Proximity stop setting distance

Using the right proximity stop is part of the puzzle, it is a trade off of high refresh rate, sensing range and it needs to be shielded. The unshielded I would expect to be less reproducible vs. shielded. Since I embed my sensors, I can only use the shielded types. I use triple sensing shielded sensors typically with refresh rates in the 500 Hz or higher and sensing distances between 12-20 mm depending on the model and if it is a 18 mm or 30 mm body. The one below has been very consistent and you should be in the +/-0.001" repeat range for a a 12 TPI or higher, but as you increase the speed it might vary to +/-0.002 at 500-600 RPM. I use a 30 mm sensor on my lathe, I would figure I could thread on upwards of 1000 RPM, but see no need when I get very clean threading at 300-600 RPM.

Other factors are the type of VFD being used and also the programming parameters, as mentioned too quick a stopping time will decrease accuracy if the VFD is trying to prevent an over voltage buss error. On the Yaskawa drives I disable this parameter, but also make sure that I do not use too short a stopping time. On heavier machines I may use DC injection for stopping the last couple of Hz, but it all depends on the lathe/system and requires some parameter tuning.

Another thing to think about is rotating mass and momentum, if the mechanical top speed is set lower say 200 RPM at 100% speed there will be more mass spinning for the same RPM vs. having the gearing set for 800 RPM and the speed pot is turned down to 25%. Similar to a mill or lathe in back or low gear vs. normal/high gear, the former takes longer and is harder to stop. Using an encoder might give tighter control, but at the end of the day there is a limit to how much energy the VFD can dissipate when braking.

I also recommend using an additional small limit switch in series with the sensor as a backup to the proximity sensor should it fail or one drives the carriage in the wrong direction. Last but not least, I have been adding chip shields over my sensors, as a stray chip or swarf will instantly trigger the sensor.
 
I’m using a Turck NI8U-M12-AP6x sensor, it has an 8 mm sensing distance that I measured to be an actual 7mm, and a 2 khz switching frequency, so ok for testing purposes, but not the best choice for long term use. It is also a NO sensor, so I would get an NC sensor if I stick with this. This is my hokey set up for testing, as you can see the lathe needs a good cleaning from the last project. I’m also using a PLC for my controls, so I will need to check and see if the scanning frequency could cause any delays in the circuit. It’s a short program, so it shouldn’t, but you never know until you check.

C0E6F542-16F6-4F3B-9E6C-AA318CCAD9D8.jpeg

I didn’t have room to add another relay in my control box, so I had an Allen Bradley Micrologix 1000 PLC on the shelf that I programmed for this. I had to dig out my 20 year old laptop to use the programming software, I’m glad that still works, no way I am paying AB prices for new software. I programmed it so that the sensor input latches the stop circuit and will only unlatch when going in reverse, so there should not be a danger of accidentally going in forward after going to neutral (I tested that several times). The downside of that is if I am doing imperial threads, I still need to go in reverse, but since just about all the threading I do is metric, that is a small inconvenience that I will figure out a work around. That also enabled me to install a forward and reverse jog joystick instead of just having forward.

My VFD is a Teco 7300CV that I got probably 15 years ago and it is still working great. I hope I did not just jinx myself. It is more robust than most current designs and can do a decent amount of braking internally without a braking resistor. I can’t thread at 500+ rpm speeds and stop quickly, but at lower speeds and less than 3mm pitch, it was stopping very close to what I predicted for the decel time.

The inertia difference between running the motor at 100% and 25% makes sense that it would be easier to stop at the lower speed, but I would have thought if the braking resistor was functioning properly that it would be able to stop that as well in the required time to be predictable. Maybe something else to add to my testing if I have time.

I’m off to shoot a few hockey games, so this will have to sit until those photos get edited, but I appreciate all the information you have given. This discussion will help many others here as well I’m sure.
 
I just tested using high range and reduced Hz on the speed pot. There is a noticeable difference in stopping speed. 10 TPI at 250 rpm with 6" chuck was .047" less travel when the prox sensor tripped. That's almost a half a rev on the chuck. It's fast. Only change was selecting a gear that would give me the speed at 30% on the pot vs a lower gear and 65-70%. My system was purchased from Mark, uses his recommended settings with a 90Hz top limit and braking resistor.

Mark, once again you have taught me something.
 
I too did some more testing, but took my chuck off to reduce the possibility the VFD needed the braking resistor, and gearing for a higher speed and setting the VFD to 25% still stopped noticeably faster. I’m glad I started this thread since I never would have expected that.

Something else that I just realized about threading in reverse, this may have been mentioned in some of the other threads I read, but I don’t recall seeing it, I am limited to about a 3” bore with the left hand tools until I run out of cross slide travel. I don’t know if I will need to thread something that large, but it would suck to buy tooling to reverse thread and then have something just larger than that I need to thread and not be able to do it. I could reinstall the compound and get more capacity, but haven’t checked how much more that would get me. For the small left hand Micro 100 tools I bought, that probably won’t be an issue, but for the 16IL bar I was thinking about, it could be.

I spent a bit of time cutting some internal threads, it took some practice to get close to the stopping depth I wanted, but it looks like once I get some more experience it will get easier to do. I’m going to play with this some more, but I’m liking this more than I thought I would.
 
Ischgl99 said:
I too did some more testing, but took my chuck off to reduce the possibility the VFD needed the braking resistor, and gearing for a higher speed and setting the VFD to 25% still stopped noticeably faster. I’m glad I started this thread since I never would have expected that.
Click to expand...
I'm glad you did too.
 
You must log in or register to reply here.
Back
Top