Where does this means of isolation need to be located, I always thought it had to be on the roof in place of a RTI but I can't see anything that indicates this
Can the device in accordance with 4.4.1.2 meet the requirements?
5033:2012 4.4.1.5 (d) Location of means of isolation for servicing
Re: 5033:2012 4.4.1.5 (d) Location of means of isolation for servicing
I see what you mean. You could interpret it as the one disconnector satisfying the requirements of 4.4.1.2 & 4.4.1.5, although I think the intention is to have a ‘means of isolation’ at the array, perhaps just badly worded.
It’s clearly required at the array in 5033:2014. Clause 4.4.1.6 ‘means of isolation at the PV array’.
It’s clearly required at the array in 5033:2014. Clause 4.4.1.6 ‘means of isolation at the PV array’.
Re: 5033:2012 4.4.1.5 (d) Location of means of isolation for servicing
I presume your reference to "RTI" means "roof top isolator". You would be better to stick to the words used in the requirement ("array switch disconnector" ) rather than sloppy slang. Partly because arrays are increasingly not on rooftops. But mainly because the writers chose to use the particular term because it has a particular meaning, which is more than simply an isolating switch. Not all isolating switches are equal; and the reason for using the different term is to be clear that the required functionality in this case is more than simply isolation. Refer definition [1.4.74].
It's common to just say "isolator"; but it isn't accurate and we always need to bear in mind the specific requirement in any particular case.
With PV arrays we also need to abandon our usual mindset of electricity being a one-way street; from source via wiring to load.
Yes a PV module / string / subarray is primarily intended to be a source, but it can also act as a load for another source; eg a battery, or another part of the array.
We also need to remember that PCE isn't always an inverter, and some kinds of PCE can act as a source back-feeding the array.
So we're dealing with the possibility that wiring ' fittings can be energised from either / both ends - like traffic in both directions of a 2-way street.
4.4.1.1 specifies the overall requirement, the ability to isolate arrays from PCE / application circuit, and vice versa.
The device required by 4.4.1.2 is for isolating the (remote) PCE from the incoming array cable.
Always required, and must be adjacent to the PCE. In the words of 4.4.1.3: at the end of the cable that is most remote from the PV modules".
The device required by 4.4.1.4 & 4.4.1.5 is at the other end of the array cable (as shown in the various Figs), to isolate the array cable from the array. 4.4.1.4 being [A]; and 4.4.1.5 being [NZ].
Where the PCE can act as a source, both these devices need to be isolated before work on the array cable can be undertaken safely.
The array switch disconnector must be "adjacent" to the array (Oz), & "at the array location" (NZ).
The main difference is that for NZ, 4.4.1.5 allows the array output switch disconnector to be omitted; provided several conditions are met.
The presence of a "means of isolation" required by 4.4.1.5 (d) is one of these conditions;
and is required for " the array panel wiring" and "for servicing purposes".
Again note the words used; a "means of isolation" only needs to provide an isolation function, and need not even be a switch.
The paragraph includes several numbered sub-paragraphs; some of which are advisory while others set specific requirements for the device(s) used.
(i) says that sub-array fuses may be used as the means of isolation (which does not exclude other options);
and if this option is taken (iii) requires a special notice WRT removing all such fuses.
(iv) covers required voltage rating.
(ii) requires a label WRT switching off the PCE isolator(s) before isolating the array.
In the 2014 edition these provisions are closely similar, but written much more clearly. I know that edition is yet to be cited by ESRs, and that we are required to comply with 2012 +A1+A2; but seeing how the words have been changed can assist understanding of the intent.
The most significant change is that instead of providing means of isolation for the "array panel wiring" (which implies wiring within the array); it's for the "array output cabling". This, along with other changes, strongly suggests that the issue is isolation of the "array cable" at both ends, by first disconnection the PCE end, and then at the array end. Because action 1 means there is no load current in the array cable, we can use the sub-array fuses for action 2 . Unless both these actions are taken; we can't be certain that the array cable is isolated from all possible sources. So we need to have isolation at each end of the array cable, but only one lot needs to be a switch disconnector.
So while you are correct that no particular location is specified in 2012 for the "means of isolation" ; clearly you can't isolate both ends of a cable using only one device.
As indicated by the Note to 4.4.1.3, there are similar issues when working on modules / wiring within the array; because each string / sub-array can act as a source for all the other strings / sub-arrays (2014 edition provides much more detailed explanation in Clause 3.3) But while the wording of 2012 may suggest the issue is related to wiring within the array; 2014 edition clarifies that the intent of the clause is isolation of the output from the array; not isolation for parts of the array. As does the fact that in both editions, the clause heading is "PV array switch disconnector"(even though no such device is actually required).
It's common to just say "isolator"; but it isn't accurate and we always need to bear in mind the specific requirement in any particular case.
With PV arrays we also need to abandon our usual mindset of electricity being a one-way street; from source via wiring to load.
Yes a PV module / string / subarray is primarily intended to be a source, but it can also act as a load for another source; eg a battery, or another part of the array.
We also need to remember that PCE isn't always an inverter, and some kinds of PCE can act as a source back-feeding the array.
So we're dealing with the possibility that wiring ' fittings can be energised from either / both ends - like traffic in both directions of a 2-way street.
4.4.1.1 specifies the overall requirement, the ability to isolate arrays from PCE / application circuit, and vice versa.
The device required by 4.4.1.2 is for isolating the (remote) PCE from the incoming array cable.
Always required, and must be adjacent to the PCE. In the words of 4.4.1.3: at the end of the cable that is most remote from the PV modules".
The device required by 4.4.1.4 & 4.4.1.5 is at the other end of the array cable (as shown in the various Figs), to isolate the array cable from the array. 4.4.1.4 being [A]; and 4.4.1.5 being [NZ].
Where the PCE can act as a source, both these devices need to be isolated before work on the array cable can be undertaken safely.
The array switch disconnector must be "adjacent" to the array (Oz), & "at the array location" (NZ).
The main difference is that for NZ, 4.4.1.5 allows the array output switch disconnector to be omitted; provided several conditions are met.
The presence of a "means of isolation" required by 4.4.1.5 (d) is one of these conditions;
and is required for " the array panel wiring" and "for servicing purposes".
Again note the words used; a "means of isolation" only needs to provide an isolation function, and need not even be a switch.
The paragraph includes several numbered sub-paragraphs; some of which are advisory while others set specific requirements for the device(s) used.
(i) says that sub-array fuses may be used as the means of isolation (which does not exclude other options);
and if this option is taken (iii) requires a special notice WRT removing all such fuses.
(iv) covers required voltage rating.
(ii) requires a label WRT switching off the PCE isolator(s) before isolating the array.
In the 2014 edition these provisions are closely similar, but written much more clearly. I know that edition is yet to be cited by ESRs, and that we are required to comply with 2012 +A1+A2; but seeing how the words have been changed can assist understanding of the intent.
The most significant change is that instead of providing means of isolation for the "array panel wiring" (which implies wiring within the array); it's for the "array output cabling". This, along with other changes, strongly suggests that the issue is isolation of the "array cable" at both ends, by first disconnection the PCE end, and then at the array end. Because action 1 means there is no load current in the array cable, we can use the sub-array fuses for action 2 . Unless both these actions are taken; we can't be certain that the array cable is isolated from all possible sources. So we need to have isolation at each end of the array cable, but only one lot needs to be a switch disconnector.
So while you are correct that no particular location is specified in 2012 for the "means of isolation" ; clearly you can't isolate both ends of a cable using only one device.
As indicated by the Note to 4.4.1.3, there are similar issues when working on modules / wiring within the array; because each string / sub-array can act as a source for all the other strings / sub-arrays (2014 edition provides much more detailed explanation in Clause 3.3) But while the wording of 2012 may suggest the issue is related to wiring within the array; 2014 edition clarifies that the intent of the clause is isolation of the output from the array; not isolation for parts of the array. As does the fact that in both editions, the clause heading is "PV array switch disconnector"(even though no such device is actually required).
Re: 5033:2012 4.4.1.5 (d) Location of means of isolation for servicing
“In the words of 4.4.1.3: at the end of the cable that is most remote from the PV modules"
Is this wording only relevant when we’re looking at things in terms of overcurrent protection, rather than isolation?
Is this wording only relevant when we’re looking at things in terms of overcurrent protection, rather than isolation?
Re: 5033:2012 4.4.1.5 (d) Location of means of isolation for servicing
Another example of how poorly "5033: 2012" is written. Better in 2014.
Agree those particular words constitute a requirement for location of overcurrent protection. Therefore while they do fit under the overall heading of 4.4 "location and installation requirements"; they should definitely have their own heading about overcurrent devices and not be hiding in a clause about isolation requirements. 2014 edition has a lot more info about overcurrent protection requirements, and under a suitable heading.
But the Note that follows them does assist understanding of the fact that we're dealing with a multi-source situation, configured such that electricity can flow both ways. Accordingly we have to abandon our usual "one way flow" thinking .
The fact that an array cable can be livened from either end means that in order to ensure isolation-for-safety when working on it, we need it to be isolated at both ends.
Agree those particular words constitute a requirement for location of overcurrent protection. Therefore while they do fit under the overall heading of 4.4 "location and installation requirements"; they should definitely have their own heading about overcurrent devices and not be hiding in a clause about isolation requirements. 2014 edition has a lot more info about overcurrent protection requirements, and under a suitable heading.
But the Note that follows them does assist understanding of the fact that we're dealing with a multi-source situation, configured such that electricity can flow both ways. Accordingly we have to abandon our usual "one way flow" thinking .
The fact that an array cable can be livened from either end means that in order to ensure isolation-for-safety when working on it, we need it to be isolated at both ends.
- Rating: 16.67%
Re: 5033:2012 4.4.1.5 (d) Location of means of isolation for servicing
Does 4.4.1.5 (d) (i) also include the 'unplugging' of an MC4 connector within the string / array as a form of Isolation? (as long as the other requirements are meet?)
It would be nice if the standard was more specific in terms of Disconnection requirements.
It would be nice if the standard was more specific in terms of Disconnection requirements.
Re: 5033:2012 4.4.1.5 (d) Location of means of isolation for servicing
Yes, I believe so. I did a short training course on solar a while back. The tutor advised us MC4 connectors are suitable for this requirement.
I don’t know why the standard only mentions ‘sub-array fuses’ as an example, I would’ve thought plug & socket MC4 connectors would be more common for this.
I don’t know why the standard only mentions ‘sub-array fuses’ as an example, I would’ve thought plug & socket MC4 connectors would be more common for this.
Re: 5033:2012 4.4.1.5 (d) Location of means of isolation for servicing
Yes I believe it's sufficient
The hard part is 4.4.1.5 (c) which due to its alteration in ESRs means you either have to run external to the building for the exception of 300mm or use an array switch disconnector because the requirements of running more than that internally to meet the requirements of 4.4.1.5 (c) can never be met
Re: 5033:2012 4.4.1.5 (d) Location of means of isolation for servicing
Isn’t the intention of the change in the ESR around this clause, to replace the use of ‘metallic conduit’ with ‘high density PVC’ conduit, regardless of length?
So running PV cabling within a building enclosed in HD PVC Conduit at lengths greater than 300mm, without an isolator adjacent the array is acceptable?
So running PV cabling within a building enclosed in HD PVC Conduit at lengths greater than 300mm, without an isolator adjacent the array is acceptable?
Re: 5033:2012 4.4.1.5 (d) Location of means of isolation for servicing
Intention, likely but they screwed it up
"after “metallic”, insert “high density PVC”'
Which makes it say "metallic high density PVC wiring enclosure that shall be earthed"
Which doesn't exist so can't be complied with so non-compliant to have no array switch disconnector and lengths greater than 300mm not external
"after “metallic”, insert “high density PVC”'
Which makes it say "metallic high density PVC wiring enclosure that shall be earthed"
Which doesn't exist so can't be complied with so non-compliant to have no array switch disconnector and lengths greater than 300mm not external
Re: 5033:2012 4.4.1.5 (d) Location of means of isolation for servicing
yes, I've never found any "metallic high density PVC" either.
And while inserting "or" would have given a choice, i still can't work out how to earth the HDPVC option.
So agree it's probably a cock-up; and we have to interpret the intent and hope to not be prosecuted.
Actually a compounded cock-up; one within the Standard, followed by another in ESRs citation.
And while inserting "or" would have given a choice, i still can't work out how to earth the HDPVC option.
So agree it's probably a cock-up; and we have to interpret the intent and hope to not be prosecuted.
Actually a compounded cock-up; one within the Standard, followed by another in ESRs citation.