What did you do Today 2018

[Anonymous]

Now the stoopid forum is whining about my post being too long! So here's part 2:

Now, to stop some of the recent noobie members whinging about off topic posts, here's some modelling.

Having fitted the new bolts to my perch bracket I simply wasn't happy with the way they looked, too chunky. So I did some experiments; skimming about 3/64" off the bolt head and redoing the chamfer looked about right to me:

I also made up a fixture to locate the chimney base casting on the smokebox while I spot through the fixing holes:

The finish on the top of the large bush, straight from the parting off tool, was pretty good, about 1.2µm Ra. That's better than the rest of the turning. Controversially I'm in the part off under power feed (minimum of 4 thou per rev) camp.

The weather is poor today, so it's a "me" day in the workshop and definitely no gardening or housework.

As an aside Little Gransden have their charity airshow today. They normally have some WWII fighters attending and the BBMF Lancaster is due to make a pass, during which the gliding club has agreed not to launch. But if it rains as hard as forecast this afternoon I can't see it happening, bit of a shame. But that's flying and UK weather.

Andrew

[robjon44Participant @robjon44]

Hi all, not today but yesterday went to BMFA Model Flying Championships at RAF Barkston Heath, a fixture in my calendar ever year since the 1960s, like to keep up with the latest developments in electric power, wallow in the nostalgia of control line flying especially CL Combat, still going strong. Never ceases to amaze with the huge quantities of models of every type, particular mention going to the gigantic models of a BOAC airliner & XH588 the last flying Vulcan both of which when sighted from the far end of the airfield were easily mistaken for the real aircraft.

Barkston Heath has been a satellite airfield for Cranwell since the dawn of time & of course still is, however back in our schooldays us young aircraft spotters used to cycle up there to the High Dyke, there were traffic lights on the road by the gate that I entered by yesterday but in those ancient times Gloster Meteors & DH Vampires would taxi across the road from the hangers that were there, no Health & Safety twaddle in them days matey! & to think that I still meet people who don't believe that its possible to die of nostalgia. All these events then & now occurring within 3 miles of the chair I'm sitting on.

Cheers, Bob.

[Gordon WParticipant @gordonw]

Yesterday I stripped a small petrol generator to service and repair the electronics. Easy job ,all laid out on the bench, two socket sets open, multi meter, etc. etc.. This morning found that a bird or birds have taken up residence just above this bench. All covered with a nice white coating. I'm going to watch F1 on the telly and leave the stuff to dry, a nice little job for tomorrow.

[RobinParticipant @robin]

Posted by Neil Wyatt on 26/08/2018 10:15:57

Wow! Big sash weights!

What's the story? Did they cannon swing too far?

And why do you have two cannons in your garage?

Neil

I was soaking the bore in rust remover overnight. I think the hoist gradually let go and it started to tilt.

Cannons made c1800 for the Reed Shipping Line, trading the South China Seas out of Cardiff. I have the original carriages which is fun

[BazyleParticipant @bazyle]

Bob do they still do the evening free for all just before dusk? That was always great fun but an HSE nightmare.

WRT rigging lathes for lifting. After putting the strap round the bed run a rope through the spindle and tie both falls from the strap tightly into the spindle nose. This is NOT to take any weight on the spindle but it shifts the pivot point from below the cofg to the spindle line well above and makes it fully stable. It also stops any risk of the strap sliding down the bed if the tail end rises above the head end.

[MuzzerParticipant @muzzer]

Posted by Andrew Johnston on 22/08/2018 22:25:34:

Posted by David Taylor on 22/08/2018 03:14:12:

Andrew, do you need to use coolant on brass? I never do. You could stand there and blow the chips away with an air nozzle which I did for the steel parts. That was just to ensure the cutter was only cutting 'new' metal.

I found with the GWizard software the feeds and speeds it gave me basically melted my cutters on the 1.6mm steel parts. Perhaps it assumes coolant. I did specify HSS cutters. I dialed things way back to basically what I'd do on the manual machine and the steel parts went smoothly (if slowly).

Normally no; I machine brass dry. Coolant on the CNC mill is primarily to wash away the swarf, especially in this case where it's a pocket and the swarf won't naturally fall away. For me the whole point of CNC is that I can go away and do something else, so I'm not going to stand around for hours with an air blast! Although I have been known to use a vacuum cleaner when machining relatively quick parts in plastic.

I tried GWizard a long time ago and couldn't get any sense out of it. The answers always seemed pretty daft to me. I now just base speeds and feeds on experience and use a calculator or do them in my head.

Andrew

David – from your album it looks as if you have a CNC mill, in which case you don't need to go near GWizard. I've never been impressed by it unless you like examples of self promotion above substance. Bob Whorefields has made a comfy life from it but you could spend your money on something more useful like actual tooling. There are better products available for free from the main suppliers (Sandik, Mitsubishi etc) including some handy smart phone apps.

The Excel feeds and speeds spreadsheet on the NYC CNC website is also generally useful for ballpark machining feeds and speeds. Worth a look and a play.

If you are using Fusion 360, you simply need to find the surface speed and feed per tooth for the cutters from the manufacturer's site (if using a recognisable product) or the more generic feeds and speeds from the likes of John Saunders' site. The CAM program within Fusion 360 will do all the rest for you, like setting the feeds etc. Have to admit that some of the latest cutters are capable of greater material removal rate than my machine can handle, so tend to do some tests and dial the loads back a bit but they are a good guide. As well as avoiding overloading the cutter, you also need to make sure you take enough of a cut to avoid rubbing etc.

I'd always use coolant for steel if I'm trying to shift some swarf. On the other side of the pond, Mercans seem to like those mist spray things. They provide very little cooling (primarily due to the air flow), so must be limiting the material removal rate and tool life. Better than dabbing oil on with a brush though, which seems little more than superstition.

Murray

[Neil WyattModerator @neilwyatt]

Well it looks like everyone has been having fun… final print of Andrew's nameplate at 0.06mm layer height and better settings. I'm not sure you could get better results with a 0.4mm nozzle on a 3D printer of any type. This is completely untouched:

[robjon44Participant @robjon44]

Bazyle, yes they do indeed still have the evening free for all, never seen one but I got the impression that these days it is beset with pretty stringent rules enforced by men with big sticks, in my experience this is brought about by the behavior of people who think that rules apply to everybody but themselves, this is how flying sites, fishing rights etc are lost, often, sadly forever.

Cheers, Bob.

[Anonymous]

I can assure forum members that no aeroplanes, or gliders, will be mentioned later in this post.

I've spent the "me" day playing with CAD, CAM and CNC. This morning I tweaked the CAD model of my inner nameplate to allow a 1mm cutter to pass between all pairs of letters. Then I changed the CAM to take account of levels of machining, ie, 0.2mm per step down. I also changed some of the small areas to offset rather than radial machining as it was much quicker. Inner and outer profiles were added for the inner and outer circles of the ring. The CAM took ages to simulate, darn sight longer than the actual machining. After simulation and a sanity check of the G-code in a backplotting program I went for it.

Here's the trial run on a bit of scrap aluminium from a client project:

Looked ok, so onto the brass sheet; CZ120, ie, engraving brass. After a lot of faffing about thinking of complicated jigs I ended up with a simple one:

And finally the two brass nameplate rings:

Depth of the engraving is 30 thou (0.635mm) and the ring profiles are 1.4mm, leaving from 0.15mm to zero to break out the rings. Total machining time per ring was 1 hour and 20 minutes. I wasn't sure how flat the plate would be so I increased the depth of the engraving, meaning that the first level was cutting air, unless the material was more than 0.2mm off flat. Turned out to much flatter than I thought it might be, so I could have saved so time. But much better than breaking a cutter.

Of course it's at this juncture that someone points out a spelling error.

Andrew

Edited By Andrew Johnston on 26/08/2018 21:58:35

[Neil WyattModerator @neilwyatt]

Posted by Andrew Johnston on 26/08/2018 21:57:06:

Of course it's at this juncture that someone points out a spelling error.

Andrew

Aren't there two 'R's in 'Burrell'?

Very impressive, and slightly faster than 3D printing too, although I don't have to cut my one out!

Neil

[Anonymous]

Posted by Neil Wyatt on 26/08/2018 22:44:15:

Aren't there two 'R's in 'Burrell'?

Very impressive, and slightly faster than 3D printing too, although I don't have to cut my one out!

Yes, better than two Rs in Charles though!

The 1mm cutter was running at a feedrate of 350mm/min, although running at 24000rpm that's a pretty low chip load. In brass it could probably be pushed harder, but no point in breaking a cutter for s two off. And of course being CNC I can go and do summat else like cook supper and watch the telly (flying programme on Channel 4 this evening).

What we really need to do is combine CNC milling for the outside of the letters and 3D printing for the inside.

Andrew

[David TaylorParticipant @davidtaylor63402]

Posted by SillyOldDuffer on 26/08/2018 08:48:29:

There but for the grace of God go I!

Lathes are often very top-heavy and much inclined to roll over when lifted. I guess you were on your own when it flipped? Two people make lifting a lathe a lot easier because one keeps it balanced while the other works the crane. Doing both jobs yourself is tricky!

Hope it's easily repaired; they're sturdy beasts.

Thanks Dave. I'm hoping the only damage is to the electrical cabinet. I was on my own and as you say that's a difficult way to do it.

The real problem was how I had the straps connected to the hoist. The hoist has a tiny hook that you can't fit the strap ends into. I must have done it a bit differently to how I usually do it, which is sketchy at the best of times.

Needless to say, later I thought of a better way I could have done it using some parts of the Tormach lifting kit.

David.

[David TaylorParticipant @davidtaylor63402]

Posted by Robin on 26/08/2018 08:59:01:

I discovered that engine hoists are not 100% solid, dented the floor, missed the car by inches…

Ouch! I'm glad you missed the car.

I reckon if the castors turned more easily it would help – I really have to put some force into mine to get it to turn or move, and that doesn't help stability.

Barrels ok?

[David TaylorParticipant @davidtaylor63402]

Posted by Andrew Johnston on 26/08/2018 21:57:06:

I also changed some of the small areas to offset rather than radial machining as it was much quicker … sanity check of the G-code in a backplotting program I went for it.

Great result!

What do offset and radial machining mean, and what is a backplotting program?

I don't have a high speed spindle. Do you think it makes much difference? You're spinning the cutter almost 5 times faster than I can.

[David TaylorParticipant @davidtaylor63402]

Thanks for the info Murray.

[Brian HParticipant @brianh50089]

NEARLY burnt myself badly. I was soft soldering on thin brass using the smallest nozzle I have on a propane torch. Suddenly, I saw a ball of flame travelling up the torch towards me! I VERY quickly turned off the gas but still had my eyebrows and hair singed as well as ending up with sore lips. Fortunately, I wear spectacles so my eyes were not affected.

Examination of the torch revealed a small split in the washer between the torch and the nozzle, quickly replaced.

So, I would advise anyone with a propane torch to check the washers and replace any that look worn or damaged.

Brian

[Michael GilliganParticipant @michaelgilligan61133]

Posted by BDH on 27/08/2018 08:35:27:

[ … ]

So, I would advise anyone with a propane torch to check the washers and replace any that look worn or damaged.

Brian

.

Wise words … Thanks for the reminder.

MichaelG.

[David TaylorParticipant @davidtaylor63402]

With the help of a club member who used to be a rigger we got the lathe back on its feet and onto the stand. I'd still like an opinion on the hardwood under the stand – will the lathe just fall off?

I brought the 5" mogul back from the father-in-law's workshop where it has been resting while my new workshop was built.

Also trying to make a house number faux cast sign. I can't get the rest machining operation to remove all the bits left over from the previous pocket operation. Plan B is to make the 2nd operation a normal pocket cutting mostly air and let it run for hours.

David.

[RobinParticipant @robin]

Posted by Neil Wyatt on 26/08/2018 16:14:47:

I'm not sure you could get better results with a 0.4mm nozzle on a 3D printer of any type. This is completely untouched:

How big is it? Have you got it as an STL file? I am not being competetive, just curious.

Black is the hardest colour to get a nice finish on, it shows every string and every smear.

[Anonymous]

Posted by David Taylor on 27/08/2018 00:37:53:

What do offset and radial machining mean, and what is a backplotting program?

I don't have a high speed spindle. Do you think it makes much difference? You're spinning the cutter almost 5 times faster than I can.

If you look closely at the bottom of the ring where the lettering is the toolmarks are radial to the centre of the plate. So the tool path tries to cut always on a radial. This makes sense for a shape like this and is quicker than other options. At each level a clean up pass is made that profiles the circular edges and round each letter to clean up areas not covered by the radial toolpath. However, when pocketing out the internals, for example the letter O, a radial toolpath isn't appropriate. I initially used a radial toolpath because the operations were cut 'n' paste from the first toolpath. The area to be removed is small so a radial toolpath does a lot of small movements as the pocket is not truly round. An offset toolpath enters somewhere in the middle of the pocket and then moves outwards in a series of parallel paths, the distance per sideways step being set in the CAM program. I set it at 75% of tool diameter. Although the change of toolpath only saved a few seconds per pocket that adds up over the number of letters that have internal cavities and over five levels to get the final depth. It probably saved 15 minutes or more overall.

A backplotting program reads in the G-code and plots the toolpath from it. That is different from CAM programs that display the toolpath they think they will use based on geometry, rather than that generated by the G-code the CAM program generates. Although not common I have had issues where the toolpath represented by the G-code was not the same as the toolpath the CAM system thought it was using. The backplotting program also gives a sanity check on maximum travels in each axis so you can double check there are no unauthorised moves outside of the machining envelope that will let the tool hit clamps or mill into the table. I use a program called NCPlot. I bought my version, a few hundred US dollars; they may do a simpler free version?

Some years ago I spent a lot of time trying engraving on the Tormach. The results were a failure and I came to the conclusion that the spindle speed simply wasn't high enough. I dithered over a high speed spindle for quite a while. Then I noticed that ArcEuroTrade had stopped selling them. Fortunately I was able to do a deal to buy a demo unit plus inverter. Mounting the inverter and wiring was simple, but I had all sorts of issues with the odd Chinese connector on the motor itself. In the end I binned it and bought a proper shielded connector of similar size from RS. So not a straightforward installation, even for a professional electronics engineer!

There are two reasons why the high speed spindle is essential, in my view for small cutters. The chip load I ran on the nameplate milling was about 0.005mm per tooth. At 5000rpm that would be a feedrate of 60mm/min. So the machining time would have been nearly seven hours per nameplate. For me that's not acceptable. When I modelled the bevel gears for my governor at 5000rpm the machine time was nearly 20 hours per gear. The other issue is to do with axis movement. I think the resolution per step of the Tormach is on the order of a tenth or so. But when setting up I find I can make several clicks on the jog console without the axis moving. Then when it does move the distance may be bigger than the nominal step size. For a small cutter running at slow speed that can be a significant variation in chip load, resulting in broken cutters. I suspect that was my basic problem with the engraving trials. If the cutter is running much faster with a consequent increase in feedrate then any small inconsistencies in the axis movement are proportionally much smaller and less likely to break the cutter. So far I've got about 40 hours of machine time with 1mm cutters, and I've only broken one, where I boo-boo'd on the vertical offset and it was trying to cope with a DOC of 0.5mm and full width.

Personally I find the high speed spindle essential for small cutters. Whether it is of use for you will depend upon the type of machining you are likely to be doing.

Now off to do an aircraft inspection, and then I'm afternoon tug pilot.

Andrew

[Neil WyattModerator @neilwyatt]

Posted by Robin on 27/08/2018 10:39:21:

Posted by Neil Wyatt on 26/08/2018 16:14:47:

I'm not sure you could get better results with a 0.4mm nozzle on a 3D printer of any type. This is completely untouched:

How big is it? Have you got it as an STL file? I am not being competetive, just curious.

Black is the hardest colour to get a nice finish on, it shows every string and every smear.

I think it's 2 5/8" diameter.

It's Andrew's STL.

Neil

[Anonymous]

The inner boss is 2-5/8" OD. The nameplate ring on it's own is 2.5" OD and 1.5" ID. If anyone wants a copy of the STL file for the nameplate ring send me a PM.

This evening I've broken away the spare brass and cleaned up the edges with needle files and wet 'n' dry. Both the ID and OD measure pretty much on dimension give or take a thou or two. That's not bad for CNC mill interpolation, and it was after filing and wet 'n' dry. I've also created a 2D drawing for the inner boss from the CAD model.

Andrew

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Michael Gilligan on 17/08/2018 21:57:52:

Took delivery of this Vertex VSI-4

[descriptive waffle is in my original posting]

Simple question: Has anyone dismantled one of these ? … am I dealing with a threadlocked screw, or just a very tight one ?

I really don't want to 'round' the socket !!

.

Today, having finally managed to get to the shop that sells high quality hex wrenches! … I removed the grub-screw, which was not threadlocked, but was extremely tight [in terms both of the thread engagement and the applied torque].

Then I removed the other grub-screw, which that one was locking [*]

Thanks to IanSC for frequently reminding us of that possibility

MichaelG.

.

[*] which required the long arm of the wrench, and an improvised handle on the short arm … all rather scary.

[Ian S CParticipant @iansc]

It will be interesting to see the works once you get it opened up. I, on occasion use an Allen Key that way round, and either hook a ring spanner on the short arm, or close the jaws of a Crescent Spanner on the short arm. It's amazing how far one of these keys will twist. Best not done with a ball ended key, it may end up not a ball end key.

Ian S C

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Ian S C on 29/08/2018 12:22:00:

It will be interesting to see the works once you get it opened up. I, on occasion use an Allen Key that way round, and either …

.

I used a small "toolmaker's clamp" [as distinct from a "small toolmaker's" clamp], which worked quite nicely.

Should be able to take some photos of the parts this evening.

MichaelG.

[Author]

Posts