DMX Cable Wiring Question (Grounding/Shielding-Related)

[Joseph]

Does the wiring on this cable look right:

It appears the shielding was grounded to the casing. Is this normal? I have a different cable that is not grounded to the casing so trying to figure out what the difference is.

TIA

 

[Yorkie]

That's not uncommon on cheap XLR3 cables but it's bad practice to connect the shell to ground as well as pin 1. Someone who knows the technical reason why can remind me...

[sandall]

Many (most) Chinese made-up cables come with the casings grounded. One potential problem is if the DMX screen is deliberately not grounded to metalwork at either end (a likely cause of hum-loops in audio). Another is if a casing comes into contact with other metalwork, which might be at a (very) different potential.

[Simon Lewis]

I seem to remember that grounding to the shell is forbidden in the DMX spec...

[Andrew C]

Seems to cover it...

Quote

6.7.2 Electrical
Cables should be protected against electrical damage and from transmitting electrical damage to
connected devices. Earthed metal conduit or separate grounding wires for earth grounding of devices
may be used to provide a separate grounding path. Under no circumstance should any cable conductor
or shield be connected to earth ground except at the transmitter as described in the E1.11 Preferred
Topology. In practice the transmitter grounding is likely to occur inside equipment, and connection of the
Data Link Common to earth ground as part of cabling infrastructure will be unnecessary. Equipment
Manufacturers should specify their earth grounding scheme in product documentation.

 

[Joseph]

Thanks for the feedback! So what I'm getting is grounding the casing is bad most of the time. I've used these cables with several par lights without issue, but it just fried the most recent lights I used. 

I guess now I'm wondering:

1. Why did it fry this time and not other lights?
2. How can you prevent this from happening to other lights?
3. Should the cable be rewired? If so, what's the best way for these to be wired? Just remove the grounding to the casing on both sides?

[sandall]

21 minutes ago, Joseph said:

 Just remove the grounding to the casing on both sides?

Yup. Either unsolder or just snip the links.

Edited March 22, 2022 by sandall

[timsabre]

This cable will not damage anything on its own - it will work fine until you have another fault present. If you have had damage then you need to be looking for another fault - probably unwanted voltage on the DMX line caused by a power supply shorting to earth or something. 

The reason for not linking the plug shield to ground is to prevent large fault currents going to earth via the DMX cable. This can cause them to catch fire in an exciting way which is not what you really want. 

[pmiller056]

A related query. Any metal lamp casing (unless built as double insulated - which I've not seen) has to be connected to mains/powerline earth/ground via the lamp plug. What is the consensus on connecting DMX pin 1 (cable screen) within a lamp to the metal casing and to earth via the power connector? This may be repeated many times in an installation, earthing the DMX cable screen in many places.

I've got some small unbranded oriental DMX controlled lamps. They work ok for the intended use. As supplied, the power inlet earth connection is unconnected because a plastic case is used. I have no idea about the integrity of the primary/secondary insulation of the power supply module used in these lamps, so I have modified them to permanently connect the power inlet earth terminal to the power supply module output 0V terminal. This terminal is also connected to the DMX cable screen. The intention is to provide a local path for fault current if the power supply insulation breaks down and prevent the DMX system becoming live or carry fault current. Is this acceptable practice and within the DMX specification?

 

[Jivemaster]

Firstly the DMX cable and connectors will likely NOT be rated to carry an appropriate fault current, nor be rated for a mains voltage.

secondarily any fault imposed on the DMX drivers/receivers will easily destroy them rendering the whole series of lanterns dead (uncontrollable).

If these plastic cased lanterns are supplied as un-earthed then that's how they should be considered, and no jury rigged earth created.  

There is no jury rigged bodge that will correct a non compliant lantern, if it complies then use it -if it doesn't comply don't use it.

[Jivemaster]

Power ground and signal ground are two very different things required to have very different properties, and hence should be designed for their purposes then integrated into the total product design.

[sunray]

I saw the words Jury-rigged and having never known it before, I immediately thought Jerry-rigged was the intended expression.

So checking my thoughts on google I find they are both valid, depending on which 'hit' it seems Jury-rigged is a potentially little less insulting than Jerry-rigged.

 

Everyday's a schoolday.

 

I haven't double checked but I believe mains earth, metal case and DMX ground are commoned together in some LED fittings and a 4 channel dimmer pack I own. Reading your post is quite thought provoking, especially the way the LED fittings are wired internally.

 

S

Edited March 27, 2022 by sunray

[paulears]

I'm not completely sold on the dire doom and gloom interpretation. You often hear the current handling scenario and the dangerousness of sharing grounds, but we don't do it for audio equipment - which also has mains voltages present and common grounding systems, so if it is so dangerous for lights, then it is as dangerous for audio equipment - and we now distribute mains and data to powered speakers in line arrays, don't we? There's no difference in risk assessment. I suspect the problem with DMX is the interference to the data stream from the circulating currents that can get superimposed and because of the daisy-chaining, so hard to fault-find.

 

Pin one problems with chassis bonding are well discussed in audio circles - why do we attach danger to life to lighting kit? Weird!

[sunray]

4 minutes ago, paulears said:

I'm not completely sold on the dire doom and gloom interpretation. You often hear the current handling scenario and the dangerousness of sharing grounds, but we don't do it for audio equipment - which also has mains voltages present and common grounding systems, so if it is so dangerous for lights, then it is as dangerous for audio equipment - and we now distribute mains and data to powered speakers in line arrays, don't we? There's no difference in risk assessment. I suspect the problem with DMX is the interference to the data stream from the circulating currents that can get superimposed and because of the daisy-chaining, so hard to fault-find.

 

Pin one problems with chassis bonding are well discussed in audio circles - why do we attach danger to life to lighting kit? Weird!

Indeed, and a similar thought went through my mind.

With audio the effects of ground loops is immediately obvious, not neccessarily so with data.

However in my experience I have encountered more 'electrical faults' in lighting than sound systems, that added to a typical DMX circuit visits more mains powered devices than a typical audio circuit, I think it's not unreasonable to have more concerns.

[richardash1981]

I wonder if it is considered more likely that lighting (being historically a much larger load) will be distributed across multiple phases and indeed mains supplies. Whilst they should still all be earth bonded to a standard which means there is no touch hazard between different items, that only means (generally) less than 50V - which would be enough to make mincemeat of an RS485 line driver which only has 7V of common mode tolerance! The DMX standard authors probably also had an international audience in mind, and earthing practices have historically been rather "different" internationally.

Certainly it's a lot harder to fault find DMX control than analogue audio, as with all digital control it works perfectly until it doesn't! Also (it seems) rarely does anyone have a scope / eye diagram tool on site to see what is going wrong. Speaking of which, has anyone tried using one of the numerous "pocket oscilloscope" devices to check DMX waveforms? You seem to be able to get 10M samples per second, two channels, for not lot of money (if they are in stock!), which I would think be enough to immediately see if you have an open leg, shorted leg etc.