What's new
What's new

Looking for advice on a new CNC mill for arts/prototype work in shared shop space

johnw188

Plastic
Joined
Oct 19, 2021
Hey all,

I've been lurking here for a couple of months now reading as much as I can, and figured I'd reached the point where I had enough context to ask useful questions. The quick summary is that I'm looking to buy a mill for making mostly one off parts and potentially some small production run work for a couple products that my friends and I have been working on.

Longer background - I studied mechanical engineering at Cal Poly SLO, got my degree, passed my first licensing exam, and then moved into software for the past 13 years. I learned how to cast/weld/machine in the school shops, working on world war 2 era manual lathes and similarly ancient mills. A few years back I started doing engineering work again with a crew of folks building large scale art installations for places like Burning Man, and I've been gradually putting more and more effort into my mechanical design and manufacturing work. I'm looking to purchase a CNC mill to build parts for various projects, generally in the range of 10-20 for the larger runs and one offs for the others. I'm a little constrained on shop space so footprint matters to me.

Jobs would range from machining nodes for assembling dome structures, building smaller parts for automated telescopes, and some toys/products that I'm looking to sell. Basically generalist prototype type stuff. I want to be able to get a good surface finish on things, but I think that's more down to the step size than it is the specific tolerances of the mill itself. I've never worked on five axis stuff before but the CAM work actually looks pretty straightforward from what I've seen and I could see it saving a bunch of effort building fixtures for one off parts. I've been looking around at some of the Haas machines and the Brother Speedios, I saw that some folks used the Brother mills for similar sorts of work with good success.

Does anyone have any thoughts on mills that I should look into for this application?

I also wanted to ask about mills in environments where there's multiple types of work going on. Our shop tends to oscillate between woodwork and metalwork, and I've never dealt with larger machines in places where there's occasionally sawdust in the air. I'm obviously not going to be machining wood, I'm just wondering whether its a "don't run the mill when there's woodwork going on in the shop", "don't run the mill when there's woodwork in the shop and make sure you carefully clean it before starting again", "cover the mill with plastic before doing woodwork so the dust is kept out fully", or "this is a bad idea find a different space".
 
Take this with a grain of salt, as it is just my opinion, but I cannot imagine that a 5-axis machine would be justifiable price wise for the type of work you are describing. Especially if you are looking at buying a new machine. If I were in your shoes, I would be be looking for a used HAAS, Fadal, or similar machine. There is nothing wrong with making and using fixtures either. It is good good mental exercise if nothing else and you can do a lot with a decent 3-axis machine and a little know how. It's what I do every day.

One thing I can say with certainty. Do NOT buy a Fryer. We have a VB-40 at work and it has been a headache since we first got it several years back.

On a side note, when you are considering CAD/CAM software, I would take a look at Fusion 360. Don't know how much FEA you do but, but if that is a consideration, I like Fusions FEA environment fairly well. We used ANSYS in school and I would often model assignments in Fusion, do the FEA there just for the sake of comparison, and then export the models to ANSYS to finish the assignment.
 
Brother Speedios, I saw that some folks used the Brother mills for similar sorts of work with good success.

If you're going to buy new, this would be my first choice. Sounds like space is a concern, and possibly power as well. These run on 30A 3PH, and are quiet. S500 is a nice small footprint. Roughly two additional feet in width will get you an S700 for not much more in cost though. The extra 8" (200MM actually), in X is worth squeezing in if you can. Especially worth it if you plan on doing any rotary work. The X2 machines have quite a few upgrades for keeping crap out of the fans, and there are a lot of aftermarket options for air filters for fans/motors from places like MariTool.

Speedios are obviously fast while they're running, but they're quick and easy to setup as well, especially if you get a tool setter and spindle probe. Out of all the machines at our shop (Mori Seiki/DMG Mori, Okuma & Brother), the Brothers are my favorite to setup and run.

If you don't mind me asking, where in CA are you located?
 
Take this with a grain of salt, as it is just my opinion, but I cannot imagine that a 5-axis machine would be justifiable price wise for the type of work you are describing. Especially if you are looking at buying a new machine. If I were in your shoes, I would be be looking for a used HAAS, Fadal, or similar machine. There is nothing wrong with making and using fixtures either. It is good good mental exercise if nothing else and you can do a lot with a decent 3-axis machine and a little know how. It's what I do every day.

One thing I can say with certainty. Do NOT buy a Fryer. We have a VB-40 at work and it has been a headache since we first got it several years back.

On a side note, when you are considering CAD/CAM software, I would take a look at Fusion 360. Don't know how much FEA you do but, but if that is a consideration, I like Fusions FEA environment fairly well. We used ANSYS in school and I would often model assignments in Fusion, do the FEA there just for the sake of comparison, and then export the models to ANSYS to finish the assignment.

Thanks for the feedback, I really appreciate it. I think your point on the five axis stuff being unnecessary is pretty spot on, it's something I was wavering on. I actually really enjoy the process of making fixtures, I felt like industrial engineers got to have all the fun anyways.

Are the FEA tools in CAD software any better these days? I know when I was studying we called the solidworks FEA view "manager mode", because it made pretty pictures which made managers happy but weren't actually super valuable for actually building things because you had little control of the mesh. I'm wondering if computing power now just allows them to blow through those issues through brute force.

I've been using Fusion360 for CAD/CAM and it's been really awesome.
 
If you're going to buy new, this would be my first choice. Sounds like space is a concern, and possibly power as well. These run on 30A 3PH, and are quiet. S500 is a nice small footprint. Roughly two additional feet in width will get you an S700 for not much more in cost though. The extra 8" (200MM actually), in X is worth squeezing in if you can. Especially worth it if you plan on doing any rotary work. The X2 machines have quite a few upgrades for keeping crap out of the fans, and there are a lot of aftermarket options for air filters for fans/motors from places like MariTool.

Speedios are obviously fast while they're running, but they're quick and easy to setup as well, especially if you get a tool setter and spindle probe. Out of all the machines at our shop (Mori Seiki/DMG Mori, Okuma & Brother), the Brothers are my favorite to setup and run.

If you don't mind me asking, where in CA are you located?

I'm in San Francisco, thus the shared shop space :).
 
What is the range of part size you're looking at? And budget?

I love Speedios, that all we have now, becuase our work fits their sweet spot very well. That said, for what you are describing they may not be right, and you might actually be better served with a tool room mill. Reasoning:

1. Limited Z travel, about 12". Totally a non issue for small parts, but for large artsy stuff it may bite you.

2. 30 taper does limit how far you want to hang out tools, again for larger parts with maybe a deep bore or such.

3. The speed isn't much of a factor on one off custom parts. The acceleration and chip-to-chip time is meaningless compared to how long it will take to design and fixture this stuff.

You might be that rare person who should get a TM-2 or equivelent machine.
 
1. Limited Z travel, about 12". Totally a non issue for small parts,

Just for clarification on this point, the typical Speedio spindle face can travel from about 7" above the table to about 19" above the table. That puts the envelope in a usable place. The R650X2 40 Tool machines have about 17" travel.
 
I run a shop pretty similar to what you are describing. 100% of everything we do is prototype, quick turn, both wood and metals.

The shop is divided physically for the reason you describe. It isn’t a sealed barrier but there are walls and doors separating the area. Do not skip this part. You will screw the machines up if you’re running wood and meta in the same shop. Ask me how I know! In a perfect world, you would have a hallway acting as a vestibule between the two areas so that any particulate has to travel through 2 doors and some distance. Our method works well enough.

If you’re going to be doing a lot of wood routing, I would suggest investing in a proper wood router. CR Onsrud comes to mind. Those machines are designed for it. Sure a VMC can cut wood but that’s like hammering a nail with a front end loader. Not the right tool.

As far as machines are concerned, something like…brother, Hurco, Haas (maybe), potentially a Mazak EZ or a Doosan DNM would work. Those are going to be not crazy expensive. A good new machine tool with proper speeds and feeds and good tooling will give you good finish quality. If it isn’t,‘it’s probably not the machine and it is probably you.

I don’t necessarily agree with your assessment of 5 axis programming. It’s insanely complex. Or at least it can be. And probably will be. There is a good bit of skill and time that goes into creating a good program, 5 axis or otherwise. 3+2 machining is a lot like 3x programming, just with rotations so that is pretty straightforward. 75+% of the 5 ax work can and should be done 3+2 or 4+1. Or at the very least, the roughing cycles.

I’m usually the one to encourage everybody and their mom to buy a 5 axis machine. But that is with a caveat - you need to utilize it. If you run it like a 3x it will lose you money, not make you money. Gotta de-program your brain from what you know as far as VMCs and throw it out the window. Well, many things will still apply but it is a totally different animal.

You have to think of this as a holistic investment. What do I mean?

Let’s say you buy a CNC lathe and 16x30 VMC. Sweet, congrats! But the cost of ownership is far higher than you might think. Workholding - invest about $10,000 for quality Workholding right out the gate. Trust me. Not just 6invise. Vacuum plate, maybe some modularity. Modular is very important with high mix environments. Quick changeover is critical. Next, we have tooling. Plan to invest $20,000+ in quality tooling. 1 VMC of that size needs 50 tool holders to be well equipped. Otherwise you’re constantly swapping holders and tools. Ask me how I know! Basically you need to add $10-30K to the price for the machine to get it all properly equipped.
 
Just for clarification on this point, the typical Speedio spindle face can travel from about 7" above the table to about 19" above the table. That puts the envelope in a usable place. The R650X2 40 Tool machines have about 17" travel.

For sure, the Speedio's are extremely well thought out. Like I said, our VMC's are Brothers and I couldn't be happier. If your parts fit their envelope they can't be beat, I'm just not sure the OP's parts do indeed fit. Need him to fill us in on the sizes of these parts.
 
Maybe "Peeps" here should give OP some ball park prices for what he (JohnW188 ) wants to get into.

esp. as most folks outside of manufacturing usually get this wrong by at least a factor 10.

Maybe OP could puke-up some rough sketches, broad tolerance (ranges) in parts and materials.

For some geometries fixtures can be cripplingly slow and sometimes - "Just the ticket".
 
Workholding - invest about $10,000 for quality Workholding right out the gate. Trust me. Not just 6invise. Vacuum plate, maybe some modularity. Modular is very important with high mix environments. Quick changeover is critical. Next, we have tooling. Plan to invest $20,000+ in quality tooling. 1 VMC of that size needs 50 tool holders to be well equipped. Otherwise you’re constantly swapping holders and tools. Ask me how I know! Basically you need to add $10-30K to the price for the machine to get it all properly equipped.

This. Work holding in particular adds up really fast.

Also 50 toolholders sounded like a crazy number. Surely I don't have that many? I went and counted: 48! 3 facemill arbors, 14 ER, 5 sidelocks, 3 hydraulic, 8 high-precision collet, 10 shrink-fit, 3 integral shank tools, a deburring brush, and a chipfan. Some of that is extraneous but sometimes I really wish I had just one more collet chuck or another slim-nose shrink fit or... it never stops.
 
I'm just barely getting started with 18 holders about a month in. My goal is to keep one of each frequently used cutter in a holder, plus have some spare holders to setup replacements for when they get dull.

I paid about $110k for a CM-1 with a fourth on it, wired for five, with a few options. That's not that much more than a minimum for a new machine, unless you're doing dead simple 3 axis only, in which case you can do around $60-$70k.
 
Yes for a plain jane 3x milling machine you can expect to spend anywhere from $50-120K for the machine with whatever options you want. All prices listed here are for new machines.

Haas has the pricing listed on the website for a VM2. THe new Doosan cheaper ones have the prices listed as well, $70K I think? 16x30 Hurco is going to run about $110K delivered and installed if you get the bells and whistles. Not sure about Brother Speedios but I would imagine it is a 6 figure investment.

Plus you'll need to consider auxillary items such as an air compressor. A good air compressor will be several thousand and up. People spend ungodly amounts on air.

Also cutting fluids - not a killer expense but something to be mindful of.

If you are actually planning to get into the multiaxis world, expect to spend $200,000 at minimum just for the machine. OK OK there are options in the $150K range yes but they are smaller machines. Hurco VM10Ui (8" platter) is about $175K delivered and installed.
 
YHaas has the pricing listed on the website for a VM2. THe new Doosan cheaper ones have the prices listed as well, $70K I think? 16x30 Hurco is going to run about $110K delivered and installed if you get the bells and whistles. Not sure about Brother Speedios but I would imagine it is a 6 figure investment.

Sorry to go off topic, but are you finding Doosan list prices online? Where?

Speedios aren't that far off Haas prices for the base machine, but options are more expensive (4th, probing, chip conveyor, etc.).
 
I felt like industrial engineers got to have all the fun anyways.

Uhh, no.
We machinists get to have all of the fun cutting the steel.
And wtf is an industrial engineer anyway? It seems like people find a way to add the engineer title to anything these days.
At this moment I'm a typing engineer, then I will become a programming engineer, while being a foreman engineer.
 
Uhh, no.
We machinists get to have all of the fun cutting the steel.
And wtf is an industrial engineer anyway? It seems like people find a way to add the engineer title to anything these days.
At this moment I'm a typing engineer, then I will become a programming engineer, while being a foreman engineer.

Yah,

and if you get really really GOOD at all that ^^^ you can become a "Sales " - "Engineer" - :willy_nilly:
 
...snip...

Ask me how I know! Basically you need to add $10-30K to the price for the machine to get it all properly equipped.

Don't forget metrology, fluids, air, rigging/transport, electrical, support tools, CAD/CAM...Add another 10-20k there. You're now at -$50,000 and don't even have a mill yet.

I'd be highly inclined to buy a cheapo CNC mill to get started. I chose a Fadal but there are definitely other options.
 
Yah,

and if you get really really GOOD at all that ^^^ you can become a "Sales " - "Engineer" - :willy_nilly:

Sweetwater (an audio sales site) assigns you a “sales engineer”. I always cringed haha. I’ve also seen marketing engineers somewhere else.
 
Thanks for the feedback, I really appreciate it. I think your point on the five axis stuff being unnecessary is pretty spot on, it's something I was wavering on. I actually really enjoy the process of making fixtures, I felt like industrial engineers got to have all the fun anyways.

Are the FEA tools in CAD software any better these days? I know when I was studying we called the solidworks FEA view "manager mode", because it made pretty pictures which made managers happy but weren't actually super valuable for actually building things because you had little control of the mesh. I'm wondering if computing power now just allows them to blow through those issues through brute force.

I've been using Fusion360 for CAD/CAM and it's been really awesome.

I really like the FEA capabilities of Fusion. It seems pretty straight forward and intuitive, although I have never done anything super complicated with it. It has pretty good flexibility with meshing, like refinement in specific areas, overall sizing, etc. All the times that I compared the results with ANSYS, they were pretty much spot on.

I also really really like the fact that you can choose to do the entire FEA in the cloud so that it doesn't turn your computer into a brick while its running.
 








 
Back
Top