New member, new Journeyman (drag chains, dust collection, spindle grounding, ...)

From what I have read on the topic, my guess would be when power (router/spindle) and signal cabling is in close proximity, such as when bundled in a drag chain. Given the prevalence of drag chain use on CNCs and other industrial equipment, I would think quality cable, appropriate shielding, and correct grounding is mitigating the problem in most situations.


Check out @RowdyRoman 's stuff. He makes a variety of nice to have accessories.

1 Like

Hey Tom,

you mean when someone comes up with the idea to combine this trim router with its unshielded household appliance cable with a drag chain?

I don’t hope anyone will do that if they have to choose a new cable anyway

Use shielded cables to avoid electrical interference. Not doing so might
result in unexpected behaviour or the unit.

Use shielded wiring for the motor cable and all analog and digital control

Allow the effective shield area of these lines to remain as large as possi-
ble; i.e., do not strip away the shield (screen) further away from the cable
end than absolutely necessary.

Omron MX2 User’s Manual

Only use shielded cables suitable for drag chains with a core cross-section
designed for the rated motor current. The cable shield must be grounded on
both sides.

Mechatron HFS Series High Frequency Motor Spindle’s Operator’s Handbook

…And the cables that Onefinity puts to the steppers, if they are already using those flimsy tin-plated Molex plugs, are they at least shielded (not that I won’t throw them away as a whole anyway)?

I think that is exactly the problem. I have learned a great deal since building my controller enclosure, and the topic of proper cabling, shielding, grounding, ground loops, etc was constantly stressed in all the industry supported documents I have read. I really enjoy learning about all of this (your contributions on this forum, and the fact that you are referencing manuals, suggests you are the same), and putting the knowledge to use in my practical applications. Not everyone has the time, patience, or interest in doing this.
Comments like Onefinity’s are probably made more from a business perspective, where it is easiest to avoid EMI issues in the first place, rather than dealing with the often difficult task of trying to identify their source after the fact.


Hey Tom,

I think a comment like “drag chains can introduce bug creating emi” is a very general statement.

I hope with my answer I have assured all readers that indeed, you can buy an Onefinity cnc and use drag chains. I assume that who plans something that is at least somewhat demanding like that will certainly also implement the cabling professionally (and in my opinion will not have to fear bug creating EMI)

1 Like

D-1 CE-EMC Installation Guidelines
Avoid parallel arrangement of low-level signal wiring and power-carrying
or noise-prone conductors.

Omron MX2 User’s Manual

Would we say, is the stepper driving signal a low-level signal. Steppers are power motors, stepper drivers are power electronics.

Do you know the answer to this question: The cables that Onefinity puts to the steppers (if they are already using those flimsy tin-plated Molex plugs) are they at least shielded (not that I won’t throw them away as a whole anyway)?

I do not believe stepper cables are considered low-level signal given the voltage and amperage they carry. I do not see shielding or twisted pair wires on the 1F stepper cables. From what I understand shielding can be used to help keep out, or help keep in, EMI. Also, for this to work best the shield would need to be connected to earth ground, at the controller end is what I see most often recommended. Even if the 1F cables had shielding, this would lose effectiveness without this ground connection.
It doesn’t help that the Internet is full of seemingly valid information that regularly contradicts itself on all of these topics.

Because I also run my homing sensor cables in my drag chain, I purchased shielded cable and properly grounded it - all shields to the same earth ground (star configuration) at the controller end. I also looked for sensors that use 24VDC to help mitigate issues of EMI which I believe affect lower voltage signals more. I feel the extra cost of shielded cable, and extra time to solder the shields to common ground, is worth it.

1 Like

Hey Tom,

I share this opinion. What I see in the Omron EMC advice is to avoid a data line close and parallel to a e.g. high frequency spindle power line. I would avoid this anyway. But it’s a very general advice.

I thought so. But the signals to the steppers, which have varying high frequencies and which go through these cables are “dirty” in terms of EMI, and strong due to the amperage. If I imagine this together with the unshielded household application cable of the trim router it cannot be a good thing. It’s clear when you build things this way, you can only warn.

I just bought a bunch of Omron E2B Cylindrical Inductive Proximity Sensors. They sell nice shielded cables with nice connectors with it (to avoid to have too much to solder). That’s the fourth thing I would put into the drag chain (after the shielded high frequency spindle power cable, the (shielded) Z Axis stepper driving cable, and the cooling hoses)

1 Like

Which specific E2Bs did you decide on?

E2B-S08KS01-MC-C1. They are 8 mm ø with connector.

Star configuration is to avoid ground loops.

– Source: Omron MX2 User’s Manual

Connection of the spindle to the frequency converter

Lines 1, 2 and 3 are to be connected to terminals U, V and W on the frequency converter. The protective conductor connection must be placed on the PE terminal of the frequency converter. On the frequency converter side, the cable shield must be laid as large as possible (using a suitable cable clamp) on the grounded plate (usually the base plate of the control cabinet) on which the frequency converter is mounted. If the screen is not connected, there may be potential differences, which can lead to EMC problems or even electric shock!

– Mechatron HFS Series High Frequency Motor Spindle’s Operator’s Handbook

If your spindle has a PE connector, the instructions say that the cable shield should be connected at both ends. Unlike flimsy no-name spindles, my Mechatron spindle has the fourth “PE” pin connected to housing which must be connected to cable shield. However if the spindle housing has contact to the 80 mm mount which has electrical contact to Z Slider which has contact to X Axis which has contact to Y Axis which has contact to machine’s feet which could have been grounded, it would introduce a ground loop. Is the machine grounded at the feet?

From the reading I had done earlier I found that the shielding requirements for spindle/VFD are different. However, misinformation is plentiful as this thread highlights:

As for the grounding of the 1F frame, I suppose that would depend on if you did this yourself - some CNC users ground their chassis for a variety of reasons. I have seen some stepper motor frames grounded by some, which in turn also grounds the CNC chassis.

Grounding is one of the most written about and misunderstood topics I have come across in my reading. Sometimes the different answers come down to ‘best practice’ vs ‘good enough’, as well as ‘industry’ vs ‘residential’ electrical codes, and so on…

When in doubt I think it best, as you are doing, to follow the manufacturer’s guidance.

As an aside, which Mechatron spindle did you end up purchasing? I have not decided on a manufacturer yet - still need time to save some money.

Butch, I use a cheap Harbor Freight 1hp dust collector with a dust boot I got off of Amazon. It required a little bit of trimming of the PVC to make sure it cleared when the spindle moves up bit this setup works great. The dust boot came with ring inserts for different sized spindles (but not one for 65mm). The rings center the spindle in the dust boot where someone with a 3d printer could make a ring to fit a 65mm spindle that was offset so the boot doesn’t interfere with the mount for the Z axis pretty easily. I’ve been lazy and haven’t replaced the cloth bag for the dust collector with a real filter yet but I will. I’ve been waiting for my Journeyman rail to ship (just got the UPS info so soon now) before finishing things up.


Hey Tom,

Yes, usually it’s a disease, that’s why I rarely approach things this way. Usually I buy literature used for engineering studies. My approach (in every matter) is to understand things. I achieve this mainly by rtfm rather than by visiting web fora. But web fora are nice if you meet someone that is exploring the same thing at the same time!

That is a common phenomenon with a lot of topics: Everyone says something about it, regardless of whether he or she has studied it or just heard something about it.

I’ve been dealing with grounding for a long time and I would say, it is not rocket science.

When I said above that it could introduce a ground loop, I am aware that it may not always be relevant. A relevant ground loop is if your cable TV provider grounds its cable in his distributor board out there a few miles down the road and then you ground it again at your cable TV wall outlet. It can produce a lot of hum that deteriorates the signal.

Even if the housing of the spindle may be grounded via the bearings on the rails of the CNC, I would not consider this as to be the reliable way of grounding it. I would think that there must be an intentional electrical connection of correct dimension. Protective Earth (PE) has to carry any current that can appear erratically at any point on the machine.

Also there is a difference between circuit grounding, shielding and protective earth. Usually shielding and protective earth do not participate in the operational current flow. There may be assymmetrical signal or power cabling where shielding is at same time functional grounding, but to avoid ground loops, usually digital and analog signals use symmetrical (balanced) connections with separate shielding (or twisted pair if one wants to avoid shielding).

A 3-phase spindle is operated with three wires using a delta circuit, there is no Neutral conductor that has to be connected to ground. The fourth connector on the spindle, to which the spindle housing and the cable shielding are connected, is protective earthing.

I bought this one: HFS-8022-24-ER20 (Datasheet) (with Cooling Station)

That looks like harbor freight unit I was looking at. It does an Ok job? I’ll probably go with 1F dust boot kit. Thanks for the info.

" However if the spindle housing has contact to the 80 mm mount which has electrical contact to Z Slider which has contact to X Axis which has contact to Y Axis which has contact to machine’s feet which could have been grounded, it would introduce a ground loop. Is the machine grounded at the feet?"

Isn’t this ‘contact’ that you speak of only available through the Z and X and Y bearings? So not a reliable electrical connection. I know it’s not a path you are wanting but it seems it could be an intermittent path and hence difficult to troubleshoot if there were ground loop issues.

Hey Bob,

yes, you are right, I came to this point in what I wrote later (above): The fourth pin (Protective Earth) of the spindle does not participate in operational current flow, so the issue of a potential ground loop is neglectible, but as being part of protection, it is clear that there must be an intentional electrical connection of correct dimension. For the intermittent path through the bearings this is not true. The instructions of the manufacturers mentioned above also are very clear about that:

On the frequency converter side, the cable shield must be laid as large as possible (using a suitable cable clamp) on the grounded plate (usually the base plate of the control cabinet) on which the frequency converter is mounted.

It is important here that if shielding serves as protective earth here, when choosing a cable, it must be dimensioned in such a way that it can divert any possible fault current that may occur on the machine.

Also for safety, I always plan to use a RCCB (in fact, a RCBO, a combination of RCCB and MCCB) in my control cabinet.

1 Like

Hey Alex,

what I like about your setup is the combination of simplicity and effectiveness. You have installed the dust collector at an ideal point in the room to hang the hose at only one point. I assume that the hose will also work with the Journeyman X-axis. You built the table so big that the Journeyman will fit on it. I always try to do everything so perfectly, but I always have to realize that of course I can’t get things done very quickly. With your setup, I immediately think about the fact that you already had a lot of fun with it because it is already running. And the cable at the other suspension point on the ceiling, it seems to me it is a cable and a transparent hose? What is the box for to which you attached the Onefinity Controller? It looks like a PC. Or is it a cooling station?

Hey Butch, @Butchbal

If you find drag chains too complicated I found this guy has a simple solution:

1 Like

It’s an old PC case. Inside it is the VFD, power supply, and the solid state relays. It’ll be mounted to the side of the table out of the way. I was going to mount the controller inside it as well and move both the power switch and the E stop button but decided not to for now due to possible electronic interference. I also ran 240v power to the case and split it up for the various things that need power (like the VFD and the SSRs for the water pump and dust collector). I have the two water lines and power cable for the spindle bundled together away from the servo cables. Like with most of us here it’s a work in progress and will change with time.