Thread: BigG CNC

Originally Posted by rodm

Got BigG running tonight. Speed was tested at 9m/min but noticed a bit of whip in the X axis balllscrew so have detuned to 6m/min

OK Rod, I have hired a hit man. Better start looking over your shoulder from now on.

Bob

Don't worry Bob it is not carrying anywhere near the same load as your machine.

Well after a good day yesterday I had it all go wrong today. Main trouble was limit switches and false triggering. I raised the Debounce interval and it disappeared but then you have a delay on triggering the switches so it started bouncing off the ends of the machine rather than stopping on the limits. Finally fitted a cap on the signal terminal back to earth and it fixed it.

A break away from the machine while I went to the markets and I realised I had created a groung loop with the shielded cable. I took five volts around the limit switches and joined the last switch back to the BOB input. Both cables were shielded to ground and I joined the shield on all the switches. Sounds like a ground loop in my books.

Ok so do I snip the ground shield at one of the joins or do I leave it as it is because it is working?
Also if I am wrong somebody point me in the right direction.

Spindle isn't working so back to trouble shoot that.

As you can see it didn't go as expected today but you have those days.

Gotta love the ups and downs. I'd cut one end of the ground loop. All cables "should" only have one grounded drain in theory. Last thing you want is weeks down the track and having issues with signals somewhere and not remembering the loop. Just one opinion.

Daniel

Hi Daniel,
Thanks good advice. Always better to do it properly in the first place.

Well it got worse. I had run the spindle on its own and no problems. I had run the drivers on their own and no problem. Run them both at the same time and bedlam. Nema motors start taking off by themselves and making a racket and spindle sounds like it is running on two phases and not three. Running the spindle through the DC06 and the only other common connection is earth.
Disconnect the DC06 and spindle is fine but no change in Steppers.
Check continuity on all earths and they look good.

Sat there pondering and the green light came on. I am using a second hand (ebay) computer which is a Dell and they come in a plastic case. Ran an alligator clip from the metal frame of the machine to a metal part on the back of the computer case and 99% of the problem disappeared. A proper screwed earth from my star earthing to the computer case should fix it. It's been a long problem solving weekend so tomorrow night for that.

Daniel am I on the right track here or can you offer anytinhg else? Should I earth the Bob to my star earthing or is it sufficient for the 5vdc and ground through the switchmode do it. I have a separate 5v switchmode for signals.

Hey Rod,

Great to hear you got most of your problems solved. The damn earthing of signals can be a pain, there are so many different things to think about to get a good quality hookup. Look forward to seeing this one in action.

As for the 5v grounding. All logic signals should be seprate from the earth of the mains potential. This is so your logic doesn't get fried if the full 240vac is applied accidently through a fault to the ground somewhere. I would use the switchmode PS as the earth point for the logic supply.

Daniel

I am going to need some help to solve this one.
Drivers work fine without spindle on. Spindle works fine.
Turn on spindle and the drives move small amounts until it gets up to speed. once on full speed it is steady as a rock. Turn off spindle and when it winds down to lower speed from there to stop the drives start jittering again.
Have tried another computer - no change.
Have added a smoothstepper - no change.
I have a CNC4PC bob and gecko 251 drives. Tossed the smoothstepper out because it made no difference.
All signal wires, motor wires and spindle wire shielded and grounded one end except for the spindle which I have ground both ends but also tried with spindle end ungrounded.
If I disconnect the parallel cable to the Bob but keep the drivers on, the motors are steady as a rock when the spindle fires up.
Have tried another spindle and same result. Have another VFD I can try if needs be.
Need suggestions and where to start.

Originally Posted by rodm

I am going to need some help to solve this one.
Drivers work fine without spindle on. Spindle works fine.
Turn on spindle and the drives move small amounts until it gets up to speed. once on full speed it is steady as a rock. Turn off spindle and when it winds down to lower speed from there to stop the drives start jittering again.
Have tried another computer - no change.
Have added a smoothstepper - no change.
I have a CNC4PC bob and gecko 251 drives. Tossed the smoothstepper out because it made no difference.
All signal wires, motor wires and spindle wire shielded and grounded one end except for the spindle which I have ground both ends but also tried with spindle end ungrounded.
If I disconnect the parallel cable to the Bob but keep the drivers on, the motors are steady as a rock when the spindle fires up.
Have tried another spindle and same result. Have another VFD I can try if needs be.
Need suggestions and where to start.

Just a thought, are you getting too much line voltage drop (from shop plug), when running the whole machine. Is your wall plug properly grounded.

Good luck on your build.

Check the logic supply again during the running to see if its exactly 5v. or fluctuating. Also if you have a spare BOB like the mb-02 try that. I Had issues with one of the cnc4pc and when searching saw a few issues online from others too.

Hope you get to fix it soon,
Daniel

Thanks Check Twice
That was one of the earlier tests. I have also separated the power to different points with no change.

Thanks Daniel
Yeah you might be on to something there. I have a Homann Bob here and will try it. Was avoiding that one as it means a lot of disconnecting, etc.

Wohoooo problem solved.

It ended up being induced noise from the VFD. Lucky for me I have a mate that is an electronics guru who tracked it down. I would still be standing there scratching body parts if he had not come to the rescue.

Fix was to put a filter on the input to the VFD. Everything was put back to the original setup.

I will put a filter on all my VFD's from now on as I tried three different VFD's and got the same noise problem with each.

So I am back to finishing off the little bits and pieces on the machine and a couple or three nights should see this done.

Thanks to all for their suggestions as it could have been anything and just having somewhere to vent the frustrations is good.

It is great to have this fixed.

It always feels so nice when something that is giving you grief is solved, great news. You have an interesting build and would love to see a video of this puppy cutting.

Glad your buddy solved the issue.

John

Its a great time to crack a beer. Which type of filter did you put in? One of the panel mount IEC ones or did you use ferrite cores or clip ons etc. I should do this with mine just in case it causes grief in the future.

Great to hear your back on track.

Daniel

Thanks John and Daniel

Well my elation was short lived. Put the VFD back in the machine enclosure and it re-appeared but only slightly. The VFD case is plastic and must be letting out EMI through the case which is being picked up by the power in cord. This runs directly below the VFD. Being much closer to it than in our testing mode has resulted in a return of the stepper jitters but only slightly.

I had a play around this morning and by separating the VFD power from the machine power it goes away. It was all jury rigged to test this so while I am confident it has been identified and am reasonably sure it can be corrected it isn't over yet.

I use both the panel mount IEC and clip ons. The panel mount IEC's seem to do a good job. I am going to investigate putting two on the same cord. I have one within 50mm of the VFD which is what is recommended but it is inducing EMI on to the lead to the power. I will put a second filter on the case of the enclosure with the theory it will contain any EMI inside the metal enclosure. The VFD is inside an all metal box separate from the other machine electroincs. If this doesn't work I will just separate the power knowing this works as per my test this morning (fingers crossed).

At least my experience is out there now in case anybody else has a similar problem.

Rod, thats a kick in the pants. Cause you have found these issues in the power, EMI can be cause from the 3 phase output also. There was a mechmate post (Finished writing plugin for Mach3 to control the VFD via RS-485 - Adelaide, Australia - Page 3 - MechMate CNC Router Forum) He has filters on input and outputs and his emi engineer seemed to know alot and it was the huanyang VFD. Might be worth trying.

Daniel