PDA

View Full Version : my project build, need help please



jpcnc
26-07-2010, 07:20 PM
Heres some pictures of a prototype machine I built last year which I am re-developing and building again.
http://img40.imageshack.us/img40/6710/newebay012.th.jpg (http://img40.imageshack.us/i/newebay012.jpg/)
http://img39.imageshack.us/img39/1272/newebay011.th.jpg (http://img39.imageshack.us/i/newebay011.jpg/)
http://img154.imageshack.us/img154/3752/newebay010.th.jpg (http://img154.imageshack.us/i/newebay010.jpg/)
http://img39.imageshack.us/img39/673/newebay009.th.jpg (http://img39.imageshack.us/i/newebay009.jpg/)
http://img36.imageshack.us/img36/9951/newebay008.th.jpg (http://img36.imageshack.us/i/newebay008.jpg/)
http://img32.imageshack.us/img32/5640/newebay007.th.jpg (http://img32.imageshack.us/i/newebay007.jpg/)
http://img30.imageshack.us/img30/8968/newebay006.th.jpg (http://img30.imageshack.us/i/newebay006.jpg/)
http://img29.imageshack.us/img29/2808/newebay005.th.jpg (http://img29.imageshack.us/i/newebay005.jpg/)
http://img20.imageshack.us/img20/4529/newebay004.th.jpg (http://img20.imageshack.us/i/newebay004.jpg/)
http://img40.imageshack.us/img40/2192/newebay003.th.jpg (http://img40.imageshack.us/i/newebay003.jpg/)


The cutting head in the picture is just a mock up so ignore that bit, what I want to do is make the 2 axis run off a plc or simular. It is basically a lathe to reduce the diameter of the tyre. It needs to take around 4mm off in .5mm cuts.
Ive found a company on ebay that sells the rails, bearings and ballscrew/nuts at a reasonable price, can anybody recomend a uk company thats as cheap?
The main issue I need help with is the plc/cnc side of it. Ive been working cnc's for about 14 years now, but I havent got a clue how that side of it works? I already have orders for machines and need to get cracking asap, and bring it in as cheap as possible.
So any ideas are very welcome.
Thanks
Jason
www.jpcnc.co.uk (http://www.jpcnc.co.uk)

GeorgeD
26-07-2010, 08:43 PM
Aside from your quest,isn't it illegal to reduce a tyres wall thickness and retread it if the thickness is reduced?

jpcnc
26-07-2010, 09:09 PM
Im not re-treading it, its for track tyres, they come new with 8mm of tread but for optimum performance track-dayers / racers will drive hard in practice to reduce them to around 4mm, but they can easily wreck the tyre in the process

GeorgeD
26-07-2010, 09:22 PM
Doesthis work? ie I would have thought it would be far to hard to skim rubber without it becoming rough in the process.

There are two places for cheap drivers and motors and cnc components and thats.......

Zapp Automation who's linkyou find here on the site.
The other is ArcEuro Trade,however this site only sells Steppers and their drivers but are cheap.

I wouldn't buy off ebay for mechanics of the build as I thought it was cheap once,its not unless you mean China?

Tom
26-07-2010, 09:32 PM
The machine looks good.

2 axes of motion, plus a motor driving the tyre.
For linear guides you could probably look up the IGUS linear guides. Cheap, and drylin W especially has little play with the low forces you'll be using to cut.
Drylin N http://www.igus.co.uk/wpck/default.aspx?pagenr=1969
Drylin W http://www.igus.co.uk/wpck/default.aspx?pagenr=1994

The more interesting part of the project is the need for the machines to be operated by unskilled people (by that I mean people that didn't build the machine). A PC running mach3 is unlikely to be robust enough.

You'll need:
1) Safety (guards, big red e-stop buttons, etc)
2) A black box to store the program, give step and direction commands to the motor drives, and interpret operator button pushes (go, stop, jog cutter, etc).
3) Some drives to drive the motors.
4) Motors (probably steppers).
5) Power supply.

1) You can do.
2) This is the most interesting bit.
A Siemens LOGO PLC is too simple (slow) for your needs.
http://www.automation.siemens.com/mcms/programmable-logic-controller/en/logic-module-logo/modular-basic-variants/Pages/Default.aspx
But this Galil PLC may be over spec (still only 300USD tho)
http://www.galilmc.com/products/rio-47xxx.php
You can find a PLC somewhere between the two. The main constraints are probably the amount of I/O (for buttons, etc), and the need to find one fast enough to send step pulses for the motor drives.
3) Easy once you know what motor sizes you need.
4) Easy once you have an idea of the cutting forces involved.
5) Easy once you know what boxes you've bought to do all the above.

Hope it helps. Basically you need a PLC. lol Give me a shout if you need a hand getting something up and running.

irving2008
26-07-2010, 11:00 PM
For something this simple a PIC based box, like Steve's rotary controller (http://www.mycncuk.com/forums/showthread.php?150-CNC-Rotary-table-lathe-head-indexing)with different firmware would be the answer.... simple display with ability to program in the number of cuts and depth and some constants such as tyre acceleration etc...

Swarfing
26-07-2010, 11:31 PM
Given that you will always have a know centre for the wheel a simple touch off to catch the diameter would mean your depth of cut will be accurate with Irvings idea also?

irving2008
26-07-2010, 11:59 PM
yes, i was thinking one of those cheap digital angle protractors would work as an arm to measure the outer diameter, easy enough to read the angle and convert to a radius as all the lengths are known and its just a case of doing the math. You could put an index pulse on the tyre spindle to measure speed directly...

Ross77
27-07-2010, 12:02 AM
Sticking with the PIC idea couldnt you just have limit switches at the end of each side? you could even ditch the cnc bit and just have a mechanical advance at the end of each travel, simple

GeorgeD
27-07-2010, 12:39 PM
No need to advance into the CNC side,a wide belt sander will alleviate all the headache. :-)

Tom
27-07-2010, 04:25 PM
Yep, this lot are right... Keep it simples.
Still, there is something to be said for a modular off-the-shelf approach, especially if in the beginning you need to get them out the door. You could always reduce selling price (or increase profit) later by making some simplifications, and adding a bit of soldering work to the build.

jpcnc
27-07-2010, 04:27 PM
thanks for all the input, as for the belt sander, this is a no-go as it generates too much heat and wrecks the structure of the rubber.
For the overall design ie cnc / plc i did think of a mechanical advance but it started getting a bit complicated, I lay awake last night and have actually come up with a totally new design which is much simpler and something I can actually pattent. I will keep you posted as I try and get it drawn up. I still need some good quality linear rails though so il look into the links provided. Cheers

GeorgeD
27-07-2010, 05:00 PM
The way I see it whichever way you do it,rubber is a sod to machine with a view of getting a smooth surface on the face...thats why I asked does it work ie have you tried a mock cutting surface of the rubber?

jpcnc
27-07-2010, 05:12 PM
it is hard to machine i agree, ive done trials on castors which went quite well, but the compound was harder, im gona be speaking to our tool rep next week, hopefully he will help out.
A simular machine in america uses a rotating carbide blade angled about 45 deg to the wheel, this just peels it of with ease. I intended on something simular in my old prototype, but im hoping to simplify it and remove the motor and cost with a single point tool, something very sharp, lots of rake etc and apparently fairy liquid and water is a good cutting fluid ( aside from freezing it)

irving2008
27-07-2010, 05:51 PM
thanks for all the input, as for the belt sander, this is a no-go as it generates too much heat and wrecks the structure of the rubber.
For the overall design ie cnc / plc i did think of a mechanical advance but it started getting a bit complicated, I lay awake last night and have actually come up with a totally new design which is much simpler and something I can actually pattent. I will keep you posted as I try and get it drawn up. I still need some good quality linear rails though so il look into the links provided. Cheers
If you plan to patent anything the LAST thing you should do it mention it in the public domain!

GeorgeD
27-07-2010, 06:00 PM
Patent for someing can cost as little as £3 or as much as £5000,thats before you do research into whether the idea as already been put on the market somewhere in the world?

In a nutshell it ain't worth the headache if you don't have the resources.

jpcnc
27-07-2010, 06:15 PM
Thats besides my point, lol all I want is advice on the build

GeorgeD
27-07-2010, 06:42 PM
Well you now have the chance to be more specific on what you want.
You said
I've worked on CNC fo 14 years.
I know nothing about CNC but by looking and listening have gained enough to put one together,it might fail it might not but surely will work even if the steppers are only dancing to a few spikes from the computer lol

I'm thinking your setup is basicallylike a mill type of CNC

jpcnc
27-07-2010, 06:54 PM
Im thinking of driving my cutting axis with a motor aswell as a manual overide handle, instead of being cnc controlled i just want some kind of variable control over the feed rate. This is now going to be the only moving/working axis. I still need very sturdy linear rails. Any ideas on the motor setup? this will all run of 240v by the way.