The Workshop Progress thread 2018

[MuzzerParticipant @muzzer]

Recently I removed the Chinese 4 axis Fanuc clone controller from my CNC milling machine. It's clearly a very powerful and capable controller but despite my many years working in China and with Chinese, I simply can't fathom the strong Chinglish in the manual and of course any attempts at technical support suffer from the same issue. Instead I fitted a Centroid Acorn controller from the US. Cost twice as much but has the benefit of documentation and support that I can actually understand. And unlike LinuxCNC it doesn't require a PhD in Linux and funny handshakes to set up.

I've now got my Renishaw touch probe, automatic tool setter and rigid tapping working after a bit of effort and support. Works pretty nicely and wasn't too painful to install and configure. Unlike some of the CNC controller outfits, they have been fitting controls to industrial machines for 20 odd years, so by and large they seem to be able to work with most configurations. I bought mine from a Danish reseller, thus avoiding import tax and duty (being in the EU: perhaps there's a lesson there somewhere).

Tonight I machined 2 test pieces. Both the same simple shape – 50mm square stock with a spherical concave surface on the top face. You generally machine these with a ball ended end mill but the disadvantage of these cutters is that the surface speed of the centre of the tool is zero and you end up with scuffing and a poor finish where that part of the cutter is used. The approved solution is to use a 5 axis machine so that you can tilt the tool towards or away from the direction of travel. That's not an option for me, as the present Domestic Manager won't stretch to that. Instead I thought I'd achieve a similar effect by tilting the stock at 55 degrees to the vertical, so that the end of the cutter never has to attempt any cutting.

The toolpaths are quite easy to do in Fusion 360. To achieve the tilt you simply define the Z axis at the required orientation and within reason, the rest follows.

Sure enough, the surface finish is a lot better, although there would be further improvement if I used a tool with a better (polished) edge – you can see the grind marks from the tool on the workpiece. And coolant would have washed the swarf away and prevented any minor recutting. I don't have the guards back on yet, so coolant wasn't a sensible option.

All in all, an interesting result. For surfaces that would benefit from a fine finish, this approach has potential. Obviously set up takes a little longer and the machining time was about 50% longer but each piece took under 30 minutes, so hardly an issue.

Conventional (perpendicular) toolpath. Look carefully and you can see the spiral surface finish in the middle of each track:

With angled tool. Every imperfection on the ground edge of the tool shows up!

Murray

[Anonymous]

Wow, that looks good Muzzer. I've done a fair bit of CNC machining with ballnose cutters, but only in the conventional sense. The finish I got didn't look too bad to me, but obviously not as good as yours. Doooh, didn't think of tilting the workpiece. I've got some press tooling to make for the strakes on my traction engine wheels so I may well have a go at tilting the work. Or may be I should just have a few beers so that I lean over to one side.

Andrew

[MuzzerParticipant @muzzer]

Rather than use a ground (ideally honed) HSS or carbide cutter, it's possible to get polished carbide inserts for indexable tooling eg a 16mm diameter cutter with a 3.1mm corner radius. That's basically a "bull nosed" shape, as our Mercan cousins would call it. So that would give a reasonable radius to work with, albeit with a cylindrical central portion. They should have a more precise edge profile and less surface speed variation across the edge.

They would allow better surface finishes on convex surfaces and for some areas within concave surfaces. The very bottom of a dished concave surface would still be a problem. But while I am saving up for my 5 axis machine, this may be a partial solution in some situations.

The AP** inserts don't appear to be available with any significant (>0.8mm) corner radius but the Mitsubishi R390 family (different insert geometry and toolholders) are available up to 3.1mm (1/8" ). These are available on ebay and Aliexpress, so further trials will commence in a few weeks….

Murray

Edited By Muzzer on 14/04/2018 11:21:01

[John HaineParticipant @johnhaine32865]

Murray, it would be interesting to have a link to the Danish importer please?

[MuzzerParticipant @muzzer]

It was these boys. Total was 3081Kr all in which is about £358. Arrived in 3 days. I bought the upgrade license direct from Centroid as it's just a license key.

Refreshing not to get ripped off on the way into the country by both the tax man, customs and couriers. CTC Tools (HK) have a Danish warehouse too so you don't pay duties etc on their Chinese stuff – no idea how that works…

Murray

[Neil WyattModerator @neilwyatt]

Posted by Muzzer on 14/04/2018 14:47:55:

Refreshing not to get ripped off on the way into the country by both the tax man, customs and couriers. CTC Tools (HK) have a Danish warehouse too so you don't pay duties etc on their Chinese stuff – no idea how that works…

Plenty of Chinese companies have UK warehouses – HMRC have figured out how it works…

Neil

[mechman48Participant @mechman48]

Started on the flywheel for my Vertical cross; casting was a bit rough looking at the start, ( unfortunately didn't get a pic before starting ),I have another casting but that's even worse, anyhow by the time I got through a section of chilled casting, what started out as a 4" casting ended up as 96mm OD, & still had remnants of a blow hole on the rim. The only option was to mix up some two part epoxy resin & blend in some CI filing dust to strengthen & colour match…

Apologies for the out of focus shot, but you can just make out the hard spots on the centre line……

Remains of the blow hole imperfection…

Mix of 2 part epoxy & CI filing dust applied ….

​Should blend in fairly easily…

George.

[mechman48Participant @mechman48]

p.s. alternate flywheel… rough as, I suspect there's probably more chilled iron in this one than the one I'm using…

We'll see eventually…

George.

[Jim NicParticipant @jimnic]

George

As a matter of interest, where do you get these castings from?

Jim

[mechman48Participant @mechman48]

Posted by Jim Nic on 19/04/2018 21:48:12:

George

As a matter of interest, where do you get these castings from?

Jim

​Receipt shows… RDG tools.

G.

[JasonBModerator @jasonb]

The RDG ones are a very good price but you have to take the good with the bad as they vary a bit, I try to buy them at a show where you have a chance to look at them and pick what looks best from the outside.

A good source of small flywheels is this seller on e-bay, he used to cast for College Engineering and Blackgates, the ones I have had from him are fine.

George, if you are still looking for a 9" spoked one then try engineers Emporium, they list 8.5" and 9.5" but don't put them on their website. Another source may be Polly/Bruce as the R*B engine has a 9" flywheel option

[mechman48Participant @mechman48]

Machined up the flywheel resin repair; blended in quite nicely albeit never will be the same as original cast material…

George.

[JasonBModerator @jasonb]

I have been working on this engine on and off for about 8 weeks and got it completed today. My usual 24mm bore and also 24mm stroke that stands about 6.5" tall. I took various features from several other models and also some full size ones to come up with something that was pleasing to the eye and not too difficult to fabricate or cut from solid.

It's got a bit of a knock from a slightly loose crosshead pin that needs attending to which is amplified by the bench but apart from that runs quite nicely. My compressor has a small 12 lts tank and it will run for just over 9mins with the regulator set at 10psi before the tank is empty, less if I rev the life out of it

And a bit of video, you can hear how the knock gets quieter as I lift the engine to move it about, flywheel looks to be on fairly straight too :LittleDevil:

[MuzzerParticipant @muzzer]

Finished the operations in the first machining setup of the steel yoke for my Bridgeport CNC conversion. This connects the ballscrew to the quill, so needs to be pretty robust. I made one manually a few year back in loominum but this is the final part. Seemed rude not to make this on the CNC, having bored out the cavity for the ballnut on the lathe. Yes, the surface finish on the bore isn't perfect but it's a very good fit on the ballnut.

Used a snazzy "V7" carbide cutter by YG-1 which is described as suitable for both high MRR roughing and also fine finishing. Have to say, I managed to shift some steel with the thing during the 3D Adaptive (roughing) and after the finishing pass (0.5mm), the surface finish is pretty impressive. Youtube CNC warriors seem to prefer machining loominum, perhaps due to the easy shiny finish but this steel (CRS) machined just fine.

Yes, I set it up at just the right height to avoid machining the vise jaws – it's about 0.4mm

Vids here and here and here if you like that kind of thing (warning – handheld footage not suitable for Neil). The last one is quite fast – the ball end mill goes like stink, even though I'd dialled back the feeds and speeds. These old machines can still perform well when driven by modern toolpaths.

Now I will need to remount it so I can machine the back and bottom features.

Edited By Muzzer on 22/04/2018 20:54:39

[Anonymous]

Good grief, that's some serious looking swarf there. Any figures for feeds 'n' speeds and DOC and WOC?

Andrew

[Andrew HutchinsonParticipant @andrewhutchinson12802]

Muzzer,

Who does your tool setter? Agree with you on the loominum bit. Thanks for posting.

TIA,

Andrew Hutchinson

[Neil WyattModerator @neilwyatt]

Posted by Muzzer on 22/04/2018 20:53:33:

Vids here and here and here if you like that kind of thing (warning – handheld footage not suitable for Neil).

Especially not after a couple of glasses of Aussy red…

[Mark RandParticipant @markrand96270]

1) Do you really need ballscrews? It seems to be climb milling quite well in the videos. Or is it to be able to take larger cuts and faster movements with the same steppers motors?

2) Are the chip shields held up with standard 1/2" Loc-Line?

3) It seems wrong to be able to produce surfaces like that without using a rotary table and a big handle*.

 

*= Just jealous

Edited By Mark Rand on 22/04/2018 22:06:35

[MuzzerParticipant @muzzer]

Mark – this is a different machine. It is a "proper" CNC machine with industrial servos, NSK ballscrews etc. I started the BP conversion a few years back, then I had a very nasty accident with a mouse when I was browsing ebay and ended up with 2 machines. I plan to finish the BP conversion anyway, as I was perhaps 80% completed.

No, the Locline is for the coolant. I had some proper guards on the machine and am part way through fitting a full enclosure. But I got impatient and did this work without them. I used some corrogated plastic sheet to try to contain it but there's swarf and coolant everywhere….

Yes, it's shocking what you can achieve. The first yoke was make by hand on the BP and took forever. Looks crap by comparison.

Andrew – the toolsetter is Chinesium but works really well. If I remove / refit the tool, shift the table all over the shop and then remeasure, I get pretty decent repeatability – something like 5um, I forget. Not enough for the willy wavers but more than adequate for the likes of me.

I've got a blog at murraye.com that describes my ongoing exploits.

Murray

[MuzzerParticipant @muzzer]

Posted by Andrew Johnston on 22/04/2018 21:16:57:

Good grief, that's some serious looking swarf there. Any figures for feeds 'n' speeds and DOC and WOC?

Andrew

3000rpm, 480mm/min, 0.04mm/tooth, 10mm stepdown, 3mm optimal load (WOC). That's about 1.2kW cutter power, so loads of margin. The cutter has 12mm dia flutes and 11mm shank, so can cut to a decent depth without rubbing. Max depth is 40mm on this tool and I needed 38mm.

Murray

[mechman48Participant @mechman48]

Machined up the flywheel; drilled & tapped for locking grub screw through boss. Broke M3 taper tap on tapping … 'Tut Tut ' I said, or words to that effect, no chance of removing it so re-drilled on opposite side, to balance out hopefully. Will fill in 1st hole with resin / CI dust to help balance out…

​Trial fitting flywheel…

​Still lots to do but slowly, slowly catchee monkey…

George.

[Anonymous]

Posted by Muzzer on 23/04/2018 14:47:12:

3000rpm, 480mm/min, 0.04mm/tooth, 10mm stepdown, 3mm optimal load (WOC). That's about 1.2kW cutter power, so loads of margin.

Impressive!

Andrew

[MuzzerParticipant @muzzer]

Made some quick (loominum) soft jaws yesterday to hold the yoke while I skimmed 13mm of excess stock off the bottom.

Would have been pretty tricky holding the yoke securely otherwise. Making the CAD model and toolpaths from the existing yoke model was pretty quick and easy. Fits like a glove!

Murray

[Anonymous]

Wow, that's pretty slick! Great finish too. Clearly I need to investigate the V7 family of cutters.

Andrew

[Bob RodgersonParticipant @bobrodgerson97362]

I've been meaning to finish two of these off for the last year but I've been busy with lots of other stuff. They are rear hubs for a veteran Triumph motorcycle circa 1911. The body was machined on the Tormach mill using the Rapid Turn Lathe attachment. It struggled with the 2 1/2" diameter material (EN8), but I got there in the end. The profiling was done entirely with a 2mm wide grooving tool. Turning without chatter initially was a problem but as the diameter reduced it became easier.

The ratchet on one end is from a damaged free wheel chain pinion that was used as thread gauge.

[Author]

Posts