[mercaytor]

Hello all, I've been contracted by a friend to build him a CNC Router table. He owns a small sign printing shop, specializing in decals and so forth, and he's been wanting to get into wood-engraved signs. We were talking a few days ago and I mentioned that I was working on a small CNC router (for PCB milling) and he said he always wanted one but couldn't afford the $40k for one. When I told him I was fairly certain I could have a machine built for him for $10k, he told me to go for it!

So here I am: I have about a $10k budget to start from scratch and get a functioning machine built. Timeframe is flexible, and so is budget to some degree (he's not gonna kill me if I spend $11-12k...) and the requirements are as follows:

- Needs to be 4x8 ready, he has some big signs he wants to work on
- Most of the material will be MDF with the occasional Oak or Cedar
- Quality is key and oversizing is good, he's fine with spending a little more to make sure it's sufficient
- We live in Canada, so bear in mind shipping can be extreme sometimes...
- He already has a shop-sized industrial vacuum system that we can tie this into
- Prefer a solution that uses a laptop that can be removed from the machine for both design and CAM
- His shop has lots of room so we'll build a room specifically for this machine
- He doesn't have much time to research the technology so he's relying on me to design/source/build and then I'll run it for him part-time until he's comfortable with it
- No need for super fancy stuff like tool changers or all that
- I am comfortable with electronics, wiring, etc

1) Anyways, in my several years of lurking and wishing I had a budget to work with, I've looked at quite a few different kits and I think I'm most impressed with the CNCRouterParts.com PRO4896 kit (PRO4896 4' x 8' CNC Router Kit | CNCRouterParts) - Good choice for the above requirements?

2) Electronics wise, I'm of the opinion that NEMA23 would work but NEMA34 is better suited, and I think Servos might be a bit beyond the budget / need of this machine. What does everyone think of this kit: 4-Axis DIY Nema 34 Kit | CNCRouterParts
Is it worth the extra $1000+ to get the 4-Axis Plug and Play Nema 34 System | CNCRouterParts ? Or are there other sites/kits you'd recommend for this machine? Or is NEMA23 sufficient?

3) CNCRouterParts seems a little expensive for the cable carrier kits (PRO Complete Cable Track Kit | CNCRouterParts) at $300+, or is that a good price for what it includes? Seems like I could source the chain and make some brackets for considerably less.

4) The steel leg kit (Steel Leg Kit (PRO CNC) - CRP413-00 | CNCRouterParts) they offer seems pricey also at $480 for the 4x8 version. Should I build a table from wood or have a metal shop make us one instead?

5) Routers: Based on what CRP offers for mount plates, what do you recommend for this use, a Bosch 1617 / Hitachi M12VC / PC 7518 / PC 980 or 892? Or something else altogether? Do I need anything more than about 2hp for MDF and/or Oak or hardwoods? (No plans for metal at all)

6) $250 for a smooth-stepper Ethernet upgrade seems like a lot via CRP, is it really worth that much? What does this allow me to do, use a laptop vs a Desktop PC (with parallel port)?

7) Software - I was planning for MACH3 CAM software, but is there another software package that would be better suited to wood routing signs?

Thanks all!!!

[ger21]

1) If your main requirements are a 4x8 that cuts wood, than yes. There aren't really any other kits for a 4x8 router that I'm aware of. CNC Router Parts are good machines for the money. But they're not the same as a $40,000 machine. From all your other questions, I'd say you should come in far under budget.

2) The answer depends entirely on what you expect from the machine. Most new machine builders have no idea what type of performance they're looking for, which usualy makes this an impossible question to answer. I'd give CNC Routerparts a call and get they're opinion.
If you know what you're doing, you can put together a servo system for close to what the Nema34 plug and play system costs. The question is how much do you know, and how much is your time worth.

3) You can probably source the parts and get every single item they sell cheaper. I'm scratch building a similar sized machine, and new Igus e-chain will probably cost around $750. You can get Chinese chain from Ebay for considerably less.

4) See #3. Do you want to bolt a kit together, or spend time fabricating and building something yourself. I seriously doubt youcan find a metal fabrication shiop to build you a steel table for less than double their price.

5) A $10K router should be running at minimum, a 2.2Kw spindle. Again, your needs are dictated by your performance requirements, which you haven't stated.

6) They sell the ESS for $200, not $250. YOU can save $20 by buying it somewhere else. The Smoothstepper allows you to use a PC with no parallel port.

7) Mach3 isn't CAM software at all. It's machine control software. For sign use, I'd recommend Vectric's V-Carve Pro or Aspire.

Prefer a solution that uses a laptop that can be removed from the machine for both design and CAM

Imo, this is a bad idea. Cad/CAM software is much more enjoyable to use with large monitors (even better with 2!). Routers create lots of dust, not a good environment for an expensive laptop. Also, unless you plan on using a Smoothstepper or other motion control device, laptops are not supported for use with MAch3, and most won't even work properly. I'd recommend a dedicated control PC at the machine connected to a design PC via network.

I'd recommend doing a LOT of research. If you're spending someone else's money, make sure you're making very informed decisions.

[mercaytor]

Thanks for the reply, Gerry. Based on your comments, I'll clarify a bit and ask some additional questions...

2) For performance, you're right that I/we don't really know what to expect or want for IPM / feed-rate, etc. What would be a reasonable expectation given this budget and the proposed materials?

3 & 4) Yeah, I think I'd go with the pre-made kit on all of this, best chance for instructions and support when it comes to building and operating the machine.

5) Okay so it sounds like a store-bought power tool wood router is not what you'd recommend then? What would be a good supplier for a 2.2kw spindle that would fit the project requirements?

6) I was seeing the with the NEMA34 Electronics Kit they offer the ESS + configuration as a $250 option/add-on, but yeah the ESS alone is cheaper for sure.

7) Guess I'm missing something here, so does V-Carve do what Mach3 would do, or does V-Carve prepare the GCode for Mach3 to then run the machine with? Actually that makes more sense to me now, if I've understood correctly... Anyways, so yeah I will need some form of extra software to prepare the basic 2D sign design into a 3D relief-cut etc before I send it to Mach3 to actually route?

8) Laptop is out, a regular desktop rig on LAN will be better. It only needs to run Mach3, so I shouldn't be needing a beefy Intel i7 with 16gb RAM or anything right?

[Little_Chips]

I would use NEMA34 Bipolar or Bipolar Hybrid steppers and run them in parallel winding configuration as this gives the best speed torque curve but at a higher current. I would also consider the Black Foot or Green Bull machines here Build Your Own CNC Router, CNC Machine, or 3D Printer . Make sure your motor drive can handle this configuration. Bipolar steppers usually have 8 wires instead of 4.

This page Stepper Motor, Stepper Motor Driver,CNC Router, Stepper Motor Power Supply, Stepper Motor Kit, DC Servo Motor, Stepper Motor, Stepper Motor Driver, CNC Router, Stepper Motor Power Supply, Stepper Motor Kit, DC Servo Motor has steppers and information on steppers (scroll down page).
This page UniPolar vs BiPolar wiring schemes for 2-phase Stepper Motors is also a good source of info.
Take your time and have fun!

[ger21]

2) I'd want to be able to cut at 300-400ipm, with rapids up to maybe 600ipm. Keep in mind that different operations may require vastly different feedrates.

5) Most buy them directly from China through Ebay. If you want to buy one in the US, you'll pay a lot more, but can expect better support. Check out www.UGRACNC.com

7) V-Carve Pro or Aspire let's you do your design work, and create the g-code. Mach3 reads the g-code and runs the machine

As far as the Blackfoot and Greenbull, don't waste your time. A CNC Router Parts machine is infinitely better in every aspect. And the price is actually very similar.

[ClintonH]

Have you looked into a XZero CNC kit? I would phone George and talk to him see what he could do for you on the size. I just bought a ViperXZ HD kit from him and it's built like a tank. Also he's in Canada so shipping was quick and a fair price.
index
Xzero CNC

Clinton

[mercaytor]

It looks like that we've decided on CNCRouterParts.com for the machine itself, building a PRO4896 kit complete with cable track, steel leg kit, and NEMA34 smooth-stepper add-on. I believe we'll go straight to Aspire for software, I've been playing around with the demo and it looks to be very much worth the investment. With shipping, we're looking at right about $11,000 USD.

So, a few more questions:

1) For router, I'm really interested in a spindle/vfd setup, but to start with I think we'll use a Hitachi M12VC and upgrade the collet(s) with a Precision solid carbide tools for woodworking and metal forming. kit. Reason being I think it will be more advantageous to learn and abuse a cheaper tool before playing with an expensive spindle, and if I understand the reviews of PreciseBits kits, they can make a router work with TIR's near the range of much more expensive spindles by only adding about $100 to the price. Or am I really missing something vital here?

2) Starting bit set... What would you recommend for an initial kit of bits/endmills/V's etc for sign-making in MDF primarily? I know there's lots of specialty bits out there and I'm sure we'll start collecting down the road, but I want to know a good basic assortment that will get us up and learning/running. From my research so far, I've found these sets from www.ToolsToday.com
18pc, $500 - 18-Pc CNC Signmaking Advanced Router Bit Set -Toolstoday.com- Industrial Quality Router Bit Sets
8pc, $250 - 8-Pc CNC Signmaking, Lettering & Engraving Starter Router Bit Set -Toolstoday.com- Signmaking Router Bit Sets

Given what we're looking to do, is the second (cheaper) kit sufficient? I don't see a planing/surfacing bit for spoilboard in this kit, but maybe I don't need one? Or perhaps there's a better supplier you can recommend, like the ShopBotTools.com package:
"13699 ShopBot Starter Bit Kit – 1 ¼” planing bit, V-bit, straight bits, upcut and downcut spiral bits, roundnose bit – 8 Onsrud Cutters $195.00"

3) How well does simulated wood grain come out on MDF signs? We're really interested in producing signs that have the sandblasted wood grain effect but by using MDF material. I created some renders using depth textures from Google Images and editing them in Photoshop to get a more defined grain pattern, and Aspire gladly turned it into a 3D model that looked great, but obviously I have no idea how well it'd actually cut out... I'll attach a render image.
Attachment 205262


Thanks again!!!

[louieatienza]

It looks like that we've decided on CNCRouterParts.com for the machine itself, building a PRO4896 kit complete with cable track, steel leg kit, and NEMA34 smooth-stepper add-on. I believe we'll go straight to Aspire for software, I've been playing around with the demo and it looks to be very much worth the investment. With shipping, we're looking at right about $11,000 USD.

So, a few more questions:

1) For router, I'm really interested in a spindle/vfd setup, but to start with I think we'll use a Hitachi M12VC and upgrade the collet(s) with a Precision solid carbide tools for woodworking and metal forming. kit. Reason being I think it will be more advantageous to learn and abuse a cheaper tool before playing with an expensive spindle, and if I understand the reviews of PreciseBits kits, they can make a router work with TIR's near the range of much more expensive spindles by only adding about $100 to the price. Or am I really missing something vital here?

2) Starting bit set... What would you recommend for an initial kit of bits/endmills/V's etc for sign-making in MDF primarily? I know there's lots of specialty bits out there and I'm sure we'll start collecting down the road, but I want to know a good basic assortment that will get us up and learning/running. From my research so far, I've found these sets from www.ToolsToday.com
18pc, $500 - 18-Pc CNC Signmaking Advanced Router Bit Set -Toolstoday.com- Industrial Quality Router Bit Sets
8pc, $250 - 8-Pc CNC Signmaking, Lettering & Engraving Starter Router Bit Set -Toolstoday.com- Signmaking Router Bit Sets

Given what we're looking to do, is the second (cheaper) kit sufficient? I don't see a planing/surfacing bit for spoilboard in this kit, but maybe I don't need one? Or perhaps there's a better supplier you can recommend, like the ShopBotTools.com package:
"13699 ShopBot Starter Bit Kit – 1 ¼” planing bit, V-bit, straight bits, upcut and downcut spiral bits, roundnose bit – 8 Onsrud Cutters $195.00"

3) How well does simulated wood grain come out on MDF signs? We're really interested in producing signs that have the sandblasted wood grain effect but by using MDF material. I created some renders using depth textures from Google Images and editing them in Photoshop to get a more defined grain pattern, and Aspire gladly turned it into a 3D model that looked great, but obviously I have no idea how well it'd actually cut out... I'll attach a render image.
Attachment 205262


Thanks again!!!

1- The rub with PreciseBits collets and nuts is that you have to have a router with low runout to take advantage of them. The only way to really do this is to get a length of ground rod the size of the collet and chuck it on an actual router, and check with a test indicator. I don't know if they'll let you open box after box at the local Home Depot. though I've heard the Hitachi routers are usually pretty good. The upside with a higher powered spindle is that it can handle higher loads at lower speeds for a longer period of time, so you really have less chance of harm, and probably better results and longer tool life.

2- I'd buy bits as you need them. I also wouldn't get a spoilboard leveling bit under 2" for your machine as even that size would be a slow-go. That is likely the most expensive bit in the bunch as well, around $250-300... I use Amana's large 90 degree insert v-bit and recommend it as it does stay sharp longer and the replacement tips are reversible, aharpenable, and cheap. I'd get a couple straight-flute bits for machining the fields on inset signs, and maybe 15 and 30 degree engraving bits. The In-Groove set is nice, but if you don't use the other profiles, it's not worth it as the solid carbide engraving bits cost as much as the inserts. I would also consider a mortise-compression spiral for cutting 2-sided material. Get every other bit as you need it. I buy from ToolsToday and they ship fast.

3- I've seen many textures done on MDF with great results though I haven't done it myself (I prefer to use wood.) There will likely be some finishing work afterwards. Also it depends on the MDF you're using, as the outer surfaces are denser than the core, and that may cause issues.

[ger21]

1) Unless you get a really bad router, runout shouldn't really be an issue. Most better brand handheld routers have runout around .001, which isn't really an issue unless you're using 1/32" or smaller bits on harder materials.

2) I agree 100% with Louie. Buy bits as you need them. Buying a set will leave you with 1/2 of the set that yo udon't use.
Get a 60° and 90° V bit, and some 1/4" and 1/8" downcut spirals. If you can afford it, I'd buy a 2"+ spoilbard surfacing bit. Or the largest 2 flute straight bit you can find, if you want to do it cheap.

3) I think you'll have trouble trying to carve a woodgrain texture into MDF, as it won't hold fine detail. Also, any carved texture in MDF can be difficult to finish, as finish will soak into the core. I like to rough cut the textures, then coat it with epoxy, and allow it to cure before running the finish pass. This gives a very smooth surface that's easy to finish.

[awerby]

You're getting good advice here, but you're hanging onto a few misconceptions. Trying to simulate wood grain in MDF would be a big waste of time. If that's what you want in the background of your signs, use real wood and not MDF, set up the router with a drag-knife, cut stencils for your letters, and sand-blast the backgrounds, which brings out the grain. The extra cost of the wood is far offset by the hours of time you won't be spending with tiny bits trying to carve wood grain in the MDF.

Precisebits makes very nice cutters, and their collet systems are better than the ones that come with most hand-held routers. But trying to use a hand router in a production environment is like entering a VW bug in a NASCAR race - it's just not up to the job. Real spindles are built much better, and they're also more expensive for that reason. The bearings on a good 3-phase European-made spindle will just be breaking in as you discard your 4th hand router. Include a real spindle and VFD in the initial plan and budget, otherwise you're going to have to rework the system later amid cost over-runs.

For a machine that size, that's expected to cut that fast, servos would be better than steppers. The torque of steppers is best when they're at rest, and the faster they go the weaker they get. Servo motors are a bit more expensive, but they like going fast. Just put some limit switches on the thing so it doesn't kill itself if it hits the end of travel by mistake.

Andrew Werby
www.computersculpture.com

[louieatienza]

...For a machine that size, that's expected to cut that fast, servos would be better than steppers. The torque of steppers is best when they're at rest, and the faster they go the weaker they get. Servo motors are a bit more expensive, but they like going fast. Just put some limit switches on the thing so it doesn't kill itself if it hits the end of travel by mistake...

I think however in this case with the CNCRP kit and their NEMA34 R&P setup, the gearing is optimized for steppers; and with the drive ratio there will sill be plenty of torque at rapids. I believe guys are getting around 1000ipm with the NEMA34 system. For servos, you'd have to gear them down considerably, usually with planetary gearheads, and that raises the cost significantly...

[vtx1029]

Check out magnate router bits on amazon. I really like them! I have the 2" planer bit and it works great. I see they now have a 3"! Their Spiral O-bits work great too!

Magnate 2706 Surface Planing ( Bottom Cleaning ) Router Bit - 2" Cutting Diameter - Amazon.com

[mercaytor]

You're getting good advice here, but you're hanging onto a few misconceptions. Trying to simulate wood grain in MDF would be a big waste of time. If that's what you want in the background of your signs, use real wood and not MDF, set up the router with a drag-knife, cut stencils for your letters, and sand-blast the backgrounds, which brings out the grain. The extra cost of the wood is far offset by the hours of time you won't be spending with tiny bits trying to carve wood grain in the MDF.

This actually makes a lot of sense, and I had already kind of discussed it with my employer (the sign maker financing this project). He says that apparently there's no one in the local area that does the grade of sandblasting he'd need to do these signs with real wood. While I find that a little strange, we do live in a small community (approx 5,000 with a trading area of only 15,000 maybe). I had actually started investigating how difficult it would be to prepare a sand-blasting booth that would accommodate this type of work as he definitely has the space, but then I became concerned that if he builds a sand-blaster booth and already has vinyl cutting machines that (I think) could be used for masking, then there's no need for a CNC machine at all (for those types of signs).

I've told him already that any texturing/wood-grain designs in MDF won't look the same as the real thing but since neither him or I have (knowingly) seen CNC-routed MDF wood-grain-effect signs, he wants to try it out.

Actually, this might be a good opportunity to ask...

Would any body with a DIY-kit CNC machine be able to cut us a sample (say 6"x6"x1") that contains a 3D wood-grain effect from MDF so we can physically touch/see the result? I'd be willing to pay for shipping of said sample to our Canadian address, I can't imagine it'd cost more than $20 in material/shipping, but then I guess there's machine-time expense to consider too. Or should I ask in a different forum area?

PM me if this is a reasonable request that someone would be able to help me with.

[mercaytor]

Check out magnate router bits on amazon. I really like them! I have the 2" planer bit and it works great. I see they now have a 3"! Their Spiral O-bits work great too!

Magnate 2706 Surface Planing ( Bottom Cleaning ) Router Bit - 2" Cutting Diameter - Amazon.com

That appears to be a very good deal, almost too-good-to-be-true lol... So you have direct experience with these? How long and any other comments on it? Will a $300 planing bit really outlast 8x of these $37 bits? I'm certainly interested in buying one to start with, that's for sure...

[louieatienza]

That appears to be a very good deal, almost too-good-to-be-true lol... So you have direct experience with these? How long and any other comments on it? Will a $300 planing bit really outlast 8x of these $37 bits? I'm certainly interested in buying one to start with, that's for sure...

The $300 planing bit uses carbide inserts, which since they are not brazed to the body, can be made of a harder grade, meaning they last longer. And they're indexable, so you have four cutting sides per insert.

This actually makes a lot of sense, and I had already kind of discussed it with my employer (the sign maker financing this project). He says that apparently there's no one in the local area that does the grade of sandblasting he'd need to do these signs with real wood. While I find that a little strange, we do live in a small community (approx 5,000 with a trading area of only 15,000 maybe). I had actually started investigating how difficult it would be to prepare a sand-blasting booth that would accommodate this type of work as he definitely has the space, but then I became concerned that if he builds a sand-blaster booth and already has vinyl cutting machines that (I think) could be used for masking, then there's no need for a CNC machine at all (for those types of signs).

I've told him already that any texturing/wood-grain designs in MDF won't look the same as the real thing but since neither him or I have (knowingly) seen CNC-routed MDF wood-grain-effect signs, he wants to try it out.

You can get textured AZEK board that has a pretty realistic wood grain on one side, and either cust out prismatic layers and affix them on top, or cut pockets for letters and inlay the letters. This would be eons faster than trying to do a woodgrain texture. You can sanblast or torch a piece of wood, then cut pockets out to accept raised lettering. Or mask the letters' area with vinyl, then sandblast, then apply prismatic letters on top.

Also consider you can buy marine-grade plywood in different species, which looks infinitely better than "fake" wood, and you don't have to paint it to look real. You might also want to consider OBO Modulan tooling board...

[vtx1029]

That appears to be a very good deal, almost too-good-to-be-true lol... So you have direct experience with these? How long and any other comments on it? Will a $300 planing bit really outlast 8x of these $37 bits? I'm certainly interested in buying one to start with, that's for sure...

Yes I have direct experience with them thus me saying "I have the 2" planer bit and it works great." I also have several single o-flute and V bits from them. I bet I've purchased 15-20 bits from them and have not been disappointed. As to your other question the Amana insert bit would outlast because the inserts can be changed. Is it worth the extra price? Guess only you can decide whats right for you...

[mercaytor]

Fair enough, at that price I will definitely be trying out a couple!