Hello, new to the community, and haven’t even run my first job yet. Just upgraded to a shapeoko5 with bitzero v2.
I noticed something that is not making sense to me when using the built in probe sequence for bitzero. In Carbide Motion when you use the bitzero sequence, after the Z probe it goes over to bitsetter and probes the tool. In my mind this makes sense if you are using the probe cylinders. But gSender does not do the tool length measurement after the Z probe. So how does it know what the actual Z offset is after a tool change? Am I missing something?
Using v1.6.4.
Aside: thanks for the wonderful software. If I knew JS i would help solve the headless issue on RPI, for now its ser2net.
I don’t have a bitzero or bitsetter so bear that in mind but I’m pretty sure you’ll have to set up two separate things.
The probe sequence for the bitzero should set the XYZ zeros for the start of the job. I think the second part needs to be setup in the “Tool Changing” section of the Config. My best guess would be selecting “Fixed Tool Sensor” and setting the appropriate values in the new section that appears after you do.
I have never used multiple tools in one file so this just an educated, I hope, guess from being around here for awhile.
Welcome to group! I hope this helps a bit… as opposed to breaking one.
Thanks for the feedback, I think even in a single bit scenario, if you are using a Vbit, you dont want to use that to probe the bitzero since the diameter is not consistent which would need to variable results. So you would have to switch from the probing dowel to the bit after the zeroing. So I think there is still something missing here since you would need to still measure the dowels length outside the collet.
One thing I did consider is maybe adding some custom gcode to the start of my file to probe the dowel first, then switch the bit. Right now vcarve does not emit code to measure until it switches to the first cutting tool. But I think that is a stop gap for me and not a long term solution for future SO5 users who also have the bitzero.
I only have a regular touch plate like the bitzero minus the hole. Whenever I use a V bit, actually any tapered bit, I have to do an XY probe with a straight bit and then do a separate Z probe with the V bit because of the reasons you mentioned.
If I understand how all this is supposed to work then I think a separate Z probe is only needed if the first bit is tapered. Tool changes should be okay because your only doing a Z probe on the bitsetter anyway.
Just to be clear the separate Z probe at the start would be done on the bitzero not the bitsetter. Of course I could be wrong because I don’t have experience with actually using these things.
Before gSender had these things built in people with Shapeoko mills used a set of macros. You may be able to glean something useful by looking at those. The site is for CNCjs but gSender was forked from CNCjs at the start so it’s relevant.
So I’ve been thinking of what I’d do if I was trying to figure this process out. You mentioned having a probing dowel so I’d use that while I was testing this out not any bits. I’m assuming the dowel is stronger and/or less expensive than bits here.
In my design software I would set up a job with two tools. Each tool would have a short job just one pass in a decent sized circle so things don’t happen too fast for me. I’d want to be ready with the E-STOP in case I saw something bad about to happen.
I wouldn’t even clamp any stock down, I’d pick an appropriate spot for XY and just set X and Y to zero there. There are buttons for manually setting a zero for an axis in gSender. Then I’d probe the bitzero for the Z. After that I’d jog up some distance and manually set Z zero there. Remember this distance for later.
Now you can cut some air by running the job. When it gets to the tool change you’ll be able to see what’s going to happen in a safer way. You don’t even need your spindle/router on for these tests. If and when you get to changing the actual bit just stay with the dowel but slide it up or down in the collet so it’s a different length.
When the job is done use the GOTO buttons to get back to XYZ zero. Now jog down by the distance used above and then try to slide the bitzero under it. That should tell you whether the bitsetter did it’s job correctly.
EDIT: You could also temporarily set the maximum travel rate for your mill to a lower value so that the moves to the bitsetter don’t happen so fast that you don’t have time to react. I’m assuming that the mill is going to use a rapid move to the bitsetter’s location. Better safe than sorry.
Just in case you don’t know, I’ll add that for probing the Z with the bitzero at the beginning you should turn it upside down so that it’s stable.
That actually sounds way easier to just do the XY separate then just do Z with the final bit for V bit carves. It unfortunately still doesn’t solve the multi bit single file problem though as the instant you change your bit your Z calculation would likely be off since bitzero only gets you the absolute not factoring in tool length, unless its smart enough to handle that calculation on first tool change if the tool matches the original Z probe on bit zero? Would make the XYZ probe flow not super effective though unless you were using an endmill maybe?
Regarding the macros I would be a bit weary of them given they are older than the SO5 itself (HEAD is 5yr old) and I think that was the first platform they started shipping automatically with bitsetter by default.
Here is a little more information on the dowel. See image 3
Includes 1/4" reference pin to eliminate tool flutes as a source of accuracy problems.
I think mine also came with a 1/8" pin as well.
That test flow is a good idea, that will tell me if there is truly an alignment issue. I just have make sure the tools sit at different lengths enough that its measurable. I should have an answer by later today. Thanks
Yeah at some point it has to use the bitsetter with the first bit, then allow you to change bit and finally use the bitsetter again so it can calculate the difference and adjust accordingly. When exactly that happens I don’t know. Maybe it does it at the start of the first tool change?
Hopefully that test flow will help to figure those things out without breaking anything. Good luck!
So I ran the test job and it seems gsender is smart enough to probe the first bit before the tool change when you kick off the job. Even though there is no M6 for T1 it still probes it. This was the part I was missing. So i skipped the wizard and re-ran my cut. It probed the pin, then I changed the tool and it probed the tool. Checked cut depth and it was exactly what I set it to.
Thanks for the help with the air test, running that solved my confusion!
One caveat I just found, this flow does not work with the remote mode.
Tried to trigger a job from my phone and no tool change wizard appeared, so I assumed the pause was because it just couldn’t render. Nope, it skipped the wizard entirely and that was the M6 in the gcode. Hit go again and the spindle spun up with the pin in it, slapped big red button to stop that nonsense.
EDIT: turns out somehow i got the main session disconnected from the CNC and the remote was still connected, if the main session is connected then the wizard pops up there.
