XY Zero Accuracy and program theory question

[foamman]

Hi,
I recently ran a 3D file with very tight tolerances and had to re-zero the XY location between tool changes. Sadly the next cut was noticeably off (roughly .040) which led me to question what the deviation from one XY zero to the next might be. Does anyone know how accurate the XY zero should be? Perhaps my machine is running out of spec.

I looked at the code and saw that it repeats each input touch twice. Why does it take two readings? In the ZZero file it says the second reading is for more accuracy. Would it help to run the XY Zero file slower? Would it be more consistent?

On a programming level, to make sure each XY zero is very consistent, could the program take say 5 or 10 readings (VA) of each axis and divide the data set by 5 or 10? Theoretically this would lead to a more consistent standard of deviation. Any thoughts on how to write this code? I am just learning the SB code and could use some guidance.

Thanks for the input.

Ben

[blackhawk]

Ben - I can't help you out on the code writing. But, when you zero X and Y, it should be much closer than .040". I have never measured mine, but I would expect that I am within .003-.004".

Why did you have to re-zero X and Y between tool changes? Unless you turn off your machine or shift the workpiece, the X and Y zero shouldn't change.

[foamman]

Hi Brad,
Actually it was an accident. I hit C3 instead of C2 and the machine went through the XY program. I have learned that if you try to stop it mid program it will generally crash and make you re-zero it anyway so I just let it run.

It is definitely off more than .004. One wonders what the accuracy is for the limit switches.

Thanks
Ben

[jerry_stanek]

I can rezero mine using the C3 command and it is within .002 also C2 is within .002

[steve_g]

Ben...

Do you have proximity sensors for limit switches? If so you might consider adjusting them. The closer they are, the more reliable the repeatability is. My accuracy is well under .005 and more likely .001-.002.

SG

[foamman]

Hi Steve
I will check to see what the gap distance is between the bolt and the prox switch. I have the prox switches sold by ShopBot.

Thanks
Ben



Ben...

Do you have proximity sensors for limit switches? If so you might consider adjusting them. The closer they are, the more reliable the repeatability is. My accuracy is well under .005 and more likely .001-.002.

SG

[myxpykalix]

Ben,
Here is a practice that i do when setting up a file that has never failed me and saved me to get EXACTLY back to where i need to be EVERY time.

I set my material on the table and clamp it down where i need it to be.

I do a C3 and now i am at Lower left, 0,0

My material 0,0 is in the center of the material on the table. So i move the bit to the center of material and WRITE MY X, Y COORDINATES ON PAPER

Then i Zero my X, Y at those coordinates

Now if you start your cutting and somehow lose your settings all you have to do is to go back and do a C3 and zero it back at 0,0 at lower left, then do a MY,MX(those written down coordinates) and you are back EXACTLY where you started.
Unless you have slop or something moves it should be right on the money. I have had to restart 3d carvings from a crash and saw no visible alteration from the carving. This has happened a few times over the years and saved my butt.
The key is writing down your coordinates and keeping them handy. Hope that helps:eek::D

[ssflyer]

But Jack,

He states that his machine is off ~0.040" after running the C3 routine, so writing it down is still going to be off.

I think Steve is on the right track, unless there is considerable slop in the machine.

[Simops]

Ben...

Do you have proximity sensors for limit switches? If so you might consider adjusting them. The closer they are, the more reliable the repeatability is. My accuracy is well under .005 and more likely .001-.002.

SG

Ok I'm a litle confused now......so there is a difference between limit and proximity switches....what comes standard out of the SB factory?

[steve_g]

Michael...
The proximity sensors are the limit switches. I asked to be sure that a home brew micro switch wasn't involved. When dealing with mechanical switches you get a phenomenon called "bounce" that can make repeatability difficult.

SG

[Simops]

Ah OK.....good! Thought I had missed something:)

Cheers