Hello Guest it is February 25, 2020, 01:30:19 AM

Author Topic: Noise Problem on a large machine  (Read 3525 times)

0 Members and 1 Guest are viewing this topic.

Offline Dan13

*
  •  1,208 1,208
    • View Profile
    • DY Engineering
Noise Problem on a large machine
« on: November 07, 2014, 09:10:56 AM »
Built a 4.5m cylindrical grinding machine. The control console is a attached to the carriage and travels with it along all the machine length. The machine uses steppers and has 4 axes. Cables running to the carriage (in a single chain) are 10m long - 2 cables to the steppers (shielded and grounded at the electrical cabinet), 2 cables to two 3 phase inductance motors (shielded and grounded a the electrical cabinet), a VGA cable to the monitor on the console, a THICK ground wire to the console.

The electrical cabinet has 4 Gecko stepper drives, a mini PC, USB SmoothStepper, VFD and a few relays and contactors. SmoothStepper has a separate 5V power supply, but also tried to power it from the USB. The USB cable shield is grounded only at the PC. Everything is grounded in the cabinet to a single star point. and EMI filter is used between mains and monitor, PC, 48V PSU and 5V PSU. The contactors have capacitors over the contacts.

PROBLEM 1: An E-stop button on the console cutting 230VAC to a three-phase contactor in the electrical cabinet. The contactor cuts off power to stepper PSU and VFD. Hitting the E-stop causes the monitor to turn off for a couple of seconds and then it returns back. Disconnecting the stepper PSU from the relay and leaving on ly the VFD still causes the same.

PROBLEM 2: Turning ON either of the two 3-phase motor - VFD or direct, causes the monitor to turn off for a couple of seconds. Turning off the motors causes the same.

These problems, I think, are due to to bad VGA cable (although shielded).

PROBLEM 3:  At times the steppers (particularly one of them) start moving on their own. The 3 phase motors are not running at this time. Disconnecting the USB cable from the SmoothStepper stops it. I am not sure, but I thing that disconnecting the VGA cable from the PC also stops it. This is the most significant problem.

Any thoughts?

Dan
Re: Noise Problem on a large machine
« Reply #1 on: November 07, 2014, 01:10:31 PM »
1) Remove the capacitors from the 3ph contactor contacts. I assume they are in parallel with the contacts. This is dangerous in the event a shorted capacitor can leave one phase energized. It's also not necessary from an electrical standpoint.   
2) Check the 230V, 3 ph line side with a voltmeter to make sure it's actually 230v between phases with the contactor open and closed.
3) Check the load side voltage of the contactor contacts to make sure you don't have a bad contact block. (See #2 above)
4) Make sure the ground connection from the star in the cabinet is going to a solid earth grounding point.

Sounds like either a voltage issue, loose connection someplace or ground looping.

Offline Dan13

*
  •  1,208 1,208
    • View Profile
    • DY Engineering
Re: Noise Problem on a large machine
« Reply #2 on: November 07, 2014, 03:30:47 PM »
1. To be more accurate, it's not bare capacitors, but rather a self contained dedicated filter with caps and resistors in series. These are usually used across contacts. Anyway, using it or omitting makes no difference.
2. Checked and, no fault there.
3. Also checked to be OK
4. The star point doesn't go to solid earth ground. It goes to the electric outlet in the same cable with the 3 phases, and the ground wire is only 2.5mm in diameter. I will try connecting the star point metal construction of the building.

Dan

Offline dude1

*
  •  1,253 1,253
    • View Profile
Re: Noise Problem on a large machine
« Reply #3 on: November 07, 2014, 04:55:17 PM »
make sure that the earth/common is sunk to ground from the machine to where the power comes into your shed if not whack a earth stack in (very important)

it sounds like a big power draw have computer and screen on its own power supply

there is a white paper written on common/shield connections where you have low and high on there own star 10mm apart it makes a differences tried it my self

also check with the manufacturer of your VFD some are meant to have shield connected at each end

above are things I have done and they worked
Re: Noise Problem on a large machine
« Reply #4 on: November 07, 2014, 06:42:43 PM »
One thing for sure, if an earth ground appears to fix your problem you haven't found the problem! I always consider this solution to be hocus pocus and possibly hiding a problem that could get you killed.  A large part of the problem is that electronic engineers don't seem to understand that electronic signal grounding and power grounding are completely different.  So they go looking for the mythical 'clean ground', which is what that earth ground is supposed to supply.  Nothing could be further from the truth.

In power grounding the purpose is to ensure that no two adjacent grounded points ever have a voltage difference that can kill.  Everything electrical leaks power to ground, all the time.  When things go bad, like a motor winding going to ground that 'leakage' may be thousands of amperes!  So by connecting everything together creating as many 'ground loops' as you possibly can you ensure that a very small voltage difference is all you see from that railing you have in one hand the handle you have in the other.

In electronic signal grounding you simply want no current flowing on the shielding because when done properly one one end is ever connected to ground.  Hence only the very small currents from signal leakage and induced currents through the air will ever flow through that shield. In order not to have ground loops you MUST check whether devices have ground connections to their cases. The single grounding point works only if no other device ties to ground through the frame of the machine.  If the ground point of the device is already grounded to the frame, insulate the shield from ground at that point, and make sure it is connected to the common ground point only back in the panel.

One reason your problem may happen when you turn the power off not on is because that is when the magnetic field in the contactor coil collapses and created a large back EMF.  You should have a suppressor on the coil in any event.  In the early days of digital meters I fried a few on contactor coils when I hit the stop button!

Opening the contactor to the VFD has a similar effect, an arc is drawn for an instant, and there is a large surge. I suspect the monitor is not going off on low voltage because it happens when you turn other devices off.  So that suggests it is going off due to seeing a high voltage when arcing or a load dump occurs.  I'd check your VGA cable to if it connects the shield to ground at both ends, because that is likely to be the case.

Hope you get your problem sorted quickly.

Offline dude1

*
  •  1,253 1,253
    • View Profile
Re: Noise Problem on a large machine
« Reply #5 on: November 07, 2014, 08:43:29 PM »
some vfd or vsd need a ground loop the one on my machine does as that is what the manufacturer demands to be done there is some very technical reason for this the traveller tried to ex-plane why buts it way over my head. its a very expensive and technical vsd it has 10000 different settings.

that's why I said to check with the manufacturer of the vfd

all my electrical work is done by a electrician who knows what he is doing
« Last Edit: November 07, 2014, 08:47:47 PM by daniellyall »

Offline Dan13

*
  •  1,208 1,208
    • View Profile
    • DY Engineering
Re: Noise Problem on a large machine
« Reply #6 on: November 08, 2014, 01:11:18 AM »
Thanks guys. Will check the points you're suggesting.

Just to make it clear - I forgot to mention it - the problem with the monitor going off happens both when the Estop is depressed and upon its release. Anyway, the problem with the monitor bothers me least at this time. What really bothers me is the steppers moving on their own while not even an inductance motor running, starting or stopping. When this happens, disconnecting the USB cable from the Smoothstepper stop it.

Dan

Offline dude1

*
  •  1,253 1,253
    • View Profile
Re: Noise Problem on a large machine
« Reply #7 on: November 08, 2014, 02:14:55 AM »
that sounds like a very bad problem go through and check all electric connections before you do anything else.

hold a meter on the machine neg to neg on a power supply not the machine, positive on the machine hit estop see if there is power on the machine frame if there is check all electric connections with the machine disconnected from its power supply.

for the steppers to move by there self there is something very wrong
Re: Noise Problem on a large machine
« Reply #8 on: November 08, 2014, 09:58:24 AM »
What is interesting is the steppers move on their own which is usually caused by electrical noise. What type of VFD's are you using? If is one of those cheap low quality Chinese drives they don't meet UL, ANSI or CE standards for RF noise emissions from the inverted section. Essentially they are great radio transmitters. Even when the drive is energized but not outputting voltage to the motors the drive still is emitting RF energy at a somewhat lower level. Try to route all low voltage cables away from the drives and motor T-leads. In industrial machines have PLC's and controls are either in a separate cabinet or have a metal divider to minimize interference from drives.    

Offline Dan13

*
  •  1,208 1,208
    • View Profile
    • DY Engineering
Re: Noise Problem on a large machine
« Reply #9 on: November 08, 2014, 12:53:14 PM »
It is a Delta VFD. High voltage cable are routed separately from the low voltage wires in the cabinet.

Dan