[Joe CNC]

My group and I are currently designing a low cost CNC gantry to fit common rotart style cutting tools (like the dremel type) for our Senior Design class in Mechanical Engineering. We are currently looking at the market and would be ever so grateful if we could get all your input on how this 'thing' should be designed. If you could send any comments on specs or things that are of concern to you (travel speeds, cost, accuracy, etc) to [email protected], we would be very grateful.
:cheers:

[cncadmin]

Read though the threads here their is a wealth of information.

[chuckknigh]

Check out http://***** for some good ideas, too. John Kleinbauer sells plans for machines similar to what you describe, and has quite a few ideas which have been proven to be VERY useful to those of us in this hobby.

I second the idea to just read this woodworking forum...a LOT of us have built machines, successfully, and have posted details of our designs, including analyses of what worked, and what didn't.

Also, while you're here, check out the "Open Source" forum on this site...many of us are participating in a volunteer project to design a set of plans for the newcomer.

-- Chuck Knight

[signIT]

I will stay with the Zoltar CNC router it is enough for most people who makes signs and small parts from all kind of soft metal and plastic material.

The Zoltar design may be improved but not less expensive, if anybody will make a cheaper machine then this design, it must be made of paper and biscuits.

[Joe CNC]

I will stay with the Zoltar CNC router it is enough for most people who makes signs and small parts from all kind of soft metal and plastic material.

The Zoltar design may be improved but not less expensive, if anybody will make a cheaper machine then this design, it must be made of paper and biscuits.

Thank you for your input and your criticism. Even though I agree with the fact that the Zoltar is a cheap design, It is BUILT rather than ENGINEERED, which is what my group is planning on doing. In addition, keep in mind that the whole thing still has to be built by the user and is not offered as a kit/preassembled unit. On top of that our current estimate is that we could hit the market at a cost of about $400-500 based on a current parts price and mark up of x2 (therefore not thorugh wholesale).

I encourage an active discussion on this though seeing as we are in need of input so that we can finalize the specs (we really need your input: CUSTOMER NEEDS!!).
thanks again,
Joe

[High Seas]

It is BUILT rather than ENGINEERED, which is what my group is planning on doing.

EXCELLENT! It would be great if you guys can through the engineering process reach some conclusions regarding the mechanical trade offs. I'm thinking as you enginneet the product, what differences in resolution are achieved - given up by using:
drawer slides
Skate bearings on pipe
80/20 and SLIDE systems
Thompson rails
THK and similar rails.
Likewise for linear motion transmissio;, Acme, Allthread, Ballscrews, COGs and belts, etc. Hearing the choices you make and why will help a lot of people I think.
Cheers - Jim

[signIT]

That's right go on, it will be very interesting to see what type of construction ideas you will use everything is worth to try.

I personally don't believe thet wood can be a serious component as it moves to much with temperature and moist.

[Joe CNC]

Thanks guys, it seems like you all are in favour of us getting some proper designs done. However, the real problem we are facing is that we do not understand enough about the market side. We need to know how big this market would be. How many people would actually buy one? are we talking 10, 100, 1000, 10'000...?? also what does the market want? Do people prefer high speed and low accuracy or vice versa? are they prepared to pay $200 but not $300, $500 but not $600 or where are the market limits?
This is quite obviously not necessarily a mechanical engineering problem in the usual sense but unless we know what the market POTENTIAL is, there is not going to be a design (since we cannot work out the specs by simply using the "force").
Would anyone of you be prepared to fill out a short form with some specific questions that we need?
Thanks again for your HIGHLY valued input

Joe

[thielert]

Joe,
You’re asking the million dollar question, and the answer is that there is no one answer. If you look at what is currently on the market for hobbyist CNC mills and routers there is a vast selection of very different capacities and capabilities.
Simple rules of the CNC hobby:
-Everyone wants one, many try to build one, only 0.5% get as far as cutting a single part.
-Everyone wants a huge machine, few have room for a machine larger than 12’x24’
-Everyone wants to cut aluminum, few need to cut aluminum, the best use for these cheep hobby machines is to build RC planes and other hobby components.
-Everyone wants to hold a 0.001”, a thou doesn’t mean squat in balsa wood b/c it expands and contracts with temperature and humidity. A hobbyist machine only needs to out perform what a man can do by hand two or three fold. So holding 0.015” is awesome.
-Ball screws are all the rage, ACME was the thing last year, you only need 1/4” allthread from homedepot with a cutting board plastic nut hand tapped.
-Don’t bother screwing with your own controller design. Current limiting resistors are a waste of time and amps. Get one of the fine 3 or 4 axis chopper boards pre assembled. Works great and no problems.
-NEMA23 steppers – EBAY.. your not ready for servos or Geckos.
-You can spend a lifetime on linear rail selection, gas pipe, drill rod, BWC, 80/20, Thompson, NSK, THK. Each more expensive that the previous. Gas Pipe can be rough, drill rod and roller skate bearings is tough to beat. Checking out the K.. B.. plans, they are worthwhile and they are inexpensive.
-MSCDirect.com is good
-McMaster.com is great
-EBAY.com is awesome!

So back to the million dollar question… the answer (for you) is..
1’x1.5’ cutting area
Dremel spindle
MDF construction
Allthread
Bearings of your choice


I have known several eng students who have started what are about to. Only one finished.. and I had to sell him a finished machine just before his senior presentation.

It took me 6 months to build my first machine from KB plans. It took me nearly 2 years to build my second machine (4’x4’x1’ all Aluminum). This is not a quick, easy, or cheap hobby.

Best of luck.
Tim T

[thielert]

I forgot to adx your question of market potential. No one will buy one if you’re just a kid who makes a single machine as a college project. If you make is come in different colours and have Ron Popiel or that Austrailian guy sell them on late night TV you’ll sell thousands.

Marketing is an enigma to most engineers. The adage “If you build it they will come” or “Build a better mouse trap and a path will be beaten to your door” are bunk. You can have the best product in the world and go broke without good marketing. A talented marketer can sell crap to millions. Marketing is an entirely different industry than product development.

Look at the history of the Tupperware company. Amazing success by full separation of product and marketing. My advice to engineers.. don’t try to be marketers. My advice to marketers.. don’t try to be engineers. It is a very rare person that can do both adequately.

Best of Luck
Tim T

[jimbo]

I am confused by post #5. You say its not a kit but the user has to build it. What do you get for 400-500?

I would like to see mass produced personal use machines. Say using even dovetail ways or some simple bearing arragement. Needs built using a stable material. Needs to be modular, perhaps allowing some sort of length extensions for upgrades. Probably stepper motors with chopper drives are cheapest solution. Allow for 1/8" and 1/4" shank bits. Cutting size would probably be 2'x14" for smallest configuration, upgrading to 2'x2' and 2'x4'.

Software is also a big portion of the problem. The process needs to be more streamlined. I would like to see it almost as simple as printing something off the computer.

Marketing is a big key. If you could get the product out to a big retail store and market it as an engraving tool, modeling, etc and appeal to the kind of people that like legos, rc models, art, even woodworking. Also things are very price sensitive in todays market. I would shoot for 150-300 at most, most people can gamble that for a product as a gift or for themselves. But I really think software needs to also be a part of it. It will be hard to make a complex process simple enough for wide acceptance. Maybe have specialized packages for engraving, modeling, sculpting.

[High Seas]

It sounds like you need to do some market analysis. I'll bet there are some folks 'cross campus that do that kinda thing or are learning how. Would certainly make the effort a systems approach. Sound engineering is great - but you're right - no market - no sell, no deal.
If this is a bona-fide college/university'etc engineering project - you might contact some of the better known builders out there, and ask them what they think the market looks like. SHOPBOT is pretty well known in the woodworking circles and first attracted me to this hobby. DOn't know how many they've sold - but its the 3000 to 6000 dollar range as I recall. I DIY'd rather than spend that much. (But then again - I'm not yet to their scale 4x8). Google the available machines and do some research, categorize, develop a market survey - less than 20 questions - then start asking. Bet you'll find a solid number in less than 100 surveys!


Tim T - I always thought of Marketing as something separte from SALES - but I know it's all gotten quite blury:
Idea ---> engineering ---> prototyping---> engineering analysis---> CAN IT BE BUILT?
Idea ----->market analysis------->cost analysis -----> DOES IT MAKE SENSE TO BUILD?
.............................Product (even crap) give it to SALES ----> sell at any cost!
Just off the top of my head!
Cheers - Jim

[Joe CNC]

Thanks for all your input guys,
I guess I am really asking the million dollar question, but it tought it was worth trying
What I would really like to get out of anyone willing to sit down and write a reply is how much YOU would be willing to pay for something like this. I've got a first answer on this stating the $50-300 range. Obviously this is somewhat optimistic especially for the wholesale market. Also keep in mind that some places sell the controller harware (the device in between the computer and the stepper motors) for $600 alone. So getting the gantry, electronics and software for less than about $200-300 is not really feasible.

Also, in reply to TT's post, I do know how much involvement this brings with it since I have my own little CNC in the garage (with electronics that I designed and built on my own, I may add .

But anyway, please keep on posting... and if possible give me your (brutally honnes) feelings and opinions.... and if possible a price you would be willing to pay.

Cheers to all you CNC'rs

Joe

[jimbo]

I am still not clear on what you are doing for your design session.

Looking back, I realize that would be way to optimistic at this point. The only reason I stated the cost that low was from looking at printers, scanners, and cdrom drives. Technically cnc machines are not much more complex in hardware and electrically just beefier. I just hoped the economy of scale could offer such a machine for the price of todays printers. To me, a price that low would be a "cant resist" price as opposed to 500 which I would have to think about.

If the machine was limited to cutting wood and plastics. I think a very low cost machine could be made.

What about making a survey?

Jim.

[chuckknigh]

$600 for the electronics? I think not. Not for a basic introductory machine.

Do keep in mind, a SIMPLE and PRIMITIVE controller circuit can be built quite cheaply. A Darlington array, a parallel port connector, some diodes, and some resistors are all that are needed. It doesn't provide much in the way of performance, but it'd allow a very inexpensive machine to be built.

Optional upgrade modules could be a secondary market -- chopper drives, or even Geckos would be a definite step up from a wave drive...but the wave drive would be cheap and functional.

You'd also have to include some software -- primitive software would work, but something like TurboCNC (shareware) would allow full functionality at minimal additional cost.

Take a lesson from the Harbor Freight mini lathe. It's a rough, primitive, but functional machine with "good bones" which people have tricked out to do real precision work...and some have even converted it to CNC. HF provides the basic machine, and the owner decides what to do with it.

Check out the circuitry in their Router Speed Control, too. It's fairly primitive, but it works reliably. Many people start with the HF units, and then move on to "bigger and better" things.

I think offering a fairly small CNC machine (perhaps based on an off-the-shelf 4x6" cross slide vise commonly available from China for around $20) with a cheap 'n dirty controller circuit, and powered by some commonly available steppers would be a good idea. For your time and trouble, sell it for $100-150, maybe as much as $200.

You'd probably sell a lot of them, especially if you did Home Depot/Lowes in-store demonstrations, like Shopsmith does.

-- Chuck Knight

[DDM]

The other aspect that you have to look at is the manufacturing cost of manufacturing the product. The major companies that make printers have it down to an art and 1 second on a molding machine is big business to them. They are doing things in mass production for a mass population and from there they can dedicate a whole production line to a certian series of printer. if CNC machines were in the mainstream like printers you would see them for 500 bucks no problem with many capabilities. This is just from the marketing and production standpoint which really doesn't cover the actual engineering of the machine.

I do agree on doing a survey on what the ability of the machine is and how much would you want to spend for one? I don't know what kind of good it is going to do you other than a little paragraph on market research. Best of luck to you guys.

[signIT]

If one could find a concept with a good standard using easy to find parts.
Like silver steel rods combined with low price adjustable bushings like oilite or the teflon type of slide bushings.

But perhaps the best of all contact Chris Reinders The Zoltar designer and buy his concept improve this machine and begin to make money.

[Joe CNC]

Thanks for your continued input. I would like to quickly reply to a few small details. First, there are companies that offer the driver electronics for $600 (I believe Sherline.com, but don't quote me on that) also we have thought about clever ways of incorporating those little wishes everyone seems to have and I am currently compiling a list of them. Also I am putting together a website that I will post later on with some questions we would like you guys to perhaps give us some answers on. It has less to do with the actual market size (even though that would still be appreciated) but more with the actual design specs that we should aim at. Once it's up (hopefully tomorrow) I'll post the link and also the link to the results once the information gathering has been completed... so please hang around for that...

Well so long and thanks again for all your expert input
Joe

[Cliff_J]

Joe,

Amoungst the many great responses some have hinted but the discussion in the open-source plans made one point obvious: There is no one-size-fits-all solution. Diverse budgets, needs, and expectations make an adaptable solution a design requirement.

The next point I'd like to address is that consumer education is part of this as well. Many people I've read about expect at first to build a 4'x4' sized machine for a few hundred or a 1'x1' dremel-based machine for pocket change. They easily underestimate the pieces that go into the machine and how quickly each seemingly small cost adds up. We've been spolied by years of mass-production bringing costs down while CNC was always a low-volume product for business where an ROI analysis would still green-light a 6-7 figure machine. Pretty conflicting mindsets.

So maybe a 3-axis mechanical sub-assembly of 2 or 3 sizes with a single motion sub-assembly of motors and controller to mate with the mechanical part. Then the end-user would have a better understanding where $200 is for the motion and $300 is for the small table and $500 for the large table. Explicitly lets the purchaser know where the money is headed. Could be expanded to high-use high-precision mechanicals for $1200 on the small table and so on.

my two cents...
Cliff

[Joe CNC]

Hello all you CNC fans...
As you may have noticed if you were following the rather interesting debates over the last few days is just what is of importance on a CNC...
well as you may assume, we've thought about this for quite awhile and have come to the conclusion that there's only one way to really find out... that's to
ASK YOU!!!
well here it is, a page built with a ton of questions that we would like to get some answers for. it takes less than five minutes to go through them all and we would ever so greatly appreciate all input.
Since this is meant to be a discussion forum rather than just an information gathering place, we will of course post a link to the analyzed replies (within a week or two).
So here it is:

http://members.tripod.com/cncgroup/disc2_post.htm

please take a few minutes and give us anything you can (else just leave it as it is...)
Thank you so much
Joe