Is it cheeper to build

[pete1089]

Hello All
I got the CNC bug a few years back but other things took up my time, now I'm looking to get into it.
I want to get or build a high precision machine with a cutting area of about 39 x 48. Before I start asking a pile of questions, I want to know if there is really any savings building it your self.
Thanks

Similar Threads:

• Build Thread- Router home build Australia. Solid steel build
• Need some direction to build a small machine to allow me to build a larger one
• Mint's Build Aluminum/Steel Build thread.
• Newbie - To build or not to build Router/Plasma Table

[Jim Dawson]

You need to define ''high precision'' and CNC machine. That term seems to mean different things to different people. For some it might mean +/- 0.010 inch, for others it might mean +/- 0.0001 inch or better. Are we discussing metal (CNC milling machine) or wood (CNC router) machining? For the most part they are not interchangeable, however wood work can be done on a milling machine, but metal work is not so good on a router.

My choice would normally be a used machine that is in good mechanical shape, maybe with a dead computer/controls which many would concider to be a boat anchor and can be bought for scrap price.

[ger21]

It's generally cheaper to build, but you'll likely spend hundreds of hours on a scratch build.

Do you have the tools, equipment and skills to build a "high precision" machine?

[pete1089]

Yes, I have the skills. I am a millwright, I also have a well-outfitted woodshop.
It would be used mostly for wood and some soft metals.
I would like the precision to be around.005
I would use wood and aluminum to build, but use linear bearings for all direction movement.

[peteeng]

Hi Pete - 0.005mm or 0.005 inches (0.127mm) ? Peter

[pete1089]

lol yes I'm old school .005 thou inches

[peteeng]

Hi Pete - Then that's easy to achieve. I'd recommend a timber machine so look up some of the timber machines in the forum. Plywood you will deal with very easily and it will build a machine that will do better then 0.1mm very easily. Peter

[GME]

One caveat about plywood. Absolutely do not use plywood from the box stores, or most lumber yards, for that matter. You should use something like Baltic Birch (BB). No to minimal voids and stable.

You don't hear about it much, but you know how hardwoods can turn concave or convex along a fresh cut line, due to releasing internal stress? Take my word for it, the same happens with the low quality plywood. I ripped some of it along its 8 foot length with a Festool track saw and a guide rail that was perfectly straight. (I confirmed its straightness after the cuts). The change of shape was astounding (in a very bad way). Some pieces off as much a 1/4". I wasted 3 sheets of 3/4". The way plywood is made, I wouldn't have thought it possible. Anyway, for something like a CNC, you will want to be sure to use the best stuff you can find.

Oh, an great alternative to BB is a product called Apple Ply. It's made by States Industries in Oregon. Like BB, it's all hardwood; minimal voids. As the company advertises, "ApplePly is widely used in retail fixtures, contemporary furniture and architectural interiors." The inner plies are 1/16" birch. There are a variety of veneers, but I believe maple is the most common stocked at places that carry it. I believe ApplePly is significantly more expensive than BB. I looked for a 'where to buy' link, but couldn't find one.

Gary

[peteeng]

Hi Pete and Gary - yes absolutely agree. Quality materials are always the best to use. Peter

[pete1089]

Ok, so just to confirm I'm looking to get .005 thou(inch)/ .0001 mm or better

I would absolutely use Baltic Birch or maple and MDF
So I was hoping you kind experts would direct me to a DIY plan that has all the engineering done, that I could use on say a 4 foot x4 foot, footprint.
As I said I would use linear rails for all travel
Build Question
1 / Can I get the precision using a belt drive system. I do not want to go the ball screw if I do not have too
2 / Being from Ontario, Canada where is the best place to search for the components

[ger21]

Ok, so just to confirm I'm looking to get .005 thou(inch)/ .0001 mm or better

.005" = 0.127mm


I'm not aware of any plans for a good machine made from wood. What you want is to make the components torsion boxes, which are the stiffest possible wood construction. It's basically just a table, gantry beam, and two gantry sides. THere's not much else too it.

Belts can give you the accuracy, but need to be fairly large, so they don't flex. Belts would not be my choice.


Motiontek | CNC Plasma Controller KIT | AC Servo Motor | Timing Pulley | Vacuum Pump and more

https://www.rovercnc.com/

[peteeng]

Hi Pete - I'm not recommending this company just letting you know they are there to do some research with. https://buildyourcnc.com Belts will be fine I'd recommend 25mm wide. My machine has 16mm belts and holds better then 0.1mm but next time I'd use 25mm so they are a bit stiffer. With Scoot (see image) all axis are belts even the Z. But next time I'd use a screw on the Z. This is to get more plunging force. Plunging requires lots of force. Scoot can push about 20kgf but this is not enough need probably 60kgf prefer 100kgf for Z with a big tool. Takes a big effort to design a router from scratch. I'm about 70% thru a plywood machine design but currently no plans to complete it for various reasons.

There are lots of images and info on timber machine builds on the net and in this forum. I suggest you build a small one 600x600x200mm or so to get through the learning curve. Its easier to do this then deal with a big machine. Once your through the first one the second is fast and easy. All of the software and electronics can be migrated to the new one if you want. There is a thread here where someone built a small MDF machine used that to build a better ply machine then that to build a good aluminium machine then that to build an exceptional machine.

Using linears is a good choice, you do need medium or high preload cars. So be careful of cheap ebay bearings as they are often "clearance spec" bearings and will have play in them. Getting a small MDF machine going is still a thrill and a challenge so take it in steps. Peter

Images
Scoot - 1/2 sheet machine 16mm "T" belts all axes. Construction stainless steel. Scoot does signage, guitar parts, all timber stuff. Plastic and the occasional aluminium
Brevis No2 benchtop machine 10mm "AT" belts. Construction mild steel. Nearly did this one one in ply. I built it as a demo machine that I can take to trade shows.
Chris - new aluminium grill for friends car. Plus a plastic roundel
belt twist - don't drill belts, the steel wires don't like it and after doing it once it's a very bad idea all round.

[wmgeorge]

Unless your also good at controls and wiring I would suggest a kit from Avid / CNCRP. They have lots of online support and when you get done you will have a working machine.
If you build a good solid wood table and skip the leg kit it will save some money.

[stuarttaylor00]

Whats your experience level with CNC and budget?

[pete1089]

My CNC experience level is ,I see them working at work ,but those were $75000 plus machineand done repair on them.
Budget Not sure but let say $2500 can

[pete1089]

Ok than will do th wider belt
Also will do the ball screw on the Z axis
You have any Pic of your machine
Thanks

[wmgeorge]

It all depends on what you want? With zero CNC and controls wiring experience its going to be a lot of learning and troubleshooting. I have a lot of those skills and decided I wanted to Use my machine to make things, not spend time building a machine. Which I have done in the past retro fitting a old mill drill machine. I have had my fun, I would rather make things.

I understand your a Millwright by trade, moving loading and unloading machines, setting in place, leveling and getting ready for the electricians and mechanics. I have worked with you fellows and helped set machines.

[pete1089]

yes millwrights do those thing. But this millwright repairs,machines meaning many hr on operating mill lathes ,grinders.I spent 30 years in the auto manufacturing ,maintaining repairing and improving equipment

[he1957]

Building or retrofitting or even repairing a machine can be both very rewarding and frustrating.

With your years of experience doing what you did you can probably also appreciate the issues when replacing parts on a machine for which the manufacturer made a slight change and the new part doesn't quite fit because of a part "change" :-)

Just had this problem with the tail light assembly on my 20 year old car. The new taillight has a different wiring harness and the bulb mounts use 4 locating pins instead of 3 for 2 of the bulbs so my option is to try to find an older tail light assembly or splice the new with the old. It would have been "nice and right" to be able to just drop it in and be done but not the case :-(

Depends on whether you want to spent a lot of time working on your machine or using the machine to make stuff (as has been mentioned). Perhaps a kit from a good supplier and then you can consider improving it to meet your needs. A reasonable machine can always be used to make better parts for the machine or for a new machine.

Personally, my goals are to make stuff, not keep fixing/tinkering with the machine I want to make stuff with. Building/fixing etc often has a lot of hidden costs that'll blow the budget with a few more $ here and there - it adds up very quickly.

[wmgeorge]

yes millwrights do those thing. But this millwright repairs,machines meaning many hr on operating mill lathes ,grinders.I spent 30 years in the auto manufacturing ,maintaining repairing and improving equipment

I guess I was speaking from a Union shop viewpoint. You certainly can have more skills than just millwright ones.Good luck.