In my view a single bonding connection to the whole rack would suffice, and in view of the extra low voltage, even that might not be required.

quite. Assuming decent construction - that is using shake proof washers or similar- I'd be happy to omit bonding between anything that passes a 'beep' test with a meter on the continuity range reading  - to within a last digit flicker reading the same as the test leads. If need be document the results.

Bonding is more important between parts that can become separated in normal use, like doors on lift off hinges, and even then personally I prefer finger stock or spring contacts, as bond wires tend to get broken off and not noticed

And of course it matters more in structural parts that carry current by design, such as in when the vehicle chassis carries return current from a light fitting via its bracket or something, which is not really what you seem to be doing.

regards Mike.

Thanks for the input, much appreciated.

Given the comments, I will fit a single wire to the overall rack and shall ensure all bolted joints have shakeproof washers, and will probably use nylock nuts too.

In the worst case scenario, I can always retrofit additional bonds.

I have access to a milliohm meter, so can easily check that all parts are equipotential (within a given number of millohms).

Tha is again for the input. 

If you are not sure how many milliohms to worry about, and actually want a figure, you can estimate if you can  decide what the worst credible fault is - perhaps a short from a battery supply fused at 10A would need to drop no more than a few % of the battery voltage at 10A, to ensure a fast and reliable 'pop' with a few times 10A, without overheating the bonded part too much in that time. 

So for an item that could credibly short to a cable fused at 100A, the bonding requirement needs to be that much better than if the requirement was to operate a 10A fuse. If you can be sure that the fault condition is incredible - if the insulation chafed off and then the wire snapped it still would not reach, sort of thinking, then you do not need to worry about that current path at all.

In vehicle battery systems where the voltages are lower and the currents higher than mains practice, the impedance can easily be beyond the lower limits of a normal meter, and then not worth trying to measure,  rather things are just bolted up nice and tight onto clean metal and the resistance calculated/estimted. Usually fixing metal work gives you lots of cross-sectional area, and even in a large vehicle the distances are quite short.

(For an extreme example a motorized tank turret I think was 200-300A at 24-28V - so to blow the 500A fuse fast needed a reasonable multiple of that, say about 2000- 3000A , so that needed (and had) a return path of less than ten  milliohms or so. (1mm2 copper is 16-19 milliohms per metre, but you should also consider the adiabatic heating of that.) The tank cables were something like 35 or 50mm2 - I am quite sure your case is much less extreme !! These figures are not exactly right, being dragged from memories of a conversation many years ago with someone who worked on the controllers for the things.)

Note that your earth wire may need to be reasonably chunky for mechanical strength, even if low current handling, unless protected. The normal 7671 rules used at 230V are not really what you need but there is some read-accross of the ideas.

Mike.

The equipment rack will have 12 Volt and 28 Volt equipment mounted on it.

What's this "bonding" there to achieve? Is it the "earth"/chassis return for the d.c. system or meant to be part of some protective measures? or for EMI reasons?

   - Andy.

I can't see any bonding is needed In the usual sense of the word,we only bond to keep voltages from being unequal between simultaneously accessible conductive parts. As you using extra low voltage this is most likely not required,unless I'm missing something or it's for other reasons as stated above

Hi Mike,

Thanks for the extensive reply.

To be honest, I'd not actually considered the DC current fault scenario at all, so I will be taking a look at that over the next day or so. The DC fault current could be in excess of 100 Amps. 

My primary concern was and still is compliance to the regulatory requirements for "protective equipotential bonding" that apply to metalwork that's fitted into a vehicle, when there is 240 Vac present in the vehicle. Note - I have just noticed that I didn't mentioned there was 240 Vac in the vehicle. However, there is no 240 Vac equipment or cabling in/on the equipment rack.

My original concern about a possible need to bond every separate piece of metal that makes up the equipment rack came from a statement that  I read in the following article: https://electrical.theiet.org/media/1187/electrical-installations-in-caravan-camping-parks-caravans-and-motor-caravans.pdf. Specifically the statement "Regulation 721.411.3.1.2 requires structural metallic parts that are accessible from within the caravan to be connected through main protective bonding conductors to the main earthing terminal within the caravan."

Steve

Hi  Andy,

Primarily, I'm including the bonding to ensure compliance to the "protective equipotential bonding" requirements. 

The bonding wire (or any part of the metal equipment rack) is not intended to carry DC 'return' current (all DC equipment have dedicated positive and negative cables)

Steve 

Hi,

Apologies, I did not make it clear in my original post that the vehicle installation DOES include 240 Vac equipment. However, this equipment is not on the rack that I'm referring to, and there are no AC cables on the rack either.

Steve

Primarily, I'm including the bonding to ensure compliance to the "protective equipotential bonding" requirements. 

In which case the short answer is probably that you don't need any bonding at all then. In general it's extraneous-conductive-parts  that need main bonding - i.e. parts that can introduce a potential (voltage) into the system that can differ from the installation's own Earth reference (i.e. the MET). So in normal buildings it's things like metallic water and gas pipes that bring "true Earth" potential in from outside (or fault voltages from other installations). Metalwork that's wholly inside the vehicle isn't going to be able to import a potential from anywhere. (If is was half in, half out, resting on the ground outside, it might be different).

Occasionally you'll come across supplementary bonding that does include metalwork that wholly within an installation - but that's to establish another equipotential zone within a smaller area (e.g. a bathroom) - probably unlikely in this case (and again if the metalwork is wholly within that area, it's unlikely to be an extraneous-conductive-part as far as that location is concerned).

   - Andy.