Acoustic Performance of Office Partitions

When we look at office acoustics and acoustic separation between offices & meeting rooms we often see lots of values quoted for acoustic insulation but what do these really mean?

We’re going to look at some terms and concepts such as sound reduction indices, indirect or flanking transmission, background noise and speech privacy to see how these interact and affect the real performance of office partitions on site.

Sound reduction index

Well, one of the most common numbers you’ll see is an Rw value. This stands for weighted sound reduction index. It’s a value for the laboratory measurement of the sound reduction performance of a particular building element. It usually only applies to the building element that has been tested, ie a particular partition or ceiling construction or type. As such its best use is to compare one building element or type of construction with another.

In relation to various partitions types some of the values you’ll see are

Partition construction

Weighted sound reduction index

Single glazed
6mm toughened


6.4mm laminated


12mm toughened


12.8mm laminated


12.8mm acoustic laminate


Double glazed
6 + 6 mm toughened


6mm toughened + 6.4mm laminated


6.4 + 6.4mm laminated


12 + 12mm toughened


12mm toughened + 12.8mm laminated


12mm toughened + 12.8mm acoustic laminate


Solid partitions
100mm wide single layer plasterboard with insulation


100mm wide double layer plasterboard with insulation


100mm wide double layer soundbloc type plasterboard with insulation


Table 1 – Typical weighted sound reduction index values (Rw) for various partition types


Speech Privacy

Makes sense so far, right?  But what do the values actually mean in real terms. The next table shows you how these numbers relate to what you might hear, or not want to hear, between two rooms in an office environment.

Weighted sound reduction index


Speech Privacy Level

25 dB

Normal speech can be overheard

30 dB

Loud speech can be heard clearly

35 dB

Loud speech can be distinguished under normal conditions

40 dB

Loud speech can be heard but not distinguished

45 dB

Loud speech can be heard faintly but not distinguished

>50 dB

Loud speech can only be heard faintly with great difficulty

Table 2 – Speech privacy levels for various weighted sound reduction indices


Right so that’s easy then – you just decide what speech privacy level you require and specify a partition construction that gives you the appropriate weighted sound reduction index.

Hmmm, guess what? It’s not that simple!

There are lots of other factors that really mess it up but two that we really need to consider are:

Indirect or flanking transmission

As we said the Rw value is a laboratory tested value but it should be obvious that we’ve never going to achieve laboratory conditions on site. All the junctions between the partition and the other building elements are going to affect the actual site performance. No matter how carefully we detail the junctions some sound is always going to pass between the partition and the ceiling or base build walls, floor etc.

Similarly grilles and air conditioning ductwork running between office & open plan areas, cable tracks grommets and floor boxes, fire alarm and security cabling will all make a difference to the overall insulation of the room.

We also need to be very careful how we detail the partition itself. Flash gap junctions, recessed skirtings, power sockets and cable ways will affect the performance of the partition in situ.

All this stuff comes under the general heading of flanking transmission but how much does it affect performance?

Well if you’re daft enough to put an air grill in each of two adjacent offices with a duct connecting them without any attenuators, you might as well forget the partition because the sound is just going to travel from one room to the other straight down the duct.

But assuming you don’t do this, and you layout out the services carefully, specify the junctions and detail the partition correctly the fact is still that you’re probably still going to lose around 5dB on your weighted sound reduction index value. This means that from a partition that you’re expecting to get Rw 40dB so that loud speech can be heard but not distinguished you actually get 35 dB loud speech heard under normal conditions.

And, more importantly probably a very unhappy client!

The actual value that we get on site is measured as Dw which stands for weighted sound level difference.

Background Noise

Whereas indirect and flanking transmission work against us when designing for speech privacy believe it or not background noise actually works for us.

Again the laboratory tested Rw values take no account of levels of background noise levels that are present in real office environments. Background noise is present from outside, from open plan areas, and from M & E installations. It’s measured in dBA and for most office environments it’s between 40 & 50dBA. Research has shown that the optimum level of steady background noise is between 45 & 50dBA to provide a good degree of speech masking but without being intrusive in open plan areas but in individual offices office & meeting rooms a background noise level of around 40 – 45 dBA is probably more appropriate.

This gives rise to another criteria for measurement known as the Speech Privacy Potential (SPP) which is a figure that combines the actual sound reduction index expressed as Dw with the background noise level in the receiving room in dBA simply by adding the 2 values together.

For example a well detailed and well installed 100mm wide double layer plasterboard with insulation could produce a Dw 45dB which with a background noise level in the receiving room of 40 dBA gives an SPP of 85. The table below shows us is a high rating and should mean that loud speech is barely audible.

Happy client again!

Speech Privacy Potential

Weighted sound level difference DwBackground noise dBA

Privacy rating

Speech Privacy Level



Loud speech barely audible & unintelligible



Normal speech is barely audible, loud speech mostly unintelligible



Normal speech can be overheard some of the time, loud speech can be heard



Normal speech can be overheard

Table 3 – Speech Privacy Potential



So we’ve seen that you can’t just take lab tested values for sound reduction indices as a way of designing for speech privacy and predicting the performance of office acoustics. You need to design and detail the complete office environment carefully to avoid undoing the benefits of specifying partitions with higher Rw values. But, you can use background noise to help speech masking so that you maybe don’t need to specify as high values of acoustic insulation as you might think.

Also there hasn’t been space in this blog to introduce the roles of such factors as office layout, doors and other openings, reverberation and acoustic absorption in office acoustics but we’ll try and cover some of these in future blogs.

This blog has been intended as a very simplified overview but we hope it’s been helpful and given you some insights. Thanks for your support!

There are a lot more indices, scales, measures and standards to consider and apply if you want. If you want to learn more either generally or for a specific project:

The AIS – Association of Interior Specialists publish a very good acoustic guide and can organize CPD sessions on acoustics |

Or you could contact us either by phone, e-mail or the chat module on our website |

Leave a Reply

Your email address will not be published. Required fields are marked *