Hello Guest it is March 28, 2024, 11:15:12 AM

Author Topic: Denford Orac problem  (Read 29011 times)

0 Members and 1 Guest are viewing this topic.

Re: Denford Orac problem
« Reply #20 on: August 25, 2006, 06:11:40 AM »
Hi John

As it happens I'm going down the RS232 front! I've got an AVR microcontroller that does all the work, Mach sends a code to it, the AVR does it's thing monitoring the sensor states and switching motor outputs and then returns a selected tool to Mach to ensure the correct one is selected. Only had it on the bench using switches but it seems to work ok so far. Now waiting for my mate to rebuild his toolchanger so I can start testing!

Cheers

Dave
Re: Denford Orac problem
« Reply #21 on: September 02, 2006, 12:31:25 PM »

Hi, Well I got the chuck and after a few problems I fitted it, It has reduced my vibration problem about 50%. Now the vibration is just the tool and not the whole machine.

I have made a video which is about 10 meg and can be downloaded here, www.jtechdesign.co.uk/orac/lathe3.mpg
Please note the video and sound my not match exactly as I had to do them separately.

I've used different tools at different heights to the center and they all do the same to a lesser or greater degree, The only difference since I changed the chuck is the shoulder of the piece shows no chatter marks and looks well turned.

I'm starting to think that it may be something other than the machine it's self.

I have noticed that there is a delay before the retract from the cut allowing at least 5-10 revs of the spindle before retracting.

On my Boxford it retracts off the cut immediately

I'm wondering if, because the tool is just sitting there, it is just rubbing on the radius of the tool nose.

Why is there a delay between commands, I'm turning only using G1 and G0 commands.

Could it be that I'm using a 800mhz computer and it's just not quick enough?



Any idea's are welcome as I am almost at whits end.


Kind Regards
John
Re: Denford Orac problem
« Reply #22 on: September 02, 2006, 01:47:13 PM »
Hi John

Just had a look at the video - firstly have you tried dropping the spindle speed - we had a chatter problem and found that dropping it cured the problem.Can you give us an idea of speeds, feeds, depths of cut, diameters, material etc?

Secondly - headstock bearings? Have you checked them - do they need tightening?

Thirdly do you have backlash enabled? If so Mach will dwell as it applies the backlash to draw the tool out - from memory you can set this as a % of the max feedrate. Also if you are running with no backlash have you got constant velocity enabled?

I may be able to try the same test on the one at work-allthough that has the original control system in it.

Cheers

Dave
Re: Denford Orac problem
« Reply #23 on: September 02, 2006, 02:37:29 PM »
Hi, I've tried several speeds and feeds, the current program is below.


Headstock bearings? I've thought of that and tightened the end nut as tight as I could, I lost 200 rpm off top speed and It made no difference to the problem. I've since loosened it off one tab notch although the top of the headstock still gets warm pretty quick. I have tryed checking for radial play, cant perceive any.

I have listened to the spindle at 1000 rpm taken the front cover off the headstock to reveal the bearings and I can hear this sound (www.jtechdesign.co.uk/orac/SPINUN.wav)

You can just hear the races jingling in the background, I don't know if this is an indication of wear or not.


I've taken backlash off totally today and it made a slight difference to the delay.


I had been working in dead stop mode and changed to con-velocity and this has made no difference.


I'm working in stainless steel but I've also put the same diameter 5/8" in free cutting steel and although it's less pronounced I still get the same effect.




Cheers for the reply.

John





part code
--------------------

G18 G90 G94 G80

M09
M06 t0101
G0 Z0.00
M08
M03 S1100


G0 X17.6 Z0.8
F200
G0 X15.6
G1 Z-34.9
G0 X17.2 Z-34.2
G0 Z0.8
G0 X15.2
G1 Z-34.9
G0 X16.8 Z-34.2
G0 Z0.8
G0 X14.8
G1 Z-34.9
G0 X16.4 Z-34.2
G0 Z0.8
G0 X14.4
G1 Z-34.9
G0 X16 Z-34.2
G0 Z0.8
G0 X14
G1 Z-34.9
G0 X15.6 Z-34.2
G0 Z0.8
G0 X13.6
G1 Z-34.9
G0 X15.2 Z-34.2
G0 Z0.8
G0 X13.2
G1 Z-34.9
G0 X14.8 Z-34.2
G0 Z0.8
G0 X12.8
G1 Z-34.9
G0 X14.4 Z-34.2
G0 Z0.8
G0 X12.4
G1 Z-34.9
G0 X14 Z-34.2
G0 Z0.8
G0 X12
G1 Z-34.9
G0 X13.6 Z-34.2
G0 Z0.8
G0 X11.6
G1 Z-34.8974
G0 X13.2 Z-34.0974
G0 Z0.8
G0 X11.2
G1 Z-34.8762
G0 X12.8 Z-34.0762
G0 Z0.8
G0 X10.8
G1 Z-34.833
G0 X12.4 Z-34.033
G0 Z0.8
G0 X10.4
G1 Z-34.7664
G0 X12 Z-33.9664
G0 Z0.8

Re: Denford Orac problem
« Reply #24 on: September 06, 2006, 05:41:51 AM »
Hi John

This doesn't seem right at all - is the chatter only at the shoulder or on the whole length?

Do you get chatter with all tools - tipped and HSS ? We had a problem where we had some tips that were slightly different and we couldn't get them to cut as well.

Presumably gibs etc have also been checked?

The previous work I have done in EN24T was at around 1000RPM, 50mm/min feed and upto 1mm cut with no coolant - this was on pieces around 12mm diameter.

I pass your video on to a couple of people and see what they think.

Cheers

Dave

How long do you estimate the delay to be? Seconds?

Re: Denford Orac problem
« Reply #25 on: September 06, 2006, 06:26:45 AM »
Hi David, I've started new topic in the main Mach3 forum as I think my problem is with the software/computer rather than the machine. The chatter only occurs at the end of a cut, The machining is actually very good until the end. This is happenning with all tools I've tried. Changing the chuck has reduced it from vibing the whole machine.

I have noticed that at the end of the cut when I command a G0 rapid off (about 0.5mm) just to clear tool, The machine just sits there for about 1 second before moving.

I have changed the G0 rapid to a G1 instead and this has reduced the problem by about 50%. Now it is at a level which doesn't offend my ears.

I presume that the delay I am experiencing is not due to software but my computer which is only a 733mhz.

I'm hoping to upgrade it to 1.2-1.4ghz and hope this sorts my problem.


Cheers


John
Re: Denford Orac problem
« Reply #26 on: September 06, 2006, 06:33:03 AM »
Hi John

That was my suspicion but it can be difficult to narrow things down - I showed the video to a friend and he made the following comments :-

"Clearly it starts to chatter when the feed/spindle speed is reducing suggesting
that it chatters when the tool load is reducing.

I would suggest he uses a tip with a smaller tip radius. What tip style is he
using? Bear in mind that some tips are designed to be set over or under centre
and will cause chatter when set ON centre.

What spindle speed is he using? It looks to me like 20mm diameter so I would
suggest a small lathe with a tipped tool would need to be run at 600 - 700 rpm.

Is the bed of the machine level and tightened to a beafy bedplate? My
Colchester used to chatter like hell until I carefully levelled it up.

A few ideas at least."

With regards to computers, my mill runs on a 1.9Ghz machine with backlash and the only delay is when the backlash is taken up as would be expected, apart from this it barely stops in 'dead stop' mode.

What version of Mach are you running? Have you checked the Pulse frequency and the 'time in int' values in the diagnostics?

Cheers

Dave
Re: Denford Orac problem
« Reply #27 on: September 06, 2006, 11:03:41 AM »
Hi Dave, I'm running mach3 at 25khz. I've tried 35-45khz but that just crashes the computer. 

I can't find the 'time in int' value you mentioned but the step pulse is set to 0 in motor tuning. I'm using Gecko's for the drives.

I can understand what your friend was saying, I'm still guessing it's the speed of the computer. The tip I was using in the video was 0.4mm radius TPUN type and the material is stainless 5/8" bar, I've tried altering speeds and feeds from 700rpm to 1200 rpm and all the tools I've used including an HSS one I ground myself make the same chatter at the end of the cut. I say chatter, It's more like vibration like the tool is rubbing that ends up as chatter. Noticing that there is an obvious visual delay between commands whereas my Boxford with it's original hardware/software will run the same program with no discernable delay at all has led me to assume that this is where the problem lies. The tool is sitting there rubbing the stock until the next command is executed.


Because my computer is a Compaq in order to upgrade I have to buy an adapter from China to fit the later processors.

I'm hoping this will cure the problem.


Cheers

John

Re: Denford Orac problem
« Reply #28 on: September 06, 2006, 02:38:05 PM »
Hi again, I have had a good idea, I have spent all afternoon configuring a copy of TurboCnC to work on the machine and I've run the same g-code. I have no problems at all with TurboCnC running, in fact I would say the machine is a lot smoother and quicker with no waiting between commands. Running in a Dos environment has some benefits.


So now I have the answer. Is it a Mach3 configuration problem or Mach3 itself or because I'm running a 733mhz computer. By the way I have no other software running other than Mach3 on Win 2000.


Cheers
John
Re: Denford Orac problem
« Reply #29 on: September 07, 2006, 03:56:03 AM »
Hmmm - sounds like you've got the problem narrowed down. Probably worth running this past Art and see what he says.

Cheers

Dave