MechMate CNC Router Forum

Go Back   MechMate CNC Router Forum > Common Assemblies & Parts > Rails & Rollers
Register Options Profile Last 1 | 3 | 7 Days Search Today's Posts Mark Forums Read

Reply
 
Thread Tools
  #1  
Old Fri 16 July 2010, 04:51
Surfcnc
Just call me: Ross #74
 
Queensland
Australia
Question Regarding The Methods Used for Shimming Rails

I have read much of the forum and wanted to ask "exactly" how people were setting the height of the rails with the thin metal shims.
Nearly all threads talk of using shims but none that I can find actually explain the process itself.
Specifically what measurement instruments / processes are used.

A dial gauge for example only measures from the top of the x beam to the top of the rail.
Given this the rail can be "very accurately" set to within parts of a mm to actually follow a bent X beam ??
Using the string line or wire method, a measurement is still required to set the rail height.
I am a little doubtful that a "movable object" like a wire or string provides a stable enough reference point to measure using a vernier (as a possible example) to an accuracy of .1mm.

The last issue I would like to inquire about is... Do builders ensure the rails are coplanar, that is, one straight and level rail is straight and level with respect to its partner on the machine.
I have seen no mention of this process at all but suspect that given the skill level of some of the builders it must have been considered before.

I am foxing to a degree as I do have a method to ensure all the above issues are fully considered but it is expensive.
In summary I would appreciate it if some of the forum gurus can enlighten me.
Enlightenment unfortunately does not include tall tales of old fashioned trademanship or bold statements about the accuracy of your setup with out a sound underlying method to support those statements!

Thanks in Advance
Ross
Reply With Quote
  #2  
Old Fri 16 July 2010, 06:19
KenC
Just call me: Ken
 
Klang
Malaysia
Quote:
Originally Posted by Surfcnc View Post
Enlightenment unfortunately does not include tall tales of old fashioned trademanship or bold statements about the accuracy of your setup with out a sound underlying method to support those statements!

Thanks in Advance
Ross
You just took out the fun....

Use the string & shim method. It works. #0~#62 + a few more fold in unreported build, is a statement good enough for you?.
Reply With Quote
  #3  
Old Fri 16 July 2010, 06:40
Surfcnc
Just call me: Ross #74
 
Queensland
Australia
Hi Ken

We both have measured the thickness of our respective countries Coke cans with our vernier calipers it seems.
Got to tell you I loved drinking all that Coke too (Diet so as to not interfere with my figure !!).
You make a valid point about how the other machines have been set up.

....But today I'm fishing for a technique or method of measurement other than sighting it in.

Regards
Ross
Reply With Quote
  #4  
Old Fri 16 July 2010, 07:04
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Used a long alu extrusion as a reference, then feeler gauged between the extrusion and the rail. (extrusion checked for straightness by seeing if all faces give the same readings). Sag due to gravity was of no concern, as the workpiece would do the same - and bounce back if needs be.
Reply With Quote
  #5  
Old Fri 16 July 2010, 15:19
Alan_c
Just call me: Alan (#11)
 
Grabouw (Western Cape)
South Africa
Send a message via Skype™ to Alan_c
Ross

I used the taut string method (actually fine fishing line) I made brackets to hold the line with a mechanism to put plenty of tension to mitigate sag. At each end I used carefully selected shims (rollers from a very large roller bearing) to space the line from the top surface of the rail and a third one to check the gap as I placed shims around the bolts to get the rail straight. (Oh yes I did the side to side alignment first to get the rail straight horizontally and then on the top to adjust the vertical alignment)
Reply With Quote
  #6  
Old Fri 16 July 2010, 17:07
David Bryant
Just call me: David #99
 
Western Australia
Australia
Hi Ross
I am in the middle of the same process. My side rails have about a mm high spot in the middle.
I got a flat Bastard file onto the beams untill the 75X30 Al extrusion sat flat in the Ydirection all along the rail. Took under an hour. I figured this would ensure my rails were vertical.
The Fishing line and magnifying glass showed me off straight but was too small and lightly loaded to allow measurement with a feeler dauge for height changes.
So I went shopping for piano wire and have not got any that is long enough yet...

Due to my bow up in the middle of the first rail the 75X30 extrusion was rocking and not allowing measurements under it, so I am just planing.
I think I will set the 75X30 along the x rail raised by spacers at the 25% and 75% points. I will then measure the gaps near each bolt hole using a set of number drill bits/feeler gauges. I will then calculate required shim heights to use between my Al Rails and the X Beams. I will set up the straightness with the fishing line and flexing the rail as I tighten the bolts to get it straight.

Not sure if I can then use the gantry with two rollers on the fixed rail and one on the second rail as the gauge to set the two rails parallel. The V roller gets a small gap between on part of the tyre and the rail if it is pushing in or pulling out.

Still thinking about how then to ensure the gantry is then equal height.

Costed the Bosch GRL150 HV set rotary laser with a sensor that shows 0.1mm I thought I would bolt both rails tight set laser up in the middle of the table. read rail height variations. Shim. Test again and set rails straight by string and sight glass.
Repeat above for gantry rails.
The laser is $1 000. And none of the local hire shops have one.

Maybe I just think too much and should just bolt it together so it runs smooth as other have done....

Cheers
David
Reply With Quote
  #7  
Old Fri 16 July 2010, 19:42
Surfcnc
Just call me: Ross #74
 
Queensland
Australia
Thanks everyone

The methods all seem to use the gauging technique with some variation in what to gauge from. All makes good sense so thanks for you input.

Any other good ideas are always welcome.

Regards
Ross
Reply With Quote
  #8  
Old Sat 17 July 2010, 02:03
KenC
Just call me: Ken
 
Klang
Malaysia
Quote:
Originally Posted by Surfcnc View Post

I am foxing to a degree as I do have a method to ensure all the above issues are fully considered but it is expensive. !

Ross
Care to share your method?
Reply With Quote
  #9  
Old Sat 17 July 2010, 02:54
Surfcnc
Just call me: Ross #74
 
Queensland
Australia
Hi Ken

I guess I'd better now I have started the discussion.

I was considering asking a surveyor to check the rails using a laser measuring instrument called a Total Station.
Due to the professional nature of the instruments usage, a working instrument is also required to be periodically calibrated to ensure their accuracy.

The Total Station has a advantage over an automatic level (often referred to as a dumpy level) or laser levels because they are able to check the rails for straightness as well as level in the one setup.
The machine can perform this magic by using the horizontal and vertical angles it records and mathematics (Trigonometry) to calculate the answers.
Shinny things like rails as especially suited to the reflector less technology in the newer machines.
In the past I have had the opportunity to see these machines set out bridges, tunnels, roads, locate underground services etc etc and have a great respect for what they can do in the hands of an experienced measurement professional .

My question about the shims is to see how accurately I can set up the machine first, then if I felt the need to then have the surveyor come in and check my results.
As I have said privately to others on the forum, if a surveyor was to check my results I am fully prepared to be humbled by the accuracy they can achieve.

Regards
Ross
Reply With Quote
  #10  
Old Sat 17 July 2010, 09:28
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
You can make yourself a small rail-rider laser to project a dot about 10 or 20 meters away.....if the dot stays in the same place then the rail is straight.....
Reply With Quote
  #11  
Old Sat 17 July 2010, 12:46
MetalHead
Just call me: Mike
 
Columbiana AL
United States of America
What are you trying to accomplish? Horzontal, Parallel and Vertical allignment?

You can do the piano wire method to get the rail straight. You can use body shims (Auto / Fender Shims) or the "beer can shims" to get the height of the rail correct. If true level on the rail is a major issue make yourself a water level out of a long piece of clear hose with colored water in it.

Get one side where you want it in relation to horizontal and vertical. Then repeat for the other side adding the parallel setup.

You could build a rail guide with two V rollers and one of the X Gantry tubes and some thread-all if you have not already welded that part up if using a marked wire or measuring tape is not close enough. You could also use a piece of tubing or a 2x4 even. Just something stiff to hold the rail width demension.

It is all relative anyway. Are you setting up true to the build? .. Ie from the bottom of the table leg without the feet installed? Because if you are doing all of this in relation to your floor (which aint level) then I hope this is the final resting place for your MechMate. A move of even an inch any direction can cancel out your whole leveling effort if your basing this from the floor.

One of the greatest things about this design is it is not a Porsche... it is a Volkswagon. Very forgiving to these types of adjustments.
Reply With Quote
  #12  
Old Sat 17 July 2010, 12:55
MetalHead
Just call me: Mike
 
Columbiana AL
United States of America
Ok on futher reading I see you are talking about veneer.

I am starting to see why your talking about this.

I still think using a water level would work. I also think one of those laser levels that you buy to do trim work or the one the pros use to do drop ceiling would also get you where you want to be.

I guess you would also want your table bed close to this plane also. I know you take a leveling pass to start with once the spoil board is in place.

I would also be interested to see if anyone is holding .001 across the who table. Hard thing to accomplish.
Reply With Quote
  #13  
Old Sat 17 July 2010, 19:11
David Bryant
Just call me: David #99
 
Western Australia
Australia
Hi Ross and Mike
I started with the water level in 4mm diameter tube (slightly wider is better)with a cross connection in the middle so I had four tubes one at each corner. This set the corners within a +-1mm. I was taught to only use an inclined manometer if I was expecting better than +- 1mm accuracy. So was sighting through the clear plastic tube. If you go this way some Black/white diagonal lines behind tube makes the level easier to see. Make absolutely sure that there are no bubles of air in the tube.
This got me roughly flat, but my table is currently on paving not concrete. So I am noting the advice about moving...
Watch the laser specs because some of the moderate cost ones are not that accurate (Worse than the water tube over 4m!) Hence why I dug up the spec above for the Bosch.
Cheers
david
Reply With Quote
  #14  
Old Sat 17 July 2010, 23:01
KenC
Just call me: Ken
 
Klang
Malaysia
Quote:
Originally Posted by metalhead View Post
one of the greatest things about this design is it is not a porsche... It is a volkswagon.
well said!
Reply With Quote
  #15  
Old Sun 18 July 2010, 03:58
Surfcnc
Just call me: Ross #74
 
Queensland
Australia
Hi All

Water you say - What a coincidence as I spent all morning in the water today, what a cracker on our Queensland coastline today with nice waves and super clear water.

I use the water level regularly and generally it gets things better than a poorly setup laser but takes a bit longer AND you got to yell at any kids/ dogs/ helpers that step on the pipe just as you are about to make your mark, haha.

David - had to Google the inclined manometer still a little vague on that one.

Gerald's idea of throwing a laser spot at a distance is appealing to me as it amplifies error on the rails into more easily managed (larger) measurement units.

Thanks everyone this was exactly the sort of options and discussion I was trying to extract from the forum users.

Regards
Ross
Reply With Quote
  #16  
Old Sun 18 July 2010, 14:51
David Bryant
Just call me: David #99
 
Western Australia
Australia
Ross
Dont try an inclined measurement for this. Very old school pressure/ level measuring. It was used by my tradesman to teach me the range of accuracy of a water manometer/level. (also dont expect better than +-1mm from the water level. Unless you go to glass bore not plastic and view the maniscus in the middle)
I liked the laser/ rail idea as well. I have the gear to do it, except I was thinking as a final test to put the laser on the gantry or the car and get an indication of the total setup accuracy.
Hmm Just thinking out loud. I have a cheap laser level with a glass vial that shoots a laser out one end, maybe a pair of bearings at each end mounted in angles to follow the v rails and a weight that hangs from the level and goes under the rail to keep it vertical.
With the target for the laser at a known distance onto a concentric circle target the overall waviness is measurable.

I might have a go at that while I get delayed for a few weeks as I have the Concrete guys comming in next monday to put concrete where the pavers are under my table.
cheers
David
Reply With Quote
  #17  
Old Sun 18 July 2010, 18:24
Surfcnc
Just call me: Ross #74
 
Queensland
Australia
David

Don't know if I'm more impressed with your understanding of the laser level technique or you having the good sense to remove the wobbly pavers !!

If you do the test with the laser, a photo of the rig would help and leave a record for others to see.
I follow everything except the weight you mentioned.

Regards
Ross
Reply With Quote
  #18  
Old Sun 18 July 2010, 18:42
David Bryant
Just call me: David #99
 
Western Australia
Australia
Hi Ross
Yes the pavers should have gone years ago.
Think of the laser being the top of a coat hanger above the rail. If a coat is on it (the center of gravity is below the rail) then it will hang from the rail and hold the laser stable. If no coat then there is the laser balancing on the rail (center of gravity above the rail and be unstable).
Cheers
David
Reply With Quote
  #19  
Old Mon 19 July 2010, 06:17
David Bryant
Just call me: David #99
 
Western Australia
Australia
Quick laser mount

Hi Ross
I roughed a car for the laser.

V roller at each end

weight is about half the weight of the gantry.

Weight holds laser upright, pulled with string, observe laser 20m away (at night due to weak laser.

Scrap Aluminium angle was used for frame of trolley.
Sorry poor qual photoes
David
Attached Images
File Type: jpg Pic_0719_643.jpg (53.0 KB, 490 views)
Reply With Quote
  #20  
Old Mon 19 July 2010, 06:23
David Bryant
Just call me: David #99
 
Western Australia
Australia
second photo on rail

Ah I had to rotate photo to fit the limit
Attached Images
File Type: jpg Pic_0719_642.jpg (65.9 KB, 492 views)
Reply With Quote
  #21  
Old Mon 19 July 2010, 06:35
KenC
Just call me: Ken
 
Klang
Malaysia
how do you ensure your datum is accurate?
Reply With Quote
  #22  
Old Mon 19 July 2010, 06:46
Surfcnc
Just call me: Ross #74
 
Queensland
Australia
Hi David

A very creative solution. I had already guessed a night time session would have been needed for a 20m throw on a low powered laser and of course on cue you have mentioned it.

The issue of where the measurements are being taken and how they relate back to the individual bolts being packed should not be as problematic as I first imagined. The rig will tell you if there is a variation and it is up to us to find which of the two or three possible bolts need a tweek. Reducing the wheel base of the rail car further and aligning the laser output directly over one wheel might further tune what seems like a already great idea.

In the end the quality of the measurements you can infer back from the target to the machine will be the acid test for the device, so if you would share those findings when you have time, it would be appreciated.

Just saw Kens post come in as I was typing, we are both thinking datums it seems. My suggestions are designed to address this to a degree.

Regards
Ross
Reply With Quote
  #23  
Old Mon 19 July 2010, 07:32
David Bryant
Just call me: David #99
 
Western Australia
Australia
Hi
Datum is jst the "average" of the path that is drawn as the trolley is pulled via a sting gently from one end to the other
I dont think the position of the laser is important as the setup Indicates waviness along Z and Y. I aimed at wheels about the width of the gantry rollers as this is the most significant waviness
Maybe this is best to test the first rail and the second with the gantry with the laser at the second rails end. Extra weight at the first rail end
Any change in "parallelness" of the rails would be detected by the height increase.

This trolley wont detect a some of the twist modes (Think two straight rails but one on a slope.)
I like the idea of a good detector measuring the height of the rails refferenced to the middle of the table.



What this effort has pointed out to me is how neat the design is. For anything thin (Mdf sheet etc) most of the errors are averaged out.
I am just thinking about how I can show my setup is within +-1mm x,y,z, just for fun
Reply With Quote
  #24  
Old Mon 19 July 2010, 07:53
Surfcnc
Just call me: Ross #74
 
Queensland
Australia
Had a quick look at the instrument side of things and the best that was around at a reasonable price is a hunting rifle 5 -10 mw green laser dot. The slightly more expensive and more powerful hunting rifle green lasers had a reported 1.2mm beam spread at 100 yards.

David I see you have quickly jumped from aligning a rail to the whole machine.
In the end it is how the machine as a whole measures up.
I can already visualise you attaching the laser to the cutter tip and driving the gantry and car around to get an overall picture of the machines accuracy.

Regards
Ross
Reply With Quote
  #25  
Old Mon 19 July 2010, 08:20
David Bryant
Just call me: David #99
 
Western Australia
Australia
Ok I had already thought about the angle block this would attach to but think about the effort to re shim.... You would know the offset but when would you move the Z, height of the gantry rails height of the x rails ARRGHHHH
So I figured
Start with one rail straight and flat....

This got me to the Bosch laser detector with 0.1 mm resolution that could be used on all four rails, and the Z axis for vertical...

Gee it would be so easy to just cut flat things and no laminates or three D

Beam spread is not a big issue If you get a say 10mm dot just put a white card up and get a 10mm disk with a small hole in the middle. allign disk with dot and mark card through small hole with sparpie. even a small laser pointer and hot glue with paperclip to hold it on.

Ps dont think about thermal expansion with AL rails and steel V guides
Reply With Quote
  #26  
Old Mon 19 July 2010, 11:20
smreish
Just call me: Sean - #5, 28, 58 and others
 
Orlando, Florida
United States of America
Okay, my 2 cents.
I leveled my machine without the rails on it. Bolted the sections down, did a horizontal alignment for the gantry to move nicely - and that was it.

I was lucky or just good fabrication method, but no shims were necessary for my first machine.

...for the horizontal alignment I used a digital vernier caliper off one face of a secondary straight edge (aluminum tube section) for setting up one straight line, then a story pole to gauge the rail on the other side of the machine.

Took about a 1/2 day for a complete set up.

Good luck.
Sean
Reply With Quote
Reply

Register Options Profile Last 1 | 3 | 7 Days Search Today's Posts Mark Forums Read

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
I have a question saas Introduce yourself and start planning 6 Thu 10 September 2009 08:36
Amperage question shelbygt500buddy 702. Power Supplies 2 Thu 09 July 2009 11:32
One grind rails question. (another alternative rail proposal) isladelobos Rails & Rollers 7 Wed 08 July 2009 08:01
Methods of grinding Z slide dmoore 40. Z-Slide 4 Sun 31 May 2009 06:48
Handmade V-Rails using a baby 4.5" angle grinder - moving rails over grinder Gerald_D Archives 23 Wed 26 September 2007 23:46


All times are GMT -6. The time now is 18:43.


Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.