Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Low Resistance Ground System

Status
Not open for further replies.

rphelps

Electrical
Dec 16, 1999
2
0
0
US
Can anyone think of a reference as to how low the earth resistance should be on a ground system. The NEC Code indirectly recomends 25 ohms with a provision for getting out of actually having to obtain that level. The Mil Spec 98 is a little more stringent. The Green and Emerald books both support this idea I believe, although I can't find it directly written. I am having a problem convincing the project team that lower is better.<br>
<br>
It goes without saying that any improvement beyond 25 ohms will have to be weighed against the cost but what is the real benefit. 25 ohms is a target that is personnel protection related (1/3 the average human body resistance). <br>
<br>
In a remedial situation, what level should be the goal to protect instrumentation communicating on an RS-485 network across large areas. We have had repeated damage and we know our earth ground resistance is extremely high.<br>
<br>
I do realize that there are many other factors involved like proper installation techniques, bonding, isolation, TVSS, etc. We have gone all through these areas and made corrections and improvements but are still getting failures.<br>
<br>
I am being challenged that for equipment protection, there is no benefit to pursue a level of earth ground resistance any lower than 25 ohms. It has also been stipulated that any improvement would be detrimental because it would make us a better target. I don't agree. Thanks.
 
Replies continue below

Recommended for you

I might not answer your specific question but here is some info which can help you along in your analysis.<br>
First of all, I am not 100% familiar with the NEC as I am used to working according to IEC364 specifications.<br>
IEC recognises two different grounding systems where the star point of the LV transformer is connected to ground :<br>
- the TT system : where the loads are connected to a different earthing electrode than the transformer's,<br>
- the TN system : where the loads are connected to a PE (or CPC=distributed Earth) cable.<br>
<br>
To which earthing system are you refering ?<br>
<br>
In any case one can say the following :<br>
- in TT, the object of the game is to get an electrode impedance as low as possible. The figure of 5 to 10 Ohms per electrode (there are two: one for the transformer, one for the network) is often mentioned. The lower this impedance the higher the potential earth leakage current, and the easier it is to implement earth leakage discrimination with RCD units, all the while offering protection against indirect contact and against fire, etc...<br>
Ex :eek:n 400/230V network, for an electrode impedance of 10 Ohms, one would get an Id of 11,5 Amps. <br>
One should remember that the resistivity of the earthing electrode can change with time. It is said that this change can be in a ratio from 1 to 3, and it can depend on the humidity of the soil, frosted soil, age of the electrode, ... so the best is to start with an electrode with an impedance as low as possible to cater for any deterioration or fluctuation in the impedance of the electrode.<br>
- in TN systems, the problem is different as the earth fault current returns through the PE conductor, and is generally quite high (in kA this time). Usually, the breakers should be set to trip instantaneously on this current (the earth electrode is not part of the return current). The earth electrode impedance will usually only come into play during HV faults, ...<br>
<br>
Regarding your comms equipment, try to keep it away for all the power conductors, shield it, and earth the shielding as often as possible (at least at both extremities). TN systems intefere a lot more with comms installations than TT systems (due to the high fault currents, in kA). In TNC (Neutral and Earth common) installations, load imbalance and triplen harmonics create currents on the PE conductor. This in turn can cause EM interference, differential voltage in cable shielding, etc... Reducing the electrode impedance will not improve the installation here, you have to reduce these currents in the PEN conductor (switching to TNS maybe ?).<br>
<br>
Good luck !
 
Lower than 10 ohms is seen as a desirable value. However this is not always possible without an uneconomic amount of cost. In situations where low resistances are unachievable it is important to evaluate the step and touch values so as to minimise risk of injury. Other ways of equi potential bonding insulating operating handles can be implemented
 
The &quot;required&quot; resistance depends upon the type of installation. For anything following the NEC, 25 ohms is a target but not a requirement. If one electrode is over 25 ohms, add another and the resulting resistance can be anything. <br>
When it comes to electronic installations (communication towers for instance), I have seen required resistances as low as 1/2 ohm - difficult to achieve in many cases.<br>
Some electrical substations installations have specified 5 ohms max.<br>
It all depends!
 
Theoretically, the resistance to the ground shall be small enough to open an immediate protective device upstream the fault to the ground fault (which is usually by connecting a phase to the equipment ground). The system ground fault is a different matter or concept related to the voltage level. It is not unusual that the quality of ground is tested by special meters and even practically tested by creating a short to the equipment ground and waiting whether or not the protective device open the circuit. New ground fault circuit interrupters can be very effective in clearing the fault to the ground and can be tested by pressing a small push button on the receptacle. Some grounds have very small resistance, e.g. ground to a copper water pipe (one ohm or less) which should be looked for first before installing any ground electrodes. Also, the steel building structure can suffice in some cases. This shall be verified by measuring the earth resistance. Some projects, e.g. nuclear power plant, have system ground, equipment ground and instrument ground which are all separate, stand alone systems for better, more sensitive protection of equipment and people.
 
Status
Not open for further replies.
Back
Top