Project: Silent Overkill

Alright, it's clearly past time I made some actual progress. So I've lopped a chunk off that bar of steel - didn't like the idea of trying to spin a 10kg block in a small machine! :eek: Also didn't like the idea of cutting it by hand so I went rummaging in the loft for the bigger bandsaw!



That's not how you use a vise stupid! Yeah, well it worked didn't it?! It did technically fit in the jaws if I removed the 3D printed softjaws but it was so damn heavy that it would only have stayed put if I'd clamped down on it hard enough to leave bit marks in the end. This way I get three points of contact and no marring.



You call that a clean cut? Er, no...I call that a got bored and tried to speed up the cut by rotating the piece (which does work) but it didn't track very well and the result is a rather hacked off lump....which totally doesn't come to bit me in the arse in a bit! :rolleyes:

Let it snow, let it snow, let it snow!



...and this is why we have a shop-vac! ...and yes, that's a 3D printed adapter to mount a cheap crevice tool and a cheap brush tool rather than whatever ridiculous amount Festool want for such things. Why Festool if price is a concern then? Well, put simply, at the time it was the only rectangular one I could find and it perfectly fit the gap between the end of my workbench and the wall. Dead useful for not dying horribly and prematurely by breathing in all the dust from various cutting/sanding/cleaning type ops. Would recommend to anyone wanting to live longer than next year!



Mounted, trued up, painted red and facing off the end. This is where it bit me. The diameter here is 100mm so with larger diameters, the outer edge (the diameter) is moving faster for any amount of rpms than a smaller diameter bar is. So you can spin a 10mm bar at 1,000+ rpm and face off the much smaller end in a flash but try that on this and things will break....in your face! So this was running just a tad over 200 rpm and because of my nasty hack-job was a horribly interrupted cut. It's flat when all the red is gone (it's easier to see when it's spinning) and as you can see, there's only a band of silver being cut at each rotation. So that basically smacks into the tool every time it comes round. That rather limits the depth of cut you can take if you don't want the carbide insert shattering in your face...which I don't! Also on the issue of paranoia vs prudence, the four-jaw chuck wasn't technically necessary for this as it doesn't need to be carefully centred or put deliberately off-centre....but an extra jaw's an extra jaw stopping several kilos of steel breaking loose in my general direction. Similarly the tail support isn't necessary....but I really want it to stay put while it's thunking against the tool!



...and done. Well, I say done but this is simply flat and ready to start making stuff with...which I shall put off to another day. Call that a nice finish?! Nope, it's a pretty terrible finish typical of lower grade steels like EN1 and EN3 (which this is) and it can be improved but there's no point until it's an actual finished surface - frankly I only cleaned up the outer edge because the rust annoyed me and to balance it as it wasn't round. Right, I shall now take bets on how long it stays 'resting' in the chuck before I make any more progress! :D
 
More progress for you then! Two lots...in one day! :eek:

I've managed to align the spindle and the gantry but haven't yet gone as far as facing the bed. Something rather destructive doesn't seem like the first thing to do when you don't really know what you're doing!
I've set up a machine configuration in Fusion 360 - seems nuts that they support that but don't provide the machine config for you. Then as a tester, I've just done a quick slot across a piece of scrap wood. Nothing challenging, nothing exciting nor even useful; just a proof of working. Andy dubbed it a "Hello world" which I rather like.





Sides are a little ragged but that's an up-spiral bit for you - 6mm, solid carbide, 1-flute, TiN coated...if you want to be specific! Point is that work location worked, things went in the expected direction by about the expected amount and nothing went p'twang! So it turns out that you CAN teach an old dog new tricks! :D
 
Looking good for an old dog :P

Honestly though part of me wouldn't face the bed, I'd face a destructible surface instead. If the bed is off as much as you say, can you be sure those lovely pre-made mounting points are truly perpendicular once you've faced it off?
 
I'm half with you already. Were I just doing wood, I'd definitely leave it alone and slap a spoilboard on top and surface that. My concern is doing aluminium and maybe even trying to thread mill the stupid thread in the res ring in steel. If you're talking sub tenth of a mm accuracy, I'm not sure an MDF board is going to be stable enough. I'm not 100% sold either way as it would mean losing the hard anodised finish.
The mounting holes, you're also spot on but it's worse than that in that I can't find any information on 'where' they are. That might sound a dumb thing as they're clearly right there but where is that in terms of coordinates and how far are they spaced apart - seemingly somewhat randomly. As far as I can tell, they're 60mm centres on the X axis but 50mm centres on the Y axis...only with a row offset by half in the middle....and 80mm gaps in places. There's not a regular logic to it to be honest. Whether they're perpendicular doesn't really matter if you're only using them as clamping points. Would be much more important for both position and perpendicularity if they were for location dowels.
But yeah, I need to work that out as it doesn't look like there's a replacement part available (at least not listed) and the material to make one is a good £50...so I'm not planning to be hasty! :D
 
Last edited:
I shall have a watch and let you know. It seems alright so far. A bit low when it comes to polish and creature-comforts but then the price is similarly low. If you compare it to a Haas or a Tormach it's going to come up wanting on (I hope!) every point...but then I saw a video of whether one of Tormach's 1500MX CNC mills was just a hobby machine and they're £25k! :eek: The supplied spindle is no good for aluminium (yeah, it'll cut it slowly but it'll look like it was cut with a spoon), the supplied probe is a bit plasticy and the clamps a bit basic...but it all works and gets you started. So far, the only thing I'd complain about is that it's awkward to align the spindle in Y axis and that all the docs cover using Candle (free open source software) to connect with a USB cable and run that way. Nothing really covers using the offline controller bolted to the thing and how to get from model in Fusion to cutting stuff.
Incidentally, I'm thinking that a nice sheet of black acetal and I can maybe make some DDC pump tops. I like the Heat Killer ones but the £20-something shipping from Germany is...unappealing. £80-odd for two isn't an attractive option.
 
I think I may just have to order some acetal and got LePhuronn's suggested route. The Z loss should be offsettable by simply raising the router in its holder. This lot arrived and was "delivered" to my doormat or my bins while I was away for new year.



That's the replacement jaws for the vise that will actually hold stuff, a thread mill that should cut G1/4 threads, a thread mill for M3 (could be optimistic as it's tiny...but there's only one way to find out!) and the touch probe I had earlier but failed to take a picture of.

Thread mills



Assembled vise.



Have also just ordered an ER11 collet extension so I should be able to stop faffing about with router collets and use a set that hold any size up to 8mm directly. Again, I lose some Z height but again I think I can just raise the router up without losing any rigidity - may even gain some by clamping it closer to the spinny bit (that's a technical term for those unaware! :p ) but on the flip side, I'll probably lose some as it'll be a tool held in an ER11 collet holder, held in a router collet. Have to see to be honest...probably should have tried this route before ordering collet convertors etc. Also will have to wait to find out if the exceedingly expensive *cough* set of ER11 collets are actually any good or if they have so much runout as to be unusable. Again, only one way to find out!
 
OK so @LePhuronn was more prophetic than I. The mounting holes in the bed are hopeless. Wherever you put the work, the holes to mount it with are too far away or too close. Just to make life more fun, you may notice (as I failed to previously) that there are NO mounting holes on this vise. Or at least, not on any surface other than the underside....which isn't massively helpful in this situation. Was going to hold it down with toe clamps...but the holes in the bed are just hopeless. In fact, there are currently a lot of irritations that I'm going to have to fix one way or another. I suspect I may have to dump the included offline controller and connect a laptop running either Candle or Mach4 - at least if I want to be able to do the touch probing and not have to swap micro SD cards about the entire time. They're way behind the 3D printer game on this to be honest.

Right, can ya tell what it is yet?! :eek:



A 15mm slice of acetal. Hole bored through the middle (roughly - no probing really yet so aligned by eye) and then thread milled. Discovered that Fusion is excellent for modelling threads but lagging massively behind when it comes to actually milling them. Despite having modelled the thread, when you switch to CAM and set up a toolpath, you have to specify the thread pitch (19TPI = 25.4/19 = 1.3368mm) manually. You can use formulae but you still have to know the dimensions and the conversions if you're working in metric but with imperial threads. Job done, can see a spiral down the bore so I took it out to inspect it. It's more a scratch than a thread. Found out that by default the "Pitch Diameter Offset" is 0 mm. Now I took that to mean an offset from the thread spec so 0mm would mean what it should be without any adjustments....but no, totally wrong. This is the depth to cut the thread and you have to supply that information manually despite Fusion actually knowing the answer on the modelling side (you can, in fact 3D print them with no trouble at all). So Major Diameter - Minor Diameter = 1.712 mm. Bore another hole...realise there's a very good chance I'm going to hit the vise jaws, hit e-stop. Move the hole and try again. The other joy with a personal license for Fusion is that you have to do each operation as a separate file - it's an irritation that you have to pay to get round but it doesn't make a lot of difference if you don't have a tool changer - I do...but he's on holiday at the moment! :D
So third time lucky. A fitting seems a little loose but I started screwing in a G1/4 tap and it's spot-on so I think the thread on the fitting is just a little undersized.

Well, that's proof of concept at least. Will have to do better for M3 threads and see if I can smooth out some of the irritations.
 
It's funny because the bed on my old machine was all sorts of crooked. The new one? is literally almost completely flat and even.

I did mess up though, so I do need to make a new spoil board myself. Thankfully I think I still have some medite left.
 
Grab a steel rule that isn't dinged up (a proper straight edge is better but they're pricey) and hold it on-edge against the bed - like a knife cutting something. Then hold a torch pointing at the far side and see how much light you can see coming under the rule. Assuming the rule is actually straight, anywhere you get any light is not flat. If it's dark the whole length, that's flat. How flat does it need to be? Well probably not engineering level flat but I've come from a manual machining start-point so in a lot of ways this is a step down in that sort of precision. My mini mill is twice the price of the CNC and to be honest, that had to have parts replaced and then the bed sent out for surface grinding. They did a lovely job mind; IIRC it was flat to within about 5 microns all over. Does it matter? Maybe. It depends what you're machining and whether you need tight tolerances. If you clamp something hard down to a bent surface, the stock will bend too so an engraving will have an uneven depth (if you don't compensate with a flattened spoilboard as you are). When you get to metals, you want as much rigidity as possible so I'm not sure MDF would be a great surface. Then, if you have any unevenness to the bed, it's going to translate to the part or it's not going to clamp so well to the bed because it's not touching it for a lot of the area. Am I overthinking it? Yeah, problably; it's certainly my MO! Trouble is, I can't not....I've always said that the cruelest thing you could do would be to give me something that measures to an extra decimal place! :D <lines>I MUST NOT USE A MACHINIST LEVEL FOR HANGING PICTURES! </lines> :eek:
 
You must log in or register to reply here.
Back
Top Bottom