Scratch Build – In Progress Project: Turn a chunk of aluminium into a case

You don't scrap that. You frame it and put it on the wall next to Grandma's portrait.
I couldn't use that on a computer because I'd be dragging a guitar pic across it all the time.

 

It's a lot stronger than I expected. But yeah, can certainly get bent out of shape by accident. I'm still thinking about adding in some supports for the fins to make it a bit more sturdy.

I'll keep it around in my pile of "stuff that looks cool, but wasn't quite good enough to go to the customer". Now I need to find a guitar pick to see if it makes a cool noise.

 

So, I've finally managed to get a break from the real work, aka the work that pays the bills. Time for an update then.

Slap it in the vise upside down to take off the bottom.


After removing the bulk of the material, I realised I was right, there was no longer enough material to support the part when doing a finishing pass. So I used the superglue and tape method to attach an offcut of 20mm plate to the inside.


Back in the machine and finishing pass complete.


First step for the rear is to drill and counterbore the case as well as tap the top panel. This allows me to put a couple of cap screws in to give the back of the case a bit more support from the top panel while machining the rear I/O.


Rear I/O all cut out. Couldn't blow the coolant off for the photo as there was a pool of it inside that was going to go everywhere (But mainly all over me) if I tried.


Another shot of the rear after machining.


Tried to tape up the inside and plug one of the threaded holes before bead blasting. Mainly to try and preserve the machined inside finish. Gave up a bit and had some over spray. But you can still tell it's a machined part which is what I was going for.


Last but not least, I got the DSLR to take a couple of nicer shots of it after bead blasting. Images are a bit noisy, the Canon 500D is showing it's age vs the new gear, looks a lot better on Instagram



So, now that the machining on this is done, it's time to test fit hardware to make sure all the critical dimensions are right. Then onto designing V2.

 

I've had suggestions on other forums to allow for an internal PSU. Looks like I'd need to add 25mm to the height of the case to fit the HD Plex 200w AC/DC unit in. That will be good in some cases as it allows a higher CPU cooler as well. I'm working on the design now to allow me to make it out of a 200mm square extrusion. The plan is to have a common top and bottom panel as well as fan filter element, then 2 versions of the extrusion, 1 for an external power brick and one to house the HD Plex 200w unit internally. Also working on ideas for other cases which could utilise the 200mm extrusion to make it worthwhile getting a die cut if I get to that stage.

This is the design for REV B of the external PSU version. Hopefully I'll get started on machining it shortly. Will get on with the paying work for the next couple of days while I think about any other possible tweaks I need to do to the design before I rip into another bit of billet.

 

Started machining the 2nd prototype over the weekend. This time I decided to be less wasteful and try keep the core of the billet. It was an interesting exercise in creating toolpaths and fixtures to do so, but at the end of the day, the extra time was not worth the value of material saved

Billet was rough machined closer to size using multiple facing operations. Much easier to remove material this way than side cutting.


Part way through OP1, outer walls finished and a heap of 10mm holes drilled through the billet. Again, drilling is a much more effective way of removing material than milling at that depth. Though, a thru spindle coolant carbide drill would have been a much better tool for the job than the standard HSS drill I used.


OP1 all done


Bottom of the part after OP1


Next I machined a solid top panel. This is used to support what will become a square hollow section in the vises while machining and prevent it from being crushed. It also has holes to allow me to attach the core of the billet to it so it's secure when I cut through and free it from the outer section.


Top panel in place and fixed to the core of the billet.


Everything ready to go for OP2. As there is no reference to pick up on here, the first step was to machine to rough height and take material off around the perimeter to allow the probe to touch off on the side walls machined in OP1.


After OP2.


Finally, the almost finished part, and the core of the billet saved for a future project. About 4.5kg of usable aluminium left. Which is somewhere around $60NZD worth of material. There is a witness line from where I machined from either side, largely present in the middle and not at the corners. The Material must have moved due to internal stress once it was parted off from the core. No stress for a prototype though.


I forgot to save this photo to my phone when I uploaded it to my instagram story. Saves me writing a caption for it.


This is the internal corner that was finished with the end mill above. Not too bad considering the length to diameter of the tool. Will clean up nicely with bead blasting.


All in all the whole thing would have been a lot easier if I had extrusion available at this point. Or access to a wire EDM machine. But I make do with the tools I have. Something I find myself saying often, and laughing about when I say it. I started my business with a CNC router I built in the garage at home. Now, the machine alone is worth somewhere in the region of $150k NZD and I still call it making do

 

The rest of us: Drill 4 corners and torture yourself with a hack saw.

 

And today I had to craft one of my contraptions to speed up the crafting process, with scrap materials to avoid spending too much .

Different point of views, I like the new part

 

No matter how super-duper your tools, you'll always be "making do". Metalwork is 20% making stuff, and 80% upgrading your tools or making new tools to make stuff.

Much prefer this case design --less waste of aluminium (I know the swarf gets recycled, but still....).

 

I'd be curious how long it'd actually take to do by hand with a hack saw. Maybe one day when I get to hiring staff and one of them annoys me I'll find out

Yeah, I definitely feel like a spoiled brat at times. Your work is incredible mate, shows you don't need to spend a heap of money to come up with cool stuff if you have imagination, skill and patience.

So true. When I got the machine, friends in the industry told me, you'll spend at least half as much as the machine on workholding and tooling. I didn't believe them, I quickly realised how right they were. For sure, the design that can be made from extrusion is more practical, and when it's sitting on the desk, you'd barely be able to tell them apart from the front. It's just slightly higher to accommodate the fixings at the back. I have ideas in mind for a number of variations that use the same 200mm box section as no one over here has a box section that size as a stock item so I'd need to get a die cut if I wanted to make more than a few.

 

Same here: so far my 7x16 mini-lathe has had:
- bigger chuck
- brass gibs
- nylon gears replaced for metal ones
- phenolic plastic wheels replaced with aluminium ones (the plastic ones kept losing their handles).
- knurled aluminium handle on tailstock lock
- quick change tool post
- magnetic work light.
Next is changing the bearings for tapered rollers, saddle stop and changing the saddle gibs (again) for a special tapered gibs arrangement. And this is not counting the number of cutting tools, boring tools, knurling tools, tap and die holder, live centre, steady.

The mill has had:
- blade spring to gas piston upgrade
- working lights clamps
- reverse switch
- plastic handles on quill feed replaces with knurled aluminium ones
Next is changing the bearings for tapered rollers again, table stops and a DRO setup. Of course it, too, has an arrangement of chucks, ER32 collets, end mills, face mill, work holding equipment (some self-made), parallels, V-blocks, rotary table.

And that is just for a small hobbyist setup...

 

@Nexxo - This is why we can't get an update out of you. You are the OCDest of us all.

 

I can stop anytime I want. I just don't want to, is all...

 

Now that the mad rush that is the lead up to the Christmas break is over. I've found a bit of time to get back onto this project. Unfortunately in terms of the how it's made photos, there's not a lot to show. A lot of what I've done is the same process as the first prototype, with a slightly tweaked design. So I didn't see the sense in going over the process again.

There are a couple of new elements for this design. The bottom panel, and power switch components so I'll include some photos of them being machined.

The bottom panel was first machined on one side to neaten up the stock surface as well as take an even amount of stock from both sides to reduce issues caused by internal stress. Though this plate stock is much nicer to work with than the extrusion I was previously working with. Here it is flipped over and roughed out.



And after the finishing operations.


Next up it was time to machine the power button and the switch hold down. These were both done as a single operation part with a t slot cutter to part them off leaving just a small 0.2mm tab that was cleaned up by hand afterwards. Potentially not how I'd make them if I were doing a larger qty. But for one off parts, a bit of hand finishing is much easier than setting up for a 2nd op.




Not entirely happy with the power button, it's difficult to say what exactly is going on, but it just doesn't quite have the feel I want. I have some springs coming to see if giving it a bit more resistance helps. Video below of the power button in action. Sorry for the vertical video, it was taken for an Instagram story.


After the power button I got onto the new top panel. The process for this was much the same as the previous, so no photos of this process. But I will include a video below (vertical again sorry) of it being removed from the fixture. It's a super satisfying part of the process to do.




Finally, some shots of the case as it is at the moment. This one has thrown me a bit, as I'm normally my own worst critic and see nothing but the flaws in everything I make. However, this, I actually quite like this. Next job is to get working on the filter that clips into the top panel. I'm struggling to find a mesh fine enough to make it, but will get straight onto it once I do.

 

Oh wow, that's a whole lot of up grades.

I had a mini-lathe for a few years. It was quite handy. But I realised, in a roundabout way, it was costing me money. I'd only ever use it when I was working at night to get my self out of a sticky situation. Then I'd spend 3 hours on something I could have paid the guys at the engineering firm up the front of our complex to do on their big boy lathes in about 30 mins if I just waited til the following day.

The pressure of being profitable really takes the fun out of a lot of things!

 

so cute , the grill is really splendid

 

Masterpiece

 

Wow very skilled it looks amazing

 

-Still- love it.
Skip the mesh, with space being this close, you can make an expanded fanguard for the cpu cooler instead. ...and a fine mesh will ruin your airflow.

 

Thank you all for the kind words

That would be a lot simpler. The filter is proving to be the most complex part of the whole damn thing, which is why I've left it til last

This particular case is destined to go out in the workshop by the mill. The mill is fully enclosed, but when I'm ripping a bit of aluminium apart, some chips to manage to escape, would hate for a bit of aluminium to end up sitting somewhere on the motherboard where it shouldn't Mine is probably a fairly unique use case though

 

Those cheap Chinese lathes have got better over time, but the cost cutting is still obvious. But with some inexpensive mods you can get surprisingly good performance out of them, considering the price.

Of course when you have access to the kind of professional equipment that you do, why bother with cheap hobby tools at all?