You know what sucks? Having a multi-run project that spans more than one day only to find out that the Nomad and Carbide Motion apparently has a hard time remembering its last zero position.
I guess I shouldn’t have assumed that zero would be kept. I ended up gouging my stock and breaking a $30 Onsrud cutter that I had just spent 15 minutes on lopping off 1/2" of the length.
Why is the Nomad and/or Carbide Motion not remembering zero? Seems like an easy thing to store and recall when things are started back up again. Now I can’t do what I was intending to do today and have had to get another cutter (ordered more than one this time) shipped overnight.
I feel like it has remembered zero before. I’m using the latest Carbide Motion build (357). I quickly realized what was happening but by the time I could hit pause and quit the job the cutter was gone. It doesn’t take much to kill a 1/16 end mill. I definitely had a few choice words with the Nomad.
Can you possibly look into preventing this in a future build @robgrz ?
In general, one should never trust that a zero is preserved after a power cycle or one has walked away from a machine. This is taught in machining school - never trust a zero to still be there.
This dates from the time when mills had to be shared within and across shifts and there were multiple operator/machinists per mill.
It’s always a good idea to validate your zero hasn’t moved, even if one is the only operator. This is doubly important when working with different machines as a fair number of machines are quite adamant that they do not and will not preserve zeros across a power cycle… or someone played with the machine and moved a zero.
Because I learned things “the old way” I never depend on a zero to be preserved across a power cycle. If it seems like it was, I check it. Of the 5 CNC machines I’ve often used, 4 of them are adamant they do not preserve zeros across a power cycle.
One can write down their zeros and use them for another run. Just remember to clear/reset the machine and set the zeros again… don’t trust that something hasn’t been done to the machine (or was done but forgotten).
If one is depending on a zero to be preserved but is changing stock, one had better have a jig that ensures the stock change places the stock pieces in exactly the same position as that of the first piece of stock. The zero doesn’t change… but the stock might. The ole machinist’s mallet and light tapping ensures positioning, as does a well designed fixture.
Here is a wonderful video - NICE JOB @FlatBaller! - that clearly shows this in action:
While powered on the Nomad most certainly maintains its zeros. One could leave it on for the length of the multiple day run?
All that said I seem to remember postings that said the Nomad does preserve zeros across power cycles - but I can’t seem to find them. I just tested my Nomad 883 Pro and it seems to have kept its zeros - and I see @FlatBaller says the same.
Was there a tool change and the tool length was not remeasured?
Now I can’t do what I was intending to do today and have had to get another cutter (ordered more than one this time) shipped overnight.
Sorry this happened. In:
I do remind people to order end mills - especially the small diameter ones - in duals or quads. This ensures that breakage or wearing out doesn’t stop production. When one starts with the last one, place another order.
I have definitely learn my lesson for this. My beef is that neither the Nomad or Carbide Motion kept zero. There is no way that I could get said accuracy by redoing the zero by hand. I didn’t write down the coordinates (I’m so new to this I can’t remember if Carbide Motion has those displayed after zeroing?).
I didn’t write down the coordinates (I’m so new to this I can’t remember if Carbide Motion has those displayed after zeroing?).
The coordinates are on the zeroing screen. One can write them down. Getting there, repeatedly, to set the zeros again, takes a bit of practice.
Remember that the stock has to be able to accommodate the zero and perform the machining. If one is depending on the corner of the stock, the zero being persistent doesn’t help much unless the stock positioning is also persistent.
Check out the video I posted. @FlatBaller demonstrates excellent use of a jig.
The zero of the corner of the jig is known, measured and recorded. Knowing that, the zeros can be entered after a machine reset and jobs can proceed. The jig ensures that the zero doesn’t move. The mallet and fixtures ensure the stock is placed on the zero.
I have definitely learn my lesson for this. My beef is that neither the Nomad or Carbide Motion kept zero. There is no way that I could get said accuracy by redoing the zero by hand.
The jig/fixture method, written down zeros, always resetting the machine and entering the zeros would solve this.
Maybe it’s just me or the way I learned machining, but I trust nothing. I never assume, I check, if I’m even slightly suspicious, I rezero. YMMV.
Mine Nomad loses zero between job loads even if power is not cycled. Even if I zero the machine, leave CM open while I’m working the Meshcam file then load the Meshcam file, it will lose zero. I don’t trust it to keep it more than a minute or so. I don’t know why but its a real pain.
Mine (an earlier version) remembers zero about 95% of the time. I learned same lesson the hard way also… got too confident with it remembering and then crash-boom goes the little endmill.
Other erratic issues (hard/impossible to reliably replicate) have led me to cycle everything from scratch (machine and quit/restart CM) before each job. It only takes a few seconds but gives better peace of mind.
Mine Nomad loses zero between job loads even if power is not cycled. Even if I zero the machine, leave CM open while I’m working the Meshcam file then load the Meshcam file, it will lose zero. I don’t trust it to keep it more than a minute or so. I don’t know why but its a real pain.
Really? I know you have a Nomad 883 Pro like I do. I don’t see this.
With CM and the Nomad both on the zero is known… it doesn’t have to be forgotten. That said, that forces old school never depend methods.
Mine (an earlier version) remembers zero about 95% of the time.
That’s good to know. My quick test mentioned above worked - one shot. Thinking back, I’m pretty sure I’ve seen the zero preserved more than a handful of times. As I said above, I never assume the zero is preserved. If it looks like it is, I check it before starting.
@warba has gone through the same machinist’s school of hard knocks I did.
I’m sorry this happened to you! I’m sure @mbellon is right about the best practices for machining, but for what it’s worth, Carbide Motion has reliably recalled zero every single time I’ve ever used it in the year or so I’ve had it. This includes turning off and unplugging the machine, unplugging the usb, and rebooting the computer. It’s recalled the zero for weeks.
This is because, as @robgrz has stated in some thread somewhere on here, CM stores the zero (or offsets) in a text document on your computer somewhere.
If this behavior has been changed in the latest version of CM, that is a major change and they REALLY need to let us know about it. But I rather expect it has not been changed. If people are losing zero often, then I think maybe there is a bug somewhere. It’s supposed to remember it!
As a side note, here is how to record your zeros (offsets, really), if you are doing a job that will be repeated and have the same zeros every time: move your cutter to the correct zero for your job, then go into the “set zero” screen in CM, and click “clear offsets,” which will show you the absolute machine coordinates for that spot. Note these somewhere safe – another Nomad user just wrote them directly on his custom spoilboard. Make sure to hit “set zero” before you leave the zeroing screen.
… Carbide Motion has reliably recalled zero every single time I’ve ever used it in the year or so I’ve had it. This includes turning off and unplugging the machine, unplugging the usb, and rebooting the computer. It’s recalled the zero for weeks.
This is because, as @robgrz has stated in some thread somewhere on here, CM stores the zero (or offsets) in a text document on your computer somewhere.
If this behavior has been changed in the latest version of CM, that is a major change and they REALLY need to let us know about it. But I rather expect it has not been changed. If people are losing zero often, then I think maybe there is a bug somewhere. It’s supposed to remember it!
Excellent point. Could this be a problem with the latest CM? Are we all on 357?
FWIW, I used the exact same computer. I didn’t even unplug it from the port when I fired it back up. Where is the location and name of the file you are saving to on a Mac?
Shame on me for assuming the machine kept zero. Given that the above quote states that zero is stored on a file on the computer initially used and that I used the same computer on the re-run, yes I do place blame on Carbide Motion for failing to keep its side of the bargain. I’ve learned my lesson and I will not trust it in the future however. That was my bad, but in this instance I really don’t believe it was my fault for relying on the machine or the program sending the code to remember zero if that is one of its features.
Just to make sure we are talking about the same thing. … the"center of the table" zero has always been right. When I rezero for the current job the Nomad (or my computer) has lost that zero many times. I set it like always, go set up a job in MeshCam, then watch it go to some other zero when I run the job. I then rezero on the workpiece exactly the same way as I previously did and it works.
I don’t think user error is to blame for what I see happening.
Digging up this ghost of a thread from 5 years back, but I’m seeing the same thing in CM #535 on my new Nomad3. I just ran a job that used a small portion of the stock at the front edge. I carefully zeroed the Z to just kiss the wasteboard. First piece came out flawlessly. Then I simply jogged my Y by 40 mm and rezeroed just Y) to make a duplicate at a new location in the same stock. But when I did, Z shifted by about 0.7mm, causing me to cut into the wasteboard (no biggie) and end up with a part that’s 0.7mm too thin (big problem).
I didn’t cycle power, shut down CM, or even re-load the file. Not quite sure how Z changed. X managed to stay the same, through.
At first I thought I was seeing things, but it just happened to me again, this time when swapping out stock. I ran a job (in foam) to be sure my Z was correct. Half way through, I was satisfied, so I aborted and loaded up my UHMW stock. I was about to hit RUN when I decided to double check my Z, and it was off (high) by about 0.7mm again.
Seems like this is a bug, and fairly consistent, so maybe the CM folks can retrace my steps and find it. For making more than 1 of something, this is a crucial need to get to even moderate production levels.
