Lighting design input in the DSM calculation (ApacheSim)

[Blackeagle]

Question : It has come to my attention that it occurs to be some discrepancies between actual design light efficacy value and software calculated values when lux method is in the UK DSM calculation. The actual design on paper appears to be more efficient than output from the model , can anyone explain me how to avoid this ?

Issues with DSM method , the following part explains my point of view and aim to raise some averseness concerning this issue
Reported power consumption(in the BRUKL ) due to lighting energy does not seem to be matching actual figures provided by the engineers . I understand that DSM tool is based on rules and guidance laid down in the NCM Modelling guide which states user should enter power density per 100 lux for the purpose of UK compliance calculation ( this was later revised by BRE and inference method also allowed for the same purpose of calculation ) . this value itself gives an indication of how efficient the lighting system is ( against notional building standard ) however degree of efficiency will be different than anticipated design values (calculated based on luminaries lighting data using more comprehensive lighting design software such as Relux etc..) an example would be designer comes up with 2.8 w/m2 100 lux power density with equivalent of 70Lumenaire per circuit watt light efficacy for the room however when 2.8 w/m2/100 lux info is entered in the model, calculated efficacy by software appears to be as 58 Lumen/watt which appears to be under the actual design value , so the BRUKL will report misleading lighting energy consumption from the actual building . I understand that DSM is not a design tool , so it is quite natural to get different wattage value in the end however when it comes to the point that you will need to advise the engineer that lighting will still need to be improved , he may argue on the basis that his lighting calculation showed light efficacy in excess of 55 lumen /watt ( national building standard )
this difference is mainly to do with Lux level used in the zones that is determined by the software based on activity type assigned which cannot be edited by user. The light consumption is the product of light power density entered in the model and the lux level assigned by the software , the final resulting light power will be different than actual design since it is based on different lux standards . the similar impact can also be seen in the light efficacy values in the actual building.
To avoid this problem and to be more in line with actual design performance , In my opinion , DSM should allow lux level ,power density per 100 lux and lumen per circuit watt information to be entered in the same area (3 set of information), if the national building is allowed to mimic the actual building in certain parameters of the design (same geometry, same operating system etc..) why not using the same lux level information in the national building ,? the following will demonstrate what I actual mean by example
Say actual design calculation is based on following parameters
300 lux at 2.5 W/m2/100 lux with equivalent 75 lumen /watt light efficacy,
Same building in the national building calculation should have ;
The same lux level (300 lux) but light efficacy at 55 lumen per circuit watt , software then calculates corresponding power density in accordance with actual building light efficacy curve , which is in this example would be= 3.4 w/m2/100 lux ( (75x2.5/55 = 3.4 w/m2/100 lux ) so it becomes irrespective of actual room geometrical information ( no need to calculate R value in the national building calculation) ,
with this method, the same power consumption from actual design will be reported in the BURKL document and will avoid any confusion between different disciplines ( due to discrepancies between actual design light efficacy value and the software reported figures)
Note:I am aware that you can put actual design figures in the form of w/m2 or watt for the zone , however if you meant to choose this route, resulting light efficacy may well be higher than the national building . I have tried this method before and used W/m2 information derived from the reasonably well efficient design , although I snapped the actual light consumption ( as expected from the design) , the building still showed high energy consumption against national building . because the national building lux level is generally lower than actual design figures.

Conclusion
As some of you may be aware or not , SBEM tool allows full design information (including actual design wattage and lux level ) to be entered in the software whereas DSM tool currently allows only power density per 100 lux or inference method , ( should the inference method be used ; reported wattage value would not match actual design values, should the PD per 100 lux information used ; reported light efficacy would not match actual design light efficacy ) to be consistent with SBEM tool, I therefore believe ,LUX method should be used for the purpose of DSM calculation in the way I described above (3 set of info ) otherwise I cannot see any benefit of using current lux method . it is better to use inference method instead

If you have any comments or would like add anything on this , please let me know as I am planning to write email to BRE to change or amend the way this calculation is carried out
Regards

[Complex Potential]

The inference method is not an equivalency, it's a substitute for the W/m2/100lux data if you don't have that level of detail. When inputting an assumption about lamp lm/W efficacy, the NCM then has to make further assumptions about how efficiently those lamps are distributed throughout the space, which is often an idealised scenario that is rarely matched in reality.

At early stages, I input a 60lm/W efficacy (for example) and then output the room by room W/m2/100lux figures and hand it to the electric guys as a target. I never tell them the lm/W figure that generated those targets because it is incidental and just causes confusion..

The EPC conventions document states that if you know the design W/m2/100lux figure then this is what you should use and only move to lamp lm/W figures if this information is not available (but you should be able to justify why).

I do agree that the inference method generally gives a more advantageous result and I'm sure many assessors default to using it instead of true calculated design densities, but they shouldn't.

All of this is due to change in the 2013 version of the regulations anyway, due for release in April. You may want to have a read of the draft NCM 2013 document if you haven't already. It's available from the BRE.

CP

[JohnLloyd]

I'm not sure it's due to change in the 2013 regs, I thought they were just implementing another method we *could* use based around the LENI calculation method?

[Complex Potential]

I think the 2013 NCM Guide talks about inputting lighting figures in terms of W/m2 instead of W/m2/100lux which is quite a significant shift.



Of course the inference stuff is still in there but there is now a measurable penalty to needlessly overlighting a room which is probably a good thing.

[Blackeagle]

Thank you Complex Potential for your response above

Yes I do agree with you , where full design calculation is available , inference method should not be used as stated in the SBEM and EPC convention guide . Nevertheless this does not answer my concern regarding the use of lux method (W/m2/100 LUX) in the calculation.

I understand that you consult the engineers by giving design target power density per 100 lux values at early stage of the projects , but when you get involved in a project ,where design was already firmed up , there is little room for amendments ( say at the end of design stage; you have been asked to produce BRUKL or draft EPC ) , you have to use calculated lighting information by a designer who knows how efficient the lighting system is based on his own version of calculation . my point here is there will be high likelihood of difference in efficacy level between the output from the model and design light efficacy values due to issue that I raised in my email below . I don’t think lux method is appropriate to use in the calculations. In order to make proper like a like comparison ; national and actual building should use the same lux level. Otherwise resulting power density and efficacy of the system may indicate reasonable level of variances which may cause an issue when it comes to make reasonable recommendations to responsible people.

I have quickly looked through the draft NCM 2013 modelling guide which, comes into force on 6 April 2014 in support of new edition of ADL2A, if I am not mistaken there seem to be major difference in the way you define general lighting at zone level, this time you set W/m2 to achieve required Lux Level , which seem to be kind of addressing issues regarding current Lux approach , it is not the same as you used to input PD per 100 lux at zone level and also they have increased national building light efficacy from 55 lumen /watt to 60 lumen/watt so it will be even more difficult to beat notional building standards

[CCaceres]

Hi!

I have been struggling with this situation for a while now. I have a project that needs to demonstrate compliance. On the early stages where no lighting details were ready I used a figure of 60lm/cw for lighting and the building passed Part L. This target efficacy (60lm/cw) was given to the design engineers as part of the specifications.

A year after, lighting designs are ready and I can update the model with the power density figures (w/m2/100 lux). From the lighting specifications these figures achieve an efficacy in excess of 60lm/w. (64 to 70 lm/c). Nevertheless, when I input the power density figures in the model, efficacies calculated in the model appear to be much lower due to the activity NCM lux levels set per room. Now, the building fails to pass Part L.

Looking at the latest NCM modelling guide (January 2013) it appears it is allowed to use both methods Power density (w/m2/100 lux) and inference (lm/cw).
http://www.ncm.bre.co.uk/filelibrary/NC ... ry2013.pdf

I don’t really understand why is it better to use, or we should use, (w/m2/100 lux) if the model multiplies this figure by the lux level set on each room which is a non-editable parameter determined by the NCM templates. I agree with Blackeagle on the three sets of information (lux levels, PD and Efficacy) to normalize both buildings (actual and notional) and run a like for like comparison.

Can someone clarify?
Thanks.

[JohnLloyd]

If you remove the "maintenance factor" from the W/m2/100 lux calculation you'll get a result much closer to that you got when using the lm/W figure.

It doesn't say you should do that in the NCM Guide, but it doesn't say you shouldn't either.

[Complex Potential]

Hi CC

Something you have to understand is that the lm/cW method is simply a measure of the energy efficiency of the individual light fitting based on the assumption of an almost perfectly lighting layout. The W/m2/100lux is the more accurate measure because it also takes into account the efficiency of the layout of those fittings on a room by room basis.

In other words, specifying efficient lights at early stage is only half the battle, you then need an electrical engineer to incorporate them into an efficient lighting layout before you see the benefit. If the difference between the two methods is as much as you say then it sounds as though your electrical guys need to smarten up their act.

[VVladinovsky]

Hi,

Recently I obtained a lighting calculation from another consultancy on Part L compliance. It contains weighted average results in lumens per circuit watts per luminaire - the total luminaire lumens per circuit is divided by the total circuit-watts.
I asked my accreditation body if this would be accepted as an evidence for the EPC audit. They replied: 'The luminaire lumens per circuit-watt option is an acceptable method.'

I am entering the data as lamp lumens per circuit x LOR = 1 as the Luminaire lumens per circuit watt field is greyed out.

I'm not sure if I'm following the initial problem CC mentioned with W/m²/100 lux but in the past I used enter lighting values in Watts. My understanding is that Watts, W/m² and W/m²/100 lux values are interchangeable and these values are comparable to NB lighting level as long as the ratio of AB lux level against NB lux level is used:
From iSBEM Technical Manual 2010 (p.73)
Where a full lighting design has been carried out:
1.Use the actual lighting circuit wattage, multiply it by the illuminace level for this zone from the Activity database, divide it by the illuminance level for which the lighing design was carried out, and divide it by the zone area.

VV

[Complex Potential]

That accreditation body wouldn't happen to start with an "S" and end with an "a" would it? I can only assume that full lighting design data is not available to you for a valid reason because the EPC Convention document is quite clear about using the W/m2/100lux method if you do have access to it.

But if you've been given the all clear to do it then go ahead, just make sure you enter the average design lux levels into your spaces wherever they are above the NCM database assumptions if you are working under the 2013 regs.