Thread: Advice on Small CNC Machine for making packers

I've also run jobs where parts haven't been the same and the ONLY reason for that is not having held down securely enough. We now write our programs differently when we run lots of small parts and everything is accurate to some tiny fraction of a mm.

I can tell you the cost and specs of our machine, but it's totally irrelevant because it's an industrial machine that could probably cut a benchtop machine in half. The OP has given the maximum required working envelope as 250x150x25, milling plastic. Given some of the things we've done with simple 3-axis milling (no fancy tilting head), some basic jigs and a straight bit, I can almost guarantee that he'll be able to make whatever he needs with a basic benchtop machine. The only thing that'll need some extra thought is a vacuum table.

We use AutoCAD and proprietry software and there's no ongoing costs.

If the OP is willing to spend a bit of time learning the ropes and tweaking, they could look at getting a cheap but decent Chinese 6040 machine and bolt down a sheet of Delrin with a bit of MDF underneath as a spoil board. That would allow multiple items to be cut from the sheet before having to be replaced, making it a cheaper entry point before going to a vacuum table. Run Mach3 for the machine and design the parts in Fusion 360, get some hands-on experience with a CNC machine, make some parts, and then be in a much better position to judge the merits of investing in a more expensive machine. That's maybe $4K?

Originally Posted by RustyArc

If the OP is willing to spend a bit of time learning the ropes and tweaking, they could look at getting a cheap but decent Chinese 6040 machine and bolt down a sheet of Delrin with a bit of MDF underneath as a spoil board. That would allow multiple items to be cut from the sheet before having to be replaced, making it a cheaper entry point before going to a vacuum table. Run Mach3 for the machine and design the parts in Fusion 360, get some hands-on experience with a CNC machine, make some parts, and then be in a much better position to judge the merits of investing in a more expensive machine. That's maybe $4K?

Hi Rusty
I agree with you somewhat but

Tony, the OP, is buying for a business. I've bought equipment for a business and as a "boss" I'm paying Tony to do his current job, not to tinker with and tweak a CNC machine. As the "boss" if Tony is doing something other than his normal job, he is "costing" the business around 3x his hourly rate in lost productivity or potential income. I'd be prepared to wear that sort of cost for a few days, but not on an ongoing basis unless it was bringing-in significant additional revenue or saving the business significant costs. And in that situation I'd be redefining Tony's job to include the CNC activity. (In another post John (ElectroStream) has spent the best part of 3 months tweaking his CNC.)

So as a "boss" I'd be looking at a CNC solution that essentially works "out of the box". i.e. I'm paying someone outside my business to "tweak" a CNC machine to make what I want it to make.
3D CAD software is slightly different, but again as a "boss" I'd be balancing the cost of Tony learning to use the software vs paying a contractor to do the design work for me, with Tony and at least one other employee being trained in selecting a numerical model from the library, loading the model into the CNC controller and milling the part.

Tony also hasn't told us much about the parts apart from there being 20 to 30 different shapes / sizes, with the largest being 150 x 250 x 25, and that his intent is to produce the parts to order.
I think this is where the real CNC milling challenge lies for Tony. For me, to order doesn't mean next week or even tomorrow, it means "like within the next hour". From what I read, the 1/8" tooling used with 6040 type machines is too small to shave 8 mm off a 25 mm Delrin sheet (to produce a 17 mm thick packer) in any reasonable time. So I'm expecting that Tony would need a machine with a larger spindle, one possibly as large as 1/2".
The other unknown for this discussion is the shape of the packers -- do they include a french-fitted recess for the hinge? is the top edge rounded over or square? what radius is required for any internal corners? -- these considerations will affect the number of cutters required to mill each packer, one or many, and in turn the desirability of including auto tool change with the machine, and also if 4 or 5 axis milling is desirable. Assuming that a typical packer can be manufactured with two cutters -- one to plane the Delrin to thickness, the other to cut out the part -- auto tool changing may become a desirable rather than essential specification.

then there's OH&S requirements and work holding. Unless Tony is never sick or on leave, using the machine needs to be very straight forward. You can't supply to order if the only person who knows how to use the CNC left on a month's leave yesterday.


I believe that all of these considerations point to a something more than a desktop 6040 type machine. And whilst I agree with Elanjacobs that the machine where he works would probably cut a 6040 type desk top machine in half, unfortunately Tony might need something approaching that size to achieve his "to order" objective.

HI Guys,
sorry for any late replies, I was away for the weekend, attached is a link to an example of a product that will need a packer
Refrigeration Hardware CRH Australia

your feed back please.
Apologies if i am not allowed to post links to my web site, not looking to sell anything just asking for advice, happy to remove it if required.

No sharp internal corners...this makes life easier. A benchtop machine and a 1/4" endmill should be all that's needed.

If you can stretch the budget, maybe something like the CNCShark is worth a look https://www.carbatec.com.au/machiner...ter-clamps-bit
It's closer to $7k, but there's no tweaking or stuffing around and you have locally based tech-support and warranty.

Originally Posted by ian

I'm paying Tony to do his current job, not to tinker with and tweak a CNC machine.

Meh. As I said, it's really up to Tony as to whether he has the time or inclination to work with a cheaper solution. And as you say, the scope to develop new skills and resources for the business probably depends on the owner, and they may well have your attitude.

The 6040 machine I have has a 1500W spindle running at 24,000 RPM - if the packers are like I imagine, a relatively lightweight machine like this will do the job just fine in a reasonable amount of time (maybe 15-20 minutes?), and if needed, with the one endmill. A commercial machine will undoubtedly be faster, particularly with an auto tool changer, but a cheap entry point that builds CNC knowledge within the business at the expense of time and experimentation may be a good fit for some enterprises.

Originally Posted by elanjacobs

No sharp internal corners...this makes life easier. A benchtop machine and a 1/4" endmill should be all that's needed.

If you can stretch the budget, maybe something like the CNCShark is worth a look https://www.carbatec.com.au/machiner...ter-clamps-bit
It's closer to $7k, but there's no tweaking or stuffing around and you have locally based tech-support and warranty.

I must totally disagree here. The Shark CNC is NOT worth looking at.More than one of our members here can attest to that.
If you do a search on this site you will find some horrible comments about this machine.

Rodm had this to say about the shark

Saisay bought one

For cutting out Delrin/Acetal of the size mentioned most of the entry level Chinese CNC gantry routers will do the trick so about $1500 for a machine, Mach 3 for about $250, Cambam for both CAD and CAM about $180 and blow the dust out of an XP pc with a parallel port takes care of the capital costs with no ongoing subscription costs.

From the tone of the OP I would assume that Tony has been tasked with coming up with a solution to a problem and that actually running the machine would be taken care of in the normal run of work in the production department where someone with experience in machining or cnc even would have their job title lengthened.

I don't think that entry level printers will do the job as spacers tend to have a lot of crush applied to them and would likely split and thick spacers take up a lot of filament at $50 a kilo and take a long time.

Acetal is great to work with, it can be cut fast without melting, run through the thicknesser to speed things up and tools last a long time and if I have lots of parts to make I just cut through halfway and then while the next sheet is being cut I roughly cut out the parts on the bandsaw and run them around the table router with a flush trimmer bit and this speeds things up as there is no need for holding tabs that have to be cleaned up anyway. I have a number of routers sunk into my table with a flush bit, couple of round over bits and an arissing bit

Most likely, this solution to a problem will become part of the tool kit for resolving future problems with product development then something like I have listed would be a cost effective way of entering the field of CNC and as the company is in manufacture then there is most likely a draughtsperson or designer on hand already. These parts would most likely be just 2D but Cambam can do 2.5 so pockets can be machined as well as 3D but 3D is really slow.

So I think that to take a 2D drawing of the packer and create tool paths and gcode and then to create the parts would take only a couple of hours for the someone who drew the part in the first place on the first go: once familiar with the CAM side, it should be possible to be making those sort of simple parts at a profit by the end of the second day.

Good luck with your project.