What Rods should read

[malcolmsanford]

Here is what a rod should read and what our old friend E54 advocates.

Fair enough it is not a single rod but 15 19mm dia x 1800mm length rods in a network around the building all joined by a 70mm conductor. So a bit expensive and also would pay havoc in Mr Browns raised flower beds!!!

It is a functional earth for data and our remit was less than 1ohm, so pretty pleased with that

 

[shanky887614]

Why does an IT system need a functional earth RA of less than one ohm?

would it not be better to have there own transformer with tns on the secondry side? than spending 30k on rod install

 

[uksparks]

@ UKSparks, I'd suggest rather than spending 30 pounds per earth pit you might just use a length of plastic drain and cut a slot each side for the wire coming in and going out. You get 225mm PVC pipe and end caps (8-inch in old measurements), you could make numerous rod covers from a single length of pipe.

That is a sodding good idea!! thank you for that, that would look good too, I could even rivet some nice hard plastic earth connection tags to the lids! I like that idea, thanks very much!!

 

[davesparks]

You made me get the drawing out now lol.

There in fact 21 (I missed one) dotted around the building They are roughly 7-10 meters apart depending on basically where it was convenient to site a pit. I wanted to get each rod approx valuation of between 10-15 ohms.

The 70mm conductor is not cheap I agree but I spec it as I needed to hit that 1 ohm target. I was thinking of using some lightening tape but I decided against that as the conductor lays bare and I wasn't sure how it would react in the soil.

I have yet to find any sort of realistic calculations on getting the "right" results for this. I guess you can take soil resitivity and conductor sizes etc, but the bottom line is suck it and see. I may have been able to get away with 50mm conductor, but it it hadn't worked, then there is a lot of red faces on the site, especially mine lol

It is better to use bare hard drawn copper conductor to link them as it increases the contact area with the soil.
If you think about it the wafer thin copper plate on the rods is fine in the ground so a chunky 70mm square conductor must be ok.

 

[Welchyboy1]

Jeeees, thats a lot of money!

blimey 3/4 rods too!!

Well ideally I would like to get a sub 1 ohm, but will see, its only a large summerhouse thing, but I would rather not accept 200ohms, I banged a rod in my garden last week, two 5/8 rods connected end to end and got 42 ohms, I think with 6 I might stand a fair chance of getting a good reading, but who knows! Ill find out in the morning.

The cost does start to creep up indeed, the pits are the best part of £30 each, so its a little hard to explain to the customer I need the best part of £100 of concrete pits when they don't even understand what a rod does.

I imagine it would be pretty tough to explain away 8x earth rods with concrete pits and 25mm earth cable for a summer house, whats the point in that????

bit OTT innit

 

[uksparks]

Ok, so if you have one rod of say 20 ohms, the reading will be 20 Ohms ok..

So suppose, I have two rods in to the ground both reading 20 Ohms, and link them together, what will the resistance be then? Does it work the same like if you were to IR a whole board, taking parrallel resistances into it etc?

What I am trying to work out, is if I have say three 20 ohm rods, hook them together, whats the end result and how do I work that out? Presumably there must be a calculation for this?

 

[davesparks]

would it not be better to have there own transformer with tns on the secondry side? than spending 30k on rod install

You want a sub 1 ohm Ra for the earth of the transformer secondary so they wouldn't be much further forwards.

 

[uksparks]

I imagine it would be pretty tough to explain away 8x earth rods with concrete pits and 25mm earth cable for a summer house, whats the point in that????

bit OTT innit

My point exactly, I don't want to go overboard! However, the bloke is not exactly poor, he has just spent £15k on his summer house, so whats a few extra rods going to hurt so that I can have some fun trying to get a good reading eh!

 

[davesparks]

Ok, so if you have one rod of say 20 ohms, the reading will be 20 Ohms ok..

So suppose, I have two rods in to the ground both reading 20 Ohms, and link them together, what will the resistance be then? Does it work the same like if you were to IR a whole board, taking parrallel resistances into it etc?

What I am trying to work out, is if I have say three 20 ohm rods, hook them together, whats the end result and how do I work that out? Presumably there must be a calculation for this?

I've never yet managed to get a satisfactory answer to this.
But it all depends on how far apart they are and the solid conditions etc. Each rod will have a 'sphere of influence' in the ground around it. If the spheres of 2x20ohm rods overlap then the overall Ra will be higher than 10ohm.

 

[uksparks]

Ill let you know how I get on tomorrow!

 

[PEG]

the earth will settle to the rod in time.

True,and this will not take as long as you think...you find this out,if you try to pull a 40mm cable out from a 50mm moled-hole... must be the same principle as a copulating dogs' appendage :yes:

 

[UKMeterman]

View attachment 29117



It is a functional earth for data and our remit was less than 1ohm, so pretty pleased with that

I ask what system needs a functional earth of 1ohm?

 

[shanky887614]

I ask what system needs a functional earth of 1ohm?

the building is at least 200M parimiter so its not small and multiple floors as well i imagine, im guesing the rods are for a supply tx and might be feeding external buildings as well we don't know.

the hospitals for example have only a handful off 11kv tx's with supplies running in ducting below ground to transformers dotted around the place (i was looking at the paperwork after a proper with one of the big panels)

 

[Marvo]

I ask what system needs a functional earth of 1ohm?

IT systems and data centres require low imp earthing for comms earth references, many comms systems use a very low voltage carrier so even small potential differences across reference points can lead to problems. Also for effective operation of surge and lightning protection, a low impedance earth prevents a rise in potential during their operation which is especially important if there's remote assets. Also for effective screening of cabling etc. The earth for these installations is highly functional as well as its usual protective role.

 

[IzzyS]

Ok, so if you have one rod of say 20 ohms, the reading will be 20 Ohms ok..

So suppose, I have two rods in to the ground both reading 20 Ohms, and link them together, what will the resistance be then? Does it work the same like if you were to IR a whole board, taking parrallel resistances into it etc?

What I am trying to work out, is if I have say three 20 ohm rods, hook them together, whats the end result and how do I work that out? Presumably there must be a calculation for this?

It's not quite as straightforward as resistors in parallel. There's a detailed formula in BS7430 for this, but practice and theory are not the same.

The following may be subject to errors, but gives an idea of what to expect.

Assuming one 8' (2.4m) x 5/8" (16mm) rod gives your 20 Ohms example, the soil resistivity would be ~50 Ohms/m.

Also assuming that the soil resistivity is uniform (unlikely), two rods in parallel, 1 rod's length apart, gives 11.78 Ohms.

3 rods - 8.58
4 rods - 6.84
5 rods - 5.73
10 rods - 3.29
20 rods - 1.64
33 rods - 1.00


My calculation for Malcolm's 21 rods spaced 7m apart , averaged at 13 ohms each comes out as a combined resistance of 0.77 Ohms, but doesn't include the interconnecting copper.

 

[malcolmsanford]

I ask what system needs a functional earth of 1ohm?

I can't really say too much as it is a military installation but Marvo is virtually spot on. The equipment is high tech IT equipment with an emphasis on...... struggling to know what to say but it's virtually an x box scenario but a little more realistic.

Put it this way the building has 7 data rooms with about 20+ racks each room all connected together to achieve the result. Unfortunately being an old sparks the data side of this and all the hardware/software is a little beyond me, but the spec was a 1 ohm "clean" earth or "noiseless" earth so it could not be go via the transformer.


The actual supply into the building is via 3 Tx and I'll show you the earth grid around that when it is finished. I was asked to help on that but fortunately we had some HV guys sort that part out