Lathe the other day, the spindle speed went right up for no obvious reason which was a bit alarming but not catastrophic. However, I think that must be what's happened to the High Speed Spindles while I've been away and they've failed. I know it can't be the frequency converters because they're separate for the two spindles and it's a common problem.
So now I'm nervous about leaving it to run on its own until I've figured out what's causing it. It's a real pain because it's a rare event that I can't just repeat on demand. I'll have to dig deep and see what I can think of to tease out what's going on.
Such is life! It's the first time I've have any significant trouble with the machine since 2008, so I can't complain.
It's probably something really simple such as a dry joint but finding it won't be easy.
|
|
JonL
Elder Statesman
WWSME (Wiltshire)
Posts: 2,993
|
Post by JonL on Feb 18, 2021 21:26:45 GMT
Intermittent problems are the worst to diagnose. I wish you luck.
|
|
|
|
Post by keith1500 on Feb 18, 2021 22:41:16 GMT
Can you make a simple speed detector to monitor the spindle and initiate a pause or stop? Perhaps infra red device reflecting off a dob of white paint on the spindle.
|
|
|
|
Post by Roger on Feb 18, 2021 22:48:58 GMT
Can you make a simple speed detector to monitor the spindle and initiate a pause or stop? Perhaps infra red device reflecting off a dob of white paint on the spindle. Hi Keith, That's a possiblilty, although I'd have to do it for both spindles. I might clip the maximum speed of the drives so that it isn't so disastrous. It's a real pain, because I don't have another High Speed Spindle and the bearings won't arrive for a while. I can finish the job slowly on the main spindle, but I want to get to the bottom of what's going on. I think I managed to override the overspeed using the slider on the Mach 4 control panel when it happened once before, so it sounds like it might be a software issue. I've noticed that the ESS has thrown an error dialog on occasion indicating that the Ethernet connection was lost. However, the machine didn't miss a beat and I just close the dialog box if that happens. I haven't changed the software in over a year, but of course Windows has updated and I wonder if it has something to do with this. I've contacted Mach and Warp9 Forums in the hope that someone will have some insight as to what this might be. It's difficult to diagnose with the fault being so intermittent. I know it's not the program because that's been run before without any issues. |
|
|
|
Post by David on Feb 19, 2021 0:28:47 GMT
Not the update anyone was hoping for, but your troubleshooting will be interesting to follow.
|
|
|
|
Post by keith1500 on Feb 19, 2021 7:32:40 GMT
Hi Roger,
I wonder if it might be something as simple as fine dust etc has got to a feedback sensor. May be just needs a clean.
After all you do a lot of very fine work which produces dust rather than proper swarf!
Good luck with problem.
Keith
|
|
|
|
Post by Roger on Feb 19, 2021 8:06:20 GMT
Hi Roger, I wonder if it might be something as simple as fine dust etc has got to a feedback sensor. May be just needs a clean. After all you do a lot of very fine work which produces dust rather than proper swarf! Good luck with problem. Keith Hi Keith, It's not that sophisticated, it's open loop speed control. Mach4 doesn't have any feedback architecture. Mach4 outputs a stream of pulses which are then converted using a PWM charge pump arrangement to give 0-10V analog output for the speed. I'm going to reconfigure the drives so interpret that as a smaller RPM request to protect the spindles. I don't need the maximum RPM, probably only about half. I can tell Mach4 about the new calibration so I get the same speeds out that I'm expecting. I use the CM106 board which has an ESS (Ethernet Smooth Stepper) interface on it. That takes the higher level commands from Mach4 and creates the pulse streams locally to unload that burden from the PC. I'll take it all apart and see if there's anything obviously amiss in the area that generates the spindle output. |
|
timb
Statesman
Posts: 512 
|
Post by timb on Feb 19, 2021 8:51:27 GMT
Hi Roger, you have probably though of this but we always had lots of issues with pickup on 0-10V analogue signals and had to go to great lengths to ensure a smooth signal. Have you plugged anything in recently that has not been there before? Could it be that the cable screen or a screen connection has failed? Harmonics on the inverter??
I dont envy you finding the cause.
Tim
|
|
