Robbo's last post raises an interesting point.

What I described previously is following the route prescribed by the compliance document for building code clause G12.

The building act section 22 says that if you follow the compliance document you must be treated as having complied with the building code. So you can give a two fingered salute to the council if they don't like it provided you have followed the compliance document (and note that both AS/NZS 3500 and G12/AS1 are in the compliance document so you can use either).

The building act section 23 says that you can come up with ways of complying with the building code other than the compliance document. I have attached the department of building and housing guide to coming up with alternative solutions.

So if you want to use the World Health Organisation guide to legionella as a basis for an alternative solution you are allowed to so long as you can establish it meets the performance requirements of the building code in my previous post. You would be proposing an alternative solution to the legionella requirement but probably using the compliance document for the no-scalding part ie: keeping temperature under 55 (as arguing for a temperature hotter than this will fly like a lead balloon at the council - at least we are better off than the Aussi's - its 50 max. over there).

There are a few practical issues with doing this:

1. Electric cylinder thermostats often bottom out at 60 these days.

2. If the thermostat craps out then the safety cut-out is often at 88.

3. Don't forget about stratification in the cylinder - just because the thermostat at the bottom of the cylinder (on the outside of the pressure vessel) has cut out at 55 there is absolutely no guarantee the water delivered from the very top of the cylinder wont at times be quite a bit more than 55.

To get around this you may have to spend quite a bit of time tuning the cylinder to make sure the outlet water temperature doesn't go above 55. And it will be slightly different for different heat up cycles and also different for cylinders of different height to diameter ratios.

4. If you are doing a consented job you will need to work an alternative solution through the council. Some will allow you to make an application for a "standard" alternative solution if you intend to use it over and over.

5. Remember if you are doing work which doesn't require a consent you still need to comply with the building code and if you use an alternative solution you would be really really really well advised to document the solution you applied and file this so that if anyone ever comes knocking you can trot out the folder with the information and not be in a position of having to justify your solution after the event which will put you in a seriously bad place. Even if you can come up with a reason why what you did is safe the fact you are doing it after the event will likely see you nailed for poor practice, perhaps with some justification.

If you still want to use an alternative solution based on Robbo's attached WHO guideline the guts of it are on pages 49 and 50. There are a couple of things to note about the graph:
1. The y-axis scale is a logarithmic scale - so each division is ten times the previous.
2. The time shown is that to kill 90% of the legionella. So you would usually double this time to kill 99%, triple it to kill 99.9% etc.

Combining these two means from this graph to kill 99.9 % at 60 takes about 6 minutes while at 50 it takes 300 minutes.

Having bored you all with all of that and no doubt having outstayed my welcome, my gut feeling is that for the vast majority of systems if you set the thermostat to 50 then everything would be just fine in a society with some degree of personal responsibility (provided you don't really need the small amount of extra stored energy). If a proper cost benefit analysis was done I have never thought the government would be able to justify the 60 degree heated water cut to 55 deal. Same goes with cold water expansion valves in areas where the water supply isn't scaling enough to stuff the TPR if it operates on every cycle. It is pretty easy for the valve manufacturers to scare the authorities though with stories and pictures of exploding cylinders and scalded kids....

(At the risk of getting barred - none of this is the PGDB's doing)