To the person who made no attempt to repair and failed to report the damage you did to zone 6…
…and left a mess behind the machine…
…you owe DMS an hour of time.
I’m starting to think suspensions from the MultiCam with remedial training may be in order.
We need to get sacrificial spoil boards implemented or the gcode precheck for max cutting depth written. Precheck would move head to xy close to front of machine and perfom drill operation to top of spoil board, Run precheck before your first step, if the bit hits the spoil board you probably missed top of spoil board setting. But damage to spoil board is only a hole in an area we don’t use anyway. Ideally log the userid logged into the machine and job name so we can track if precheck was skipped.
We also need a public record of who logged into machines with timestamps. There are way too many issues reported that start with "Whoever did X damage…
Determining the who isnt really an issue
In an area who doesn’t use? I’ve used that area before.
This. In the very near future there will be a requirement that no cutting be done directly on the machine vacuum boards. They will only be used to hold down your own sled/jig/board to which your work is attached. This uneven/damaged table stuff needs to stop.
It may be worth adding something to the Multicam training and check-out covering the specifics and mechanics of spoil board repair process, if not already incorporated.
It pains me that we have to add extra steps and make it more difficult for users that are not having creating this issue. It seems like we should be asking those that continue this issue to fix it or retrain. I think some of the cuts in the spoil board would have easily cut through any sacrificial spoil board and into the existing.
Having shared resources means that each user should be expected to maintain the tool to a community standard. Accidents will happen [and should be corrected], but continuing the same mistake should have consequences. I think the enforcement of the consequences [do we have any?] has been our real problem.
What are your thoughts on having it a separate (remedial) class?
Speculating is not necessary. A non-zero number of people have cut into the bed.
In this case…
I / chair will enforce whatever rules the community/ SIG comes up with.
What I had in mind as suggestion was, at a minimum, just making sure Bert’s video on how to repair was viewed as part of training and, if we are still doing in-person sign-off, have the person talk the signer-offer through the process and show them they know where the materials are located, etc. Seems an ideal opportunity to do this while job is running or they are switching bits and checking max-depth again : “do you get why we check this again; and what happens if you don’t check/reset; and if you screw up the spoil boards, what you need to do then?”
But suspension of privileges until remedial training occurs is certainly always an option, IMO, as/when warranted.
The precheck job would minimize repairs because the very worst thing that could happen is you would have a single hole to fill. And a hole entirely covered with the workpiece wouldn’t affect the vacuum
Currently there is no way to know you are cutting into the spoil board until the job is done.
not really. setting the correct tool height/cutting depth would resolve most of it. stuff like the bit coming loose not included
We have new collets and nuts, bits shouldn’t slip unless not torqued. Forcing or encouraging a precheck hole test would at least verify the spoilboard depth setting was correct, and rule that out if the spoil board gets damaged. I don’t see a downside other than adding a minute or two to the job.
Please consider the extra problems requiring everyone to have their own spoil boards will cause. Large item storage is shut down. So it would be a bring it then take it home again every time. Also, size would be a problem for those with smaller vehicles.
It appears something has changed. IMO it would be better to fix that, and as has been discussed, reiterate the Z-axis setup and repair procedures in training.
Todd, I’m not sure things have changed in the grand scheme. When I was SIG leader, I saw the same kind of issues and I initially tried to prod for better results on Talk. After figuring out that Talk wasn’t very efficient at reaching those who were making mistakes, I took a different approach. I watched the CNC frequently and spotted users who were having problems. Most of the time I could identify them, seek them out individually, and help them figure how they got in trouble and how to avoid same. To the person, almost every one was interested in learning how to be a better operator. My having the time to devote to mentoring was a key factor. If you have a day job, there is not enough time left to devote to monitoring and mentoring. So couple less oversight with most DMS users being rusty after a looong COVID absence, stuff happens. Users don’t know what they have forgotten. My opinion is that making mistakes and not repairing damage is more about not remembering how than it is about not caring.
The DMS Multicam is serving a diverse and infrequent user group. It is a sophisticated enough machine that it is easy to get rusty during periods of non-use. Whether it is one person or a squad of folks … if the Multicam “owners” have to contend with the demands of a day job, there are going to be things that go sideways because there won’t enough free time to find and help those who aren’t avid Talkers. To a big extent, posting on Talk is singing to the Choir. Reaching the folks in need is more like missionary work … you have to go to them.
Your mileage may vary.
I have no problem believing it’s infrequent use that’s the root cause.
I’ll make the same suggestion I made for the HAAS: We need a setup checklist. The setup operations should be similar enough that a list of the important steps should be possible. Works for other complex machines. Like airplanes. You memorize the immediate emergency procedures, and use the checklist for the follow ups and the routine.
I agree with a checklist. It works for things like airplanes where your life depends on not making mistakes. It might even work with the HAAS if @nicksilva was still in charge to scare people if they mess it up. (@TBJK is too nice a guy in comparison!) My observation is that folks chopping wood are not similarly motivated.
I think you are closer with
but for the Multicam, the “emergency procedures” are “how to screw up”. I am a big believer in training focused on what NOT TO DO. Folks, especially makers, are good at figuring out how to do things most of the time … they just need to know where the lines are between OK and NEVER DO THIS.
The NEVER DO’s are a shorter list and can be made more memorable with some creative training and use of the TV monitor and signage. Or, you could just get Nick to volunteer to scare them!
Any policy that relies on people to do the right thing is a failure from the start.
I’ll say this much - I’ve always contended that one of the main problems (besides people) is the software tools (or lack of) that drives the MC. If you’re running a single part, it is in fact no different than running a prototype. At the point of running you have never run this part before. So just hitting GO and watching it run into the table is not just a consequence of person but of the software. Here’s what I mean: when running a prototype on the HAAS, there is always a safe height defined (same in vcarve) which perhaps 0.25" above the work piece. We teach them to SINGLE STEP to the point where it is going to move the spindle to the safe height. Next is to set the rapid rate to about 5 or 25% speed. At 100% (200 IPM) it is to fast to stop if you just let it go. As it comes down at a slow rate you PAUSE it before it touches the surface. You look at the DISTANCE TO GO. If you’re above the surface, say 1/2 inch, and the DTG reads 2" logic should say to you that there is something wrong and continuing would be disasterous. If it’s good you let it continue at the slow speed until it starts cutting. IF and only IF it looks right you take it out of single step and hit full speed and GO. Anything after that is a programming error.
This is a brief synopsis of how we run a first (or only) part. After the first part and nothing crashed you’re safe to reload a new part and GO. Unfortunately (and not that anyone would follow this anyway on the MC) there are no such tools available. You got GO and you got PAUSE. But pause doesn’t help you if it quickly drops into the table.
For a professional router table, the MC is severely lacking in the control area. If you look at a GREAT cnc like from C R Onsrud - it comes with a FANUC controller which is an industry standard in 3 axis mills and gives all the features as I’ve described above for the HAAS.
The MC team should contact MultiCam and ask if they have any better control software available. At the very least it should be suggested that they create some. Even Chinese CNCs allow for the use of MACH3/4 software which gives great control. As it is - I personally would NEVER recommend MultiCam to a client.
2 pennies worth. CHEERS!
Any lack of policy enforcement that allows people to screw things up again and again without consequences is a failure as well.
