Scratch Build – Complete ⭐ WING X99 - A CNC-milled Scratch Build! (Benchmarks,temps and wallpapers posted)

Welcome to Bit Tech, Gareth

And the home of theeeee braaaaaaaaave

 

Lovely bend work and it seems you were conscious of bend allowances too,I like what im seeing here

 

@Cenedd - Some scratch more than others.

 

Vault-Tec: Cheers. Shall endeavour to post some useful (or interesting) content while I'm here!

Cheapskate: Very true. My metalwork efforts are more 'scratch' than 'build'. Just not getting that perfectionist accuracy....but in my defense, I'm at hack saw, Dremel (contrary to my title, I not only know what a Dremel is, but also which end is the dangerous end!) and cordless drill level rather than CNC mill - let alone hand built CNC!

Andreas: It's really interesting to see the development of your mill. I keep trying to persuade myself that I can justify investing in a small mill - and failing. I have trouble with spec-creep: you start off looking at a drill-press ...and then it may as well be a mill ...and then it may as well be CNC ....and then it may as well be <insert next expensive thing here>. Before you know it, you're spec'ing a metalwork shop that will need its own building. At least with graphics cards there's a best you can buy at the time! Still, it's useful to know the information for interest and in case I do actually take the plunge....what does a divorce cost these days?! ;-)

 

Well I always go full retard when I do stuff

and land of the free?

You should go for it! Building the CNC was amazingly fun, rewarding and absolutely frustrating at times. But all in all really fun and great experience. You also then have a CNC when you are done witch is really useful!

I realize not everyone have the same access to tools but you can still build one with basic tools. You have to start somewhere and I think that the shapeoko should be looked into if you would like to build one. Was one click away from ordering one when my dad asked me to design and build one for their company instead. Little did I know It would take about 2 years go build that one with univercity studies and everything

 

Same as the back and the top, but for the sides.



It's kinda neat, the mill produces these very thin rolls of aluminium when facing the stock.



As you can see these are very very thin. Nothing to do with the build but still kinda neat


Its very nice that these are on all the passes since this tells me that the machine is very consistent in the passes.




Left side done! Time do the the other side.






Right side done!


Next update! I will be making the slots for the air to pass through. Here is the fastening for the next milling operation. Had to use this many fasteners due to the 2mm thin edges that is left unsupported on the edges from this direction of the plate, so each have a light clamping force and I really do not want them to get lose and ruin the part.

 

I have one,well worth the money,accurate and,with some effort with setting up tooling,fast.
Go for a 1.5kw 3 phase spindle with it and it is golden....and quieter than those stupid routers people sling in them.

 

So at the risk of littering the thread (please tell me if/where there is a better place to take this), if I were to give in to getting a CNC, you're saying the Shapeoko is good but to put a better motor on it than the router they suggest. Three phase mains is not an option in my shed so I'd need a single-phase VFD and a 1.5kW spindle?
Does the spindle just spin at fixed speed all the time and get moved about by the Shapeoko via USB control? (USB rather than parallel is part of the attraction here)
How much of a problem is the lack of enclosure of the Shapeoko vs the Nomad? I've got half of a 5x3m carpeted office space (where scattered metal chips would be unwelcome!) and a small workbench in 5x2m shed side which wouldn't be too bad as long as I can just hoover up (It's a VAX, it can handle rubble!).

 

The dust shoe on my mpcnc gets the bulk of the chips, if you can, vent the vac outside. The fine dust will escape, even on mine with the cyclone before the vac I get a fine coating of dust on everything when milling wood. I've seen folks hang welding curtains to keep the chips contained as well. Especially around carpet!

 

As said above, a water cooled spindle is much more quiet than a air cooled router spindle or regular wood routers you find from dewallt etc. Having some kind of dust collection is also a must if you are going to use it in a office type environment. I have a big cyclone dust cleaner and that works wonders, BUT it is really loud that it even drowns out the sound of my 2.2kw air cooled spindle. But given that everything in my workshop sounds a lot (lathes, cutters, mills, etc) it does not matter at all. You are not disturbing anyone and you still need ear protection anyways.

But I imagen a quiet watercooled spindle with a regular vacuum cleaner would do a okay job at a okay sound level

 

Slots for the air to pass through.


All clamped down and ready to go!


Damn, this was going so well...


Well, some damage to the collet and the holder. But everything was okay, just some minor scratches but still no fun...


It's my first mill that I break, I was running the end-mill to lean on the cutting fluid. Problem was a combined chip clearing (due to the lower air pressure with less coolant flow) and to little fluid lubricating the aluminium causing it to stick to the end mill and rubbing instead of cutting against the material.


This end mill costs about 30 USD. **** happens and stuf breaks. Luckyly this is a small end mill so they are cheap compared to bigger ones so I guess that is good. But the smaller the end mill the more likley they are to break so I guess that evens out.


Coming along nicely now after that small hick up, very time consuming though. But low and slow is the way to go with such a thin diameter end mill.




About 1 hour later



Finally done!




Really like the results. I have nothing to complain about. Very happy that I could save the piece even though the tool broke. But since it snapped of during the roughing pass there was still material to be removed so it took away the damaged part. There is not even a mark on the piece. Lucky, but the chance of the tool breaking is at the highest when doing rough passes.


Same procedure for the other side. This operation takes about 2 hours with setup and material removal for each side.

 

Yes, but to be fair, I've had far worse wanks.

I mean, just, look at this



It's sooo satisfying.

 

Yes! Super satisfying, I pull these to long strings when I am bored and waiting for the CNC to complete the passes

 

Time to mill the front of the case. Do not want to mess this part, lots of work going in to it.


Will be using this precision grinded bar to line it along the Y axis and this squared piece of metal to align the X axis and also clamp down the front.




Tried my toolpath on the test piece I made to make sure that I did not get any vibrations when doing the real part.


Will be using this 8mm end mill and no dust shoe so I should be able to get some good video. So look out for that!


Using these clamps to remove any vibrations.



B is for bottom



First cut done! Worked out great. "Just" have to flip it around and do the same thing on the other side.


Time for the top part of the front.


Top part toolpath done! Turned out great. Now on to the front and the logo.


Done and done! Was very nervous about this, with all due respect. Some hitch in the program caused the updated toolpath to not register. Took the cut down to 2mm instead of 3mm AND it tried to do the whole profile on depth 0mm instead of 3mm. I was very lucky to be able to stop it in time and fix the issue. Only thing that changed was that the cut is now 6mm deep instead of 5mm that I intended. Will make no difference in the end, but I was about 0.3 seconds from ruining this piece...


Will be fitting the logo and button here. Will be a magnetically dampened button. Hope it works out nicely! Otherwise some springs will have to due!






Last "major" piece on the outer shell for the assembly to be possible is next. The case foot!

 

Looking good. Interesting to see how you're approaching things - from a problem-solving perspective...that and I've got you to blame for looking at CNC more seriously now!

 

Vibrations? On THAT massive bit of metal? You humble madman. I like how you left extra during bending to mill to fit.

@Cenedd side thread - Figure out what you want to do in the future with one and research heavily. For example: Shapeoko's major flaw is a lack of Z travel. You wouldn't be able to do what Andreas did in that last post with one.

 

I take no responsibilities!

Yes, that is very good advice. Z-travel usually matters a lot on how stable CNC-routers are also. Less travel = more stable. Usually though, you don't need that much Z-travel if you are doing panels, some engraving, contours, etc. I have 200mm in travel or about 8 inches.

 

Making the foot! This will have a piece of plexiglass sandwitched between the bottom part and this foot to try and give it a floating effect. I think it will look nice!


Had to do a very tight clamping to be able to use this sheet I had left









First side done! Time to flip it around.


Had to use some scrap pieces to support the outer edges





All done!


The side facing the bottom part of the case.


The part that is going to face the floor. The cutout will have a 2mm neoprene rubber glued it. Next up is threading and drilling to be able to assemble these flat pieces to a case.

 

Floating sounds good. Neoprene also good to not scratch what it's under.
Found a hackspace vaguely near me so going to see whether I can get a bit more clued up there before I do anything rash...which is my usual MO!

 

Definitely worth doing the research first. Even if it is tempting to just buy one on a whim



Well, there have been a few, I have a small graveyard of parts that I have to throw away. So far I have made a new, bottom frame, top frame, PSU-holder, side frame and an extra front that I was going to use if the bend did not work out as planed on the first try. All in all about 20 hours lost/wasted I would say. But that is the nature of prototypes and one offs. Here are some of the errors and the new parts fixed!


New side fram (the one that got shatter marks is being replaced)


New side frame in progress with the updated design



Here you can see the added material on the sides. I just added a few millimeters of extra material. I should have done this from the start. Part got much stiffer and nicer looking. Mabaaaaaaad




Done!


Not a screw up here I just did this after I fixed the left side so this is in this update. But here I did the last machining pass on the back side for the LED strips on both frame sides.







Both turned out great. Really pleased.


Here you can see the first bottom frame. It has this quite deep (0.25mm) dwell mark due to the PID controllers on the motors not being tuned correctly. I fixed this after I did that part so it bothered me so much I made a new one just for that silly mark. I just knew it would end up bothering me so much If I left it in. So the OCD costed about 2 hours of extra work. Not much in the grand scheme of things though. Really pleased with the new one also


Reluctantly found out I had messed up on the top fram, this was not very fun since that is a very time consuming part to make. I had a bad toolpath that had taken about 4 mm to much on where the "top line" should be. Did not notice this at first so that was a bit bummed out about this since I thought it turned out good. But this allowed me to do some redesign on the fastening of the plexi sides and the plexi top so all in all, turned out good after all!


These are the slots that will secure the plexi sides in place



Done! That's all the bad parts fixed. Now it is time to drill and tap all the holes so I can assemble this into the outer case. This should be fun and very nerve wracking at the same time. Do NOT want to remake any more parts now, sick and tired of doing the same work twice