[smackdaddy]

I need to build a simple three axis machine to apply glue to a product I manufacture. Tolerance and speed are not that crititcal,but cost is. I want the most simple cost effective way to do this. The part is less than 6 inches long and 3 inches wide and 1 inch deep, so size can be relatively small, and there is no need for beef, there will be little if any resistance in the motion of the axis. Any and all advice would be greatly appreciated! Would consider buying existing machine but I need every thing, computer,software the whole schamolly! Can't keep up with production!

[epineh]

http://www.praga.com/spatula3.gif

Sorry, couldn't resist

Are you sure you can't just get a better applicator ? As I see it you will need not only need 3 axis of movement, so we are talking PC, control software, driver electronics, motor's, linear rails of some description but also some kind of metering for the glue distribution, and to make it worth it in terms of speed some kind of job changer - electronic control to interface to the PC to run the job change logic, the list kinda keeps goin on...

I hate to be a naysayer (spelling?), and don't know what kind of units per minute you want to do, but seems a little complicated way to do something simple.

Of course if you are talking serious volume then all of this may be justified, can you give us any idea of the numbers ?

You could skip the PC control and program a PLC or Microcontroller to control the electronics, but this means learning to code or paying someone to setup the logic or program.

Russell.

[Switcher]

This may sound like overkill, I would start with Mach3 as your control, then get something like this:

http://www.hobbycnc.com/hcncpropkg.php

3 Axis Package
With (3) 127oz-in Steppers
$245USD + $16 USA S/H

Kinda hard to shop, when you don't mention a set dollar amount?





.

[smackdaddy]

Thanks for your time Epineh. We hope to ramp production up to around as many as 5,000 a month, Also let me mention I and a partner do this in the evenings and weekends so we need to make the best of our time. I assume that if I purchase a machine. I will need to add all the hardware you mentioned. If I build one that is custom built for this one job which requires just one program could I possibly use my laptop and have a company with a cam package write the software on to a disc then send it to a low end microprcesser and avoid alot of head aches or am I dreaming. If it is not obvious I am very limited in my grasp of what are and what are not limitations when it comes to cnc.

[smackdaddy]

Hey Switcher. I am looking to spend hopefully no more than a thousand dollars on the machine.

[sansbury]

Thanks for your time Epineh. We hope to ramp production up to around as many as 5,000 a month, Also let me mention I and a partner do this in the evenings and weekends so we need to make the best of our time...

I'm with Epineh on the spatula.

I don't know much of anything about CNC/CAM but from running my own software company I've learned some hard lessons. One of the biggest is that the problems you anticipate at this stage are often not the same ones you hit once you go live. A lot of businesses fail because they invested too much time and money in solving problems that they ended up not having.

I'm guessing I could apply glue to a few hundred pieces an hour by hand, and you could always set the TV up in the garage and have friends and family pitch in. Unless the product has a ton of parts, I can't imagine it being too hard (though maybe a bit boring) to make 5,000 in a month that way. And it scales down well, so if sales are weak, at least you haven't invested a bunch of money in a machine you end up not needing.

Maybe after a few months of that, if demand continues to be strong, and your boss asks you why you come into work every morning smelling like glue, then you consider alternatives. One I'd include then is outsourcing the entire production process to a contract manufacturer so you can focus on selling more of your product--and then you can quit your day job

[epineh]

While CNC control would be nice for this, I think you could do it even simpler and cheaper by using a good old fasioned electrical/mechanical arrangement, your travel is small, and every job is the same.

I'm thinking a motorised bed of some sort to suit your part, simple DC motor with gearbox, easily found at an electronics type outlet and a couple of limit switches, and some very basic logic with a couple of timers and reversing switches, the glue can just pump out when the bed is making a pass.

Even simpler a jig of sorts to load the pieces quickly and appy the glue, using your own power to provide motive force, when done, push or pull a lever that pushes the glued part out and gravity loads the next one... and so on...

Russell.

[ZipSnipe]

Yep I like Epineh's idea, some type of assembly line type deal would work great, but if you insist on going the cnc route then go for a large router type that way you can make a jig to hold multiple parts and glue away.

[smackdaddy]

This is a great forum and I appreciate all the good advice. I think Epineh is on to something. I can purchase a glue dispencer that has a turntable with it (somewhat reasonable) then I can build a camming device that will allow the part to stay under the needle while it rotates. This could be an affordable fix. Thanks Epineh I will post as I try this concept out! If anyone has an idea on how to build this device I would again be thankful....The part is a fishing lure. It is a half of a fish body. It has shiplap all the way around it. You can see one at www.smacktackle.com. I'm fired up again!

[epineh]

I'm glad I could help

There is a little irony here, the whole reason I got into CNC in the first place was to make quality fishing lures myself, CNC providing good repeatability for timber lures and no plastics to mess around with...now I have built a wood router that is quite capable of doing this, and am now looking at selling my reef fishing boat to pay for a machining centre to retrofit and make more CNC stuff - this stuff is quite addictive

Russell.