NSW 'B' class mogul

[David]

I did the outside profile with a 6mm rougher, then finished with a 6mm not-rougher. I drilled the 'eye' of the hook 2mm and just used the 2mm endmills as a 'rest machining' operation to remove the resulting web. They were not taking heavy cuts at all. It was all pretty uneventful aside from my usual CNC cockups. I initially had a 0.5mm doc with the 2mm EMs, and after one broke I reduced it to 0.25mm and the next one lasted longer but still broke so the doc wasn't the only problem.

[Roger]

Nov 14, 2019 9:59:56 GMT David said:

I did the outside profile with a 6mm rougher, then finished with a 6mm not-rougher. I drilled the 'eye' of the hook 2mm and just used the 2mm endmills as a 'rest machining' operation to remove the resulting web. They were not taking heavy cuts at all. It was all pretty uneventful aside from my usual CNC cockups. I initially had a 0.5mm doc with the 2mm EMs, and after one broke I reduced it to 0.25mm and the next one lasted longer but still broke so the doc wasn't the only problem.

I don't think 0.5mm deep cuts are excessive, but certainly on the high side. If you're going that deep, the feedrate probably needs coming down quite a bit. I tend to go with more shallow cuts than less deep ones, even though it punishes the end more. A finishing cut with a new cutter soon cleans up anything where the cut isn't parallel. The problem with deeper cuts and small diameters is clearing the swarf. It doesn't take much to break them, so I prefer to make smaller chips and make it easier for them to get out.

[David]

I was taking 0.25mm axial cuts and the chips were like dust. The problem must have been the radial width of cut. It's shown in yellow here and it shows the whole tool will be cutting on the way in, and probably half or more on the way out too. Perhaps that's too much or it was flexing and springing back at the beginning and end of each pass.

[racinjason]

David I would have drilled the eye first.
Cheers Jason

[David]

Hi Jason, I did drill first. The order was spot, drill, outside profile rough then finish, then last of all the 2mm EM to open out the hook. That picture of the simulation shows the 2MM em path happening last, although it's a bit hard to see the hole is already drilled - the only clue is the blue colour for that operation.

I started on adding the threaded sleeves to the clack valve castings today. The longer one was easy and went well. The shorter one that also has a 60deg cone in it is not going so well - I'm on my 4th attempt! It has a short flat bottom hole that needs to be a good fit over the casting, a cone at the other end, and is only 6mm long. I may change the casting so the inlet leg is shorter and at closer to 90deg as I can't get the current one to screw down far enough due to the cladding. I'd checked in F360 that it would work ok but in practice it's not so good.

[Roger]

Nov 14, 2019 22:10:02 GMT David said:

I was taking 0.25mm axial cuts and the chips were like dust. The problem must have been the radial width of cut. It's shown in yellow here and it shows the whole tool will be cutting on the way in, and probably half or more on the way out too. Perhaps that's too much or it was flexing and springing back at the beginning and end of each pass.

That looks fine to me, it's probably just too high a feedrate or even too high spindle speed. There's a temptation to run small cutters too fast so that they rub rather than take a cut. If you slow down the feedrate you need to slow the spindle speed down too. I'm a fan of slowing everything down when it comes to small cutters. There's no real hurry when I'm making something, so giving the cutters an easy time means they last longer and I don't have to live with thrashing spindles and screeching cutters like you see on YouTube where everyone is trying to impress you as to how quickly they and rip through metal.

[steamer5]

Hi David / Roger,
Just for my curiosity sake. Having drilled the 2 mm hole & given that the hook part opens out would it have been possible to drill a 2nd hole either 2 mm or maybe a bit bigger such that it left a small amount of metal between them & thus removing the bulk of the metal inside the hook? Would have to be done before outside profiling. Then using the 2 mm EM to remove the remaining bit, followed by only have to do the finishing?

Cheers Kerrin

[Roger]

Nov 15, 2019 21:34:36 GMT steamer5 said:

Hi David / Roger,
Just for my curiosity sake. Having drilled the 2 mm hole & given that the hook part opens out would it have been possible to drill a 2nd hole either 2 mm or maybe a bit bigger such that it left a small amount of metal between them & thus removing the bulk of the metal inside the hook? Would have to be done before outside profiling. Then using the 2 mm EM to remove the remaining bit, followed by only have to do the finishing?

Cheers Kerrin

Hi Kerrin,
Absolutely, as long as there's some meat between the holes there's no reason why you can't do that and make life easier for the cutter. The problem is that long series cutters are frighteningly expensive unless you but PCB ones which are much more fragile since they're not intended for cutting Steel. Still, given a light enough load, you can use them but it does take ages. What you've suggested would speed the process up somewhat, but it's still a bit painful. I don't overly worry about that though these days, I just set it off and come back some hours later. I keep an eye on it on the Web Cam to see if the cutter has broken though.

[David]

I got one of the clacks finished today. The short threaded sleeve is a bit of a fiddle as it has a short flat-bottomed hole and a cone, but once you know how to do it it's ok.

This is the setup for soldering the short threaded part on. Lots of tippex and less than a ring of thin silver solder. The steel nut is to protect the threads and to help finish forming them when it's taken off because the I couldn't get a nut to start on the outside. Next time I'll change the order of flat bottom hole and cone to avoid this problem.

I shortened this leg by about 3mm before soldering the sleeve on to help with the clearance problem and also bring it closer to prototype length.



Everything in place:



I realised why I didn't catch the clearance problem. In F360 I had the clack sitting on the boiler bush and clearing the cladding but that was just pushing it straight down, not screwing it. So as it goes around it starts to foul the cladding everywhere other than where it is meant to finally end up. It's only a few mm so I'm just going to put a washer under it to make it look like the boiler bush is standing proud of the cladding.

[David]

A few small jobs today. I soldered the second clack casting together yesterday. The very thin silver solder I've bought from our local ME supplier is brilliant for this sort of job. Paste would be even better if I could get it for a reasonable price but I haven't managed that yet. I'm really pleased with how the bolts on top don't show any solder at all.



I've loctited the handbrake clevis onto its shaft in preparation for drilling and pinning.

Tonight I spend a few hours cleaning up a pair of curved angles I cut from 3mm steel for the cab and soldering them on. It was tricky to hold these in place and get them tight against the cab sides but a machinists clamp and some gardening gloves did it. I also filed the curtain rail bracket threads flush with the outside of the cab and fixed the one I had going through a clearance hole before I realised the cab side was meant to be threaded. I figured while I was soldering it was a good time to get to that.

I couldn't figure out how to take a clear photo of the angle soldered on but I hope this gives an idea.

[Jim]

Looking good David. You should be feeling pretty happy with the work you've done so far I certainly would be if it were me.

Jim

[David]

Thanks Jim! I am pretty happy with it and if it looks anything like as good as your loco I'll be very pleased. How's the beam engine going?

[Jim]

The beam engine is coming along nicely thanks David though I'm quickly discovering it's putting on weight faster than me. It's not a light weight model that's for sure.

Jim

[David]

I've spent all day (and stayed in the workshop past my bedtime) to get these horrors out of the way. 3 wins, 1 draw, 1 loss, and hours upon hours tweaking models and toolpaths and 1 metre of 3mm x 100mm steel for what looks like a simple part (although the steel was also used for 4 front angles which were more economical being able to get two in the same space as 1 rear angle). Glad to see the back of them! They're the rear angles for the cab sides. They take an hour each to machine.

I apologise for the photos, it's quite difficult to capture them because they're pretty small and it's steel on steel.

The day started like this. You can see the double sided tape let go despite the clamp. Of course I had to use a chisel to get the part off the wood after this - the tape was sticking plenty well by that point.



Then this happened on another one. I think the endmill picked up some of the tape when it was cutting at the full depth because there was a plug of goo and chips at the end of it. Thank goodness it happened right at the end and this part was recoverable.



That prompted me to adjust the toolpaths to not cut all the way through the stock to avoid the tape. It's a bit of extra cleaning up but better than risking another scrapped part.

I was a bit worried about the soft soldering but it was easy with bakers fluid and a very mild flame. They fit very well.

[Roger]

Hi David,
Holding some of these things can be a challenge, at least you got the job done in the end.

I don't use double sided tape and I avoid thick backing material. They both have the same problem in that they are too soft and they have a tendency to deform during the cutting operation and reduce the clamping forces. The best all round backing material is hard faced board between 2-3mm thick in my opinion because you can clamp down on that really tightly and it doesn't compress significantly. It also resists absorbing cutting oil.

It's hard to judge how thick the material is, but it looks like you might be being a bit ambitious on the depth of cut. I would't go any deeper than 0.5mm deep on anything like that simply because the cutting forces can overcome the clamping force. It doesn't matter if it takes longer if it saves you having to recover or discard a damaged part. Patience in these things is well rewarded. It doesn't bother me if a cut takes 6 hours or more, as long as I can just let it trundle away and get on with it unattended. I never stand over the machine waiting for it to finish. Exually well, I rarely scrap anything because I don't push my luck.

[David]

It's 3mm plate and the resulting angles have 0.5mm thick walls so they're 3 x 3 x 0.5.

Interesting notes about backing boards and I'll have to see if the local big hardware store has something like that.

I did have to keep an eye on these because I sometimes saw chips sparking when they were recut in full width channel so I had to keep blowing them out. If I kept the channel clear there were no problems.

I was using up to 1.5mm depth of cut on the edges, down to 1mm in the channel. A 6mm carbide endmill should be easily capable of that. The spindle speed was 1400rpm and the feed was 30 mm/min on the edges and 25mm/min in the channel. The finish was pretty good so I don't think the cutter itself was struggling.

I was thinking last night perhaps the way to do these would be to cut a rough shape out of the stock and clean it up then solder it to a backing plate.

Would a magnetic chuck hold steel and a sacrificial steel plate?

[springcrocus]

Nov 20, 2019 22:37:29 GMT David said:

I was using up to 1.5mm depth of cut on the edges, down to 1mm in the channel. A 6mm carbide endmill should be easily capable of that. The spindle speed was 1400rpm and the feed was 30 mm/min on the edges and 25mm/min in the channel. The finish was pretty good so I don't think the cutter itself was struggling.

I must be missing something here but this seems way too slow a feed rate to me. 30mm/min divided by 1400 rpm equals 0.02mm per revolution of the cutter, or one thou in old money. Assuming a four-flute cutter, this is one quarter of a thou cut per tooth per revolution. I would expect the cutter to blunt itself, through rubbing, in the first half-inch or so. Or maybe I'm just getting old and confused.

Regards, Steve

[Roger]

Nov 20, 2019 23:02:32 GMT springcrocus said:

Nov 20, 2019 22:37:29 GMT David said:

I was using up to 1.5mm depth of cut on the edges, down to 1mm in the channel. A 6mm carbide endmill should be easily capable of that. The spindle speed was 1400rpm and the feed was 30 mm/min on the edges and 25mm/min in the channel. The finish was pretty good so I don't think the cutter itself was struggling.

I must be missing something here but this seems way too slow a feed rate to me. 30mm/min divided by 1400 rpm equals 0.02mm per revolution of the cutter, or one thou in old money. Assuming a four-flute cutter, this is one quarter of a thou cut per tooth per revolution. I would expect the cutter to blunt itself, through rubbing, in the first half-inch or so. Or maybe I'm just getting old and confused.

Regards, Steve

Hi Steve,
I think you're right, to me it's the spindle speed that's too high rather than the feedrate too slow. There's a tendency for all CNC users to run the spindle way too fast in my opinion, I don't know why.

[Roger]

Nov 20, 2019 22:37:29 GMT David said:

It's 3mm plate and the resulting angles have 0.5mm thick walls so they're 3 x 3 x 0.5.

Interesting notes about backing boards and I'll have to see if the local big hardware store has something like that.

I did have to keep an eye on these because I sometimes saw chips sparking when they were recut in full width channel so I had to keep blowing them out. If I kept the channel clear there were no problems.

I was using up to 1.5mm depth of cut on the edges, down to 1mm in the channel. A 6mm carbide endmill should be easily capable of that. The spindle speed was 1400rpm and the feed was 30 mm/min on the edges and 25mm/min in the channel. The finish was pretty good so I don't think the cutter itself was struggling.

I was thinking last night perhaps the way to do these would be to cut a rough shape out of the stock and clean it up then solder it to a backing plate.

Would a magnetic chuck hold steel and a sacrificial steel plate?

Personally I wouldn't go that deep with the 6mm cutter at that feed, the cutting forces are getting to be quite large and you can't clamp it that tightly. I'd use a smaller cutter and go down in 0.5mm increments so you're not removing as much metal and the cutting forces are lower. 20mm/min would be plenty for say a 3 or 4mm cutter. Steel plate is much harder to machine than free cutting Mild Steel bar.
If you ever see sparks, you're probably using too high a spindle speed. If in doubt, slow everything down and take lighter cuts. Trying to do things in a hurry ends up making more work in the long run in my experience. If the machine isn't cutting almost silently, I experiment until it is. The happier it sounds, the better the finish and the longer the cutters will last. It's not a matter of what the cutter will cope with. You can use that philosophy with a commercial machine that's exceptionally rigid, using flood coolant and parts clamped so there's no way they can move. That's not the situation that we find ourselves in. We have to coax the machines to get the results we need because they're less than ideal for the job.


This is similar to the sort of backing material I use. Maybe you can find a PCB manufacturer near to you who uses this. That's where I get mine from. It's got holes in it from the production process so it's scrap as far as they're concerened, so they'll give it to you.

[David]

It seems hardboard isn't a term we use down here. Do masonite or melanine ring a bell? Australian ebay has no results for hardboard and searching for 'white faced' just brings up watches.

I often get confused by these terms. Sundela is another one I read a lot in UK magazines and have no idea what it is!