I’m running the Avid moveable rack ATC and have the AV40S spindle.
Interesting thought on the “spit out” force, and I discussed this with support yesterday. I wasn’t sure if the tool holder was supposed to “eject” into the tool bin fingers or not, but that is what it is doing. The spindle stops in Z short of the tool holder fully engaging with the fingers of the tool bin, then spits out or ejects with some force into the fingers.
The video I got yesterday of the tool holder going too deep into the fingers was after about an hour of cooldown on the machine, so I’m not sure heat is ultimately the issue.
I’ll report back when I hear something from Avid on a solution or next steps.
That is something that I haven’t tried! I’ll have to give this a go when I can reproduce the problem on my spindle.
I need to test Avid’s new release of Centroid and check out their code – I haven’t had time to look into it, and update my scripts/app to work with it.
Cool - I’m definitely interested to hear what people find out.
I was thinking of 3D printing a really soft TPU insert that goes into the bottom of these “hole” style holders to alleviate the problem, and experiment with timing (per @djdelorie 's suggestion). The problem with something under the collet is that it is dependent on the collet size and tool holder length (ie: ER32 is different than ER20, and there are different length tool holders in a given size). It also doesn’t work with wide bits (ie: 2" surfacing bit), but those probably won’t fit in a “hole” style holder anyways.
RE material: I’ve tested PLA, PETG and Bambu’s new strong PLA. PETG is pretty brittle, and still fails along layer lines just as easily as PLA, based on some simple “break it” tests I did. “Strong PLA” seems stronger for layer adhesion. I don’t have a lot of tool changes in my “hole” style holders, so I don’t know how they’ll hold up long term.
IMO, the tool holder sockets should not rely on the tapered rim to hold the tool holder in place. That should just be to keep them loosly in place when the rack is moving around.
The plastic is pretty soft and can stretch out pretty easily and just from manufacturing variation. I also have a couple that don’t hold the tool holder very tight. If they get too loose, or the spindle is a little sticky and then spits out the tool holder harder, then can fall through.
There really should be a flat hard stop under that that the tool holder can rest on to provide a precision height for pickup, as well as preventing the tool holder from dropping through.
This discussion got me thinking. The photo below is a quick prototype that I will 3D print to mount to the moveable beam on the ATC. It can be bolted to the top of the beam with some slot nuts and screws. The top with the larger hole will slide in just under the black plastic tool holder fingers and I set the diameter of the hole a bit smaller than the ISO 30 diameter so that bits can go through but the holder can’t, preventing dropping to the floor if something happens.
I’m going to print this prototype this afternoon and see if it works and I’ll report back with some photos.
That will be interesting. I have seen vertical tool holders for sale that are essentially a pocket hole that solely depends on gravity and no springforce from the side. I’m assuming there is not actual need to retain the tool holder because it my experience the ATC spindle either releases cleanly or it fails to release and hangs on with a death grip (which no piece of plastic is going to pull out).
I think one thing thats nice about the AVID design not having the resting ledge under the tool holder is that it maximizes the diameter of the tooling you can use. However, it might be worth giving up a little working diameter to make sure it doesn’t fall through.
Yeah, this is kind of my “safety net”. It will slide in just under the black fingers and bolt to the main beam. I printed the first one and screwed up a dimension so I’m currently printing another. It will allow the tool holder to drop about an inch or so before this catches it. The ISO 30 diameter is 1.97" and I arbitrarily made the hole 1.7". This is not a solution for larger bits but rather something that wont catch a lot of chips and dust such as a foam catch tray.
With any luck they will never need to be utilized for their purpose, but if something happens it should keep expensive things from hitting the floor.
I appreciate all of the work going on here… However the correct way save tools from not getting dropped is to pick them up/drop them off correctly the first time. We did a TON of testing on this, and if it’s not working right on our AV spindle and machine we should solve this.
Tool holders are wear items. They should last tens of thousands of cycles, but in our experience when you put a tool in and out enough times when the tool placement isn’t spot on, the tool holders wear out faster.
Not being “spot on” can happen if you square or tram your machine AFTER finding the tools. Both of those things can affect where the tools are placed.
For those experiencing this I’d pay very close attention to that.
Another thing to be mindful of: If you leave your control box on, but shut down the software the Z axis brake grabs, and when you start the software back up (without power cycling your box) the brake will release when you move our Z axis… this is a theory but you might lose a bit of Z position when that happens. When the software exits the “no fault out” powers down the steppers/servos and that’s why the brake kicks in (brake is ON when there’s no power to it)
I am wondering if maybe there’s slight Z movement in that scenario… best practice is if: Software is off, box should be off.
If you do bring up the software without power cycling, E Stop and re-home.
For those that have loose tool holders I’d check these things and get some new tool holders so they’re tight. (Personally I have a couple that I smashed up during testing, and despite bending them back they’re still a little loose)
I just had a video call with Corey and we have some testing procedures I’ll go through to try to root out the problem.
In the mean time, I did 3D print the “safety net” and installed it and it seems it will work fine so I’ll make more for the rest of the holders as a “just in case”. No down side to installing them to be on the safe side. My larger surfacing bits I keep on the wall next to my manual change location so not having the maximum diameter available in the carriers is not an issue for me.
I know there is lots of good discussion about the underlying causes, but I wanted to talk about the 3d printed part of your question. I describe my experiences in the attached post, but the TL;DR is petg and abs broke quickly if they were used for the tabs. What worked well was having a TPU insert to get squished. You will have to experiment with your slicer settings. (My post talks a bit about what worked for me.)
Of course this was all done before avid had come out with their solution, which is likely better than my 3d printed one.
To answer your questions, yes I have recalibrated the location of the bins 4 times now, and no I still have the issue from time to time. I am working with Avid’s support team and they are awesome. We had a video call yesterday and have developed some testing procedures to try to figure out the root cause.
I never intended for this 3D printed part to be a solution, but rather a safety net. Each incident drops a $90 bit pointy part first into the floor. I thought these would be better than building a foam lined box or shelf, and I had the hardware and a 3D printer at my disposal so I never had to leave the shop or buy other materials. I shared this as others on this string seem to have the issue as well.
Here’s what I’ve been doing - augmenting my fork-style rack with “hole” pockets. See the foam underneath to prevent the extra drop. I can easily remove these when I need to use my full table width (a square indicates them in).
I just dropped an ER32 tool holder into my setup and it works. The collet nut just touches the 3D printed part but it fits. Would be best to reprint to give a 1/4" or so of additional clearance for use with ER32’s.
Ah, good point. I don’t have any ER32 tool holders yet. I didn’t notice the nut was that wide. I guess you could still put a safety catch down below to catch the nut surface, but it probably wouldn’t be a good Z height reference surface, unless the nut tightens to the same place across all collets and tool holders.
Awesome.
It gave me an idea to modify above. This is with ABS. This one is not sanded - Still have to sand the inside ridge 400+ grit to have some slide effect. Just my impression that the ABS gives a little which makes a much softer delivery and retraction. More forgiving for homebrew dust boots.
Eric - still testing what caused the damge - just installed the Beta and ran out of time to map the tool holders.
For those of you that have experienced tools getting pushed through the tool holders, in particular @ckporto123 and @Pertech
Have you ever done the following:
Powered on the machine, homed it and used it. Then did you power off the software but not the control box, and then later started up the software and continued using the machine without re-homing it?
