Reply with QuotePMDX.COM - Products for CNC and motion control applications
These say the have a schmitt trigger built in.....although I don't know what that really means.
yeah, I have come to the same conclusion talking to some folks. I have some schmitt triggers that I'm going to use and see how that works out.
Vibration may well be an issue, as these little optical switches are pretty flimsy and they wiggle with very little pressure. I may need to build some little firm covers for them
Ian
PMDX.COM - Products for CNC and motion control applications
These say the have a schmitt trigger built in.....although I don't know what that really means.
A regular logic gate will have a threshold voltage that causes it to flip from 0 to 1 but if you have a slow rising signal with some noise and a fast logic gate you can get a period where it flaps back and forth. Anything above say 1 volt is considered 'on' anything below that is considered off. But if you have a rising or falling signal that stays near the line for more time than you want then a noise can cause you float over and under it a few times before it transitions far enough from the line. Kind of like a night light that flickers on and off in a dimly lit room.
The Schmitt trigger has a band threshold instead of a single line. A rising signal crosses the trip point and sends the output to 1 but now the device is made so that it would have to cross a much lower threshold to flop back to zero so even if some noise is on the line the output goes from 0 to 1 and holds there instead of flipping back and forth with the noise. Works the other way too. Once it flips to 0 you have to cross a higher threshold to get it to transition back to 1.
It acts to filter out noise and also to 'square up' signals that look more like ocean waves than square waves for other circuits.
CNC: Making incorrect parts and breaking stuff, faster and with greater precision.
That's a very helpful explanation......cool.
So is it possible that the optical switch is falsely tripping and causing inconsistent homing in this case?
that's the idea. Frankly, I truly hope it helps, because I don't want to have to re-think my homing
Ian
Schmitt triggers made a big difference. Now my homing is accurate to within about 0.002". I turned a couple diameters, homing in between, and they were as bad as 0.003" under diameter or 0.001" over. I can live with that for my purposes.
Ian
That's good to hear about the improvement.
I intend to use an optical switch in the normal fashion, but only as a coarse homing switch. Once the coarse switch is triggered, it will enable another optical on a wheel fixed to the rear shaft of my stepper.
I'll be using 1/8 microstepping, so at a 3" diameter.......one step is about .006" of travel for the outer edge of the pickup disk. So in theory, the system should home within one step each time.....which will be near .0001" for my microstepping setup.
Its unlikely that I will need such accuracy.....but it won't hurt. I'm sure there will be other variables that will detract, but at least the switches shouldn't be the weak link.
Not being terribly electronics savvy, I'm still fumbling my way through how to make it all work in Mach. I believe that I will just use the coarse switch to enable the fine. I think I will probably need a long enough vane interrupter for the coarse to keep power to the fine until it does it's thing.
I would certainly welcome some greater wisdom on how to make this work.
I know enough to be pretty sure it can work in Mach ....... mostly from a confirmation of the theory from Steve at PMDX, but I'm gonna have to refine it a bit more.
That's the direction that I'm headed as well. I have encoders on order from digikey.
I dunno how to do it i mach, I'm using EMC2 for this machine. EMC2 handles switch + encoder homing pretty well. Also, I'm using a Mesa 7i43 card for my IO, which handles my encoder resolution as well as my stepgens.
Ian
That's interesting.
I'm inclined to use Mach simply due to my familiarity with it from my router. Do you feel that EMC2 is better suited for this? I really have no idea at all about EMC2.
Do you need to use Linux?
I wonder if Mach can use encoders?
EMC is definitely more flexible. Plus, it'd be pretty rough trying to get steppers and encoders wired up via parport for mach3.
When I get my stepper encoders in, I'll be using two stepgens, three encoders, and around eight GPIO.
Ian
I got my motor and VFD today, and got them installed. No computer interface to the VFD yet.
First impressions are that the motor doesn't seem to have much torque, even at full speed. I haven't done anything to configure the VFD yet, so it may just be a config issue.
Ian
Hi Ian......do you have any updated news on the motor and VFD?
which motor and VFD did you get? On my 1/2 HP motor I can't stop it with my hand even at 5 hz turning like 30 rpm. I'm using the 3HP Teco cheap VFD. Their wasn't much to change in the VFD except the accel and decell times and max freq. Also had to change the input from the control pad to the potentiometer for easy speed adjustment. I'm using .5 sec accel time and at 60hz it can hit the target. At 120hz start up it over shoots by maybe .3 sec but is acceptable. Decell is also about the same meets at 60hz and slightly overshoots at 120hz.
Jeremiah
PM45 CNC Build in Progress
I haven't been able to work with it all wkd as I am in Pittsburgh for the NRA convention.
It's the leeson 1hp iec motor and 1hp teco vfd. I set the max freq to 120hz, and it won't even turn the motor that fast. The max speed depends on the belt ratio that I have, but even with no v belt (just turning the counter pulley via the timing belt) it wont spin up to full speed. Stalls the motor if you turn it up too high.
I'm hoping I'm just doing something wrong.
Ian
I got the motor sorted out. 3 phase wiring is complicated! The motor's box cover has a diagram inside with four different wiring setups, marked Low/High voltage and NEMA/IEC. It was wired for high voltage from the factory. I guessed that maybe low was 230V and high was 460V, and I rewired it to the Low voltage diagram. I suppose I guessed right, because now it runs as I would expect it to.
Ian
So can you pretty much run one one pulley setup and still have the broad rpm range from the vfd......does it seem like it will be powerful enough?
I suppose the motor makes plenty of torque, but is it enough without the further drive reduction?
What pulley are you using?
It depends if your running a 2pole or 4pole motor. 4pole motors have twice as much torque as a 2pole but they have half the speed. You never get something for nothing.
Also if your using a Vector drive you will have much better low speed torque then just a V/Hz drive.
michael,
I was running it last night on the 720rpm pulley, which I believe is a 2:1 reduction. It's an 1800rpm motor, 1725 with slip, so if I run it at a range of 30-120hz, I should get 430-1725rpm on that pulley.
It seems to have plenty of low speed torque now that it's wired correctly. The TECO VFD I have is a sensorless vector drive, and running at 5hz I can't stop the motor by hand. However, since the motor is cooled by its own fan, I will need to rig up external cooling to run it at less than 30hz or so.
Ian
Ian......how did you mount your x axis ballnut?
I originally deepened and widened the carriage for clearance. When I went to mount everything, I discovered that the V way pokes through into where I had intended the ballnut to run, so I had to pocket the underside of the X slide. The slide has a couple holes drilled and tapped in it. I made a nut mount block that threads onto the nut and screws to the X slide with a pair of 10-32 SHCS.
Ian
Posting Permissions
