What are things about a Desktop CNC Machine you would like to see?

Hello everyone!

In the past few months, I (student mechanical engineering) have been looking over the internet in a search for a functional CNC machine that could fit in a fairly small place. After finding barely anything useful or affordable, I decided to make a small CNC Mill myself. As it seems like I am not the only one trying to find a capable machine, I would like to create a new product: A desktop CNC Mill designed with great tolerances and repeatability in mind. The machine will be designed with strength and rigidity in mind.

To gain more insight for this product I created the form below and would really appreciate your feedback!

Desktop CNC Mill

Thanks a lot and have a great day!

Hi Matthias:
There are DOZENS of offerings in the desktop and small home shop toy CNC market; everything from bare bones "Maker" kits to fully functional machines.
If you want a decent machine you don't need to make one yourself...you just have to find the money to buy one.

If you really really want to make and sell a new product, you have a hard road ahead of you.

I have a Defiance VTX-1 benchtop 3 axis CNC mill.
Light Machines Inc made comparable models.
They cost anywhere from 20 to 60 thousand dollars new, and were capable enough to find homes in proper machine shops who were making the kinds of parts a little machine like this can do well.

Neither company exists anymore...they lasted maybe 10 years and died...not enough profit in it to make it worthwhile in the long term.

Making the machine is the least of it...making it PROFITABLY and finding buyers for it is the real problem.

So while I hate to stomp on your dream...

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

implmex said:

Hi Matthias:
There are DOZENS of offerings in the desktop and small home shop toy CNC market; everything from bare bones "Maker" kits to fully functional machines.
If you want a decent machine you don't need to make one yourself...you just have to find the money to buy one.

If you really really want to make and sell a new product, you have a hard road ahead of you.

I have a Defiance VTX-1 benchtop 3 axis CNC mill.
Light Machines Inc made comparable models.
They cost anywhere from 20 to 60 thousand dollars new, and were capable enough to find homes in proper machine shops who were making the kinds of parts a little machine like this can do well.

Neither company exists anymore...they lasted maybe 10 years and died...not enough profit in it to make it worthwhile in the long term.

Making the machine is the least of it...making it PROFITABLY and finding buyers for it is the real problem.

So while I hate to stomp on your dream...

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

Click to expand...

I always wanted one of those light machines mills. They were the ultimate benchtop machine 20 years ago. I settled for a Taig instead and thats what got me into this stuff.

I filled out his survey, but what I could really use for my small medical device parts would be a Haas CM-1 but with significantly more X and Y, and ideally high pressure TSC.

MatthiasBeek said:

After finding barely anything useful or affordable, I decided to make a small CNC Mill myself.

Click to expand...

Define usefull or affordable?

MatthiasBeek said:

CNC machine that could fit in a fairly small place.

A desktop CNC Mill designed with great tolerances and repeatability in mind. The machine will be designed with strength and rigidity in mind.

Click to expand...

I'm sure you could make what you require, but your going to have to compromise somewhere.

Want to make it affordable? well then it's probably going to be made from aluminum, rolled leadscrews, and will use steppers, therefore will likely lack rigidity and repaeatility.

Want rigidity and repeatibility, so will likely require Cast Iron castings, AC servos, ground leadscrews, therefore not affordable.

Are you going to make your own spindle? Is it going to require a tool changer?

-------------------------------------------------------------

Figure out your budget, go to Alibaba and look at prices for all the COTS components, look at all in one controls from GSK, LNC or Masso. (don't waste your time with Mach3/4 or Linuxcnc) After you have prices for all the sourced parts, what's left over to buy and machine all the components?

-------------------------------------------------------------

And don't make the mistake that was made with Swissmak. I think the real mistake he made was listening to the peanut gallery. He started with what could have been a simple machine, asked the peanut gallery for suggestions, and the know nothing hobbyists (assuming it was their advice he followed) who said, "it would be nice if it could do this, or that, or the other, had these axis, or this spindle taper etc etc."

The end result being a long way from the original intended design and market.

-------------------------------------------------------------

Buy something like a Tormach, and see if you can make it better and cheaper. If you've never designed a machine before I wouldn't start with a clean sheet of paper. Take a similar machine apart, see how it works, there's probably a lot you can improve on, and a lot you can learn from the design.

The story is that Adrian De Caussian of Fadal used to help out Haas occasionally, that was until somebody went around to Haas one day and found the Haas employees had disemmbled a Fadal to see how it ticked. That was when Fadal realised Haas was getting into the VMC market.

-------------------------------------------------------------

Good luck

triumph406 said:

Figure out your budget, go to Alibaba and look at prices for all the COTS components, look at all in one controls from GSK, LNC or Masso. (don't waste your time with Mach3/4 or Linuxcnc) After you have prices for all the sourced parts, what's left over to buy and machine all the components?

Click to expand...

This much is spot on. To use a car analogy, you aren't going to build a car for less than what it costs to buy one and have comparable results; economics of scale weigh heavily. Tesla made it work by first making a product that's better in many measurable ways than the competition, at least for certain use cases, pricing it high, and then over time bringing the cost down. So look at the competition and decide what you want to improve on, and be sure to charge enough to make a profit. Don't forget to figure in costs for support and replacing the occasional lemon.

IMO, how the Swissmak failed was in NOT taking advice right off, and insisting that how he had planned to do it from the beginning was better than any suggestion, until he was too deep in and changing direction to do it right required basically starting over.

implmex said:

....
Neither company exists anymore...they lasted maybe 10 years and died...not enough profit in it to make it worthwhile in the long term.
Making the machine is the least of it...making it PROFITABLY and finding buyers for it is the real problem.

Click to expand...

I did a whole bunch of work and such thinking to get into this... I wish you luck but this a very brutal market with very few survivors.
It looks easy and a fat pig from the outside.
Take all the costs of bought and made stuff needed plus labor, double it and this sell price will still not be enough to survive on.
Bob

Thats the dirty little secret of the "maker" scene. The makers dont know what to make, but want to seem like they're doing something, so they waste a bunch of time making barely useful machines that give them yet another excuse to do nothing.

markp said:

Thats the dirty little secret of the "maker" scene. The makers dont know what to make, but want to seem like they're doing something, so they waste a bunch of time making barely useful machines that give them yet another excuse to do nothing.

Click to expand...

Yabbut their hopefully learning something usefull along the way, whether it succeds or not.

And regardless if it was succesful or not it got them off the couch.

markp said:

so they waste a bunch of time making barely useful machines that give them yet another excuse to do nothing.

Click to expand...

So how is it a waste of time?. Not sure I understand that.

markp said:

Thats the dirty little secret of the "maker" scene. The makers dont know what to make, but want to seem like they're doing something, so they waste a bunch of time making barely useful machines that give them yet another excuse to do nothing.

Click to expand...

There's a small maker space near me. A group of guys have a 800sq foot space, they all have different skills, they share the machinery, they work together on projects, they have fun, they have developed what's likley to be lifelong friendships. I wish makerspaces had existed, or I knew about them 20-30 years ago.

markp said:

Thats the dirty little secret of the "maker" scene. The makers dont know what to make, but want to seem like they're doing something, so they waste a bunch of time making barely useful machines that give them yet another excuse to do nothing.

Click to expand...

That seems rather harsh. Perhaps it is about learning or doing something you never have done.
One could say the same about people who are build car crazy or so many other endeavors.
I embrace such even though cringing inside at things. I like the "maker" scene off track or not.
I will let people use machines to do or make things off time that I know darn well are not going to work.
Remember "brown gas". If some of your people on this would you allow free access to your machine tools off the clock. Would you eat the tooling costs?
How about chasing a patent on a new cylinder head/valve design?

So low cost table top mill...go for it.
What is unknown to a maker or machinist is the business side of a company.
Bob

To build this small desk top mill you will need a larger floor mounted mill. Yes? Or make it with files and a hacksaw? OK, add right angle grinder.

triumph406 said:

Want rigidity and repeatibility, so will likely require Cast Iron castings, AC servos, ground leadscrews, therefore not affordable.

Click to expand...

I'd make the base a slab of granite. Pretty cheap, stable as hell, definitely good damping properties. And there's inexpensive linear ways.

At least a few pieces could be made better than what's commonly available and still be inexpensive.

Scruffy887 said:

OK, add right angle grinder.

Click to expand...

Better use a Bosch, the HF versions won't be accurate enough.

Desktop cnc mill.
Many here know things so what are best ideas?
Price point is huge, tradeoffs will be needed.
Screws, slides, base design and material, overall geometry and abbe, bearings, motor type and coupling, feedback, software?

Who rules this market space and why? If wanting to enter understand that.
Bob

I think I'd find a desk mounted to real machine more useful.

Fundamentally, it's a scaling issue. The control is roughly a fixed cost whether it's driving a 20lb machine or a 20,000lb machine. It doesn't matter if that cost is $80 (raspberry pi driving bargain basement steppers open loop) or $100,000 (high end Heidenhain with options) for a given performance expectation you need a certain amount of control. Going small doesn;t make it cheaper.

Going small means you'll want a high speed spindle. A high speed spindle often costs more than a higher power low speed spindle. At the very least it doesn't really save anything.

You'll need the same quality screws and rails whether you're big or small. Trucks cost, rails are cheap. You only save yourself some rail by being smaller.

Smaller means you need to fit the same amount of mechanical shit in less space. That takes more engineering and design time, not less. Things like interlocks, doors, handles, etc. You have the same number of moving parts, which are the expensive bit. Hardware costs. Sheet steel is a rounding error in the cost of these things, and that's all you save making it tiny.

Castings would be a tad easier to machine making a small machine, I'll grant you that.

Are you developing something here? If you have a better way to make a useful and affordable toolchanger, say, then great. Make that and sell that. But don't delude yourself that you can bodge together some bargain basement off the shelf parts, call it a machine, and end up cheaper or more functional than the scores of machine assemblers in China and elsewhere that are doing the same thing, with the same parts. They have experience and economy of scale on their side.

Beyond that, people with champagne taste, tap water budgets and no space are shitty customers to have.

Comatose said:

Trucks cost, rails are cheap.

Click to expand...

This may be true, but for the sake of argument I feel confident I could make my own trucks. They might be at the lower end of acceptable accuracy, but for most practical purposes will still suffice for a "regular" grade of mill (not a Roku Roku).

At the same time, I don't think I could make my own rail. The form requirements, parallelism, straightness, etc. would all be beyond my capacity at this time.

And to grab prices from McM, their generic smaller profile rails are $.40/mm, which would be ~$560 for just a pair of 28" rails for (say) 16" of X travel. Yes, I know Hiwin or cheaper options exist.

Beyond that, people with champagne taste, tap water budgets and no space are shitty customers to have.

Click to expand...

This is a concise truth.

Confused on rails and trucks.