Tool changer, bad crash, broken sensor?

This morning I CS'd my 4020 machine. I went to change tools away from my probe to a small tool so I could warm up the spindle. I have a new spindle so sometimes the tools get stuck in the taper pretty good.

The machine moved the tool umbrella over to the spindle, I heard the tool break loose from the taper, but then the machine seemed to put the drawbar back in, and moved DOWN. this pushed the tool holder right down thru the tool carousel. The tool stayed in the spindle. There was a Z axis overload and e-stop.

I pressed jog, the tool changer moved back home, I pressed tool in/out to get my probe out of there, and I then CS'd the machine. Now, the carousel doesn't line up with the spindle. It is bent. You can see that some portions of the carousel are higher than others if you index through all 21 pockets. its bent maybe 0.125"?

I'm going to take the carousel off and inspect it.

what the hell happened here?

Some good news, the carousel plate itself is flat to within about 0.015" seems OK

It is possible if the spindle did not release your tool or if the Z comes down on the tool carousel to bend you turret spindle ITS: Turret Spindle, ATC-0027 . There is a switch on the turret spindle to stop that from happening majority of the time. Pictures would be great. if you can get an indicator on the turret spindle you will be able to tell if its bent. How this can happen is if the switch for the tool in and out isn't adjusted correctly (meaning to close to the tool in and out cylinder reading the magnet prematurely) the machine thinks the tool was ejected and begins the Z axis up but if the tool wasnt fully ejected it triggers the switch on the Turret spindle and at the same time the tool is popped pulled from the taper lock. But a Emergency stop was triggered and the tool in and out stops and Z axis Falls. That is a possible scenario.

And the way the switch is adjusted for the turret spindle. The switch is positioned as if there was any pull upward on the carousel the machine you fault almost immediately. Technically the switch is already compressed slightly. But as i said the tool in and out switch is very important.

Quote:

---

Originally Posted by rodney247

It is possible if the spindle did not release your tool or if the Z comes down on the tool carousel to bend you turret spindle ITS: Turret Spindle, ATC-0027 . There is a switch on the turret spindle to stop that from happening majority of the time. Pictures would be great. if you can get an indicator on the turret spindle you will be able to tell if its bent. How this can happen is if the switch for the tool in and out isn't adjusted correctly (meaning to close to the tool in and out cylinder reading the magnet prematurely) the machine thinks the tool was ejected and begins the Z axis up but if the tool wasnt fully ejected it triggers the switch on the Turret spindle and at the same time the tool is popped pulled from the taper lock. But a Emergency stop was triggered and the tool in and out stops and Z axis Falls. That is a possible scenario.

---

I agree with this assessment after futzing around with it all day.

here are some things that I looked into:
z resolver (its 2 years old i think) - it looks in good shape
DI screens - clock card test checks out fine
DI DS screen - switches all work (now whether they're setup right is a different question)
one thing about this that is irritating is that the tool in/out prox has a short lived blip on the DS screen - it is 00 when you press the button and hold it, then when you release the button, you get a 01 just as a blip. is that normal? its the same on my other fadal vmc tho - so i bet its OK?

for now, I took the turret spindle out of my other machine, as im all setup on this machine that crashed this morning. the old spindle is bent - about 0.03" as measured near the top, about 0.05" as measured near the bottom on the big fat part below the tool carrier. I overnighted a new turret spindle and new atc clips - will do the clips tomorrow and put the new spindle into my other machine (which is in pieces at the moment)

i've got the other machine's turret spindle installed - its a world better already. i'm not sure what to do to prevent this happening though.

do you know the procedure for adjusting the tool in/out prox?

thanks for your help!!!

Check these photos on pages 120 -123 page 122 tells how much shim and adjust for the turret spindle http://itscnc.com/images/Automatic_Tool_Changers.pdf as for the tool in and out adjust if a tool is stuck (you can simulate it by slightly with tool holder in spindle press it down onto a piece of aluminum) and then with switches on absolutely should not see it read the switch (tool is released). By the way the blip you see is because if i remember correctly while the tool in and out is pressed the screen is not Live its when you let go of the button you see if it was read. so with a tool pretend stuck you should not see a 1 blip on the screen. Thats why if the spindle happened to be in the correct orientation and started the tool change and the Air pressure was disconnected it wouldn't actually try to change the tool but would report "Waiting On Air Valve" because its looking for that switch at that point to proceed. I am looking for an adjustment for that tool and out switch having trouble locating an actual document. But if you looked at you other machine and remove the tool and out cylinder you would probably be pretty close to the exact adjustment as that one.

Quote:

---

Originally Posted by rodney247

Check these photos on pages 120 -123 page 122 tells how much shim and adjust for the turret spindle http://itscnc.com/images/Automatic_Tool_Changers.pdf as for the tool in and out adjust if a tool is stuck (you can simulate it by slightly with tool holder in spindle press it down onto a piece of aluminum) and then with switches on absolutely should not see it read the switch (tool is released). By the way the blip you see is because if i remember correctly while the tool in and out is pressed the screen is not Live its when you let go of the button you see if it was read. so with a tool pretend stuck you should not see a 1 blip on the screen. Thats why if the spindle happened to be in the correct orientation and started the tool change and the Air pressure was disconnected it wouldn't actually try to change the tool but would report "Waiting On Air Valve" because its looking for that switch at that point to proceed. I am looking for an adjustment for that tool and out switch having trouble locating an actual document. But if you looked at you other machine and remove the tool and out cylinder you would probably be pretty close to the exact adjustment as that one.

---

I read in the manual section 08: spindle drive systems there is a note about the tap tap cycle. in there it says to swap the air between the orientation and drawbar so when u program M19 it will hold down the drawbar. it then says to move in the sensor until you see 01 on DI DS, then back it off 1/2 a turn.

I gotta believe my sensor is OK since it looks like it has never been touched (also the wire is crusty bent into position like it was never moved). I did loosen the sw a little, held down the tool in/out button, and rotated the sensor - i didn't get it to got to 01 in about a 1/4 turn (can't do a full half turn due to wire length). that should be the same as the air line swap idea - thats just so u can do it without a helper.

I watched the tool change routine many times with no tool in the spindle by using TC,1, then manual to close, and repeated this a lot. the sequence looks good to me where the drawbar is clearly down for a fraction of a second prior to the Z rising up. now, about a tool being stuck. thats a bit a of a mystery to me. I have to say, I have never seen my machine do a 'tap tap' cycle. i definitely have seen it do the 'waiting on air valve' thing if i forgot to turn on the air at startup. but the tap tap cycle - dunno.

I have -4 cards and sys 101.3 processor proms


I can say that with the other machine's turret spindle in place, the tool changer adjusted, the tool changes are better than i have ever seen them in 4 years of owning this machine. it wasn't super well maintained before my time, and i fix it when things happen (and pretty much everything has happened so im hoping im somewhere near a well maintained machine at this point).



I did notice something bad a few weeks ago. I just moved my shop to a bigger building and am just now back in the game (until today that is). I had to take off my Z motor to get the machine out of the old shop and I put it back on here in the new shop of course. At that time I hadn't marked the location of the love-joy coupler and i had to guess at which 120° rotation the CS would be setup correctly on. I got that wrong, moved it the wrong direction, then got it right by taking the motor off, rotating the lovejoy and putting it back. thought things were good to go. then a few days later, I went to make parts. I CS'd the machine, and hit TC,1 to load up some tools - and bam - no go, the tool change was stuck on the carrier like the CS was wrong. So, confused i moved the motor again, and re-CS'd and have been off and running for a few weeks.

This should have been a big warning sign. when I bought the machine the tool changer was poorly adjusted and the carrier is all damaged up from crashing into the dogs on the spindle. A few months back before my move, I adjusted the tool changer and it was a lot better - but i guess not - the turret spindle was likely bent years ago when the prev. owner had the machine and they just f*cked with it until it mostly worked rather than buy new parts and solve the problem.

So now i've got a fresh turret spindle in there. its looking great, but I'm not confident that this issue is gone. how the hell did the machine go crazed, and move down - i still don't fully grasp that one. I like your idea of there being some momentum and after the e-stop it just kept going til it quit. but, why down? at this point i dont remember if i hit the estop or if the machine just faulted out on its own. the error messages were there tho - atc fault, and z axis overload.

hopefully this doesnt happen again. its $350 a pop on the turret spindle, and a whole day down, wasted.

about the shims and bellvilles. on both my machines (1992 ish era) i have 7 bellville pairs per side, and on my 4020, there is a 0.036" shim. maybe I ought to put more shim stock there to get it to trip sooner. looks like more bellvilles gets you more total tool load on the umbrella. I will add some more shim and see if that helps on a future failure.

Quote:

---

Originally Posted by mflux_gamblej

about the shims and bellvilles. on both my machines (1992 ish era) i have 7 bellville pairs per side, and on my 4020, there is a 0.036" shim. maybe I ought to put more shim stock there to get it to trip sooner. looks like more bellvilles gets you more total tool load on the umbrella. I will add some more shim and see if that helps on a future failure.

---

so i put some more shim on the atc failure switch for a total of 0.057" up from the stock 0.036"

now everytime you try to change tools - the switch blows and e-stops the machine. so, its too sensitive now. I will note that the tools stick in my spindle pretty well right now and pop out pretty hard when you operate the draw bar. I also should say that I pushed the sensor forward while tightening the screws on it. I checked DI DS to make sure it wasn't tripped all the time and its not. but now the vibration from blasting a tool out of the spindle trips the failure switch.

so, a thinner shim and i think this will make a nice improvement to the atc protection.

Quote:

---

Originally Posted by mflux_gamblej

so i put some more shim on the atc failure switch for a total of 0.057" up from the stock 0.036"

now everytime you try to change tools - the switch blows and e-stops the machine. so, its too sensitive now. I will note that the tools stick in my spindle pretty well right now and pop out pretty hard when you operate the draw bar. I also should say that I pushed the sensor forward while tightening the screws on it. I checked DI DS to make sure it wasn't tripped all the time and its not. but now the vibration from blasting a tool out of the spindle trips the failure switch.

so, a thinner shim and i think this will make a nice improvement to the atc protection.

---

on taking a closer look at the issue, the limit switch for atc failure is actually right at the top of the turret spindle's groove which would trip the sensor. this would allow for maximized protection against downward movements. if the turret got pulled upwards, there is like 0.25" travel before the switch would get tripped. this doesn't add up in my opinion. you may actually want to remove the shim. I did laser cut a couple 0.005" shims to fit in there - two are installed to maximize the height of the lim sw. this no longer trips atc failure on a sticky tool bashing out of the spindle, but it also sets it at the farthest possible point from detecting the tool changer getting pulled upwards. i dont get it. (please note that the picture is shown upside down for some reason on cnc zone)

Attachment 366592

Quote:

---

Originally Posted by rodney247

Check these photos on pages 120 -123 page 122 tells how much shim and adjust for the turret spindle http://itscnc.com/images/Automatic_Tool_Changers.pdf as for the tool in and out adjust if a tool is stuck (you can simulate it by slightly with tool holder in spindle press it down onto a piece of aluminum) and then with switches on absolutely should not see it read the switch (tool is released). By the way the blip you see is because if i remember correctly while the tool in and out is pressed the screen is not Live its when you let go of the button you see if it was read. so with a tool pretend stuck you should not see a 1 blip on the screen. Thats why if the spindle happened to be in the correct orientation and started the tool change and the Air pressure was disconnected it wouldn't actually try to change the tool but would report "Waiting On Air Valve" because its looking for that switch at that point to proceed. I am looking for an adjustment for that tool and out switch having trouble locating an actual document. But if you looked at you other machine and remove the tool and out cylinder you would probably be pretty close to the exact adjustment as that one.

---

Quote:

---

Originally Posted by mflux_gamblej

I read in the manual section 08: spindle drive systems there is a note about the tap tap cycle. in there it says to swap the air between the orientation and drawbar so when u program M19 it will hold down the drawbar. it then says to move in the sensor until you see 01 on DI DS, then back it off 1/2 a turn.

---

Thanks for this guys. My machine is pulling stuck tools out of spindle and I was just about to start a thread about drawbar sensor adjustment and happened to read this thread first and found my answer. So I should be safe to adjust sensor with no tool in spindle until it changes state and then back off 1/2 turn. And i can verify adjustment is correct by blocking an empty toolholder on a piece of AL and releasing tool, without seeing switch change state right?

https://www.youtube.com/watch?v=bU4jSS97rEA&t=9s

not for the not so patient person... i'm supposed to be making parts you know. one day i'll buy a robodrill anyone got a spare $80k? i dont..

Attachment 366618
Attachment 366620
Attachment 366622

It took 4 hours from start to finish to replace all 21 tool retention clips. that includes taking off the tool carrier, etc. Oh MAN what a difference!!! I'm surprised the tool changer even worked before. one downside is that you have to muscle the tools in a bit and could cut yourself on a tool. before the new clips some tools would be more or less freely loose in the tool carrier - also the tools would scrape the inside of the tool umbrella (not anymore!). I stopped using two pockets entirely because of it. now its freaking awesome. super worth the tedious pressing of roll pins.

I like the new tool retention clips so much that I am replacing the clips on my smaller VMC20 also. it has a 16 position ATC. What are the purpose of the pressed in 3/8" diameter dowels on the tool carrier? my 4020 machine has the holes on the carrier, but no pins. the vmc20 has the pins.

what for?Attachment 366992

those were for the the tap-matic tapping heads high rpm like Self-Reversing Manually Controlled Tapping Heads - Tapmatic Corporation there was a custom braket for Fadal that used the pin . they dogged off the spindle, there used to be a square dog pin in the face of the spindle. And the pins were there to keep the tapping heads oriented so when they head came down they were in the correct position for the bracket on the tapping heads. Or any other thing that you need to keep in the correct orientation so they didnt drift off. Not used much anymore so they elimated them unless needed. The pin holes are there if the customer needed them or used those type of heads.

Super annoying, my machine did the same thing. all the tips here are very helpful. The one difference with my machine is that i have a Numatix controller that was retrofitted which makes it a little more complicated. It is supposed to speed up the tool change time but at the same time it doesn't leave enough time for the sensors to realize that the tool is still in the spindle.

Quote:

---

Originally Posted by Thad92

Super annoying, my machine did the same thing. all the tips here are very helpful. The one difference with my machine is that i have a Numatix controller that was retrofitted which makes it a little more complicated. It is supposed to speed up the tool change time but at the same time it doesn't leave enough time for the sensors to realize that the tool is still in the spindle.

---

bummer on the timing.

i'm very interested in modifying my tool changers to either do servo turret (with either oem, or custom parts), or simply to increase the speed of the geneva gear. getting the tool changer to go faster in that regard is where there is time savings to be had. if you speed up the sequencing with regards to tool in/out and z up / z down, i could see how there could be potential for failure.

the geneva gear turret positioning is glacial. it totally sucks. one big issue that I see with driving faster is that the tools tend to wobble then settle down on every index. you could put on a drag brake, or an electric brake of some kind to stop it in the right place and lower the required settling time.

it seems easy enough to increase the speed of the geneva gear motor, but the mechanical settling time is the challenge there. i just get pissed off because if you watch a haas umbrella style changer it moves like 3x the speed, and it has spring loaded flaps that keep the chips out of the tool changer. im sure haas umbrella changers f up too, but at least when they work they dont eat up 50% of your cycle time.

The Numatix is great for 3D profile finishing but it is an error in the programming that allows the machine to essentially self destruct when changing tools. I am getting the programming fixed for $1800 so it hopefully wont happen again as this is not the first time it has happened. I had a guy from Iscar look at our production Fadal about a year ago and he had alot of ideas about indexible tooling for counterboring as we are using some step drills, but the he watched a tool change where the next tool was all the way across the turret and it took about 40 sec. so he then saw the reason for the step drills. We are hoping to replace all three of our Fadals in the next two to three years with something a little faster. They have been serving us well for 25 years and new machines will pay off quickly.

Quote:

---

Originally Posted by mflux_gamblej

bummer on the timing.

i'm very interested in modifying my tool changers to either do servo turret (with either oem, or custom parts), or simply to increase the speed of the geneva gear. getting the tool changer to go faster in that regard is where there is time savings to be had. if you speed up the sequencing with regards to tool in/out and z up / z down, i could see how there could be potential for failure.

the geneva gear turret positioning is glacial. it totally sucks. one big issue that I see with driving faster is that the tools tend to wobble then settle down on every index. you could put on a drag brake, or an electric brake of some kind to stop it in the right place and lower the required settling time.

it seems easy enough to increase the speed of the geneva gear motor, but the mechanical settling time is the challenge there. i just get pissed off because if you watch a haas umbrella style changer it moves like 3x the speed, and it has spring loaded flaps that keep the chips out of the tool changer. im sure haas umbrella changers f up too, but at least when they work they dont eat up 50% of your cycle time.

---

Easiest solution is to organize your tools and programing so it doesnt look all over the place. :) Tool change time is not too bad tool is close by.

If you increase the geneva motor speed, you will definitely need to add some kind of brake and it will need to be consistant. There is a sensor that is read halfway between positions and when this halfway point is reached the motor is switched off and everything coasts to stop at the next position. So coasting to the proper spot is important. And the stopped position is not even verified! It just assumes it stops in the section when the genenva drive locks. Amazing this doesnt cause more problems then it do.