Page 1 of 1

If anyone's interested, this is the kind of designing I get paid to do

Posted: Wed Apr 11, 2018 8:42 pm
by Mechanist
I think I've already mentioned that I'm an engineer at a major manufacturer of CNC machines.
A substantial part of my work involves designing machines/parts in SolidWorks (3D CAD software).

Now, obviously I can't show these particular designs here, for reasons of IP and trade secrets - but I also have my own designs in that field, and these I can show.

Here's a particularly pertinent example.
It's the (largely incomplete) design for a highly capable (for its size, at least) bench-top "mini mill" of cast granite (aka "epoxy granite") construction, which for a time I had seriously contemplated building.
As it stands now, the design is little more than a proof of concept - in that yes, it can indeed be done, while still meeting the original specification in most respects.

StoneBox.JPG (135.61 KiB) Viewed 1946 times

Obviously, a lot of the parts have been rendered transparent for clarity - no, it's not actually made of glass :wink:

To give a sense of scale: the side of the whole cube is approx. 70cm (27") long, and the machine would've weighed approx. half a ton if built as designed. (yes, very heavy for its size - but it pretty much needs to be, if it's to be capable of roughing prehardened steel, as intended)
The working volume is 390x280x155mm (X*Y*Z).

The total cost of materials, machining, off-the-shelf parts, and R&D/NRE expenditures is in the range of $15-20k.
That's including a refrigeration system for keeping the machine's temperature stable, which helps a lot with dimensional accuracy.

Here's a better view of the Y axis components:

StoneBox - Y axis subassembly.JPG
StoneBox - Y axis subassembly.JPG (112.11 KiB) Viewed 1945 times

The X axis design is largely just a placeholder, but the overall idea was along the same lines as for the Y axis - that is, a cast-in prefabricated steel subframe to hold the rails and ballscrew bearings, which would then have the relevant surfaces machined only after casting the whole granite panel.

Re: If anyone's interested, this is the kind of designing I get paid to do

Posted: Thu Apr 12, 2018 11:40 am
by yangez93
You have very interesting job, where main of objective is creating something new (new designs). I had in past some lessons in school with free QCad Computer software to make basics technical drawing.(if everyone is interested to make some projects to learn something I recommend this :) )

I'm of course not specialist in your profession but I presume it is hard to make such project as you do, all elements must match each other, use some imagination and overall make that machine works :) It would be very interesting to make sketches of cars, transport machines, planes or ships. The devil is in the details->every screw,parts, bearings must match before it will be sold on a global scale.

Your works look that are very advanced, working full-time job I also suppose you make incredible designs :)

Re: If anyone's interested, this is the kind of designing I get paid to do

Posted: Thu Apr 12, 2018 3:40 pm
by Mechanist
What you see above isn't even remotely close to being complete - and most likely it never will be, because the whole project got dropped.
Designing it to that point took roughly a hundred hours, most of which was spent on preliminary design and validation.

Actually creating the model shown here only took around 10 hours in SolidWorks. IIRC it's around 300-400 parts, but only about 100-150 unique parts.
It'd take maybe another two hundred hours (at least!) to reach the point where there could be any hope of building a working machine based on the model. You're right, it's mostly about the finer details.

What I do at work is really quite unusual: out of >200 employees where I work, there are only 2 who do both mechanical design and machining/assembly work on a regular basis (including me). Not at all unexpectedly, my boss is the other one.
The whole operation is also constantly being scaled up. For a sense of scale - over the course of last year, my subordinates processed over $1mil worth of parts.

I've also started out with QCad. It's very simple and rudimentary, and 2D only - but hey, what's there to expect at that price tag (free) :)
But designing anything more complicated than a few parts in 2D CAD nowadays is like trying to headbutt a reinforced concrete wall into submission :roll:

Re: If anyone's interested, this is the kind of designing I get paid to do

Posted: Thu Apr 12, 2018 5:40 pm
by Arokhs Twin
Looks very interesting; we do a lot of our fabrication in house but the most complex things we get done by a local engineering company that has laser cutters etc. A desktop milling machine like yours would be ideal for some of the smaller things we build. We did look into buying something but for the small amount of things we make it would not be worth the expense. It's kinda difficult with our company director; he just does not understand engineering or electronics at all.

We have a project that is just lacking a case which has been put on the backburner (again) till the Winter but we are only looking to build about 200 which is too small to have a custom built case made. Our director insists on a triangular plastic case with various cut-outs which just isn't available off the shelf.

Re: If anyone's interested, this is the kind of designing I get paid to do

Posted: Thu Apr 12, 2018 6:23 pm
by Mechanist
The company I work at does sell a very light duty benchtop mini-mill.
Last time I asked, the price was roughly $12-15k, depending on the options. Yes, we do sell worldwide.

You mention 200 cases. At that quantity, it'd cost roughly as much to buy a mini-mill, as it would to just outsource the machining - but the latter option doesn't leave you with your own mini-mill after the job's done!

I use a slightly modified version of the BlackBox for >95% of my machining at work; and I'm currently building a second unit for that very purpose - a substantially modified, heavy duty version.

As-sold, it is only really intended for machining soft materials - ie. plastics/wood, and taking light cuts in aluminium. That doesn't stop me from using it to machine hardened steel - and with decent accuracy to boot! But it does take the skill of a master machinist to not wreck both the cutter and the whole machine in short order when doing that, since this would normally constitute severe abuse.

Nominal working area is 400x375mm, with approx. 100-120mm of useful Z-axis range (after accounting for the length of an average cutter). Several milimeters more are possible by raising the spindle in the T-slots, but then the cutter can't reach the table level due to the limited Z-axis travel available.
500x375mm working area is available as an option, but it's more expensive, and there are some limitations in that version.
The BlackBox is slightly bigger than a typical washing machine (not counting the optional base/stand), and weighs just under 200kg.

As for my own design above, I ended up cancelling that project.
While it would've been quite useful indeed, and in many ways a lot better than what I use at work, in the end I've decided that it simply isn't really worth the effort - at least in this particular case.

There are several reasons at play here:
  • for the occasional minor "odd job", I have the permission to do that at work, after hours;
  • just building a machine like that would be a massive undertaking, easily on the order of 3-5 years at a realistic rate;
  • I live in a flat, and it's rather cramped already - adding even a small mini-mill & all the required aux. equipment wouldn't improve that one bit;
  • even more space would be required while it's under construction, and it'd be a major logistical PITA;
  • instead of solving my problems, it'd invariably cause even more of them;
  • I already have another, arguably more fun plan for spending a similarly large amount of money.