Odd! What’s your printer?
i have a heavily modified ender 4, but for some reason my soft limit was set to 200mm Z when it should be 300mm. but i adjusted the config when i put a new board in.
but just an example of soft limits on the machine being a problem. I would love it if CM would implement soft limits within the software itself, or check the gCode before letting you run it to check if the travel would exceed the bounds of the machine travel.
would be a neat feature. or just show the max travel, like gSender does.
Interesting!
Even if it was just in carbide motion it would be amazing.
First, yeah it sucks to ruin a project for any reason.
I would suggest you keep looking for a different problem. The reason I say this is you describe the holes getting progressively deeper.
If your g-code was in fact “out of bounds” the missed steps would likely only happen once.
Once the z reached its limit and collided with an end stop and lost steps, that would be the end of it, each cut following that would be cut to the “new” depth, but the z would no longer collide with an end stop as that discrepancy has been physically altered by the previous missed steps. The rest of the holes would be the same deeper depth, not progressively getting deeper.
2 cents…
7 Likes
Maybe.
(I know you have a lot of experience with other types of CNC machinery, but bear with me here…)
It is easier for me to believe that the x-y plane (the real-world wasteboard) you are thinking is level is actually tilted upward towards the front of your machine, so the back is ‘deeper’ than the front. Meaning:
-at (X-100,Y0,Z0):
–the machine believes the very top of the wasteboard surface is at Z===0;
–AND the very top of the wasteboard surface is actually at Z===0;
meanwhile,
-at (X-100,Y-300,Z0):
–the machine believes the very top of the wasteboard surface is at Z===0;
–BUT the very top of the wasteboard surface is actually at Z===0.8
So, not perfectly level back to front, this also fits your outcome and data.
Like @SLCJedi suggested above, try it with a test piece first. And, no harm in setting your retract height to be 10mm for this job (really, until you find out why this is happening). You could also run it piecewise, doing a quad of holes per program, re-zeroing “Z” before each one. It really sucks to have an important piece messed up, but it has taught me to be overly cautious with this machine that gives me superhuman abilities.
If this is the case, either loosen all hardware and re-square the machine, starting by leveling the table, then the endplates, then the Y-axis rails, then gantry, lastly carriage, or tram the MDF so that it is in-plane w/ machine movement.
As I stated early on, I already tested the back to front height by running the bit along from front to back, just over the surface of the work (in situ) It gets ever so slightly higher at the back by about a mm.
I’ve also surfaced the wasteboard a couple of times (also mentioned earlier) and haven’t moved the machine. So it would be bloody unlikely to be untrammed.
But again, I did actually test that early on It’s definitely not that.
Consider that for your hypothesis to be true it would have to drop 2mm not over the course of the distance from front to back, but over the course of 20mm (the diameter of the hole where the gouge started) else it would have gouged either side of the hole.
This would be very obvious to look at.
There IS harm in setting the z hop to 10mm. Because, again, the Z axis soft endstop does nothing, and errors exactly like what I’m seeing here are a natural consequence of running the machine outside it’s z axis boundary.
I’ve not only got experience “with other types of CNC machines” I also wrote and facilitated classes for the state on CNC routers, and helped members of the public use them all the time as part of my job. 
This ain’t my first rodeo.
Or just resurface the waste board, right? Way easier. Way less likely to mess other stuff up, and which I’ve already done before… But it’s definitely not this issue.
You’re right. It could be skipping steps with each plunge perhaps. Due to the wood being so hard. It didn’t sound at all like it was strained though. Hmmm. I’ll look at again at the mechanics.
Ok. Truly with all due respect, I don’t know you. I don’t know if you are who you say you’ve been, I don’t know what you know, I don’t know if your “good enough” is the same as my “good enough”, so, at this point in the discussion so far, I am not going to assume these things. I am also not going to assume you are a guy who knows too much to need any help when you are here asking for help.
I hope you have found the solution, and if you haven’t, I hope you do soon. YOU are the one who has an issue with the machine not doing what is intended, so regardless of what you have taught, know, dealt with, built or otherwise understand, your problem won’t be solved, or at least predictable, until you find a way to gain new insight. Again, I don’t know you, so my suggestions will be based on what I have learned and seen others do.
I only ask that you please don’t snap back at us unhelpful peons that just wasted our time trying to read and understand a problem that has nothing to do with us.
I’m frustrated because I’m being repeatedly told the same nooby things. Things I’ve already said I tested. So many people are insisting my conclusion is wrong without any evidence, and just telling me it’s what they’re thinking, even though their hypothesis doesn’t fit the symptoms (and I’ve already tested them).
Or they’re just acting like I haven’t any idea of the solution when I have stated several times that I was pretty sure I narrowed it down.
Surely you can understand how it’s frustrating to be on the recieving end of this?
Not being listened to.
Being told I made rookie errors (when I didn’t) or didn’t do my due diligence (when in did).
Being spoken to like I know nothing even though I’ve demonstrated prior that I do in fact know the things I’m being spoken at about.
Every time someone has stated something I haven’t considered I do actually respond well, see my reply to corpse above for an example.
Yikes! 4 days of "crying on the floor? (?) and I am not sure this post has solved anything. I don’t think anybody knows what caused the trauma as a resolution has never been mentioned by the author. For the record, I don’t cry or get on the floor when I bury a bit or things go sideways.
I’ve repeatedly stated I thought it was the z axis grinding on the hard end stop ever so slightly due to moving out of bounds…
Corpse (and I now notice nate above them) have made me consider I might need to look again, but before that I had stated several times I was pretty sure this was it.
The solution would have been a shorter bit.
Now it might be slowing down the plunge.
Congratulations on not crying? I dunno what to tell you man.
The resolution is me putting an inlay in, I posted a picture above of that and I know you saw it because you replied asking what I was doing to fix the job just after I posted the picture then you removed it a minute after (presumably because you saw the picture).
This makes a lot of sense and agree that if you we’re out of bounds and crashed the Z-axis at the top this would have either happened even before milling the first hole or at the very least I’d think that gouge would have happened on the first hole and would be visible all the way through your project.
You mentioned the holes are getting progressively deeper throughout the project so I’m thinking another possibility is that because the wood is so dense and some other factor (ie. toolpath feed/speed/plunge rate/…, maybe a dull endmill, …) that you are possibly loosing Z-steps but rather because it’s getting pulled downwards and if you’re not loosing steps the endmill may be getting pulled down from the collet (slip in the collet). This could explain the holes slowly getting deeper and in the last hole before the gouge there may also be a knot in the wood meaning it all of a sudden got even harder/denser and pulled down on the Z-axis or endmill slipped in the collet a few millimeters by the looks of it.
Now only you would know if you Z-axis was crashing since you where there but I feel this would be another reasonable possibility. So I’d inspect my collets, clean them, … I’d also suggest revisiting toolpaths but sound like you already thought of that one.
1 Like
Hey everyone,
I’m sorry.
Clearly I have been overly emotional here and misinterpreting your contributions to this issue as less noble than the reality.
It is clear that everyone here is trying hard to help.
I’ve been responding poorly to comments I’ve found frustrating, but I’m gonna put the frustration down to stuff going on inside me (I did cry over a piece of stupid wood, after all), and not you all.
I’m sorry for not responding better here, getting defensive and letting my frustration bleed through in my tone and behaviour towards you.
I’m gonna take a big ol break from this thread now.
Again, I’m sorry for my behaviour here.
3 Likes
Thank you, yes, I agree, it does sound more like that, than what I was thinking before that.
I’ll see about slowing down the plunges.
I’d think feeds/speeds/plunge rate is one possibility, but I’d also look into your endmill sharpness and if your collets have some wear, they are consumables afterall. I’ve also seen some threads here people having slipping in the collet because they didn’t clean them between all tool changes. If working with this species is not in your typical day your usual routines are likely good enough but working with materials 2.5x harder than white oak then all variables come into play. Thanks for sharing, if I ever get anything like this I’ll have to remember to pay attention to all the little details.
2 Likes
Good thoughts. I’ll do those things too. Thank you. 
2 Likes
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.