PDA

View Full Version : advice on upgrade to 5 axis control



alanb
24-07-2012, 09:13 PM
Hi,
I am currently running my (X,Y,Z,C) cnc based rose engine using a 4 axis breakout board connected to the parallel port. I am in the process of adding at least another axis to my machine and possibly a further axis again to make a 6 axis machine. So need to upgrade the electronics. I currently use a standard pc running emc2. Can anyone suggest a good 6 axis board, bearing in mind that if possible I want it to be fairly future proof. I have an inclination to try running the whole show from a raspberry-pi (when I get one ) or alternatively something like the intel DN2800MT motherboard.

thanks

Alan

motoxy
24-07-2012, 09:35 PM
Hi Alan

Bit cheeky but I would love to see a picture of your rose engine. I am a woodturner hobbyist and ornemental turning was one of the reasons I decided to build my cnc router.

Regards
bruce

alanb
24-07-2012, 09:44 PM
go to www.otinexperiences.co.uk for info and click the about this blog link for a picture of my setup

alan

wilfy
25-07-2012, 12:05 AM
that website doesnt work right now.. dont know if you currently working on it or something ?

alanb
25-07-2012, 10:11 PM
that website doesnt work right now.. dont know if you currently working on it or something ?

I was updating some stuff for a few minutes. You must have been unlucky and tried it then.
Alan

m_c
25-07-2012, 11:08 PM
you need to add '/blog' to the above url to get around the Site under development/ads.

alanb
26-07-2012, 12:38 AM
you need to add '/blog' to the above url to get around the Site under development/ads.

Thanks for that and Sorry everyone, I thought I had auto redirection on the main page to redirect to the blog. Dont know where it went but will reestablish asap.

By the way anyone able to answer my original question and give me advice on a good 5-6 channel board to get.

Thanks again
Alan

mocha
26-07-2012, 12:56 AM
you've got some beautiful items there Alan, very nice indeed.

Silly question if I may, on the home page of the blog, the pictures showing the cannibalised cast iron x-y table, does the cast iron assembly not flex the unsupported round rail at all??

JAZZCNC
26-07-2012, 01:06 AM
By the way anyone able to answer my original question and give me advice on a good 5-6 channel board to get.

Have a look here they sell some very nice BOB's and the PMDX 126 is a great board. http://www.pmdx.com/PMDX-126

Edit: Sorry just spotted you use EMC and this boards mainly designed round Mach3,thou it should still work ok.

OR here

http://www.cnc4pc.com/Store/osc/index.php?cPath=33 (http://www.pmdx.com/PMDX-126)

alanb
26-07-2012, 01:18 AM
does the cast iron assembly not flex the unsupported round rail at all??
I have not really tried it out yet together with the new tool post. And it is true that were I starting again I would use 25mm diameter rail and not 16mm as I did, I am sure it must a little. The table will be locked solid when cutting patterns. If you look carefully you will see that I have placed a screw based locking clamp between the two linear bearings on each side. The bearings and rail are just acting as a rapid positioning mechanism for the table. My reason for going with a heavy tool post is to dampen the vibrations I was experiencing. It may be that my real problem is not with the tool post but with flexing of the rail. I will not know until I have run a few test cuts. If there is still a problem then I will consider separating the headstock from the tool post rail and re-engineer tool side using 25mm rail as a minimum. However compared to standard milling, cutting OT patterns is a relatively gentle operation. The max cut you would take is 1mm and more often much much smaller than that. Also the work rotates very slowly up to about 5 rpm max and often at 0.5 rev per minute on finishing / polishing cuts which can be 0.01mm at max.

Alan

mocha
26-07-2012, 12:44 PM
I'd expect that the round rail would to be bowing with all that weight on it and in the bottom edge of image 0277 it looks like it is :-) Of course that might well be the lens of the camera giving that effect?

What sort of length is the rail and weight is the table?

irving2008
26-07-2012, 06:10 PM
Assuming those rods are about 500mm between end supports (based on pic 0068 and the size of the platter) the deflection in the centre is around 2um (2 x 1e-6m) per kg

Won't bow as much as you think because the supports are fixed not simple.

I'd guess that table weighs 25 - 30kg with the slide mounted - so deflection <0.1mm in the middle...

mocha
26-07-2012, 06:47 PM
Assuming those rods are about 500mm between end supports (based on pic 0068 and the size of the platter) the deflection in the centre is around 2um (2 x 1e-6m) per kg

Won't bow as much as you think because the supports are fixed not simple.

I'd guess that table weighs 25 - 30kg with the slide mounted - so deflection <0.1mm in the middle...

Thanks Irving, I was thinking 900+ (based on ruler in pic 0263) and 40-60kgs! The other thing that has thrown me is the feet are not secured to the base...

Seem to recall seeing a spreadsheet somewhere that someone did for unsupported rail, was it yours?... (of course I cant find it!)

irving2008
26-07-2012, 08:34 PM
Thanks Irving, I was thinking 900+ (based on ruler in pic 0263) and 40-60kgs! The other thing that has thrown me is the feet are not secured to the base...

Seem to recall seeing a spreadsheet somewhere that someone did for unsupported rail, was it yours?... (of course I cant find it!)

Yes it was probably and no i cant find it either...

900mm would be 12um per kg, so 60kg = 0.7mm assuming the legs dont rotate which is a good point, but they're quite wide . My 300mm was the length the slide is on from the feet furthest from the headstock to the next set, not the whole length. Its a whole new ballgame if you want to include the full length and the other sets of feet and their rotation. But even then it'll be less than 0.7mm which is worst case (unless you assume simple supports but that would be massively incorrect as well).

alanb
26-07-2012, 10:51 PM
A bit of general information. The distance between the two supports on the headstock is 300mm and between the headstock and the end is 600mm. I just weighed the tool post and it is approx 25kg including the xy table top. The rails are clamped in the supporting 30mm deep x 25mm wide solid aluminium rectangular section. The support legs are solid ali 25mm diameter and bolted into accurately 15mm diameter housings bored in the rectangular section. I thought when making it that I could run some threaded rod between the legs for support, but it has not seemed necessary. The feet are anti vibration pads and sit very firmly on the work surface top. I made it like that because I wanted it to be fairly portable so I could move it into the house over winter as it gets too cold to work in the shed. There at least I can write the gcode and test the outcome using a pencil instead of a cutter (would not dare try to machine something in the house!!). It has got progressively heavier (and me weaker) and this has become a 2 man lifting job. The original design idea was to make the headstock and work holding parts as separate unconnected pieces (and I may do just that at some stage in the future). But I later decided on the current setup cannot remember why. Its final incarnation will be as a six axis machine used to cut patterns in wood and pewter and etch in brass plate. The headstock will move linearly in the X and Z directions and rotate around Z. The tool will move linearly in the Y and Z directions and rotate around Y. I want to keep the two Z movements because the aim is to cut patterns on a profile outline such as the side of a bowl. I think this will simplify my code as the profile outline will be traced by the Z on the headstock leaving the Z on the tool for depth cuts.


Alan

mocha
27-07-2012, 12:04 AM
A bit of general information. The distance between the two supports on the headstock is 300mm and between the headstock and the end is 600mm. I just weighed the tool post and it is approx 25kg including the xy table top.

snip

Alan

Alan, what's the overall length of the round bar between the feet? 900?

The spreadsheet I alluded to earlier is here; http://www.mycncuk.com/forums/showthread.php/2017-Unsupported-Precision-Round-Rail-Calculator/page3

The figures Ross mentioned in and around that post were for "25mm dia. 1500mm span and 250N (25 kg) load the best case deflection would be just over a mm but worst case is 5mm depending on the restraints."

alanb
27-07-2012, 12:50 AM
The tool post is on a 600mm length supported at each end. The headstock is on a 300mm length supported at each end.
Alan

irving2008
27-07-2012, 07:29 AM
The tool post is on a 600mm length supported at each end. The headstock is on a 300mm length supported at each end.
Alan
So, using the actual figures of 600mm and 25kg, deflection is between 0.147 and 0.073mm...