[Barefootboy4]

I'm at the end of my skill set. I built my own DIY CNC 5 years ago and it's been great. Lately I am getting false limit switch shutdowns. If I run the program with the router OFF there is no issue. With the router ON it get false limit trips. This tells me I have a NOISE issue. I am running Mach 3 and the debounce is set at 3000 which s HIGH. I have replaced all six limits and all wires are connected correctly. The wire is shielded and a separate drain wire connected in a standard star connection to GROUND. I did a Ohm test on the limit circuit and it is 1 ohm.

I'm at a loss as what kind of test I can do to figure out how to solve this issue.

Any ideas out there?????????
Steve in Maine

[Al_The_Man]

Have you ensured the Motor frame etc is grounded, if it is a double insulated 2 wire router make sure the gantry etc is grounded.
If still a problem, you could take a ground wire from the 25p common to the star point, this is grounded internally in the PC, but it pays to ensure it is grounded to the star point also.
http://www.cnczone.com/forums/phase-...se-issues.html
Al.

[Barefootboy4]

Thanks,
I will try this and get back to you.

[DonKes]

Put some capacitors across teh switches.

Check other posts for the size of capacitors.I don't remember the size I used.

Don

[Kenny Duval]

I feel your pain. I just got done tracing down a noise problem for the last 3 hours. Turned out to be high frequency noise on the probe tool wire. I was getting false triggers during probing routines but never during the measuring of tools on the tool plate or during actual cutting. Just holding the probe wire in my hand while it was probing was enough to fix the problem. I cut a ferrite bead out of an old VGA cable and looped the probe wire through it. Problem solved.

[wizard]

If this showed up after 5 years of running fine you might have wiring breaking down. This happens often in drag chains and can be problematic to debug. Another possibility is the router itself has become noisy and is overwhelming what ever noise reduction you originally had in place. You might try to hang another motor off the wiring feeding the router to see if an alternative motor causes the same trips. With 5 years of use you could have worn out brushes or a commutator that is worn out creating objectionable noise.

I'm at the end of my skill set. I built my own DIY CNC 5 years ago and it's been great. Lately I am getting false limit switch shutdowns. If I run the program with the router OFF there is no issue. With the router ON it get false limit trips. This tells me I have a NOISE issue. I am running Mach 3 and the debounce is set at 3000 which s HIGH. I have replaced all six limits and all wires are connected correctly. The wire is shielded and a separate drain wire connected in a standard star connection to GROUND. I did a Ohm test on the limit circuit and it is 1 ohm.

I'm at a loss as what kind of test I can do to figure out how to solve this issue.

Any ideas out there?????????
Steve in Maine

Does the router trip the limit switch circuit upon power on, while idling or during actual cutting?

[mactec54]

This tells me I have a NOISE issue. I am running Mach 3 and the debounce is set at 3000 which s HIGH. I have replaced all six limits and all wires are connected correctly. The wire is shielded and a separate drain wire connected in a standard star connection to GROUND. I did a Ohm test on the limit circuit and it is 1 ohm.

You can not connect a separate drain wire to the shield, your shielded cable for your limit switches, will have no noise protection

Ground the shield where your cable ends, the extended wire is unshielded & will carry any noise there is in your system, shields don't have to go to a Star Ground Point

You must keep their termination as short as possible

Like this photo or similar

[Al_The_Man]

When grounding a shield other than a star point, it is imperative that equi-potential bonding is done, I see Siemens have changed their site so the PDF link does not work anymore unfortunately?
This is a little from Belden, one of the major cable manuf.

"The real solution to this is to design your facility so that all grounds everywhere are the same potential. This is sometimes called a "star ground"."
Ground Loops
Al.

[wizard]

When grounding a shield other than a star point, it is imperative that equi-potential bonding is done, I see Siemens have changed their site so the PDF link does not work anymore unfortunately?
This is a little from Belden, one of the major cable manuf.

"The real solution to this is to design your facility so that all grounds everywhere are the same potential. This is sometimes called a "star ground"."
Ground Loops
Al.

That is a great link! Belden has another one here: Telescopic Shields PART 1 ? Looking at Different Shield Options discussing shielded cables. GE gets into it see: http://store.gedigitalenergy.com/FAQ...l/GER-3205.pdf. An interesting thread: Grounding screened control cables - PLCS.net - Interactive Q & A.

In a nut shell there is no perfect answer, the preference is to ground at both ends these days. In some cases the cable may require grounding at several places along its run.

In the case of this question I kinda doubt that we are dealing with a grounding issue/ I'm wishy washy were because these sorts of faults are a bear to resolve at times.

One thing I've made a habit of doing is to go through the machines electrical panel and tighten all connections when dealing with intermittent problems. Making sure all the wires are actually in the terminal blocks is another thing, I've seen machines sit in production of a decade or more before a wire that never made it into the terminal starts to cause problems. I feel for the owner of this machine because it is hard to say what is happening here.

[mactec54]

That is a great link! .

Yes the Information is great, if you understand, how it needs to be implemented

In the case of this question I kinda doubt that we are dealing with a grounding issue/ I'm wishy washy were because these sorts of faults are a bear to resolve at times..

Not at all, most Hobby machine wiring, is simply bad, because they do what they need, to get there machine running

Barefootboy4 has used shielded cable for his limit switches, but has not Grounded the Shield correctly, he could of used regular twisted pair wire & may have had a better result, you can not add a wire to a shield as he has done, this becomes an antenna & picks up any noise, that may be coming from the spindle

[wizard]

Yes the Information is great, if you understand, how it needs to be implemented



Not at all, most Hobby machine wiring, is simply bad, because they do what they need, to get there machine running

This is certainly so and does make finding faults a bit of a chore.

Barefootboy4 has used shielded cable for his limit switches, but has not Grounded the Shield correctly, he could of used regular twisted pair wire & may have had a better result, you can not add a wire to a shield as he has done, this becomes an antenna & picks up any noise, that may be coming from the spindle

Back in the day that was fairly common practice, you would wire an extension on the shield and ground it to a common spot. I'm not saying this is a good way to do it but I've seen it even on professionally built machines. Old practices are sometimes hard to break but some of those references linked to make it clear why you might want to change practice.

The problem is I'm not sure he has a grounding problem. Sounds more like a cable breaking down. We run into this all the time at work with machines that have been in operation for awhile. Even high quality cables eventually fail.

[Barefootboy4]

After reviewing this picture I realize my motor is not grounded proper. I will address this and get back to you. As far as my limit switch wiring is concerned All wires are shield and there is a seperate drain wire encased which grounds at every switch. This was done years ago from forum suggestions. I feel it's good. This issue has been a problem off and on for years.

Thanks for the ideas
Steve

[mactec54]

After reviewing this picture I realize my motor is not grounded proper. I will address this and get back to you. As far as my limit switch wiring is concerned All wires are shield and there is a seperate drain wire encased which grounds at every switch. This was done years ago from forum suggestions. I feel it's good. This issue has been a problem off and on for years.

Thanks for the ideas
Steve

If you have added any drain wire to any of the Shields, then that piece of wire will be an antenna to collect any noise that is present, the fact that you have a debounce of 3000, tells you there is a problem, a zero setting for your debounce is when you have your system wired correct

For general shield termination it needs to be at the source, meaning at the other end of where you have your limit switch shield Grounded, and done the same way as the photo

Because you are using a router, you have to do the best you can with the rest of your machine, like rewiring your router,is possible, with a shielded cable, and adding a Ground wire to the metal body of the router frame

Just remember if your shields are not Grounded/Bonded, to Earth Ground, they will not be working as they should, this will cause your system to have unexplained switch trips

Are you using a USB Breakout Board ???

[Barefootboy4]

After reviewing all suggestions I realized my Router was NOT GROUNDED at all. Picked up a proper 12 gauge 3 conductor and grounded the motor frame to ground plus a bonding strap as detailed in the picture. Checked all connections and ran 30 minute cut file today. WORKED PERFECT !!!!!!!!!!!!!!!

Thanks for the help. When I took the router plug end apart found the 110VAC connection completely loose which was most likely the root cause of the false trip issue. I love this forum and back in business making money.

Thanks again for all the great advice.
Steve in Maine

[UA_Iron]

Yes the Information is great, if you understand, how it needs to be implemented



Not at all, most Hobby machine wiring, is simply bad, because they do what they need, to get there machine running

Barefootboy4 has used shielded cable for his limit switches, but has not Grounded the Shield correctly, he could of used regular twisted pair wire & may have had a better result, you can not add a wire to a shield as he has done, this becomes an antenna & picks up any noise, that may be coming from the spindle

The twisted pair is great because it ensures that both wires are equally exposed to any noise. If the receiving end manages the signal in a differential amplifier type setup then the noise is effectively ignored.

I agree that most hobby machines have poor wiring, especially when it comes to grounding ideas and methods. There's enough questions about this kind of thing, and even enough experts around this forum that a Design Guide could be created for wiring and grounding techniques.

I am a fan of the schneider publication for wiring practices (more pictures, the better):
https://www.google.com/url?sa=t&rct=...drit_ciaXkp1MA

[Al_The_Man]

I am a fan of the schneider publication for wiring practices (more pictures, the better):

I am impressed with one thing in that, covers a pet peeve of mine.That they are one of the few documents out there that actually use and describe the correct terms and symbols for Earth Ground and Frame ground.
Just about every circuit you see drawn show a earth ground symbol, whether it is actually meant to be at earth ground or not, instead of where frame or chassis should be used.
This often leads to alot of confusion.
One book considered the Electronic bible by many, The Art of Electronics by Horoitz and Hill. commit this error all the way through the book.
Rant over!
Al.

[wizard]

The twisted pair is great because it ensures that both wires are equally exposed to any noise. If the receiving end manages the signal in a differential amplifier type setup then the noise is effectively ignored.

I agree that most hobby machines have poor wiring, especially when it comes to grounding ideas and methods. There's enough questions about this kind of thing, and even enough experts around this forum that a Design Guide could be created for wiring and grounding techniques.

That would certainly help and frankly tutorials on proper technique with respect to working and assembly would be in order. A standardized panel might be helpful also with electrical prints and so forth. Some of the electrical work I've seen is down right scary.

I am a fan of the schneider publication for wiring practices (more pictures, the better):
https://www.google.com/url?sa=t&rct=...drit_ciaXkp1MA

Wow! This is fairly well done and I have to say even now professionals built panels don't always follow all the recommendations in this guide. It is interesting that one practice I was taught very early in my career, the attaching of pigtails to cable shields, is frowned upon.

Some other things the guide points out are completely obvious even if it isn't common practice. For example bonding spare wires in a cable at each end makes perfect sense even if it is seldom done (at least in the machines I've worked on recently).

There are a couple of things that apparently got lost in translation. They reference for example the bottom panel but seem to confuse it with the bottom of the panel box. Maybe I'm just not reading it right. In another place they use the word "insulation" when I really believe that they meant "installation". The description of the various service configurations leaves a lotto be desired too.

One other thing that seems odd is the lack of mention of star lock washers in bolted up bonding connections. I was always taught that such washers are mandatory and in fact it is widely practiced around here. So I'm not sure if the value of star lock washers has been disproven or this is an oversight. At least they stress the importance of torquing down the connections. Loose connections are a killer!

In any event the document is well worth a read.

[mactec54]

This is fairly well done and I have to say even now professionals built panels don't always follow all the recommendations in this guide. It is interesting that one practice I was taught very early in my career, the attaching of pigtails to cable shields, is frowned upon.

This is one thing I always check for, in most of the Hobby builds, most terminate shields incorrectly, it is also not what you said in a previous post

Back in the day that was fairly common practice, you would wire an extension on the shield and ground it to a common spot. I'm not saying this is a good way to do it but I've seen it even on professionally built machines. Old practices are sometimes hard to break but some of those references linked to make it clear why you might want to change practice.

Even back in the day this was not an acceptable practice, just bad instruction, and people not knowing what they were doing, common sense tells you it is wrong to add a tail to a shield

One other thing that seems odd is the lack of mention of star lock washers in bolted up bonding connections. I was always taught that such washers are mandatory and in fact it is widely practiced around here. So I'm not sure if the value of star lock washers has been disproven or this is an oversight. At least they stress the importance of torquing down the connections. Loose connections are a killer!

They show a lock washers in use, under the Connection section, the star washer is only to keep the connection tight, it does not serve any other purpose, it can be any type of locking washer

[Al_The_Man]

When I got into panel building in N.A. I inquired locally about a standards Reference in this area and most of the answers I got were vague, I had noticed certain common trends in ready built panels so I figured there must be some sort of standard apart from NEC/CEC etc.
I then stumbled across NFPA79 which was a great help, not only in practices, drawings and example schematics but applicable NEC regulations etc.
For those that do not want to shell out the few $$'s for a copy, there is a couple floating around the internet in PDF.
Al.

[wizard]

This is one thing I always check for, in most of the Hobby builds, most terminate shields incorrectly, it is also not what you said in a previous post

I'm not sure what post you are referring to (maybe the sentence wasn't read right), but I was indeed taught to put a pigtail on the shields and ground everything to a common point. It is interesting that the guide refrenced above even mentions grounding power supplies in a star configuration to a common point.

Even back in the day this was not an acceptable practice, just bad instruction, and people not knowing what they were doing, common sense tells you it is wrong to add a tail to a shield

It isn't a question of what is acceptable but rather what was common practice. As far as common sense well that is debatable because the pigtail is far less of an issue if you are operating in a low frequency environment.

They show a lock washers in use, under the Connection section, the star washer is only to keep the connection tight, it does not serve any other purpose,

Again this is an issue of what one is taught and what is consider acceptable in different locations. I was taught that the star washer is there to maintain good electrical connection. It is very common to see the use of star washers recommended in the various manual that automation suppliers put out.

it can be any type of locking washer

The specifications are almost always for star lock washers. Here is another document: http://literature.rockwellautomation...n041_-en-p.pdf, that goes into detail. If you look around you will find references to star washers enhancing electrical conductivity. True or not, many manufactures specify the use of star washers in electrical connections.