TT System High Ze leading to possible Zs fails

[North88]

Doing a lot of EICRs at the moment and have came across what I'm sure is a common enough problem. In an off grid rural cottage I have a TT system, Diesel Generator that runs and charges batteries when they are below 10 percent or so.

Ra is about 150 Ohms... was about 3 ohms for Ze before main earth was disconnected from electrode.

Now table 53.1 tells me that a 500mA RCD would be acceptable in this instance, not that that would ever be the case as everything is protected by 30mA RCD albeit one main one for everything. ( An exception to this is the circuit from generator to Invertor IN circuit and Invertor OUT i don't believe is on an RCD. Invertor out is a 25A MCB which limits the entire supply which is a second 2 way DB in the shed that house the generator doing 2 sockets and then onto an 8 way main RCD board in the house.

My question is basically most mcbs 20 amp and above are failing max Zs readings ( In most instances not by much ).

Because it is a remote holiday let it has a very small amount of sockets and no kettle or any other appliances supplied and all heating cooking is gas/oil. I maybe have the luxury of sticking a couple of rings onto 10A mcbs in which case they pass but not sure if table 53.1 negates the need for this?

SO long as the RCD trips within 200ms can this be passed so long as deviations from BS7671 are noted?

Example readings are 16A MCB 2.5 Ohms which is 2.2 max and 32A MCB 3.5 Ohms which is 1.1 max. all other 6A circuits pass.

Appreciate advice as its not something i have dealt with too much over the years.

( Sorry for waffling on also cheers! )

 

[timhoward]

He also advised that the only protection the invertor provided was anti-surge san that was to protect the invertor itself and would not disconnect the load side feeding the house.

The manual says "The AC HOT OUT terminal is to be connected to the AC loads through 30A branch rated AC circuit breakers."
I've just read your first post again and you say there's a 25A MCB doing this job, so that bits all fine.

It's slightly bending my mind that there's a TN-S generator then feeding a TN-S inverter. i.e. basically two N-E connections. My brain is wondering if they should really be separate earthing systems (i.e. not linked) so a fault on one of them is contained.

@Rockingit knows his stuff much more than me in this area by the way. But I tend to go back to first principles. What would happen at each point in the system if there is a fault to earth, or a dead short.
As pointed out the generator would stall on a dead short, if the C40 didn't get there first. It looks like a fault to earth would remain undetected at the generator (which is one reason I think best if the inverter is earthed separately)

The inverter output is similarly protected by a 25 amp breaker. Dead short is no issue. Fault to earth, no chance of the 25 amp breaker tripping.
From that point onwards you are exactly the same as a house with a TT supply, everything is RCD protected for fault protection.

To my thinking the safety of this lot simply comes down to the method and quality of the install between the generator, inverter, and the first distribution board. If well installed and mechanically protected it doesn't feel any more dangerous than an average TT house.

 

[North88]

The manual says "The AC HOT OUT terminal is to be connected to the AC loads through 30A branch rated AC circuit breakers."
I've just read your first post again and you say there's a 25A MCB doing this job, so that bits all fine.

It's slightly bending my mind that there's a TN-S generator then feeding a TN-S inverter. i.e. basically two N-E connections. My brain is wondering if they should really be separate earthing systems (i.e. not linked) so a fault on one of them is contained.

@Rockingit knows his stuff much more than me in this area by the way. But I tend to go back to first principles. What would happen at each point in the system if there is a fault to earth, or a dead short.
As pointed out the generator would stall on a dead short, if the C40 didn't get there first. It looks like a fault to earth would remain undetected at the generator (which is one reason I think best if the inverter is earthed separately)

The inverter output is similarly protected by a 25 amp breaker. Dead short is no issue. Fault to earth, no chance of the 25 amp breaker tripping.
From that point onwards you are exactly the same as a house with a TT supply, everything is RCD protected for fault protection.

To my thinking the safety of this lot simply comes down to the method and quality of the install between the generator, inverter, and the first distribution board. If well installed and mechanically protected it doesn't feel any more dangerous than an average TT house.

Cheers Tim,

On Paternity leave now so just sat down and got a look at your reply. Everything you are saying makes sense to me.

What I'm wondering now is how you guys would explain why you demand the installation is "Satisfactory"? as far as EICR paperwork goes. Basically what i am asking is how would you fill out the paperwork? List departures from BS7671 in observations? Note that the Zs values are too high but explain why that is ok?

 

[North88]

It's what we're here for!

So, the generator requires an earth regardless of what's downstream of it - BS7430 is the guidance on this but you'll find an easier to digest summary in section 2.4 of the OSG.

The 1667Ω max Zs is applicable to any circuit* downstream of a 30mA RCD irrespective of TT / TN status because that makes for safe touch voltages (search on other threads on here for that).

ADS basically doesn't work in this situation - if you do some rough maths a 9KW generator is capable of producing 38A (if we ignore some finer technicalities which happen incredibly quickly then diminish). Consider then the time/current curve of your C40 breaker and it's fairly obvious that will NEVER trip because even if you had a Zs of 0.001Ω (pfc = 230kA!) you can't get more than 38ishA out of the supply. However, an eg B6 device will activate for obvious reasons - so long as the fault current remains active long enough for the time curve.

*there are some special exceptions to this that don't apply here

Cheers,

Had a read of that section and while it seems to reference portable generators the tail end of 2.4.3 pretty much sums up what I'm understanding about it all I think. Basically saying the Ra value is hard to achieve so RCDs are used to provide ADS. Is this what you would advise to note on the EICR?

I know its not for you to say as you don't have to sign the install off it is me, just after a bit of advice.

 

[davesparks]

Basically saying the Ra value is hard to achieve so RCDs are used to provide ADS.

No, Ra is not part of the earth fault loop in this installation.

 

[North88]

No, Ra is not part of the earth fault loop in this installation.

Could you elaborate a little Dave? What is your advice in this situation?

 

[davesparks]

Could you elaborate a little Dave?

It is a TNS system, the earth rod is providing an earth reference to make on epile of the supply neutral and is not part of the earth fault loop.
The earth fault loop is from the source of supply out via the live conductor and returns via the earth conductor back to the neutral point via the N-E link. Standard TNS arrangement except the source of supply is within the installation instead of being external.

The issue you have with ADS is that the source of supply cannot provide enough fault current to operate protective devices in the required time (if at all)

What is your advice in this situation?

Engage the services of someone who understands the type of installation.