Fixed z probe location

Right now this would be tough as my Mpcnc does not have endstops. But I am thinking that if I were to always start my machine in a FIXED position. I can then have the workspace coordinates set to 0.0.0

I am thinking that in the upper right corner of my mpcnc I can have a FIXED probe for Z.

for the below I will say that my probe is at 255 x and 255 y

Can I then move to g53 255,255
Then g28 z
and have it work? I am thinking this would work. But it would set my position to the top of my bed, not the top of my work.

Can Estlcam work with z 0 being my bed, or would I need to change the z probe thickness to accommodate the difference of my workpiece?

If I am getting in the weeds and anyone wants to tell me to disregard, please feel free, as I think I have a good grasp. I just want to get past connecting a probe. I would like a set probe and it just works.

Also if I am correct, using g53 is a one time deal and does not harm any workspace coordinates, correct?

Oh Wait, can we set the zero on the machine coordinate and have the Workspace coordinate z be on top of the workpiece??

How would you program z to reset the Workspace? Maybe I can google!

There are both low-complexity solutions and high-complexity solutions to your ā€œproblem.ā€ On the high-complexity end, take a look at my bitsetter solution for my Primo.

Iā€™m not sure of your goals here, but you likely can get what you want without going to the complexity of workspaces. You donā€™t specify why you need to ā€œhomeā€ your X and Y. Without endstops, if you want a repeatable home, you will need some stops that you can push the axes against. Once the axes are against the stops, you will need to engage the steppers. The steppers can be engaged by any XY movement, or you can use an M17 to engage the steppers.

For Z, you can use either a fixed placement touch plate as you suggest, or you can have a moveable one that you use on top of the spoil board. Note that the plate does not have to be level with the top of the spoil board. You just need to know the relative difference between the height the plate is triggered and the top of the spoil board, and your script can make the adjustment.

Personally, I find that sometimes I want to use the top of the stock as my reference, and sometimes the spoil board. For contour cutting, using the spoil board assures me that I cut all the way through the stock without gouging the spoil board. V-carving is very height sensitive, and I always home against the top of the stock.

There are at least two ways of dealing with top of the stock vs. top of the spoil board. The first is to specify which you want to use in the CAM. The CAM will automatically adjust for the height of the stock. To make this solution work, you cannot have a G92 in the start code.

The second way is to make a manual adjustment. Personally, I make a lot of mistakes in details, so I must keep my process consistent. I always author my CAM files with the top of the stock as reference. If I want to use my spoil board as reference:

  • I write down the stock thickness used in my CAM.
  • I home to the spoil board
  • I electronically move up the authored thickness.
  • I use G92 Z0 to set the current position as Z=0.

When I use this process, I typically set my stock thickness in CAM to 0.2mm or 0.3mm thicker than what my calipers read on the physical stock. This gives a bit of slop in case my stock is uneven with the only ā€œdownsideā€ being that the first pass might not cut the amount of material specified in CAM.

Note that you can have multiple different touch-plate-like solutions using the same circuit and logic for your existing touch plate. For example, you can have both a moveable touch plate and a fixed touch plate. You just need to wire them in parallel.

2 Likes

I work in tooling. I.t. but tooling. And i know what the big cnc s do and now it seems the lil cnc s can do amazing things tooo. No needs just wants or desires :wink:

Will review when i get home. It looks like you may be doing what i am looking 4

Ainā€™t that always the case for most things? And guess which ones are more ā€œfunā€? :wink:

May have to steal this as an alternative to ā€œit dependsā€ as an answerā€¦

Anything you want. The more changes you make, the more you need to keep track of where it is supposed to be and the more you need to pay attention in CAM.

How most of use do it now is Z probe to top of material. From there you only need to know how deep you are going, that is about it.

If you probe and use the bed as your reference, you need to know how thick your material is and program from there. I only ever do this if I have a Z dimension I care about from the other side. (intermediate instructions in the docs are shown this way). This would also include a stationary probe as you asked.

Any other height, say Z max, then you need to know how far the table is and how tall the workpiece is. I see no reason for this style.

Workspaces are best if you are moving your XY coordinates to say fixed jigs, then you do not need to keep track of several Zeros in CAM. Not really a common thing for just Z.

1 Like

I mostly wanted to get away from probing under the router and the mpcnc pipes or around them. Tha ks this gives me waht i wanted between you and robert.

The reason I do not like fixed toolsetter probes is the extra set it needs in cam (workpiece height). Rarely if ever do I care about actual Z thickness, just that I get through the material or my pocket is a perfect depth, not that todayā€™s sheet goods are 1.2mm thicker that last time.

Touch probe off the surface is what is needed in my uses case 99% of the time. If I need a tool setter style depth, I just measure my material and set the cam accordingly. I do not want to have to do that every time.

1 Like

Oh, I will add if you use a tool setter you need to surface your bed, otherwise it will be wrong every time.

1 Like

Thank you for your responses both of you!

I use aluminum tape for my z gage and use a lalagator clamp on the tape and a magnet on the millšŸ˜Š its not fixed but very mobel and easy to place

1 Like

I think you should trust the public. People will be responsible for their own understanding. If this is over their head, they will either have to understand that it isnā€™t typical, or they will get confused (and thatā€™s ok).

Also, the requirements for getting into the lounge are not knowledge based. You just have to post a bunch and visit often to be in the lounge. So thereā€™s no guarantee you arenā€™t confusing people here.

But most importantly, I visit a lot of public forums (mostly stack overflow and reddit these days) and I find my solution 9/10 times without participating. Even if youā€™re developing new stuff, do it in public so those other people can find it (even 10 years from now).

A lot of features we have developed started with someone knew asking a question and one of the regulars saying it wasnā€™t possible. That fresh perspective and debate is invigorating.

I hope Iā€™ve beat the dead horse enough. Public forums, FTW!

Ok, here we go. I changed it to the mpcnc Advice. :slight_smile:

and pulled the first line :slight_smile: