![]() | |
| Home Page | Mark Forums Read | Today's Posts | My Replies | Classifieds | Reviews | Photo Gallery | Web Links | Share Files | Advertise With Us | Ad List |
| |||||||
| General Electronics Discussion Discuss basic electronics, power supplies and anything else electronic related here. |
| This forum is sponsored by: |
![]() |
| | LinkBack | Thread Tools | Search this Thread | Display Modes |
|
#1
| |||
| |||
I bought a used cnc router about a year ago. It was old and out dated. So I stripped the electronic from it. Alot of it didnt even work anymore. I bought a 3 axis cnc stepper controller custom made for my steppers. I bought a 1.34 hp Spindle and a VFD to control that. I bought a new computer to run the thing. Bought various other little things. But I have ran into all kinds of problems in the wiring of it. I get alot of random Limit Switch triggers while not even close to limit switch. I lose or gain steps ALOT. But this only happens when the spindle is running. When the spindle is off everthing works great. When i am cutting metal objects i seem to have all kinds of problems. Like if i put a peice of alum directly on the table and try and mill on it. Things get ugle. But if i put a peice of wood under the peice of alum and mill it seems to work better. But I still have issues alittle bit. I have tried all kinds of things. I tried running ground wires to the table in various spots. I tried running a ground wire to the spindle. I tried having a earth ground. I changed over to all shielded cable. I though maybe my cnc controller needed to not touch anything so I put a peice of cardboard under it and around it. So it was not touching any metal at all. But still things seem to go wrong. Right now I have my cnc controller which is a alum box sitting inside the old steel control box that I took all the old controls out of. And above it I have a peice of fanolic with 3-4 other electronic pieces mounted to. and I have my computer sitting in a box made of MFD mounted on the side of that box. here is a little diagram of how i have it wired up. If anyone has any ideas please let me know. Thanks Eric, |
|
#2
| |||
| |||
| Eric, I suspect that you may be getting noise and trash from the VFD and spindle into your 120 Volt electrical source, and it is present in the voltage going to the computer and other electronics. It can cause weird problems. Get an EMI/ RFI filter and install it in the 120 VAC power source to the spindle. Also, a simple test is to get the 120 VAC power for the computer and drives from a different branch circuit or different location. A better test would be to get the VFD and spindle power from your neighbor or a portable generator for determining whether the VFD is the cause. A Portable generator would be the best test. Get a portable AM radio, tune it to a weak or medium AM signal, and place it near the machine. See if the radio gets a lot of static or noise when the spindle is turned on and running. Tune the radio thru the entire AM band a few times and see if it is stronger at some locations. If so, EMI /RFI is most likely your problem. Hope this helps. |
|
#3
| |||
| |||
| You want a filter between the VFD and spindle or between the electrical box and the VFD? Also a quick question. I had a electritian install the VFD and spindle. and they did not put a starter on it. I would think something like this would require a starter to keep the spindle and VFD safe. Do you guys think it needs one? Thanks Eric, |
|
#4
| ||||
| ||||
| A contactor should only be needed on the input side to the VFD, it is usefull for dropping the Motor out when and E-stop occurs. I would say most of your problems stem from poor grounding practice, You need a common ground plate set up where the service ground and all Metallic objects, table, motor case, and if possible all power supply commons and shields are taken back to. it must be a common point. Al.
__________________ CNC, Mechatronics Integration and Machine Design. “Logic will get you from A to B. Imagination will take you everywhere.” Albert E. |
|
#5
| ||||
| ||||
| 1st thing I would look at is the active state of your home/limit/estop switches. I'd bet they are wired normally open to run, normally closed for limt, etc. These configurations are very noise prone electrically. You can do all sorts of things to reduce noise in the enviornment as the previous two posts list and should, but ultimately it still is a noisey enviornment. Normally closed switches during operations and open for an event is the way to go. Now that won't help your miss step count, so you need to clean up the noise source for those lines as well. Physically make sure the wiring from the computer to the step controller is not run with any motor or power lines. You don't show how you handle the 0V side of motor power and the stepper control's, which is important. What kind of motor drivers? A picture of the wiring might also help.
__________________ Phil, Still too many interests, too many projects, and not enough time!!!!!!!! Vist my websites - http://pminmo.com & http://millpcbs.com |
| Sponsored Links |
|
#6
| |||
| |||
| The first step should be to follow Al's suggestion for grounding and the E-Stop contactor. Step 2. After completing step one, test the unit for proper operation. If you still have the same problem then supply power to the VFD / Spindle from a portable generator, and test with this configuration. If you have EMI/RFI problems, then step 1 or 2 above should either cure the problem or reduce it significantly. Step 3. If step 2 solved the problem then install a hefty EMI/RFI filter from the electrical box to the power going to the VFD. This should isolate and reduce the noise, spikes, and radiated interference from getting back into the other circuits in your electrical box. EDIT: If step 2 did not solve the problem, look at the limit switches and go back and verify all grounding and wiring. . Last edited by CJL5585; 06-06-2007 at 12:57 PM. |
|
#8
| |||
| |||
| You said you tried shielded cable. I have known people to ground both ends and create a ground loop. I believe the correct method would be to ground the power source side of the shielded cable. I would say remove the limit switches for testing purposes. Whatever you do only perform a single change at a time, so you can get a handle on the problem vs solution. Worst case scenario: It is sometimes necessary to add a reactor to the motor circuit, you could discuss this option with an application engineer from the manufacturer of the VFD. They will generally need to know much information about the motor and wiring type/length. |
|
#9
| |||
| |||
| Thanks for the reply guys. Sorry I did not get back sooner. I was on vacation for the last week so. I see some great points and hopefuly they can help me fix this issue. Something I did the other day must of made my issue's worse because now I cant run a short program w/o getting limit switch errors ever 2 mins. So I just shut the limits off for now. I had to get some stuff cut out. do you guys think it would help if I used 1 service for the entire machine and spindle instread of 1 service for the spinde and 1 for the controller. |
|
#10
| ||||
| ||||
| Some systems suppliers tout the practice of isolating everything from each other (apart from case grounding etc), myself, I prefer to common all sections of the system, i.e. ALL power supplies and commons of all systems to be connected together, preferably at one point. This includes PC power supply, motor drive power and common of VFD control signal etc. When it is done in a controlled way, it has always worked for me. Al.
__________________ CNC, Mechatronics Integration and Machine Design. “Logic will get you from A to B. Imagination will take you everywhere.” Albert E. |
| Sponsored Links |
|
#11
| |||
| |||
| Smackre, It is common practice to use one service for the entire machine. In cases where the machine requires multiple voltages, I even go as far as using a transformer that is machine specific. By using one service, the machine can be locked out for service more effectively and limits line problems to one power system. It is probably okay to troubleshoot using multiple services, but the interference is likely coming from within your machine itself, not from outside. Isolate the problem, don’t just turn off functions such as software commands that look for inputs, actually remove them for purposes of troubleshooting. Do not run signal wires in proximity to power wires, use separate conduits and raceways. Keep even shielded cables several inches away from each other, avoid coiling cables up, especially together. You may find that using ferrite torrid’s on your control wires will help considerably, these are likely still available at Radio Shack. |
|
#12
| |||
| |||
| I switch the machine over to one service. I just kinda wanted it like that. And I made a common ground plate and added afew ground wires to diffrent parts of the machine. Still with no luck. I herd someone suggested added a filter to the line going to the VFD. Can you maybe give me a link to a filter like this. I am looking into getting ferrite torrid's. Radio shack seems to sell them. At which end of the contoller wires do you think It would be best to put them at? Thanks Again for all the help. I tring to get a picture of the control cabinet so you guys would have a better idea. But there is a object right behind the machine and it would be hard to get a pic. I would need to move the heavy object with the fork lift. I probly will do that. And find my digital camera. |
![]() |
| Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
| Thread Tools | Search this Thread |
| Display Modes | |
| |
Similar Threads | ||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| Trouble | extreme | DIY-CNC Router Table Machines | 36 | 07-11-2010 04:14 PM |
| The trouble with SE | split63 | SolidEdge | 5 | 10-23-2008 11:59 PM |
| 100 in/min X, Y, Z with IH CNC Kit. Any Trouble?? | haysys | Industrial Hobbies (Support forum) | 10 | 02-01-2007 08:24 PM |
| More PIC trouble | robe_uk | PicStep Controllers | 5 | 10-10-2006 03:13 PM |
| Arc trouble | Graeme | TurboCNC | 9 | 06-24-2005 03:31 AM |