500w chinese spindle interference

[GreenCameleon]

I'm in the process of building a DIY cnc machine. It already runs, but I noticed it has some interference problems (steppers act weird on occasions, the full graphic smart controller LCD has trouble displaying the contents), due to the noisy chinese spindle I'm using. To remedy this, I've read I should ground everything in a star-like topology.

The hardware consists of a 12V chinese PSU which connects to an Arduino Mega with Ramps 1.6 and 3 DRV8825 drivers. There is also a spindle controller, which generates the noise and doesn't have the ground symbol. I was told on another forum, that since it doesn't have an earth symbol, I should just tie the earth to one screw of the chassis. The PSU and the spindle controller are connected to the same extension socket.

I'm using shielded cable to connect to the stepper motors from Ramps, but the cable is also not grounded yet (just found out it has to be).

Can somebody describe how does a star-like topology look in my case? I've also read, I should ground the motor drivers, but I can't seem to find anything on the internet regarding this.

The attached images are outdated - I have since connected the earth on the psu and the spindle controller.

Thank you.

Similar Threads:

• Need Help!- Upgrading my 3040 with a 500W spindle (replacing the 300W one)
• Need Help!- 500 watt DC spindle creates interference on arduino
• Power 500W Spindle from 1/5kW VFD
• Spindle Limit Switch Interference
• LU 15W sub spindle/ Zb axis interference parameters

[Johnboy251]

I'm in the process of building a DIY cnc machine. It already runs, but I noticed it has some interference problems (steppers act weird on occasions, the full graphic smart controller LCD has trouble displaying the contents), due to the noisy chinese spindle I'm using. To remedy this, I've read I should ground everything in a star-like topology.

The hardware consists of a 12V chinese PSU which connects to an Arduino Mega with Ramps 1.6 and 3 DRV8825 drivers. There is also a spindle controller, which generates the noise and doesn't have the ground symbol. I was told on another forum, that since it doesn't have an earth symbol, I should just tie the earth to one screw of the chassis. The PSU and the spindle controller are connected to the same extension socket.

I'm using shielded cable to connect to the stepper motors from Ramps, but the cable is also not grounded yet (just found out it has to be).

Can somebody describe how does a star-like topology look in my case? I've also read, I should ground the motor drivers, but I can't seem to find anything on the internet regarding this.

The attached images are outdated - I have since connected the earth on the psu and the spindle controller.

Thank you.

“Can somebody describe how does a star-like topology look in my case? I've also read, I should ground the motor drivers, but I can't seem to find anything on the internet regarding this”.

Get a good size (CSA) piece of connecting wire from mains earth (or ground) to each PSU (the same size as the mains power feeding the PSU’s) connect one end of the screen cable to the steppers at the control card end to a common point - (i.e. psu case ground) leave the stepper motor end unconnected to the screen. Do the same to the spindle motor. That ‘should’ help! Hope that makes sense!
John


Sent from my iPad using Tapatalk

[GreenCameleon]

Get a good size (CSA) piece of connecting wire from mains earth (or ground) to each PSU (the same size as the mains power feeding the PSU’s)

Ok. I think I've already done that (like displayed in the attached pictures. I know the I should put suitable connectors on the wire and I will as soon as this issue is resolved ). The power cables, which I connected to the psu and the spindle controller have an earth wire, which is connected to ground earth via an EU power plug (https://www.google.com/search?q=eu+p...EAW4FwIhHLm3iM).

connect one end of the screen cable to the steppers at the control card end to a common point - (i.e. psu case ground) leave the stepper motor end unconnected to the screen. Do the same to the spindle motor. That ‘should’ help! Hope that makes sense!
John

You've already lost me here.

The LCD screen has two flat signal cables, which are not shielded and have a length of 30cm (this is the lcd and the cables: https://www.aliexpress.com/item/3D-P...7-4f11d0a33bd9).

Where do I interface the screen cable if the connectors, on both sides are, are connected to the control card (ramps board) and the lcd screen respectively? I just watched a video where someone wrapped alluminium foil around the flat cables inserted. Then he inserted a metal part, which bonded to a wire to serve as a drain wire. He then wrapped the cable again with another tape. Ok, I understand that. But where does the drain wire go since the cables are using DC voltage? Should I connect it also the the earth (ground) terminal as the PSU/chassis of the spindle controller earth?

Thank you!

[CNCMAN172]

Your issue is EMI (electromagnetic interference). The most common reason is improper grounding and the lack of shielded VFD cabling. The cable that runs from the VFD to the Spindle needs to be shielded. Look for a high quality VFD cable, they will often have a foil shield with a drain wire and a braided shield. The cable needs to have four conductors. Three for the motor phases and one for ground. The ground needs to be attached to the ground at the VFD ground terminal and Ground at the spindle end. You need to measure if your spindle actually has ground meaning the body of the spindled connected to one of the pins in the spindle connector. Many spindles which have 4/5 pins have one designated for ground but often do not attach the ground wire to that connector inside the spindle. Take a meter and verify one of the pins connects to the body of the spindle motor. If you have a three pin connector on the spindle you will need to use a clamp to connect the ground wire to the body of the spindle.

What is a star ground configuration. What that means is you have a single point in your control where all the ground connections are attached. The Ground from your supply in your business or home panel needs to also connect to this point. The idea is to avoid ground loops so making all the ground connections coming to one point helps ensure you avoid that issue. Further filtering can be added by using EMI filters one the power line coming into the VFD. Normally that will be a single phase 220VAC line and you will run that line through a double pole switch and then an EMI filter and then to the power inputs for the VFD.

Finally it is really important to keep all your AC lines in your control cabinet as far as possible away from all the low voltage lines. You can prove to yourself you have EMI issues, just turn off the spindle and run a simple gcode program and you will notice your machine runs around fine. If you are still have errors without your spindle running or powered, you need to check for AC lines near low voltage lines.

Hopefully this helps.

Russ

[GreenCameleon]

I'd like to preface this by sincerely thanking you, Russ, for taking your time and describing what is occurring and how to remedy the situation.

Your issue is EMI (electromagnetic interference). The most common reason is improper grounding and the lack of shielded VFD cabling. The cable that runs from the VFD to the Spindle needs to be shielded. Look for a high quality VFD cable, they will often have a foil shield with a drain wire and a braided shield. The cable needs to have four conductors. Three for the motor phases and one for ground. The ground needs to be attached to the ground at the VFD ground terminal and Ground at the spindle end. You need to measure if your spindle actually has ground meaning the body of the spindled connected to one of the pins in the spindle connector. Many spindles which have 4/5 pins have one designated for ground but often do not attach the ground wire to that connector inside the spindle. Take a meter and verify one of the pins connects to the body of the spindle motor. If you have a three pin connector on the spindle you will need to use a clamp to connect the ground wire to the body of the spindle.

Ok, I understand that. But I'm using a spindle, which runs on DC and only has two unshielded solid core wires (power and ground). Does this mean I also have to ground it? Would that mean I would have to create a drain wire, attached to the body of the spindle and run it to the earth terminal of the spindles controller, which is earth grounded via the chassis (since it doesn't have a dedicated earth ground terminal like the PSU)? This is the spindle, which I'm using: https://www.aliexpress.com/item/Daed...a-4badb867b33b

What is a star ground configuration. What that means is you have a single point in your control where all the ground connections are attached. The Ground from your supply in your business or home panel needs to also connect to this point. The idea is to avoid ground loops so making all the ground connections coming to one point helps ensure you avoid that issue.

Thanks. I think I understand this also now. What I don't yet know is, if I am allowed to connect the spindle body/signal cables of the lcd/stepper motors shielded cables (which use DC voltage) to the earth ground terminals of the psu/spindle controller (the PSU and spindle controller are powered by AC).

You can prove to yourself you have EMI issues, just turn off the spindle and run a simple gcode program and you will notice your machine runs around fine. If you are still have errors without your spindle running or powered, you need to check for AC lines near low voltage lines.

I am 100% sure, that the cause is EMI. I noticed, that the interference also affects the operation of the LCD. If I turn the spindle fully on, before booting up the controller, the displayed text is corrupted. If I first boot the board and then turn the spindle fully on, the lcd/controller works without any problems. I've made a video of this test: https://www.dropbox.com/s/cxi2bi1q3t...F3798.wmv?dl=0

[CNCMAN172]

Green Cameleon,

Ok, you have a DC spindle motor and controller. I could not tell from your video but it appears that small gray unit is your DSP cnc controller. Do you have some pictures of the entire setup as that might help me pin point your issue. DC spindles do not have as much EMI interference as three phase AC spindles but they still have it. There are different types of DC motors, induction motors have no brushes so they have less noise than a DC brushed motor. That controller needs to be well away from any AC lines. The DC spindle still has an AC line going to the spindle power supply. Make absolutely sure that AC line is not near any low voltage signals including the ribbon cable that goes to the LCD display. You also need to make sure there is a common between the low voltage that controls the cnc controller and the power used by the spindle drive. Post some pictures and I believe I will be able to help you determine how to resolve your problems.

Russ

[GreenCameleon]

Green Cameleon,

Ok, you have a DC spindle motor and controller. I could not tell from your video but it appears that small gray unit is your DSP cnc controller. Do you have some pictures of the entire setup as that might help me pin point your issue. DC spindles do not have as much EMI interference as three phase AC spindles but they still have it. There are different types of DC motors, induction motors have no brushes so they have less noise than a DC brushed motor. That controller needs to be well away from any AC lines. The DC spindle still has an AC line going to the spindle power supply. Make absolutely sure that AC line is not near any low voltage signals including the ribbon cable that goes to the LCD display. You also need to make sure there is a common between the low voltage that controls the cnc controller and the power used by the spindle drive. Post some pictures and I believe I will be able to help you determine how to resolve your problems.

Russ

I've made some more images. There are some others in the previous posts.

The spindle is connected directly to the spindle controller now, because it was previously routed through the cable chain, where the signal cable for the z axis was also ran. Then I read, you shouldn't mix power cables (AC) and low voltage single lines (DC), so I've removed it to see, if it would make a difference.

Yes, the gray box is the lcd and an sd card reader, which interface to the ramps board (the brain) via the two ribbon cables. Others have also suggested, that the length of these two cables could be the problem for the lcd display issues (but doesn't explain the stepper motor issues), but like it can be seen in the video, the problems only occur when the spindle is fully on.

What I find strange is, that I didn't experience these issues the first few days of running the machine. To me, it looked almost like that the spindle "deteriorated" in some way and started to produce more noise, if that makes sense, because I think the cables weren't moved around that much.

[CNCMAN172]

OK, this is a little hard without a drawing of how you have things connected, but let me describe what I think I see. You have a DC power supply with is outputting power on the Brown & Blue wire. That appears to be feeding the Spindle driver. The photo was not quite large enough to read everything on that driver but this is what I could make out. There are five screw terminals on the left with three wires, but above them is a legend that appears to read 220AC? Then two connections for Speed, two connections for ON/OFF, two connections for Motor Black/Brown, and two DC power leads, Blue/Brown. Is the DC power supply also powering the dsp cnc controller? You might want to us a separate power supply to power the controller. If you have a sample wiring diagram or a link to the manual that might be helpful. The ribbon cables are long and will pickup noise but the way you have it wired right now the spindle drive is right next to the CNC controller not a good idea. On your bench setup try moving the cnc controller and the display well to the left of the DC power supply making sure they are not near the AC line that is feeding that power supply. You have the makings of what appear to be come good components for your machine. Not sure how dedicated you are to the Ardrino controller solution, but you can certainly upgrade after you get this working. The spindle controller might send out pulsating DC to control the spindle impossible to tell with out info on the controller and spindle. I would still use a shielded cable, but in your case you will only need three conductors. Spindle Common, Spindle Power, Ground. That common will not be ground with the rest of the grounds would be my guess.

Russ

[GreenCameleon]

I'll try to make a simple wiring diagram. Give me 15 minutes.

[Al_The_Man]

The first thing I noticed is that your AC supply to the DC PS has been marreted off, (insulated).
Your PS case needs to be earth grounded, the star ground is just that, a star point where all grounds that require to be earthed are connected to, a set of terminals or a plate with holes tapped to take ring terminals etc.
All motor frames should be grounded as well as PS negative terminals if possible in order to avoid random triggering.
Do a search here for my posts on equi-potential bonding. Many over the years.
Al.

[GreenCameleon]

OK, this is a little hard without a drawing of how you have things connected, but let me describe what I think I see. You have a DC power supply with is outputting power on the Brown & Blue wire. That appears to be feeding the Spindle driver. The photo was not quite large enough to read everything on that driver but this is what I could make out. There are five screw terminals on the left with three wires, but above them is a legend that appears to read 220AC? Then two connections for Speed, two connections for ON/OFF, two connections for Motor Black/Brown, and two DC power leads, Blue/Brown. Is the DC power supply also powering the dsp cnc controller? You might want to us a separate power supply to power the controller. If you have a sample wiring diagram or a link to the manual that might be helpful. The ribbon cables are long and will pickup noise but the way you have it wired right now the spindle drive is right next to the CNC controller not a good idea. On your bench setup try moving the cnc controller and the display well to the left of the DC power supply making sure they are not near the AC line that is feeding that power supply. You have the makings of what appear to be come good components for your machine. Not sure how dedicated you are to the Ardrino controller solution, but you can certainly upgrade after you get this working. The spindle controller might send out pulsating DC to control the spindle impossible to tell with out info on the controller and spindle. I would still use a shielded cable, but in your case you will only need three conductors. Spindle Common, Spindle Power, Ground. That common will not be ground with the rest of the grounds would be my guess.

Russ

I've made a simple diagram of the connections. The 12V PSU only powers the Arduino/Mega/Drivers board. The spindle is powered via the controller (the other box, which looks like a dedicated PSU).

I've also found an image of the terminals on the spindle controller. The MACH3 terminal is an analog input, used controlling the spindle automatically, which I don't use in this case, since I control it via the potentiometer.

[GreenCameleon]

The first thing I noticed is that your AC supply to the DC PS has been marreted off, (insulated).
Your PS case needs to be earth grounded, the star ground is just that, a star point where all grounds that require to be earthed are connected to, a set of terminals or a plate with holes tapped to take ring terminals etc.
Al.

The earth terminal, on the 12V PSU is connected. I've also connected grounded the spindle controller via its chassis, because it doesn't have a dedicated ground earth terminal like the PSU.

All motor frames should be grounded as well as PS negative terminals if possible in order to avoid random triggering.
Al.

Does grounding PS negative terminals mean, I just have to bridge the ground and the earth ground terminal connections on the PSU and spindle controller?

Don't know what you mean with grounding motor frames? Should I attach a drain wire to them and wire them to the earth ground?

[CNCMAN172]

Check with an ohm meter if the common terminal for spindle motor output is connected the the spindle controller case. That case might not actually be a ground connection. Do you have link to spindle drive manual.

Russ

[Eldon_Joh]

is this a brushed dc motor with pwm controller?

[GreenCameleon]

Check with an ohm meter if the common terminal for spindle motor output is connected the the spindle controller case. That case might not actually be a ground connection. Do you have link to spindle drive manual.

Russ

What exactly should I check? The voltage between the negative terminal of the "motor" output, on the controller, and the spindle controller case? Or just the continuity?

is this a brushed dc motor with pwm controller?

Yes, it is. To quote one user from another thread, on this forum, this is a "a brushed DC motor with a crappy controller" (https://www.cnczone.com/forums/diy-c...64070-cnc.html)

[CNCMAN172]

No, leave the power off and get an ohm meter. You want to measure between the motor output common terminal the negative side to the spindle controller and see if you get zero ohms. Most likely this is a floating common and you might need to connect that lead to the common ground terminal. Let me review the manual and see if there are any clues. Check the measurement with the power off. In fact disconnect the motor from the controller for this test as well.

Russ

[CNCMAN172]

OK, just looked at the ebay ad. I also want you to measure with an ohm meter between the spindle driver case and the Negative side of the DC power coming into the spindle controller. You might have to ground the negative power input lead. There was no real manual so we have no schematic of the driver. A brushed DC motor will generate a good deal of noise. Why? Think of a time when you might have helped someone jump a car battery with jumper cables. When you connect the leads often you will see a few sparks. Brushes inside of a motor have power traveling through those brushes so they often let off little sparks which are generating lots of noise. Think of a battery operated drill an older one that has brushes, look at the opening in the case as you pull the trigger you will see small sparks. Does this mean you can't use that spindle. No you just need to minimize the noise emitted.

Russ

[Al_The_Man]

Surely you are not going to leave all the controls etc out in the open, it should be all contained in a suitable enclosure, This contributes to any spurious faults you may be getting.
All motor leads should be twisted conductors, this goes most of the way of preventing or containing EMI. Especially if the one you have is PWM, If Triac Bridge control, then it should have a series choke.
DC motors have been used on CNC machines for many decades, and were the only types initially, in many cases, many HP.
Al.
.

[CNCMAN172]

Al,
He is not leaving it in the open he is just doing bench testing. As soon as he turns on the spindle the cnc control is instantly impacted. He has excessive EMI, as you can see the spindle cable is not shielded, even though it is a DC spindle. I also question the spindle control they had no ground terminal on that controller, he used the case but I question if that is grounded either. The first step would be move the spindle controller as far from the cnc controller as possible and see if the EMI is significantly reduced.

Russ

[Al_The_Man]

Al,
He is not leaving it in the open he is just doing bench testing. As soon as he turns on the spindle the cnc control is instantly impacted.
Russ

OK but if doing 'Bench' testing, which when in and of itself introduces problems, it is not really a conclusive way of testing.
Al.