Machsupport Forum

Mach Discussion => General Mach Discussion => Topic started by: JonGard on July 27, 2012, 06:13:22 PM

Title: Spindle Motor is the cause of all my problems
Post by: JonGard on July 27, 2012, 06:13:22 PM
After hours of sitting and watching my cnc mill go back and fourth and up and down twenty times, measuring at three different points I have found that with the spindle motor turned off the machine is accurate and precise. With the spindle motor on the x,y and z axis will move a lot, the longer it runs the more it moves. So now that I know what the problem is I do not know how to fix it. The spindle motor is a 24V brushless motor. I have removed the speed controller and placed it out side the control box and it did nothing other than waste my time. I’m thinking of putting the machine into the truck and driving to the middle of a filed somewhere, has anyone seen “Office Space”? then you know what’s next.  >:(
Any suggestions?
Title: Re: Spindle Motor is the cause of all my problems
Post by: Hood on July 27, 2012, 06:19:52 PM
Shielding and properly (one end and all shields tied to a single terminal) grounding  the wiring, especially low voltage signal wires such as Step/Dir will help greatly.

This is the reason I try to make all signal wiring on my machines 24v and signals such as encoders or step/dir that must be 5v I make sure they are differential signals. Both of these methods make it very rare for noise to be a problem. Obviously going that route for you would be a major task in rewiring and interfacing so go with the shielding/grounding and I am sure you will get it sorted.
Hood
Title: Re: Spindle Motor is the cause of all my problems
Post by: Vogavt on August 03, 2012, 08:01:03 PM
Just for a little humor here I can't tell you the number of times my workshop's electric garage door opener get's triggered when bringing Mach3 online! Seriously!

I know there must be a stray signal coming somewhere from my control box that happens right when Mach3 gets through loading. Happened some many times I've come to ignore it and just walk over and hit the button to close the door.

Just remember... it can always be worse!  ;D
Title: Re: Spindle Motor is the cause of all my problems
Post by: budman68 on August 04, 2012, 08:43:06 AM
This is why it scares the heck out of me when folks use wireless devices like pendants, keyboards, mice etc.

I know they obviously work for some, but I don't mind the cable enough to play research and development  ;)

Dave
Title: Re: Spindle Motor is the cause of all my problems
Post by: Hood on August 04, 2012, 09:24:02 AM
Must admit I think the same, maybe its because I have bigger machines but even a wee one can do serious damage to flesh.
Hood
Title: Re: Spindle Motor is the cause of all my problems
Post by: Vogavt on August 04, 2012, 09:37:22 AM
Nothing's wireless on mine other than my internet connection and I only use it (have it plugged in to the usb) when I'm not milling.
Title: Re: Spindle Motor is the cause of all my problems
Post by: angel tech on August 04, 2012, 10:17:14 AM
i've seen a wireless keyboard and mouse cause chaos to mach. Not worth the trouble they cause, and you have fun when the battery runs out.
Title: Re: Spindle Motor is the cause of all my problems
Post by: RICH on August 04, 2012, 10:38:45 AM
I agree with Dave and Hood. I don't use any wireless device ...other than an XBOX 360 when using CopyCat ( which is not very often).
A few friends use wireless and seem to have no problems. Only use over time can tell but if there are problems from the start than would seem
like one is asking for trouble.
RICH
Title: Re: Spindle Motor is the cause of all my problems
Post by: budman68 on August 04, 2012, 04:08:26 PM
Must admit I think the same, maybe its because I have bigger machines but even a wee one can do serious damage to flesh.
Hood

Indeed, but when a machine starts eating itself, I'm sure that's no fun to watch either, lol!  ;D

Dave
Title: Re: Spindle Motor is the cause of all my problems
Post by: JonGard on August 06, 2012, 02:08:39 PM
Thanks Hood!
I have rewired the entire machine. E-stop, home sensors, spindle motor and steppers all have shielded wire. I have a control box that houses everything, power, speed control and stepper control. The stepper motors have just wires coming out with no shielding. My question now is do I ground both sides of the wire and shield the steppers or ground just the control box side? Or am I in for more R&D?

Thanks Vogavt, that’s a great story.
Title: Re: Spindle Motor is the cause of all my problems
Post by: Hood on August 06, 2012, 02:11:11 PM
Current thinking on shielding is to bring all shields at the control end to a single point  and ground there.
Hood
Title: Re: Spindle Motor is the cause of all my problems
Post by: Hal on August 08, 2012, 08:57:10 PM
Ground only one side of the shield. I use the control box side. I would enclose the power supply and speed control inside the main box if you have trouble.

No wireless for me. Garage door may open if one of the jets fly to low. The battery in my wireless keyboard and mouse loves to die when I need to finish.

Hal
Title: Re: Spindle Motor is the cause of all my problems
Post by: Bob La Londe on August 11, 2012, 02:04:31 PM
Single end grounding of shields is the standard for most (all?) low voltage signals wires these days.  I have seen all kinds of weird things caused by ground loop when somebody mistakenly gets a second ground point.  Audible noise on audio cables.  Rolling bar distortion on video cables, etc etc ... 

Trust me.  Ground shields on only one end at a common central point. 

Title: Re: Spindle Motor is the cause of all my problems
Post by: Hood on August 11, 2012, 04:13:59 PM
Not always true that grounding at one end is the way to go. For most things Mach related it is but I can assure you that only grounding the shield at one end of the encoder cable on the motors that utilise sick stegmann  sine/cosine smart encoders will not work, they need the shield grounded at both ends.
Hood
Title: Re: Spindle Motor is the cause of all my problems
Post by: Bob La Londe on August 11, 2012, 04:52:28 PM
Not always true that grounding at one end is the way to go. For most things Mach related it is but I can assure you that only grounding the shield at one end of the encoder cable on the motors that utilise sick stegmann  sine/cosine smart encoders will not work, they need the shield grounded at both ends.
Hood

Then they are using the shield as part of the signal circuit or using an electrical charachteristic of the shield for some purpose other than just shielding if it actually must be "grounded" at both ends.  Alternatively they may require it be "connected" at both ends, but it might still only actually be grounded at one end.  Not being familiar with that particular piece of equipment of course you can argue the actual application more effectively than I could.  

An example I am more familiar with:
I know in fire alarm signal circuits often the shield must be connected through from device to device and often connected to the device (depending on the application), but its still only connected to ground at the control panel.  Accidental mechanical grounds are avoided by make the body of the device out of plastic.  

P.S. The main reason for the use of plastic is price.  Avoidance of accidental mechanical grounds is incidental. 

Title: Re: Spindle Motor is the cause of all my problems
Post by: Hood on August 11, 2012, 05:36:09 PM

Then they are using the shield as part of the signal circuit or using an electrical charachteristic of the shield for some purpose other than just shielding if it actually must be "grounded" at both ends.  Alternatively they may require it be "connected" at both ends, but it might still only actually be grounded at one end.  Not being familiar with that particular piece of equipment of course you can argue the actual application more effectively than I could. 

An example I am more familiar with:
I know in fire alarm signal circuits often the shield must be connected through from device to device and often connected to the device (depending on the application), but its still only connected to ground at the control panel.  Accidental mechanical grounds are avoided by make the body of the device out of plastic. 

P.S. The main reason for the use of plastic is price.  Avoidance of accidental mechanical grounds is incidental. 



Dont think they are, its simply a shield as far as I can see, below is a pic of the signal and power wires and shield is additional to them. The motor is grounded via  the Earth wire, the drive is also grounded so both ends are, In fact Allen Bradley recommend the shield be connected at both ends even on standard encoders. On the cables I have made up myself I have on some and not on others and they seem fine either way but the smart encoders need to be, if you attempt to run the motor with only one end connected it faults almost right away. The first cable I made up for a smart encoder only had the sheld grounded at the control end and I couldnt even get it running, in fact it faulted out as soon as I enabled the drives. I ended up swapping out the encoder for a standard one and it was only later that I discovered, by chance, that the shield needs connected at both ends.

Hood
Title: Re: Spindle Motor is the cause of all my problems
Post by: Bob La Londe on August 11, 2012, 05:41:29 PM
Perhaps there is some internal shielding that needs to be connected.  At the encoder end something would have to be connected to the body of the machine for it to be "grounded." I guess the easy way to tell is disconnected the controller end and put a meter on it between the shield and ground.  

I am not saying its impossible that they are creating a ground loop, just that its more likely they are connecting the shield to the device.

We may be quibbling over the difference between connected and grounded, but there is a difference.  On a device if there is a place to connect the shield or ground wire that does not mean its grounded at that point.  
Title: Re: Spindle Motor is the cause of all my problems
Post by: Hood on August 11, 2012, 05:55:07 PM
The shield connects to either the metal body of the encoders cover or  bolted the motor body itself. The motor power cable  has the three phases and an Gnd wire the Gnd wire connects to the motor body and at the drive end it is connected  to the Earth connection on the drive.

It was something I found unusual after all the talk of only grounding the shield at one end but as it turned out it was the exception to the standard practice. I do remember reading a bit about low voltage high frequency signals need the shield grounded at both ends but it was a long time ago and I dont recall any details.

As has been said though it is normal just to connect the one end and this for some reason seems to be the exception.
Hood
Title: Re: Spindle Motor is the cause of all my problems
Post by: rrc1962 on August 11, 2012, 06:43:16 PM
Shielding on both ends works fine as long as EVERYTHING is grounded properly.  There is no room for error.  I know that the newer versions of the ProMotion, Burny and Hypertherm controls ground on both ends.  There are closed loop AC servo systems with high voltage signals.  Grounding on one end DOES seem to work best on Mach systems though.
Title: Re: Spindle Motor is the cause of all my problems
Post by: RICH on August 11, 2012, 07:27:17 PM
Shielding is a rather diffcult to put in simple terms but I will try and confuse .....
 
The shield is nothing more than an barrier to keep something out, something in, or take something away.
The barrier can be solid, like a metal wall, or it can be open, like a fence and still act as barrier. Thus electrically
it can be 100% or less. It can be open, and be effective, but the actual opening size is frequency related.
The best shield is 100% as it totally shields what is inside from what is outside and vise versa. Additionally it can can conduct.
So what happens to the shield at the ends is important.

If one end is connected to ground and the other not connected, then it acts as shield along it's path but additionally it conducts
and provides a path for signals to the ground. It won't  transmit all signals since depending on it's length is acts as a short to
transmission.

Shields are normally taken to a single point and that point goes to ground. Ground is not ground. Ground is just a reference.
In the case of shielding, ground is taken to a single / common point and usually that single point continues to a specific point not
associated with other points of an electrical device. Simple said, the shielding is run to a point which provides almost no resistance to
the signal and separate from any other grounds.  If grounds are connected then signals can can travel along to what that other ground
are connected to.

BTW, you twist wire pairs as the signals running on the outside of the conductors have difficulty mixing or traveling since an inductance is
seen by the signal.  

Now it gets more interesting as one must consider how signals are received, mixed, transmitted, etc, etc.
A conductor can receive a signal via conductance or inductance. Any conductor which cuts an electrical field will acquire a  signal.
The shield will keep that signal out from combining with or riding on the inside wires.

I have used signal rather loosely. The unwanted signal is usually some kind of field  created by say a motor, vfd, etc. and that field is
made up of numerous signals of varying frequencies. Noise as it is called, because we can't understand the meaning as compared
to speech or represent it as a simple wave.

How intense a field is rather difficult to envision as you can't see it,but, for description purposes  just envision the field as a balloon
filed with signals and the source of the signals is at the center of the balloon. The noise weakens in intensity as the distance increases.

So use the above to understand how shielding is used when running wires to a motor or whatever.

FWIW,
RICH