[Inor]

As I mentioned in my intro post I am only a month or two away from finishing a big construction project at my home and am starting to look for my next adventure. I have been looking at building a CNC router for a couple years, watching YouTubes, reading blogs, reading forums (such as RF), etc.

When I started looking through sources for parts to build one from scratch the choices quickly became overwhelming. Since I have absolutely ZERO experience with CNC anything, I was not really sure even where to start. Then I discovered a WorkBee kit from Bulkman (dot) com [This site won't let me post links yet - GRRR!]

I went through their configuration questions:

Size: 1000 X 1500
Motors: 4 - Nema 23 High Torque
Controller: Mach3
Air cooled spindle

The price came out to about $1000, which is less than what I added up the parts when sourcing the parts and building it from scratch.

So, the first question is: Is that even a reasonable setup for my first foray into CNC-land?

Secondly, on software...

My plan is to just use Sketchup as the CAD software. I found several plug-ins for the free version of Sketchup that claim they can output file formats compatible with most popular CAM software. I even found one plug-in that claims it can generate G-Code directly. (I have no idea if any of these plug-ins work as advertised, but I have to start somewhere and I at least know the basics of Sketchup.)

I did find a web site: makercam (dot) com that claims to be able to take Sketchup drawings (output through one the aforementioned plug-ins) and create G-Code.

Then, of course, I would have to purchase the Mach3 software.

So, my second question: Does that software stack even seem like it would work or am I completely off base?

I do not want to drop $1000 on a machine only to find out I need to spend a bunch more on software that I would need to learn before I could even start to use it.

Thanks in advance for your advice!

 

[difalkner]

First question, Mike - what are your expectations? Are you going to be doing deep 3D carvings that take hours on end? How rigid do you need the machine to be? How much room do you have and what power sources are available? There are a ton more questions but these will get us all started.

David

 

[Inor]

First question, Mike - what are your expectations? Are you going to be doing deep 3D carvings that take hours on end? How rigid do you need the machine to be? How much room do you have and what power sources are available? There are a ton more questions but these will get us all started.

David

Hey David

I doubt I will be doing any deep carvings. I am thinking of mainly using it for cutting out complex parts for furniture - think things like the top rails for a Pennsylvania chest-on-chest dresser. I also have an idea for a Massachusetts chest with scalloped drawer fronts that it might be useful for.

In terms of storage, my shop is a 3 car garage (that has never had a car in it). I do not have a great deal of extra room, but I do have a shipping container for auxiliary storage.

For power, I have plenty of 110 and 220. I designed the garage specifically as a wood shop so I have 4 dedicated circuits and more outlets than I need.

 

[ger21]

Then, of course, I would have to purchase the Mach3 software.

So, my second question: Does that software stack even seem like it would work or am I completely off base?

I do not want to drop $1000 on a machine only to find out I need to spend a bunch more on software that I would need to learn before I could even start to use it.

1) You can use Mach3 in demo mode for free, for up to 500 lines of g-code. Most simple 2D parts would be much less than 500 lines of g-code.
If you do decide to pay for it, I highly recommend against it. Mach3 is obsolete, with development having stopped over 5 years ago. I'd recommend UCCNC. It's cheaper than Mach3, for both hardware and software. They are currently offering a free license with the purchase of their UC100 controller.
UC100 USB motion controller + free UCCNC software license key - CNCdrive - webshop

2) I'm not a fan of Sketchup for CAD, but plenty of people use it. Fusion 360 is free, has both CAD and CAM, and is far more powerful.
There are also plenty of free CAM programs out there.

3) Be aware that $1000 is very little money for a CNC. It would be like buying as new table saw for $40. A decent quality 4'x4' machine would normally be in the $4000-$6000 range.
As David says, your expectations will dictate what you think of the machine.

The machine will definitely cut your parts, but you'll need to make multiple, light passes. $1000 is small price to pay to get a feel for it. If you like it, chances are you'll upgrade to something better.

 

[4DThinker]

If furniture parts are your end goal, make sure your CNC/setup has allowance for vertical/angled clamping beneath the bed. We use 3 Probotix CNCs (probotix.com) with open bed in our fine furniture lab to handle everything from complex joinery to inlays and 3D carved surfaces. A rotary axis on one lets us to simple tapered legs or complex cabriole versions.

4D

 

[gdonham1]

If you are looking at $1000 dollar CNC Router you are going to get frustrated with a POC. Many of the folks on this forum use industrial strength CNC machines and their prospective is from that vantage point. I am a hobbyist CNC Router user and have a Shapeoko XXL. It has a 32x32x3" cutting area and does what my pocketbook and skill level are comfortable with. The Shapeoko and XCarve are both belt driven machines but make very good hobby level projects. If you want to go into furniture manufacturing then you will likely spend more like $10,000.00 for that level of dependability and repeatability. But which ever level of machine you choose consider this:

I have a Shark Pro HD that I hate. The Shark has propriety licenses and does not come with any CAD/CAM software. You can buy Vetric for a discount but you have to buy it.

I bought a used Shapeoko XXL and just love it. The Carbide Create and Carbide Motion software is free and is quite useful for what I do. As you get more advanced you may want to move up to different software but most people can use the CC and CM forever.

I have given this advise to new or prospective buyers.

Have a purpose in mind for the machine. If you buy one and then try to find a purpose you may never really get much use out of it.
The CNC Machine is only the first 1/3 of your cost. The second third is investing in bits, tools to prep and finish your projects.
The software to feed your machine if you choose to go with pay software can be 1/3 of your overall budget.
Most importantly is the time it will take to learn how to effectively use and product something useful with the CNC Router. There is a learning curve that is quite steep if you do not have a background in design. There is also a steep learning curve in learning how to operate the CNC Router itself.

i do not want to discourage you but if you are not ready to make a financial and time commitment this hobby may not be for you. There are countless CNC Routers sitting in garages and basements languishing in obscurity because the commitment to learn how to use the machine was not there.

Lastly before I retired I was in the high tech industry. Many of my colleges would chase the latest and greatest usually to find themselves on the bleeding edge of technology. i emphasize the Bleeding part of my last statement. Later in my career I did not even want to customize the desktop or spend countless hours finding cute little applications to waste my time on. I just wanted to turn my computer on, use it then turn it off at the end of the day. Maybe that is why I like the Shapeoko, you just turn it on, create and turn it off. There is maintenance to do on any machine but my Shapeoko is like the Energizer Bunny, it just keeps going and going.

So please consider the above advise and decide what you really want to do with a CNC before you make a large financial investment. But if you go cheap I promise that you will not like the results. Many of the cheap Chinese kits are just exercises in frustration and disappointment.

 

[Inor]

If you are looking at $1000 dollar CNC Router you are going to get frustrated with a POC. Many of the folks on this forum use industrial strength CNC machines and their prospective is from that vantage point. I am a hobbyist CNC Router user and have a Shapeoko XXL. It has a 32x32x3" cutting area and does what my pocketbook and skill level are comfortable with. The Shapeoko and XCarve are both belt driven machines but make very good hobby level projects. If you want to go into furniture manufacturing then you will likely spend more like $10,000.00 for that level of dependability and repeatability. But which ever level of machine you choose consider this:

I have a Shark Pro HD that I hate. The Shark has propriety licenses and does not come with any CAD/CAM software. You can buy Vetric for a discount but you have to buy it.

I bought a used Shapeoko XXL and just love it. The Carbide Create and Carbide Motion software is free and is quite useful for what I do. As you get more advanced you may want to move up to different software but most people can use the CC and CM forever.

I have given this advise to new or prospective buyers.

Have a purpose in mind for the machine. If you buy one and then try to find a purpose you may never really get much use out of it.
The CNC Machine is only the first 1/3 of your cost. The second third is investing in bits, tools to prep and finish your projects.
The software to feed your machine if you choose to go with pay software can be 1/3 of your overall budget.
Most importantly is the time it will take to learn how to effectively use and product something useful with the CNC Router. There is a learning curve that is quite steep if you do not have a background in design. There is also a steep learning curve in learning how to operate the CNC Router itself.

i do not want to discourage you but if you are not ready to make a financial and time commitment this hobby may not be for you. There are countless CNC Routers sitting in garages and basements languishing in obscurity because the commitment to learn how to use the machine was not there.

Lastly before I retired I was in the high tech industry. Many of my colleges would chase the latest and greatest usually to find themselves on the bleeding edge of technology. i emphasize the Bleeding part of my last statement. Later in my career I did not even want to customize the desktop or spend countless hours finding cute little applications to waste my time on. I just wanted to turn my computer on, use it then turn it off at the end of the day. Maybe that is why I like the Shapeoko, you just turn it on, create and turn it off. There is maintenance to do on any machine but my Shapeoko is like the Energizer Bunny, it just keeps going and going.

So please consider the above advise and decide what you really want to do with a CNC before you make a large financial investment. But if you go cheap I promise that you will not like the results. Many of the cheap Chinese kits are just exercises in frustration and disappointment.

Excellent advice Guy! Thanks! And I have been thinking on that. I too, have made my living in the IT world for the last 33 years. I have been asking myself if I really want to spend my limited woodworking time sitting in front of a computer or would it be better to spend it in the shop making sawdust. On the flipside, I do still travel quite a bit for work, so that might be something woodworking related to occupy my time while sitting in a hotel room in BFE. I don't know...

Like I said, I am not sold on it as a great thing yet. But it sure sounds like a fun project to build.

 

[ger21]

For a lot of people, CNC is the hobby. For others, it's just another tool in their woodworking shop.

If you don't have any specific uses for it, then there's a good chance it'll sit unused. If you don't have any CAD/CAM experience, there may be a significant learning curve as well.

 

[ButchCassidy]

My Son and I are in the same place !! We are looking into getting a CNC. We are looking at the Shapeoko 3XXL and have searched info everywhere. We also want a Laser to complete what we want to do as well. You have probably just given us the final recommendation for the Shapeoko. I have the experience in the woodworking part, with a complete top of the line equipped shop. My son has no CNC experience, but has software experience from being in the large Printing Computer industries. We " think" in will be about $3500.00 to $4,000.00 to get set up and that includes the Laser.From there you can spend much more in software.

 

[ger21]

I have the experience in the woodworking part, with a complete top of the line equipped shop.

For someone with a well equipped shop, with high end tools, I would expect you to quickly outgrow a ShapeOKO. But it's a good starter machine.

 

[difalkner]

My Son and I are in the same place !! We are looking into getting a CNC. We are looking at the Shapeoko 3XXL and have searched info everywhere. We also want a Laser to complete what we want to do as well. You have probably just given us the final recommendation for the Shapeoko. I have the experience in the woodworking part, with a complete top of the line equipped shop. My son has no CNC experience, but has software experience from being in the large Printing Computer industries. We " think" in will be about $3500.00 to $4,000.00 to get set up and that includes the Laser.From there you can spend much more in software.

Welcome to the forum! Go to the introduction sub-forum and introduce yourself. Add your location to your profile and add your first name to clear the N/a in the side panel so we'll know what to call you.

You can show us photos of your shop, tools, projects, etc. in your introduction or whenever you're ready.

David

 

[gdonham1]

I too am a traditional wood worker. I have 40+ years of experience. I bought my Shapeoko to embellish box lids for jewelry and other types of boxes. I have really enjoyed the addition of the CNC to my shop. By having the CNC in my shop I have been able to make many templates very precisely. You can cut out a pattern and file and sand it but you never get it really straight or a smooth curve without a lot of work. The CNC makes patterns easily and quickly. When I got the CNC I had not even thought of using it for templates.

Many folks get a CNC and then try to think of something to do with it. Most are kits and a lot of people are not mechanically or electrically inclined. So if the install does not go well many of the machines just sit. Most people are too proud to admit failure and the poor machines just rusts in peace.

Additionally many of the same people that get a CNC have no experience in woodworking. The combination of ignorance and laziness comes together in a perfect storm for failure.

Ignorance can always be cured with education. Ignorance is not a sin but failing to do something about it is.

Good luck sounds like you will be successful.

 

[DesertRatTom]

I think the advice has been very good. I don't have a CNC, but find it interesting. However, for me it would have to make money to pay for itself, and I'd consider $6,000 as the entry level.

I see signs that are sold to businesses as what I would pursue. To that end, I've written a guide to how to make money with a CNC. I have taught marketing as a consultant for almost 4 decades, and do woodworking as a hobby for about 12-13 years, with lots of DIY and carpentry for most of my life.

Note in the attached pdf that I emphasize the design and marketing aspects of a CNC business.

Finding a niche market is another thing to consider, one of our members makes Polish eagle plaques and signs and does a pretty good buiness in that community. Gaffboat (Oliver) is possibly our best designer and has written two books on CNC for beginners under his business name, "Professor Henry." If you search for his posts in the last couple of years, you'll see some really well done commercial signs. Another user here has a nice flow of business making the wooden backing for engraved brass placques for a trophy shop. Making furniture, as someone noted, is going to be very complex and you will need a really good machine to do it. And unless you are willing to sit and wait as a cheapie machine takes hours to cut a simple piece, you're going to soon want to trade it in for a much more serious tool.

What keeps me from jumping in is that I don't think I would be happy with much of anything you might call a starter machine. I mean, I used to be able to get work done on an old 386 desktop, but I sure don't want to work on a machine that slow anymore.

All of which is to suggest some places to start thinking about how you'll use the machine. Personally, I think it will take a year of study and making simple things to start with, then more and more complex stuff. And the CNC and software you select, you'll want to be the same as what you'll wind up with.

So here's the pdf. I should probably add bars to the list of possible customers, but many hotels have those. Note that I'm not talking about the big chain operations, but one off botiques to small regional chains. You won't be able to handel the demands of a large company.

 

[BalloonEngineer]

I have the experience in the woodworking part, with a complete top of the line equipped shop.

My seat of the pants rule is that a CNC router runs about 5-6 times the cost of a similar quality table saw. If you’re happy with a $200 job site saw, you’d probably be ok with a $1000 CNC (piranha or X-Carve). If you like your $1k grizzly, you’ll probably be ok with a $5k axiom. If you like a $2k unisaw, your looking at a $10k CNC for equivalent quality & performance.

Curious how well this matches with others experience.

Richard.

 

[honesttjohn]

Hmmmmmmmm

I've got $300 in my used (solid as a rock from the early 90's) tablesaw and about $20k in CNC machines, software, and accessories. Seems I must have figured wrong! I think I need a new cabinet saw. lol

 

[ger21]

No, you just should have found a used CNC for $1500.

I paid $1500 for my Unisaw in 1995 (or so). I'll probably have somewhere around $7000-$8000 for my completely scratch built machine, if I ever get it finished. A comparable ready to run machine would be closer to $15k.

Since I use a $150,000 machine at work, my expectations are a little higher....

 

[whimsofchaz]

If you want to build one check out this site. Dave has a great package and a forum for questions and runs a regular Saturday YouTube presentations with all sorts of subjects:

Gatton CNC

 

[Inor]

Gerry -

I am not ignoring your post. Actually, I am extremely thankful you took the time to write it because it gave me a BUNCH to think about. Thank-you!

1) You can use Mach3 in demo mode for free, for up to 500 lines of g-code. Most simple 2D parts would be much less than 500 lines of g-code.
If you do decide to pay for it, I highly recommend against it. Mach3 is obsolete, with development having stopped over 5 years ago. I'd recommend UCCNC. It's cheaper than Mach3, for both hardware and software. They are currently offering a free license with the purchase of their UC100 controller.


UC100 USB motion controller + free UCCNC software license key - CNCdrive - webshop

I am certainly not wedded to Mach3. (Actually the fact that Mach3 only runs on Windoze is a real turn-off for me.) I initially thought about going with that option because that seems to be consensus for controller software for hobbyist CNC guys.

I did discover an open source software package called GRBL that interprets G-Code and acts as a quasi-controller for the stepper motors. GRBL does not have a user interface per-se, but it does provide a text-based interface to be able to do setups and download G-Code.

I did download it, compiled it and installed it on an Arduino and it seems to work fine as near as I can tell in the lab. (Obviously I am not actually controlling stepper motors but tracking the outputs on the Arduino pins, it seems to be working correctly.) There are also several open-source GUI applications for controlling GRBL, but none of them impressed me a whole lot. If I do go down that path, I am pretty sure I will be writing my own GRBL UI at some point.

That option also means I do not need an old junker Windoze computer to run the whole thing. I am thinking I will be able to get by with just a Raspberry Pi running headless. That is a HUGE benefit in my opinion since it will not require a dedicated monitor, keyboard or mouse in my shop. I can just VNC into the Raspberry from a laptop, tablet or even a cell phone!

2) I'm not a fan of Sketchup for CAD, but plenty of people use it. Fusion 360 is free, has both CAD and CAM, and is far more powerful.
There are also plenty of free CAM programs out there.

I am also not wedded to Sketchup. But I do know it (sort of, better than I know any other CAD software). Fusion 360 does seem to have a pretty steep learning curve, but I do have some time. So I will download a copy and see what I can figure out with it. Any advice on good tutorials for Fusion 360 would be greatly appreciated.

3) Be aware that $1000 is very little money for a CNC. It would be like buying as new table saw for $40. A decent quality 4'x4' machine would normally be in the $4000-$6000 range.
As David says, your expectations will dictate what you think of the machine.

The machine will definitely cut your parts, but you'll need to make multiple, light passes. $1000 is small price to pay to get a feel for it. If you like it, chances are you'll upgrade to something better.

I am okay with dumping $1000 on something to play with for a year and decide it is not my cup of tea. I am not okay with dumping $4000 for the same experience. Plus I figure, if I do enjoy it (and I am pretty sure I will once I get a few successes under my belt), I will have a better idea of what features to focus on when I do upgrade.

I sure hope you post was meant as encouragement rather than discouragement because I took it as the former. :grin:

Thanks again for your insight! Your post has me digging deep in the well and that is fun stuff!

Regards

-M-

 

[BalloonEngineer]

GRBL and an Arduino is cheap, but no where near the performance, capabilities or features of UCCNC, Mach4 or even LinuxCNC (free and no windows involved!).

Note that with GRBL, you still need a computer to run a “sender” program to send the g-code file line by line to the Arduino and provide a user interface. You can use a raspberry pi, but somewhere a fuller featured machine will be necessary, and you will want one for design work as well. I highly recommend a little intel nuc type computer - no moving parts, no fans, dust proof to run one of the above listed controller software programs. GRBL will severely limit your possible speeds, at best it can generate 30,000 steps/sec, some hardware controllers like the ESS can do 4 Million steps/sec.

 

[ger21]

I am certainly not wedded to Mach3. (Actually the fact that Mach3 only runs on Windoze is a real turn-off for me.)

Other than LinuxCNC, just about all other controls run on Wndows exclusively.

The biggest issue with GRBL (and similar open source controls) is that it's very limited in features compared to the Windows controls.

That option also means I do not need an old junker Windoze computer to run the whole thing. I am thinking I will be able to get by with just a Raspberry Pi running headless. That is a HUGE benefit in my opinion since it will not require a dedicated monitor, keyboard or mouse in my shop. I can just VNC into the Raspberry from a laptop, tablet or even a cell phone!

Other than Mach3, every other Windows control IMO should be run with a modern Windows 10 PC. AS Richard said, you can pick up one of those tiny PC's for about $200, and they fit in your hand. You can also run UCCNC from a Windows tablet, if you so desire.

I am also not wedded to Sketchup. But I do know it (sort of, better than I know any other CAD software). Fusion 360 does seem to have a pretty steep learning curve, but I do have some time. So I will download a copy and see what I can figure out with it. Any advice on good tutorials for Fusion 360 would be greatly appreciated.

Lars Christensen on Youtube.
https://www.youtube.com/channel/UCo29kn3d9ziFUZGZ50VKvWA