This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Local Isolation For A/C Internal Units

Hi

Doing EICRs, and the remedials resulting from them.


An issue had been raging as to whether an internal unit needs to have a local isolator.

There have been 2 schools of thought over this issue with others I am working with.


First one:

It is a an electromechanical piece of equipment and needs a local isolator even though it is being fed by an external unit that has it's own isolation.

Second one:

It is fed by the external unit and they are both one piece of equipment even though they are split with the two parts in different places. Turning off the isolator to the external unit isolates all the equipment.


In my opinion a local isolator is still needed as there is no way of knowing if the internal unit is definitely part the the external unit being isolated. It may just be off at the controls.


I have come across many A/C units that have been installed by A/C engineers and they have not put an isolator on the internal unit. I'm wondering if there is a reason that they don't or if it's just ignorance of the regs on their part. I would have thought their training would have included that. Is there something that they know that means they don't need to install an isolator to the internal unit?


Anyone have any thoughts?


Thanks

  • The EICR form has a space for allowing the release of the report. However the regulations make it very clear that the duty holder is the Inspector, and it is entirely his responsibility. If one wishes to argue the fine details the OSG is not the reference document, BS7671 is. Any Inspector must have and understand the regulations throughout, there is no leeway for things like "We have always done it like that" or similar.
  • Isn’t the version of the model IET EICR in the OSG just a completed example of the one in BS7671.


    The authors of the OSG chose to give an example of a report completed by one person, but there is definitely provision for a QS to authorise it, therefore the QS has to have an input and agree the observations and the coding of them.


    Andy Betteridge

  • Ajjewsbury

    You misunderstood what I meant.

    That is exactly the way I would do it to make sure I was operating the correct device for isolation.

    What I was saying was that if a series of machines were already isolated and locked off at remote isolators then the one you was working on would already be dead. In that case you couldn't test live first, isolate then test dead to confirm. Re-read what I wrote. Having an isolator local to the machine you would be confident that you have isolated the machine being worked on.



    But isn't that down to an unsuitable procedure - you could, for instance, decide at the start of the day (or shift or hour or whatever) which machines you'd need to isolate, prove live, isolate and prove dead all of them, then start work on each one. I'm not saying it'll be easy or convenient - but it would be physically possible. (see next point as well)

     

    Regarding the second bit you commented on. Yeah! If you want to be hung, draw and quartered. Were talking labs, workshops etc. You can't go turning off other equipment just to work on one.



    That's where "service conditions allow" come into it - i.e. just what the customer's requirements are. In some places it would be acceptable, in others clearly not. If having local (or extra) isolators makes work easier or more efficient than that's absolutely fine (more than fine to my mind), but that's all a matter of specific customer requirements rather than any blanket requirement of BS 7671. Either way you need a suitable combination of procedure and isolators to be be able to be confident that things are reliably isolated.


        - Andy.
  • Just got home from work and saw all the new comments.Been a late one.

    Thanks everyone for commenting.

    I have now got it more clear in my mind.

    I will be discussing this with my supervisor as to the wrong coding of 2 when it should not be coded but only reported as to clients request.


    Ajjnewsbury:

    Regarding your first comment on my comment. I am talking about a situation where these items may have been locked of for an extended period of time and you have had no control over that. I have come across these situations. I normally termporarily re-energize the circuit, if possible, to confirm that the remote isolator is the correct one. Problem is, if the equipment can't be enrergized due to the fault. You then have to disconnect the supply cable to the equipment terminals first. With a local isolator - no problem.


    Regarding you second comment on my other comment. I was only referring to practicalities, not to requirements of the regs.
  • Although the regs is wooly I think the key is in the wording "Switching for Mechanical Maintenance". I would always include a local isolator adjacent the equipment. You wouldn't expect a mechanical servicing engineer to go searching for the external rotary isolator when all they've come to do is service an internal unit.


    I've just had 3 external units installed serving 9 internal units all with local isolation - for the sake of a tenner a unit I'd always install them.


    On another note a consultant just designed some internal AC units for me on a large public building all having local isolation but suggested wiring the AC units on local Lighting circuits which then became a totally different argument. They argued that they were low wattage units so its perfectly acceptable. I argued that if the circuit needs isolating safely for maintenance then you lose the lights in the floorspace areas which adds to the risk. Reply was you have emergency lighting for that. The argument rumbles on.
  • Similar to the issue with having electric gates wired directly into a lighting circuit I mentioned earlier.


    Is not any maintenance on machinery actually mechanical maintenance, even if the parts being worked on are electrical components?


    Andy Betteridge.