[cncadmin]

Attached pdf courtesy of High Seas.

[mrdovey]

Thanks for responses to the extrusion idea. You're right - without some notion as to how long that string might be, it would be too risky.

[pminmo]

Excellent Jim, I had pre written a post before seeing your piece. I had a surmized a lot of the same thoughts ->

There is such a dichotomy of needs.

These are facts for thought->
1. You can't go buy a reasonably priced DIY router, but you can the CNC software and electronics for one, so the open source machine(s) should be a focus at this time on the plan set(s) for a machine(s). There is an electronics Forum on the Zone.
2. There is a dichotomy of tools available to the DIY'er. (i.e. no tools to a full machine shop)
3. Metric and US component availability/familiarity makes plan(s) for both highly desirable.
4. There are several different plan forms of machines. Moving Gantry, Fixed Gantry...
5. There is a dichotomy of price needs.
6. There is a dichotomy of level of technical ability for those who want plans.

From that you have to assert one size fits all won't do the job.

From reading this thread I surmise->
1. There are needs for entry-level plans (cheap) as well as more accurate/costly machines.
2. There are expressed desires for moving gantry machine, which for the beginner is more involved than a simple fixed gantry.

Recommendation ->
1. Forget about the electronics and software at this point.
2. Create two threads, one for a moving gantry machine plan set, one for a fixed gantry.
3. In the design phase, shoot to design for both easy, fast and cheap for one version but with an eye towards a more refined version if the builder want to move up, or wants to start at the more expensive machine.

For example a moving gantry machine might be done on the cheap with gas pipe and roller bearings, but a second set of plans might use all the same elements but use Thompson precision ground shafting. In the design phase once a pipe size was determined, the holding method of the pipe would be dimensionally the same so that if removed, precision ground shafts would fit in with minimal changes.

What I believe would occur, is a couple of plan sets would wind up with many improvements over time, that you might have a bunch of plans available for downloading. If one thread or the other dies out, there wasn't enough need. No need to debate plan form, just go to the thread that is the machine you want to participate in. By creating two threads the focus starts to sharpen some.

Phil

[jgro]

FYI, I uploaded my plans in dxf format for those that are interested. I have the SolidWorks files also, but the zip file is 26 mb and there is a 6 mb limit to uploads. Have fun!

jgro

[BENHILL]

Couldn't agree more, it's the mechanical side that needs the attention of the experts on the group.

The electronics is easy to design, or buy.

[High Seas]

jgro - THATS a pretty complete set of plans! So can you also add some details - (I just did a very quick look at them so it maybe covered in the parts inventory)

Details like:
what motors are you running? what controller and power supply?
what are the x,y,z specs? how about max feed (positioning) rate typical feed milling? an estimate for cost of building the system (less software, controller, motors, ps)

If those fit the bill of the Opensource - the job may be done?


BENHILL - (not the chubby Brit comic/actor?)

The electronics is easy to design, or buy

But the integration is the stumper - if you get the wrong motor with the controller, and attempt to drive it with the incorrect power supply - its all toast! More than a few posts on the Zone 'bout that!


pminmo -

1. Forget about the electronics and software at this point.

Again - maybe the answer is a third thread that just focuses on the integration of motors, controllers, powersupply?

Your insight about separate threads is a good on etoo - keep the lengths to something readable from end to end!
cheers - Jim

[mvaughn]

Originally posted by BENHILL

The electronics is easy to design, or buy.

I completely disagree. From what I've seen on this forum, there are some fantastic CNC designs. It seems that everyone buys the electronics.

Everyone takes so much pride in building the machine why not the electronics as well. That way you can say you have a 100% DIY machine.

I know there are more that a few people here that are talented enough to design there own electronics and build them.

There are also those that don't want to touch a soldering iron and would rather purchase.

But don't forget those that are tinkerers. They don't have the know-how to design a complex circuit, but know how to follow directions and want to build the electronics.

I fit into the third category and bet there are others too.

[Hack]

The PDF document is definitely a very good start as to what we should be looking towards for this thread. Thanks for posting it.

Personally what I was wanting to develop as a group when I first started this thread was a few things.

I wanted to start out with a basic tutorial for beginners. In this turorial it would include a basic dictionary so to speak of what each component does and a picture of it as well. Add to that a breif outline that has a step by step breakdown of how a cheap machine is built (cheap being anywhere from 0 - 500 dollars - Probably designed for a dremel or laminate trimmer). This outline should discuss design types (fixed gantry, moving gantry, etc and the benefits / pitfalls of each. From here a newbie could decide which type is best for himself / herself and move on to the rest of the outline pertaining to the type of design they chose. This step by step process should include exactly that, and when followed a newbie should have complete working machine and understand the hows and whys things work. Include lots of pictures. For a summary to that project, I would like to incorporate ways to "upgrade" the machine - For examples THK's, Lead screws instead of all thread, etc. Give cost estimates for these upgrades.

I am a firm believer that it is better to teach someone the hows and whys and not give them the designs alone. With the knowledge that is gained from the inexpensive system, one should be able to design a system for thier needs. A cheap machine is also more likely to get finished.

Secondly, develop a of plans for a medium priced machine (1000 - 3000) that a more advanced person can build.

The idea behind the medium priced machine is for the non beginners to build (read, people who know the basics from building the cheap machine) and want a nicer machine whether size, accuracy, etc is their motivation. This machine should be designed to be flexible in size, speed, accuracy, etc so the user can adapt them to thier needs. Include with this, parts lists, supplier lists, people willing do some machining if needed (or standardize some of the parts and sell them as a kit - maybe the cheap machine above should be used by someone to do the building of the second machine?)

Dan

[pminmo]

Originally posted by mvaughn
I completely disagree. From what I've seen on this forum, there are some fantastic CNC designs. It seems that everyone buys the electronics.

Everyone takes so much pride in building the machine why not the electronics as well. That way you can say you have a 100% DIY machine.

I know there are more that a few people here that are talented enough to design there own electronics and build them.

My point wasn't to exclude anything, but to bring focus, without focus this will all die out because of going in circles. There is an opensoucre software forum and there is an electonics forum on the zone.

Phil

BTW, I have designed and built my own controller, so I do concur the 100% DIY

[BENHILL]

I would have thought that the 'integration' is common sense, but then I would, I design electronics..lol.

Sadly Benny died quite a few years ago now. Although I have lived not a stones throw from him all my life I never met the gentleman.

OK, break the project down into major component areas. Motors are both mechanical & electrical/electronic, but everything else is either one or the other.

If a general spec is outlined for a small (what is small?) machine & one for a large (same question), could not those 'in the know' break into groups to sort the design problems for each major component?

The danger surely, is the designed by commitee machine which instead of being all things to all men, is nothing to nobody!

[jgro]

I'm using a Xylotex controller and 116 oz stepper motors that I bought off of Ebay from Deepgroove1. These are the same steppers that Jeff at Xylotex sells. As for the dimensions, they are 15" x 30" x 6" cutting area.
The cost I would have to sit down and figure out. I estimate with the controller and motors, I've got about $400.00 stuck into it. Currently I'm using Turbocnc and getting 30 ipm jog speeds. Next I want to try Mach2 and see what I can get with that.


jgro

[teilhardo]

High seas, sorry to take away the glory
I made this before I saw that you had made a better one
its pretty disorganized but it can be made much better with proper software
It can be seen here:
http://www.teilhardo.com/cnc.png

[pminmo]

Can the administrator put up a poll like:

What do you want to see in a free set of Open Source downloadable DIY CNC router plans?

What kind of Machine?

a) Moving Gantry
b) Fixed Gantry
c) Other

How much are you willing to invest in the total project cost minus a computer?

a) Less than $500
b) $501 to $1000
c) $1000 to $ 3000
d) More than $3000

How big of a cutting area?
a) Less than 2' Square Feet
b) 2 sqft to 7 sqft.
c) Larger than 7sqft

Select the on closest to the accuracy you are willing to pay for in component cost?
a) +/- .01 inch
b) +/- .002 inch
c) +/- .0005 inch

Would you consider that you have the tools to build a DIY CNC Router?

a) yes
b) no
c) no, but willing to invest in some

[High Seas]

tei - nice chart! And no worries.
If you are looking for an app to do the chart with, I offer that a few years back we used "Inspiration." It worked real well for flow charting, decision tree building and particurally well for building cause-effect diagrams (trees).
Why don't you add yours to the downloads section? It graphically gives a good idea of the decisions to be made. It could go in the tech articles section - How tos home build?

BENHILL - small, Medium, & LARGE!

Phil - Nice survey input Especially the "accuracy" vs cost descriptors!

cheers - Jim

[chuckknigh]

How about a "proof of concept model" project, that would show the person that computer control is actually possible, for them.

One of the guys on the wooden clocks site has put up a pair of interneshing gears, as a proof of concept model. That way you don't have to worry about whether or not you can "do it."

A stepper motor, a super simple driver (4 lines on the parallel board, and a few transistors), and maybe a vise, would prove the concept of computer control to the first time builder.

-- Chuck Knight

[teilhardo]

I think that those survey questions are great and they should give us some idea of what people want to see

hey Paul (CNCadmin), could you set us up some kind of survey like that??

Thanks,
Tei

[teilhardo]

High Seas, thanks for the compliments- I'll look for that software and put my .png file in the downloads section...

[owhite]

I wonder if people would agree that if we simply had _one_ good set of plans describing a two rail system, a gantry, and a z-axis, even a high end version, that modifications could be made to those plans to go up or down in price.

eh?

Owen

[pminmo]

Originally posted by owhite
I wonder if people would agree that if we simply had _one_ good set of plans describing a two rail system, a gantry, and a z-axis, even a high end version, that modifications could be made to those plans to go up or down in price.

eh?

Owen

While I personally don't think one set of plans would satisfy eveyone that has shown interest, I do think there needs to be a popular starting point, thus my request for the survey (or similar) above. The more I search the web, read all the DIY CNC stuff, one thing becomes obvious, there are some really creative people out there. Some of the machines are so simple it's truly impressive. But back to your thought, there could be a ton of commonality in one or two fundamental designs that are at opposite ends of the cost vs accuracty spectrum. For example, 3/4" gas pipe is 1" in diameter. Around my neighborhood 4' is about $5. As is, could probably build a system that is in the .01 inch tolereance region. Now the other end would be hardened polished shafting, roughly $140 for 4' of 1" dia., and 4' of matching support $180 with considerably better accuracy and longevity. A 2' x 4' piece of 3/4" MDF for a base would be less than $10, but a 1/2" precision ground 6061 tooling plate the same size would be a couple of hunder bucks. I could see a person starting with the MDF and gas pipe, and at a later date slip in the precision shafting and tooling plate for example. If there was a simple effective planset based on components that had logical precision counterparts on the ZONE in DXF format, I bet peole would download and make design changes to meet their pocketbook/need and upload them. So on and so forth. Next thing you have is lots of plans available. I'd like to see all that creativity put to use to make the simplest, cheapest that amazes the user on accuracy. Group Think, KISS principle (Keep It Simple Stupid), "build it and they will come". BTW the stupid part of the principle, I see in the mirror every morning when I shave. :-)

Phil

[BENHILL]

Phil,

For what it is worth (about two tenths of naff all...lol), I think you are not only hitting the nail on the head, but driving it through three feet of concrete!!

I want to be able to prototype PCBs, none of which will ever be more than 12" on a side, I don't need much axial thrust, but I do want accuracy. I may go stepper or I might go servo.

At the size I need, stiffness is much more easily achieved. Motor control just won't be a problem for me.

Now the guy who wants to cut wood three feet on a side at a reasonable rate is, I suppose going to have somewhat different needs.

Hopefully not too many of my remarks will be quoted back at me this time....lol.