Messages

61

General Mach Discussion / Re: Wire Alignment

« on: November 26, 2008, 06:58:07 AM »

Thank you Ian

I made up a grinding device and ground it. Having a known reference line and levels allowed me to work to those. This lathe has double v’s for the saddle. The tailstock uses a V and flat. Total run out for the tail stock ways was about .005 end to end not bad for 30 years. I used that for the grinding head guide. The important point was knowing the error I could allow for it as I ground to compensate. It was a slow process.  It is not perfect, but good enough for most of the work I give it. The saddle was bedded with Moglice. Epoxy bearing material No scraping there. Sending it out for a regrind in Melbourne Australia would almost cost more than it would cost to replace the whole machine. With an Asian built machine.  Don’t forget freight both ways for a crane truck. No I would not do another but learnt a lot in the process. 

Cheers

John Mac

62

General Mach Discussion / Re: Wire Alignment

« on: November 26, 2008, 06:30:34 AM »

Hi Hood

In retrospect I was a bit hasty in posting. It would be better if the references to a lathe were left out.

For the record I did use some Roslau piano wire .029 diameter. It appears to be available in most countries. Google search follows: http://www.google.com.au/search?hl=en&q=Roslau+piano+wire+germany&btnG=Google+Search&meta=

I strained the wire between two points and carefully aligned them to the Lathe bed using gauge blocks and used it as a straightness reference.  To load it up a large gear maybe 25 pounds was hung just off the floor to tension it the wire did not break! Ideally it should be tensioned just short of breaking.

 To measure the error, a measuring stage made from a 10 thou micrometer and a back lit slot only a couple of thou bigger than the wire. The stage was then mounted on a kinematicly correct frame two pairs of points on the v and one point of the flat of the lathe.
By moving the stage (carefully) until the slot was centered optically and then reading the micrometer it was possible to get repeatable measurements within about .0002. that cancels out any thickness error of the wire itself but The rosleau had no measurable error with a brand new micrometer. Obviously the slot and wire is not visible to the naked eye but no problem for a cheap student microscope.  Set to about 50 power. Found one second hand for 10 bucks in a trash and treasure shop. The best way is to cut off the base and mount it on a steel support to fit your needs. (Keep the focus adjustment) The slot must not touch the wire. The bed is about 10 feet long with the headstock removed. Straight edges are out.

The horizontal plane is a different matter to check that a 10 second precision machine level (0.0005"/10 ") engineers level was used to level the machine. Buy working from both ends and averaging you can get a good result. For leveling machinery on a tight budget the only way to go. I have seen them on the internet for a couple of hundred bucks.
If you calculate the sag of the wire using the Catenary formula you can also use it to define a horizontal plane.

From reading some of the posts here there appears to be quite a lot of members building their own machines.
The wire method is not new as a browse of the internet will show, it produces outstanding accuracy.

An autocollimator would be nice but for 20 bucks a lifetime supply of wire is pretty cheap. I don’t reuse it. (When unwinding it off the coil make sure it is not kinked.)

The MIT paper Synchronous Proximity Detection for Stretched Wire Alignment Systems By J Paridiso, Link : http://web.media.mit.edu/~joep/papers/WireReadout.pdf
 is very interesting as are many other posts on the internet using wire for the CERN linear accelerator for instance. It deals with reading the position of the wire electrically. That got me thinking? Could we sense the wire using a instrumentation amplifier bridge and use Mach 3 to read the error and position a tool? Even in the x plane if the Catenary formula was used?

Maybe this forum is more software oriented. And the practical side is of less interest to members. This post was made to possibly assist machine builder members on a budget. This method will allow sub thou accuracy over long distances.  Not fast to set up but it works.

Cheers

John Mac

63

SmoothStepper USB / Re: Smooth Stepper board with encoder feedback

« on: November 25, 2008, 10:03:57 PM »

Hi Rbickle

Hi

I found these boards and am considering them? They are stepper in but use brushed DC motors, The encoder is connected to the drive board. The board will accept step and direction instructions from Mach 3. when i phoned and asked if there were any lost pulse issues they said no.  You would have to change the stepper drives as well.

http://cncteknix.com/portal/content/view/77/146/

Has anyone used these boards ?

Cheers

John Mac

64

General Mach Discussion / Wire Alignment

« on: November 25, 2008, 09:26:09 PM »

Wire Alignment
________________________________________
Hi All

I posted this on CNCzone a while back. hence the references to a lathe.
I kept copies of all the docs so if anyone is int drop a line.

This note is about getting things straight…….Very straight.

Some time ago an old but looked after Heidenreich and Harbeck lathe came my way. The bed length is about 2 Metres and weighing at three and a half tons, this is a serious lathe!

Over time there is slight wear on the bed and I am contemplating scraping her back to perfection.

One day I went into a bookshop and found a new copy in the second hand section of Connelly’s book for 20 bucks. Yep I grabbed it. For restoring old machine tools I know none better. I have already used it as a reference. Scraping is not that difficult just hard work. and common sense. I found a couple in the US here

 Amazon.com: Machine tool reconditioning and...
One of the mysteries of alignment is using a tight wire as a straight line.

How to do this this proved to be a bit harder to track down

However the following sites have helped a lot to understand the theory.

http://www.slac.stanford.edu/cgi-wra...-pub-11465.pdf    (EDIT the link is down at the moment)
it is now here:
http://www.slac.stanford.edu/pubs/slacpubs/11250/slac-pub-11465.pdf

http://www.physicsforums.com/archive.../t-220848.html   (EDIT the link is down at the moment)
It is now here:
http://www.physicsforums.com/archive/index.php/t-220848.html

http://www.spaceagecontrol.com/calccabl.htm    (A good calculator)

I guess before the Egyptians they were using a string line

On a more 21st century note here is a link to a MIT
J. Paradiso is still there.

This would make an inexpensive electronic readout system for a tight wire alignment system. As used on linear accelerators. So it is quite accurate.

http://web.media.mit.edu/~joep/papers/WireReadout.pdf

There are other references on wire alignment on Google scholar

http://scholar.google.com.au/scholar...ment&hl=en&lr=


So now I have a plan!! The next bit is to get the time to do it, in the mean time. I would be most grateful if the CNC crowd had any further Ideas.

Cheers

Macka

65

General Mach Discussion / Re: Mach 3 Sewing Machine

« on: November 25, 2008, 07:12:30 PM »

Thank you Hood

The Bosch extrusions are nice.

Alas the straightness is not quite there.

http://www13.boschrexroth-us.com/catalogs/mge50/sections/section%2016.pdf
Page 2. Lengthwise profile straightness tolerance
That shows an error of around 2 mm using a 2 metre lenth, it would be a lot more using a 4 metre length.

Using the deflection calculator Bosch provide on their site. For a 4 metre length of 100 x 200mm section I got .39mm compared to the .159mm for steel (but overall a lot lighter frame)

Easy cheap and accuracy can never appear in the same sentence.

Still there must be another way: Maybe precision laser cut plates assembled into a box frame and welded?  Maybe with small slots cut in line with where the rail would sit to enable you to place the rail against the edge for location. The laser could cut the mounting holes as well.

Has any one tried that?

Cheers

John Mac

66

General Mach Discussion / Re: Mach 3 Sewing Machine

« on: November 25, 2008, 09:52:39 AM »

 Hi Hood

The catch is for a given section size the Aluminium will be lighter but the deflection will be higher. Most of the time by the time you have the stiffness sorted strength looks after itself.  The depth of section would need to be increased to gain the same deflection number.

The other catch is the straightness of the material when I checked the steel supplier the straightness from the mill is specified as within the section length divided by 500.

That means 4000mm /500 = 8mm possible camber. That needs to be corrected before the rails are placed.

Unless the Bosch rails were perfect they would have to be machined also.

Careful inspection at the steel suppliers yard should find some better than standard. (If they will let you in!!)

There is a jobber near us that has an Elgamill I will have to contract out this part
http://www.factoryhub.com.au/DesktopDefault.aspx?ItemID=881026&TabID=202700
Gee I wish I had one of those, very nice.

The rest I can machine myself.

So the frame can be mounted and machined for the rails.

If straight stock is not available then 200x100*9  could be used to allow machining the mounting point That has a deflection of .159mm Still OK  and only 7kg metre heavier slightly better than the 250x150x5 section above actually.

Another way would be to weld thicker pieces on and machine those.

All these compromises are a bear.

Cheers

John Mac

67

General Mach Discussion / Re: Mach 3 Sewing Machine

« on: November 25, 2008, 08:05:52 AM »

Thank you Rich and Hood you guys are quick to respond!

I have been working on the calculations for the x axis.

It will need to be about 4000mm wide
Made from 2 lengths 250 x 150 x 5 RHS spaced 300mm and joined with 150 sq tube and the ends welded over and fully joined.

There will be 2 type 25 rails mounted on each beam giving 4 in total.

Compared to machining or routing sewing has the need to have a tool on the bottom and the top of the cloth hence the need for the large vertical O frame to pass the work through.

I calculated the deflection (Simply supported at the ends only) to be about .211mm with a 40kg load in the centre  (The moving head) and the distributed load imposed by the rails about 10 kg per metre.  The frame has to be really stiff the load is small but there will be resonance. I wish I had better software, and a younger brain.

Not bad for such a large span.

I used a handy tool to do a quick deflection calk called Beam boy.
I found it at:
http://www.engineers-international.com/freeshare.html
It works quite well I am sure there are more advanced tools so if you have one please post a link.

250 x 150 x 5 RHS weighs about 29.9 kg per metre add to that the rails and the sewing head and we are moving about 350kg to 400kg

For a standard 4000rpm sewing machine we need a feed rate of about 233mm per second max. That is a fair amount of weight to push around; it will need some meaty drives. (I think I will drive it in tandem from the ends). Unlike metalworking the positioning accuracy required is around .5mm, (approx 20 thou imperial).

The frame with the sewing heads not included there we need a couple of thou max error along the x axis.

Sizing the drives is not going to be easy for me. If there is software out there in cyberspace to assist I would really appreciate hearing about it.

Usually the first iteration of a new design is doomed to be changed, but so far its looking good.

Cheers

John Mac

68

General Mach Discussion / Mach 3 Sewing Machine

« on: November 24, 2008, 08:33:56 AM »

Hi All

An exciting time.. Over the last couple of years I have kept an eye on Mach 3 from the numerous forum posts both here, the CNCzone and on You tube. Mach 3 has become irresistible.

I work in the rag trade so for me the use of CNC will be a little different to metal working or routing. My interest is in stitching, while the overall system accuracy for stitching is a lot less than precision machining. There are other problems that must be overcome. The system I intend to create will be in the form of large Router table. With the added complication that the sewing axis will require 2 synchronized drives to drive the needle and the hook and thread bobbin separately, (The mechanism under the table of the machine) a simple 1:1 ratio but not constant speed and with a speed range of 500 – 4500 rpm, and this speed must be relative to the tool path speed. Not an easy task from my enquiries so far.  Yes it can be done mechanically but not without a lot more hardware.

My previous experience was retrofitting a CNC sewer. It took a lot of time. The hardware had to be hacked to enable user programming rather than going back to the supplier it was encrypted, or rather obfuscated. (Not uncommon for textile machinery) I also wrote a VB program to feed it with code its native code not G code. The software reads a DXF file using polylines and converts it. The math routines may come in handy again.

Today I would chuck out the hardware and start again.

I would be pleased to hear from any fellow Rag trade members, it would be fun to discuss some of the issues that relate to stitching.

Cheers

John Mac