My CNC Router is Plagued with RF Electrical Interference and Triggering False Limit Switch Contacts
Вставка
- Опубліковано 12 вер 2024
- I cannot run my home built CNC router without using Mach3's software settings to help filter out all the noisy lines that gives Inputs false positive triggers and halting the machine. The input lines going to the parallel port are for the limit and home switches. I have already grounded all the parts of the machine and that has helped some but it is still necessary to use the De-bounce function in Mach3 control software to run properly.
In this video I show the symptoms of false triggers and how you can use software to help alleviate the problem. A optical isolator breakout board is also wired and tested as a possible means to resolving this problem.
If you enjoy these videos, please consider helping support our channel here (THANKS):
www.paypal.com...
The Guru Brew is a weekly entertainment show that is posted on the internet every Sunday. Join us as we document our techie adventures where something fun and new is always brewing.
Please subscribe to our UA-cam channel and ring that notification bell next to any of our videos and you will notified when a new video is available. / gurubrew
Music Courtesy of Incompetech.com & AudioNautix.com
Visit us on the web at: www.GuruBrewSho...
Have a question to ask the Brew? www.GuruBrewSho...
*Our Second Channel / gurubrew2
*Twitter: / gurubrewshow
*Facebook: / guru.brew
You earned a subscriber. Finally got my plasma table operational. Trying to determine the best configuration for limit switches while avoiding emi issues. Discovered opto isolators and now this video. Very informative thank you for a beneficial explanation
A lot of break out boards are already using opto couplers (I’m not sure what one you are using). I came here looking for a solution as I’m having the same problem and your video convinced me that moving normally closed instead of normally open is the bigger fix.
Steve glad to see you back on the UA-cam scene, very good information in this video
New sub, I really like the range of topics and style of the videos. One thing (there's always one thing, right?) you could really use some better lighting in a lot of them. Nonetheless, great videos.
Hey Steve, over here working on my homebrew machine. I have trouble I cant even start to understand because I an so stinking new to electronics. Thanks for giving me a few ideas to research to start trying to work out my kinks.
We miss you mister gurubrew. I'm glad that you are back!
Glad you're back! Also, get the tick tested to see if it's one that carries lime disease...
Great to see you back Steve.
Similar problems, old solution, CNY17 single optoisolator. Can replace an LED, can drive an LED. But it has got a maximum frequency of some 20 kHz.Does great denoiseing if that frequency is fast enough. I actually use it for some 2 kHz. Good old computers used to have current loops not rs232. the CNY17 is a phototransistor and an led. You can either pull an input up or down. pulling down is inverting the signal.
Thanks Steve I have the same problem and thanks to you I knew the reason
Hey Steve, glad to see you're back. I've always loved your videos, can't think of any you've done that made me wonder why I'm subbed to your channel. This video piqued my curiosity, I recently built a tachometer for one of my drill presses which has a 3 phase motor. The tach was built using an arduino and when testing on the bench it worked perfectly, but when I mounted it on my drill press in proximity of the motor and VFD I use to power it, it gave random results. I tried shielded cables and it didn't seem to help. I did find a solution using an EMI filter for my A/C line but now I'm wondering if perhaps I should have looked into an optocoupler as another solution. Anyway, looking forward to how your experiment ends up going. Thanks for your Vids!
Larry, thanks for your kind words and experiences. I'm wondering if an AC filter fixing your problem wasn't due to fluorescent lights on the same circuit. I'll keep you posted what happens but right now I'm not going to work on it. Got other stuff to do.
When I was younger my older brother built (modified, actually) a pinball machine and faced the same problem - he solved with twister pairs, aluminum foil and .... optocouplers. It would be really cool if fiber optics could be used to do the opto coupling - for the distances needed even a fishing solid nylon string is a fiber optic. BTW, the noise also comes from the PWM controls and pulses for the stepper motors.
False limit-switch triggering can also be caused by vibration at the switches - often cheap switches will have a resonant frequency where they will "bounce". Better switches are more immune to vibration.
From your first demonstration, I would say that yes, this problem is most likely RFI.
Hello Steve, Your vast knowledge and skills are amazing. Thank you for your content.
Thanks you're always very kind
something to watch out for when i finally get around to finish building my CNC
Hi Steve, nice video! Do you use a Breakout board? Breakout boards have opto isolation on them.
I had the same problem, specially induced by a VFD, I grounded all and I used a Icstation 12V to 5V 8 Channel Optocoupler Isolation Board (I had also the wrong inductive sensor :)). It fixed the problem.
this is something i worked on professionally for a long time. one quick fix is to ground your pc chassis to your control box chassis. ie get all your chassis grounds to a common point. also be cognisant of fluorescent lighting fixtures and their proximity to the pc/table/control box. another fix is to pound a grounding rod into the floor next to the table and attach it the table to it. there are several other historically successful remedies, reply back if need be.
I did ground everything that was metal including the motors to the chassis on the controller and computer and that didn't seem to help much. You do bring up some other interesting ideas, thank you!
@@GuruBrew somewhere in the mach1 software there is a command to invert you step signals from rising edge to falling edge....switch to falling edge. when the drivers respond to rising edge it will make them less noise immune as the signal itself looks like AM/noise.
@@2007rgallo this is something I could try that would be easy thanks again
Great video as usual! Would you know what the problem(s) would be with setting the debounce setting too high? You had mentioned 200 and another person had theirs set to 1000. (What is too high?) I poked around a bit but can't find any useful information. Any guidance would be appreciated... Thanks Steve!
VEry professionally looking build.
Hey Steve, thinking out loud here, would it help if your switches were actually wired normally closed ?
Also wondering are you using shielded cable for your limit switches?
Not sure but that's an interesting experiment I could try. Thanks
Hey Steeve thanks for all this info. You have been the best help for me since I start working with my hobby CNC.
Interesting video. I was wondering though if you could not use twisted pair wires on your limit switches with some sort of shielding and ground the shield to the frame at both ends. Optocouplers is good idea as well though.
good information. Android playstore has an emi meter!
Hey thanks for reminding me I have one of those I used to use it for ghost busting. I'll get it out and play with it
have you thought about using shielded cable for your limits?
Thanks to your videos im scrounging around part for jgro router. Thanks!
I feel kind of stupid and I'm not using shielded cable now that you guys mention it
@@GuruBrew dont feel stupid. We are all here learning and sharing info with one another! With what ive seen in your videos you are far from stupid!
So what's the best solution for false limit triggers? I get the same issue with normally-closed switches, an arduino and cnc shield. Adding optocouplers would probably work but just increase the complexity and I'm trying to keep it as simple as possible, same goes for adding those RC filters from grbl's documentation. What would you suggest?
Thank You ..
Hi Steve, in a 24 v limit switch system , in a very noisy environment , do you recommend to put the limit switches on a normally close state wired to the + side of the 24v supply while the ground of that 24v supply wired straight to the controller ( of course , via the optocouple
Great video, thank you!
I had a problem with my cnc when i first built it and i ended up running a copper ground wire through my dust collection hose and grounded it on the dust collector end, haven't had a problem ever since.
Great idea. Air flow and dust can raise quite a bit of static electricity.
You're the best
Hi Steve
Have watched your videos on homing and limits setup I have wired my machine the limits work OK but when I try to home it go's to the limit switch hits it and starts to travel the other way and wont stop I have disable the limits and just let the homing enable with no luck have you or anyone of your follower have had this problem most likely there will be a setting I have not done
This is my first time using Mach3 I am only using the X & Y having a fix Z it is a plasma table the plasma is not setup yet trying to get the table to work first
So spindle EMI stsrted to attacked my USB and only the USB. To fix it i had to add a USB isolator and also wrap the cord in an EMI jacket and bond it to the machine ground bus.
All that to figure out i had created a ground loop with my laptop using a different outlet than the machine was using.
Anywyas i over killed the issue, but it meets requirements 2x 😅
Are you clear the problem using this opto-isolators
use power conditioner from amazon to eliminate your issue. The vacum and spindle is the contributor to your issue.
Hi, can I prevent interference if I put a ferrite core on the cables?
I have same issue with my home brew machine...my homing switches work but when i enable hard limits they fault immediately....did u fix ur interference?
Hey Steve what about changing your signal input to low that triggers the end stop. Normally closed switch that opens when the end stop is triggered.
Sry just seen the other post suggesting this very same idea
Is the quality of the China boards inferior to the real deal? I’ve had some China made hardware burn out easily, especially LEDs
That does seem to be a theme with knock-offs you're right