Messages

1181

General Mach Discussion / Re: XY zero

« on: June 04, 2009, 02:37:53 PM »

Thanks for your answers.  I will try that.

But one more thing.  Is the Offset screen in Mach3 working? I don`t know anything about this VB scripts, but I`ve seen that the Edge Finder buttons has some scripts assigned.

I use the circuit bord as a Z axis zero tool which is connected to GND and port 15 - it works great.

But can I use the same tool to find the X and Y edges using the Edge Finder screen?   And how?

Adam

If you look in the Files section of the group, there is a file "RayLs Probing Macros", or something close to that, that contains all the macros, and a modified screenset, to enable edge, corner, center and mid-point finding, and tool length touch-offs using that page.  With the stock screenset, most of those buttons don't do much more than zero a DRO.

Regards,
Ray L.

1182

Show"N"Tell ( Your Machines) / Re: Success! Mini Machining Center under Mach3 control - Video link

« on: June 04, 2009, 10:27:54 AM »

That is very cool!  And just think how much money you can save by making all your own bolts from now on!  :-)

Regards,
Ray L.

1183

General Mach Discussion / Re: Pulse Frequency

« on: June 01, 2009, 03:43:39 PM »

Thanks Ray! That's the answer I was looking for 

One more question.  You say "most modern PC's will handle 45kHz pulse rate". How do I know what MY PC will handle?  Will I just get bad results from the driver test if I try to go too high?

Bill,

Yup, driver test is the only way to know, and even that may not be 100%.  If driver test says it's OK, try it.  If you have wierd problems, back off and see if they go away.

Regards,
Ray L.

1184

General Mach Discussion / Re: auto backlash compensation setting using home switches?

« on: June 01, 2009, 11:09:14 AM »

Hi,

I have a machine i'm retrofitting (Dyna 2400) which has limit/home switches specifically designed to have no hysteresis, for use in an automatic backlash compensation routine that the original controller had.

Can Mach automatically set backlash numbers during the homing routine using the limit/home switches?  if so, how?

Thanks,
Mike

You *could* write a macro to do that, except I don't believe the backlash compensation values are accessible.  AFAIK, they are stored in the XML, and read in at startup, but there is no programmatic way of changing them on the fly.  Besides, there's no reason to want to set them each time you startup the machine.  Backlash should be stable farily long-term, so you can set and forget.  And, the value you set is usually a compromise, as backlash is rarely constant across the machine travel, so you typically come up with an average value, or measure it at the location where you do most of your work.

Regards,
Ray L.

Regards,
Ray L.

1185

General Mach Discussion / Re: Pulse Frequency

« on: June 01, 2009, 11:05:01 AM »

Bill,

The pulse frequency is the maximum step rate Mach can generate.  If your machine has 2000 steps/inch, and you want to run 100 IPM rapids, you need a step rate of 2000 * 100 / 60 = 3333Hz.  So, 25kHz is more than plenty.  However, if you have a machine like mine, which has 20000 steps/inch, and want to run 100 IPM rapids, you'd need 20000 * 100 / 60 = 33,333 steps/inch, so you'd need to increase the pulse rate.  Most modern PCs will handle 45kHz pulse rate, some will go higher.  If you need higher still, the SmoothStepper will go up to 4MHz.  In general, you want to run the lowest pulse rate that meets your needs, to keep from just wasting a lot of CPU power on the PC.  There is no advantage to running faster than required.

Regards,
Ray L.

1186

General Mach Discussion / Re: Resetting Encoder Count Via a BRAIN

« on: May 28, 2009, 11:28:05 AM »

I apologize for not communicating clearly. My ultimate goal is to detect racking of the gantry and stop the machine before damage can be done. The gantry is moved by a lead screw on each side. My plan is to have a rotary encoder at the end of each screw. When I detect that one screw is not keeping up with the other, then issue a stop command. My problem is establishing the inital conditions for each encoder. I am sure there are many ways to solve this problem. My first thought was to detect when the machine is in an e-stop state and then zero each encoder count via a brain. When the machine is taken out of the e-stop state, then have the brain compare the two counts and stop he machine if they differ by more than a few counts.

If there is no way to zero the counts, then my next thought was to read the counts into a variable when the machin is in an e-stop condition. Call these the initial count variables (a variable for each encoder). Once the machine is out of the e-stop condition, subtract the initial count variables from the encoder readings. Again, if these readings differ by more than a few counts, stop the machine.

I am open to a completely different approach. Perhaps I need to assign the encoder to an MPG or axis DRO. Perhaps my newness to Mach is making this a much more difficult problem than it should be.

You could zero the encoders using an initialization macro, and use that same macro to re-zero after a fault, as it simply executes as an M-code.  You could also use the macropump to do the detection, rather than a brain, if you wanted to.  I think the only downside would be the slower update rate - about 10 times per second, IIRC,

Regards,
Ray L.

1187

General Mach Discussion / Mach3/XBOX 360 Controller Question

« on: May 26, 2009, 11:29:06 AM »

It appears that most, if not virtually all, XBOX 360 controllers are wireless.  Anyone having any problems due to interference from all the other "emitters" on a typical CNC machine?  What kind of range do you see?  How about battery life?

Regards,
Ray L.

1188

General Mach Discussion / Re: how can I set the a axis to zero within a part program?

« on: May 26, 2009, 12:23:00 AM »

I have a part program that rotates the a axis about 10 times during one cut. I would like to reset the a axis DRO to zero after this cut without stopping the program. I takes too long to drive the axis back to zero as it is a very slow axis.

If nothing else, you could write a macro (executed as an M-code) to reset the A axis DRO:

Create the file M1000.m1s containing the single line "SetOEMDRO(803, 0)" in Mach3\macros\YourMachProfileName.  Executing M1000 in your G-code program will then clear the DRO.

Regards,
Ray L.

1189

General Mach Discussion / Re: First Post - Help with setup issue?

« on: May 25, 2009, 07:11:39 PM »

Looks like it's backlash.  I'm surprised to see that much backlash in a ball screw.  I've got the lower cost Nook ball screws.  I've checked the table mount and the ball nut is screwed into a bracket.  I guess I'll have to tear into the table and see if the bracket is secure.

0.006" backlash in a non-pre-loaded single ballnut is not at all unusual.  You can minimize, but not totally eliminate, this by loading the nut with oversize balls.  To get to very low or zero backlash requires spring-loaded double nuts.

Regards,
Ray L.

1190

General Mach Discussion / Re: screw mapping backwards?

« on: May 20, 2009, 11:26:27 AM »

FYI - It is my understanding that screw mapping works only when running G-code, not when jogging.  Basically, it's not particulalrly useful in its current state....

Per Art:

"Screw mapping only works in Gcode. Since Mach3 is a pulse buffering program, its not possible to map for Jog modes and such, so the mapping was written only for program runs in the interpolator. Its rarley used, and I havent looked at the code for a couple years on it. So much work has been done on the interpreter over the past year or so I cant guarentee that its still functional, but you can only expect a result in Gocde run's. Last I looked it was still working fine."

Regards,
Ray L.