Yay what a legend! Thanks. I'll definitely try this out. Sharing video to DIY CNC group on FB - Ken

Awesome, good luck! Let me know if you run into any issues, or have bug fixes :) The tool change macros seem pretty rock solid so far, but I've run into a few sporadic issues with the tool setter and incorrect offsets (may or may not have plunged an end mill through a part 😅 ). So beware fast rapids to start if you use the touch off code too :)

@@BreakingTaps thanks. I like that 3d touch of sensor you have. I might buy 1.

Thanks! Yeah I think it's going to take me a while getting used to the fact that clearance is clearance and it doesn't matter by how much. :) Makes me clench each time it does a tool change haha.

Yeah, I forgot to mention in the video but that's an aspect I'm still debating with myself. The current setup (with the spindle slightly higher than the bottom of the Z-axis plate) is configured such that the bottom of a loaded tool holder is just under the plate. Meaning it gives the most Z clearance possible on the machine. It has the side-effect of requiring the "stair step" tool holder and really complicating the clearances. If I move the spindle down, I'd lose maybe an inch? of available Z height to work on parts. But as you said, it would let me set the tools much lower (or allow longer tools without drilling holes through the table), and simplify the whole rack setup. I'm not sure the best tradeoff to be honest. Still mulling that one over :)

@@BreakingTaps Sorry, not sure you understood. If you cut the table plate (aluminum) where you have the bits set currently then you can then place them lower in the table. You don't lose any Z height but your table is cut. Your idea makes me think of something really cool. If you ever add a A axis to your Z, you can then "tilt" your tool holder and set it off the table. The spindle can swing to a 45 degree and pick up tools. Tools ---> /__________

Ahhh, I see, yes definitely misunderstood. That would definitely work! Although depending on how the spindle is situated on the Z-axis plate there is a limit to how low it can go. E.g. right now with my setup, the spindle nose can't get lower than 3-4". I set it up that way on purpose because most of my work will be on top of fixtures or a vise so I wanted extra clearance, but if you work mostly flat to the table it'd be fine to lower the spindle on the plate, and then can grab tools "sunk" into the table Re: tilting tool rack... check this out! Just stumbled on this the other day: kzfaq.info/get/bejne/opiGhZalnNLeo4E.html Pretty cool! Agreed re: 5th axis. I considered it briefly as well (with a trunnion style setup) but there's just not enough Z axis to make it worthwhile imo.

@@BreakingTaps That tilting tool rack is pretty sick! If 5th axis then you won't need more motors and boards, but it's still super cool idea to not lose cut room.

Oh boy, thanks for the warning! That sounds terrifying. I don't intend to run this "lights out"... but I definitely don't want to babysit it either. I'll spend some time noodling over a tool release sensor. A proximity sensor seems like a good idea, simple and robust assuming it can be mounted somewhere that doesn't interfere. An optical infrared/laser sensor might work too. Hmm. Appreciate the tip!

@@BreakingTaps FWIW the KUKA machining center I use to run used optical sensors to see if a tool was in the rack. It provided the necessary feedback and allowed for lights out operation.

@TheSeaShadow did it have a sensor per-tool-pocket, or a sensor somewhere on the spindle that looked at the loaded tool holder (or maybe looked down at the tool pocket)? Thanks!

@@BreakingTaps There was a sensor per pocket, the machine used software logic to determine what was in the spindle itself. So it kept track of what tool was called and made sure that spot in the rack was still empty. If any of the individual sensors tripped without being part of an intentional tool change it would cause the machine to do a new touchoff when that tool was called.

Great, thanks for the info. Will mull over the options and see what I can come up with.

I've been pretty happy with it so far! I definitely understand the reservation, the checkout is a bit sketchy as well (a russian-based credit card processor. Legit, just gives off strong 1998 ecommerce vibes :) ). But the price was so good I couldn't resist trying. Placed order, got a receipt, it showed up 4-5 weeks later. I get the feeling it's a one-man operation and he manufactures them on-demand + overseas shipping, hence the long lead time. I had a small electrical issue (bad solder connection), but the owner (Serguei) was super responsive over email helping to diagnose and fix.

Yep, that's the plan (theoretically anyway haha). It's not ideal at 3", but I already owned the facemill and inserts from my manual machine so it was relatively cheap to get a BT30 holder and try. Calculators say it wants to spin at ~3k, which is dipping awfully low on the spindle torque curve. I'll have to play around to find a reasonable low speed that still has enough torque, and maybe try the torque boost setting. On the upside, it'll be balanced unlike a flycutter. I haven't cut anything yet so haven't dealt with the fogbuster, but I'm planning on relocating it. I was already having some issues with it's location anyhow (bumping into clamp studs, etc). Current plan is to bolt above and to the right of the spindle, then "reach" down over the front. Should avoid any obstacles and the toolholders.

I have not, and it might be a complete failure :) Definitely can't run it with all the inserts, no way the machine has the torque, feed or rigidity to handle that. But I'm hoping with a single insert as a flycutter (for shallow passes, say 10-50thou) it'll do a passable job. If not I can always reclaim the location for more tools, or a real fly cutter. I already had the facemill and inserts from my manual machine, so it was a relatively cheap investment to grab a BT30 holder for it and try.

Beautiful Face Mill but, as you said - probably not enough spindle power to drive it with all the cutters in place.

Breaking Taps I hope it works!

Thanks! Can't take credit for the probe, although I do like it a lot. Probe and toolsetter are from vers.by/en/ (no affiliation, just a happy customer)

I like them a lot! I have the wired "TS" tool setter, and the wireless, non-rechargeable "WL" probe. The TS connects to the wireless receiver, so they can share the same cable/input pin which simplifies setup. The rechargeable model wasn't available when I purchased it (3-4 months ago), although I would definitely get the rechargeable version if I were getting it today. I plan to swap the battery and add a recharging circuit to mine in the future. Haven't done any formal accuracy testing yet, or even really dialed them in much (e.g. the tool setter has grub screws to adjust tram so it is completely parallel to the spindle). But from casual usage they seem good. I did have a small electrical issue with the wireless receiver, but Serguei was super helpful over email getting it sorted out. For the price honestly I don't think you can beat them, especially for hobby usage. The next cheapest I could find (Drewtronics) was hundreds of dollars more. Only downside is that I think he manufactures them on demand and shipping is overseas (for me), so it took a few weeks to arrive. I did consider an optical laser toolsetter, which you can get pretty cheap from places like AutomationDirect, but since I was getting the probe from Vers anyway I went ahead and picked up the toolsetter too.

Will do! Seems a lot of folks are curious about the probe and toolsetter, especially given the relative lack of info on the internet about it. I need to fiddle with my macros a little, but will record some footage and give an overview of the system. Thanks for the suggestion!

No problem :) To be honest I haven't done any rigorous testing of the probe accuracy and if it's better or worse than the machine's positioning accuracy. If I had to guess though, it feels like the probe is accurate enough that positioning error of the machine is the main limitation. I did find that there is a certain position that the probe isn't very accurate in (it uses three pins/springs to allow deflection, and at just the right angle it doesn't actuate smoothly) but that's generally not an issue. Considering the price it was a nice upgrade from the contact block kind of probe that's typically used, without selling a kidney :)

I cleaned the code up a little and tossed it onto Github: github.com/polyfractal/Mach4ATC Hope that helps! Unfortunately the screenset is saved as a binary format, so you can only view those scripts by going through the Mach editor :( My screenset is a mix of default and Avid, I'm mainly using Avid but grabbed a few things from the default (e.g. probing tab, some tool DROs, etc). There are a few wonky things though, like corner probing can behave strangely at times so I may not have moved it over correctly/fully. I have my drawbar solenoid on "Relay 1" which Avid maps as coolant, so you can see in the code that I use the coolant signal: github.com/polyfractal/Mach4ATC/blob/master/Profiles/AvidCNC/Modules/ToolChangePositions.lua#L127-L131 (Relay2 is my fogbuster, which they map as mist) Goodluck! I'm still super new to Mach, but happy to help if you have questions...maybe we can stumble our way through it :) I wish Mach had better documentation, most of my scrips were cobbled together from reading other scripts/modules/macros :'(

Thanks for the help getting started. (downloaded the GitHub stuff). Newbie to Mach also. get most of my programming help by reading and tweaking existing code too. slowly becoming familiar with Mach Screen editor. Cut and past screen components from one screen set to another still alludes me. Haven’t found any Documentation help but then again - it may be technically impossible.

There's an import/export mechanism for individual components. Find the component (or group of components, like a tab or panel group) in the treeview on the left, right click and there should be an "export" option. Then close the screenset and open the destination screenset, find the component in the treeview where you'd like to import the components (e.g. if you exported a tab, find the destination tab group) and right click -> "Import". It's a little fiddly but seems to work for the most part. Unfortunately if there are global scripts elsewhere (like in the PLC script) that the imported components need, those don't come along for the ride and might break things. I'm not positive, but I suspect that's why my imported corner probing screens don't work quite right.

Basically, yeah! I have one and use it on occasion, when I'm feeling lazy or Mach is not behaving. It has two sides: one side is flat and lets you take a Z measurement (either to find the z-offset of a surface, or to measure a tool). The other side has a corner notch which lets you put the probe on a corner of your work. Avid's Mach screen includes some probing routines to then touch off the sides of the plate, which will give you the corner position. The side with a corner notch can also do Z measurements too, so you can probe height and x/y of a corner all in one go. Only downside is that it needs a corner to work on, and it can be a bit fiddly (e.g. if you forget to attach the lead to your spindle when probing, it'll crash :) ). But it's super simple and easy to use otherwise!

My Avid hasn't seen as much use as I'd hoped over the years (various unrelated issues) but I'm still pretty happy with the probes! The toolsetter works great, and the wireless probe is good other than battery life. Mine is the older, non-rechargeable design and I find it doesn't do a great job auto-sleeping. So battery dies pretty quickly and replacing it is a pain and requires re-calibration. I think the newer version just has a USB charging port so much easier. All in all pretty good value for the money I think. Shame they discontinued the side-mount. It's the same body as the wireless probe, but I guess they decided to consolidate the models even more by only having vertical option.

Awesome, congrats! Did you get the s30c avid "bundle", or just the individual parts? The bundle wasn't available when I purchased but I think it is pretty straightforward and includes most everything you need. If you _didn't_ get the bundle, there are a few extra bits to acquire, and wiring Avid to the VFD can be a bit tricky but not insurmountable. What VFD did you get?

@@BreakingTaps Did get the bundle, which uses the same VFD Avid includes with their standard spindles.

Ah ok, yeah, if you got their turnkey bundle it should be pretty easy. The main difficulty was translating the Avid wiring pin naming scheme into the CNCDepot pin naming scheme when wiring them together, and programming the VFD (I still have a few things wonky that I haven't been bothered to fix yet...). But the turnkey bundle should take care of all that for you. Otherwise it was super straightforward: mounting the spindle was easy (I used the Avid's tramming mount), hooking up pneumatics wasn't too bad even for a first-timer like me, etc. Random things that I remember from getting started: - Bearings will make some "rattle" noises at high speeds when you first get them, it's normal. They "wear in" after a little while and the noise goes away - I like to let my spindle warm up for a while before really using it, the high-speed bearings are fit loose (relatively speaking) because they expand at their working temperature. So you'll have less accurate parts if you use the spindle before the bearings are warm - The spindle is case-pressurized to keep dust and other stuff out. Since I mainly work with metal, I've dialed the pressure down pretty low (~30psi iirc?) which helps save air, as it is constantly venting to atmosphere. If you're doing woodwork though, I'd definitely keep it at the recommended pressure (~100psi) so that wood dust doesn't get inside the spindle. - I have a ~27gal air compressor setup, but my compressor is a "silent" variety and only does ~3 CFM. It runs at around a 60-70% duty cycle. A bigger compressor that has higher CFM would definitely be nice but the one I have works well enough. Feel free to ping me if you have questions while setting it up! Both Avid and CNCDepot's support are great too, they were very fast in responding to my questions. :)

Yep! The macro will go check the tool length on the toolsetter probe and records the offset for for future use (although right now the other macros also re-probe each tool before use so it's not actually being used, but someday will :) )

That's a good idea! I'm actually not sure what would happen right now. There is a defined probing distance in the script, and it will throw an error if the probe isn't triggered in that distance. But I'm not sure if that is a "fatal" error like an e-stop, or if it just displays a warning and continues with the program. Definitely a good idea to kill the program (or at least pause and wait for the operator) if it doesn't detect a tool.

Aha, so I realized after reading Rod's comment that I misread yours. I don't currently have any kind of sensor to guarantee it released a tool before moving to the next. Will put some time into considering how to implement. Thanks for the tip!

Toolsetter and probe are from vers.by

Sure, no problem! I'll dig them out this weekend and tidy them up a bit, feel free to ping me again if I forget :) Fair warning, they haven't worked out as well as I had hoped, and I'll probably redo them or replace them at some point. The main issue is that different toolholders and tool sizes can cause differences in height which affects ATC.

@@BreakingTaps Hey, I'm trying to configure your module to use with tool height offsets while I wait for a probe (VERSBY probe is back ordered for a while). Would I just remove the "function to probe tool" section from your ToolChangePosition.lua file, and then set the height offsets in mach4 ? Thanks for any direction here, your videos are super helpful!

Hmm, it definitely _can_, but I'm not sure how reliably it can. Whenever I've needed a good sliding or press fit I rough it on the machine, then bore it out on my manual mill. Cutting steel is definitely at the edge of this machine's capability... it can do it if you're careful, but repeatability is a challenge.

I think that is a translation typo. E.g. the WL probe has the same "±4mm" spec, but that page describes it as "Permitted deviation stylus in XYZ directions". That feels correct based on handling the probe: the travel in any direction is ~4mm before the internal mechanism bottoms out. I think the tolerances of the probe is instead the "Unidirectional repeatability" spec, which is < 0.003 mm (3micron). I haven't tested with an indicator but it is definitely more sensitive than ±4mm. I'll reach out to Sergeui and see what he says about it :)

I wired it up with a 4-pin aviation connector like Avid uses for it's own touch/proximity sensors, so it just plugs right into one of the AUX ports. The wired tool setter is connected to the wireless probe receiver, and they both send the output to AUX1. :)

@@BreakingTaps thank you

Breaking Taps just received my tool setter, any more details on wiring it up? It came with a little resistor and I bought some aviation plugs. Just confused on how which wires go where and what to do with the resistor. Thanks

@@LoST5757 Heya :) I'll pop out to the shop today and take some photos, don't quite remember how I have it wired up. I have both the toolsetter and the probe wired together, so my setup might be a little different depending on what you got. Details to follow!

@@BreakingTaps Thanks, looking to do it similar to you, just a traditional wine rack style instead of a 3d printed tool holders.

Hey, no worries, no offense taken :) It's a good question and to be honest I didn't even think about wood, probably because I'm not really setup for wood routing at the moment...despite having a router, it's geared towards metal right now: no wood router bits, no dust collection, etc. 3D printing was also just easier, in the sense that I could knock up a model, hit print and check back in a few hours. Whereas I'd have to supervise the router while working on wood. That said, it's not a great reason and this part definitely could have been done in wood. Not tight tolerances, simple pockets that are suited to big block of pine or maple or whatever, etc. I did print in PLA which is at least biodegradable _eventually_ (I don't buy the "compostable" line about PLA either, studies show it will still take 50+ years to degrade... but at least it's faster than petroleum based plastics which are 1000+ years) I'd agree that 3D printing for "green" aspects is overhyped. In my eyes it's useful as a quick prototyping tool, and a process to build parts that are hard/impossible with subtractive methods. For this part it was mainly the first reason; convenient and hands-free.

This I got wrong about 3d Printer's too. Almost anything is printed in PLA, this is a BIOplastic made out of lactic acid. The lactic acid comes from corn starch. I think for this special case this is way more eco friendly than wood or metal.