fine tuning - x-axis resolution - HELP!!
You have to admit it is pretty cool just to watch the machine work.
My first goal was to make our company signs, and that led to building a cnc router and that led to EMC...
The math should be pretty simple, you know what your scale is now, then just take your new ratio (10 / 48)
and use that in your formula. And when push comes to shove just make some small moves and measure how far the machine moves and correct your scale as needed.
You also want to try the same with the acceleration.I tried slowing down the x-axis by a multiple of 10
Rick G
Please Log in or Create an account to join the conversation.
I tried slowing down the x-axis by a multiple of 10 and man I have seen the postal workers move faster than that
It was only intended as an experiment, and reducing the acceleration is more important than reducing the speed.
The question which you didn't answer is, did it make things any better?
Please Log in or Create an account to join the conversation.
Sorry Andy I did change both the acceleration and speed settings and found that the granularity did get much better. The .1 read as .097 to .093 after a few runs the 1 and 2 inch check came back with .97 and 1.97 so I would call that much better at least it was consistent. The only issue was anything under .1 was still off hence the decision to reduce the gearing and hope that the problem will be solved. I did notice when I tried .01 the measurement was between .76 an .83 consistently and forget .05 the stepper didn't even move.
Thanks again.
Please Log in or Create an account to join the conversation.
The new gear ratio will give you more torque and better resolution, but unless your problem was lost steps (which seems unlikely to me if long moves were good and consistent) you probably need to look at the existing mechanics.
Also do you have backlash compensation turned on?
If you make a 1" move back and forth say 10 times do the results keep jumping around or do they just keep getting smaller/larger each move?
Rick G
Please Log in or Create an account to join the conversation.
This is the update I have been waiting to send, after moving the pillow blocks to the outside of the shaft as suggested, and changing the ratio on the stepper motor for the X-axis I believe we have success. I created a new g-code file with circles of 4 different sizes and lo and behold all the circles look round and no flat areas. I have enclosed an image so you can see that your advise was taken to heart and the machine appears to be working much better now. The biggest problem I had was the probe top image is in the lower right of the picture. The small circles were oval before and this is where the .25" ball bearing will be so oval would not work. So even tho my calipers didn't show perfection when I tried to check the linear movement of the x-axis the .250" bearing indents were cut close to perfect.
Thanks again for all the help....
Joel
Please Log in or Create an account to join the conversation.
a quick question I believe I have some slop in the y and z-axis using wood for holding the rails is not the most efficient or the most precise. Does anyone know of a supplier or a company that sells pre-made z and y-axis parts? If not maybe a good supplier of extruded aluminum to make my own so I won't have the problem with the wood becoming weak and the roller bearing moving about causing slop. I am thinking of a 16" x 7" piece with 4 slots if possible. When I searched on Google all I got for hits were the typical aluminum pieces. I have come across some the pieces I thought would work but can't seem to find the company again. I believe the slotted aluminum will help with the roller bearing adjustments allowing for better pressure to be applies to the bearings thus removing the slop I am experiencing. I also believe that a good solid piece of aluminum for the z-axis will also allow for better control of the roller bearings on those rails as well, and possibly some hardened rails for the z-axis... Also I am including the pic of the changes made to the pillow block bearings that were suggested. its not the nicest looking fix, but it works and the 1/2" shafts are no longer bowing due to the stress from the timing belts. After looking at Big John's machine I believe that it is the better design.
Thanks again,
Joel
Please Log in or Create an account to join the conversation.
This company do a lot of extruded aluminium section for small router gantry machines etc.
They are in UK, but might lead you in right direction if unable to supply quite what you need.
www.marchantdice.com/
or their ebay shop
stores.ebay.co.uk/Marchant-Dice-Ltd?_trksid=p4340.l2563
regards
Please Log in or Create an account to join the conversation.
do a search on BOSCH REXROTH should also find some one more local , but again here's another in the UK , may give you ideas
www.kjnltd.co.uk/
Please Log in or Create an account to join the conversation.
You might try looking at ...
www1.mscdirect.com/cgi/nnsrhm?cm_re=tpnv-_-home-_-home
and
www.mcmaster.com/#
But there is nothing wrong with using wood with the proper design.
Rick G
Please Log in or Create an account to join the conversation.
John
Please Log in or Create an account to join the conversation.