Controlling Noise from ground source heat pumps

Acoustical Control Engineers Limited is 50 years old in 2022.  As part of the celebrations the company is preparing a series of articles describing some of the technical challenges encountered in finding practical solutions to noise problems. 

Ground source heat pumps (GSHP) are becoming an increasingly common response to the requirements for sustainable and efficient sources of heat, but they are not without problems, noise being one of these. In this article, Richard Collman and Mike Hewett describe some of the issues with finding solutions to noise problems caused by GSHP installations.

Background

Over the past couple of years, Acoustical Control Engineers and Consultants (ACEC) has been working at numerous high quality multi-occupancy retirement living premises to control the noise within some apartments due to GSHP installations. Most of the affected apartments are directly adjacent to the plant rooms but some are elsewhere in the buildings. This work has involved analysis of the problems; identification of the most appropriate solutions; and then design, manufacture, delivery and installation of acoustic engineering solutions and specification of other work as appropriate.

Being high quality premises the residual sound level within the apartments is generally very low, providing little masking sound, and the residents’ expectations are understandably relatively high.  The elderly residents may also have reduced higher frequency hearing sensitivity (presbycusis) potentially making them more sensitive to the lower frequency plant noise. Sound from the plant exhibits both tonal and amplitude modulation characteristics and can also excite modal responses (standing waves) within rooms. The buildings are relatively acoustically live lightweight concrete construction, with the plant mounted on what are in effect suspended (beam and block) concrete floors.

In order to avoid or minimise any disruption to the residents, it has been necessary to carry out the installation work while the plant has remained operational i.e. without altering pipework, plant location, etc.

The GSHP system distributes hot water directly through the building’s radiators and indirectly provides the building’s domestic hot water via a heat exchanger. 

The system comprises:

• Two GSHPs
• Two ground loop pumps (one per GSHP) to pump the fluid (brine) through the ground loop and GSHP
• Various pumps to circulate the water through the GSHPs, storage cylinders, and the building, heat exchangers, pipework, storage cylinders, valves and other fittings

Much of the equipment and pipework is supported in the plant rooms by a support frame constructed from an MFMA (Metal Framing Manufacturers Association) standard strut channel system (e.g. Unistrut). At the different sites this was found to be connected to plant room walls floors and ceilings with a variety of connection methods.

Prior to ACEC’s involvement the client had previously:

• Fitted inertia bases to GSHPs at some sites;
• Fitted resilient pads between the support frames and building surfaces to which they were attached;
• Installed false ceilings in some plantrooms; and
• Relocated some plant items to a new external plant room building.

However, for various reasons, this appeared to have provided little if any improvement.

Preliminary investigations

To start with, several of the sites were visited to gain an initial understanding of the situation. The visits included some preliminary testing such as:

• Discussion with residents to establish the nature and scale of the problem and the most significant times of day/night;
• Measurement and subjective observation in affected apartments simultaneously with controlled plant operation to evaluate the level and character of problematic sound, identify the relative significance of different items of plant and assess likely sound transmission paths;
• Indicative airborne sound insulation testing between the plantroom and adjacent apartments; and
• Indicative impact energy transmission tests to the pipework, using a mallet, to assess the relative strength of some structure-borne energy transmission paths.

This initial research established the following:

• The predominant acoustic energy transmission path between the plant and apartments was structure-borne, with the building providing adequate airborne sound insulation, probably without the need for the false ceilings that had been fitted in some cases;
• The most significant sources of plant noise within the apartments were the ground loop pumps not the GSHPs themselves;
• The use of inertia bases for the GSHPs was inappropriate. The sole purpose of an ‘inertia’ base (i.e. mass of concrete attached to the source) is to reduce the amplitude of vibration of the attached plant, which, in this case, is not particularly significant. However, the vibration isolators on which the inertia base was mounted were appropriate and a more suitable solution for the GSHPs would be to mount on the vibration isolators using support frames;
• It appeared that significant energy was being transmitted into the structure via pipework connections to the plantroom floor, walls and, in some cases, ceilings as well as directly from the plant supports into the floor;
• The energy in the pipework through the remainder of the buildings was generally relatively insignificant;
• The building construction was relatively live and provided an efficient structural energy transmission path, exacerbated by the beam and block spanning over void ground floor construction, which did not allow the ground underneath to provide damping to vibration transmitted from the plant above;
• There were strong modal responses in some rooms, typically in one or more of the 100 Hz, 250 Hz and 315 Hz one third octave bands;
• On some occasions there were significant interactions between the two GSHP systems adding a beat frequency to the sound;
• Effective vibration isolation of the relevant parts of the GSHP system should be able to provide significant attenuation to the sound within the affected apartments without the need for any additional airborne sound insulation; and
• The strut channel frames used to support the plant and fittings could provide a useful mounting system for appropriate vibration isolation in most cases. However, it was not sufficiently stiff for such an application with the ground loop pumps without significant alteration or replacement with stiffer structural support frames.

In addition to several sites with existing noise problems, three new sites were under construction.  This gave the opportunity to incorporate appropriate mitigation during the design and installation phase of the GSHP systems. However, this could not delay construction, so decisions had to be made based on the information available at that time.

Remedies

At the existing sites, the plant and pipework were decoupled from the building structure.  The resilient pads that had been fitted were unsuitable both because they would be expected to provide little deflection (and associated vibration isolation) under the loads and in the locations where they had been applied and because the frames were bolted directly to the structure through the resilient pads, bridging any isolation they may otherwise provide.  It was determined that the necessary isolation could best be achieved with steel spring isolators.

Given the very limited space, difficulty of supporting the existing strut frame sections, and large number of connection points that required steel spring vibration isolation, it was beneficial to develop some low cost vibration isolators that could relatively easily be installed, provide good vibration isolation (high deflection spring, effective ‘noise’ pad, and height adjustment) without incorporating unnecessary materials or taking up more space than necessary, which was particularly important as many parts of the plant rooms were very congested.

Results

The initial results were mixed; with reductions large enough to satisfy the residents at some sites and significant remaining levels and little improvement at others.  There were several reasons for this:

• It was difficult to identify all structural bridging points due to the often too crowded plant rooms preventing access to view all areas;
• For the same reason it was very difficult to access all of the identified bridging points particularly with the need to keep system running in the occupied buildings; and
• The ground loop pipes exited the plant room through the floor and had been ‘concreted in’ creating a rigid connection to the floor. This rigid connection was generally in a part of the floor close to an external wall providing a short route for connection to the rest of the building.

At several sites it became apparent that the issue of ground loop pipe coupling to the floor was significant and we advised that the concrete around the pipes should be broken out (by others). This work was also undertaken while the systems remained in operation and was successful in achieving further reductions of varying significance at different premises. This also allowed yet further reductions to be achieved by adjustments and refinements to the other isolating elements.

 

Graph 1 shows the time varying levels of overall sound and the three most significant one third octave bands measured in an apartment at one of the premises during various stages of the work.  The first series was measured once the initial vibration isolation work had been completed.  Although this had achieved some improvement, the overall level remained very high, with significant amplitude modulation, particularly in the 500 Hz band. This was when the ground loop pipe concreting was identified as a significant transmission path. The pipes were therefore cleared, following which, the temporal modulation virtually ceased leaving the overall level steadily around 35dBA, due primarily to the 100 Hz band. Further adjustments were made to the vibration isolation systems which reduced the 100 Hz level by around 7dB. The corresponding overall dBA level fell but was then more affected by residual sound so did not show as great a reduction. Finally, one further fixing was found and removed providing a further improvement. It should be noted that the final 100 Hz band shows some temporal amplitude modulation, although the level is sufficiently low that this is not significant.

Other sites

At another site, there remained a problem after similar work had been completed. Initial investigations had already included checking that pipework outside the plantroom was not a significant transmission path. It was then found that a steel column within the plantroom may have been compromising the attenuation system’s performance.  The concrete was broken out from around the column and the level and character of sound in the apartment became demonstrably suitable. However, the resident remained dissatisfied. It was identified that sound from the plant was most noticeable close to one wall, so the wall lining was removed and replaced with one that was decoupled from the structure, providing a further slight improvement for the resident.

At a different site, although the sound level was low, its tonal nature meant it was still slightly intrusive, although the resident was happy with what had been achieved. Further testing showed that the GSHPs were generating standing waves within the plantroom, resulting in relatively high levels of tonal sound that was then breaking through the ceiling/floor slab into the apartment above.  Rather than trying to increase the slab’s sound reduction index, a better solution was to install some strategically placed absorption within the plantroom to reduce its modal response.  This achieved a further improvement as predicted.

Some of the residents are still disturbed by sound which has been reduced to well below standard criteria (BS 8233, NANR45 etc).  This may in part be due to the very low residual sound levels in the flats but also to a degree of hyper-sensitivity and sensitisation in the residents.

As the isolation/attenuation schemes were installed, adjusted and refined the sound reductions achieved were significant and most of the residents felt that the results were satisfactory. However, inevitably at some of the sites, plant noise was still present to some degree. Options for further reductions, within the constraints of the sites, are limited as there are potential transmission paths and issues which cannot be practicably addressed. For example; there is a possibility that there are further transmission routes between the underground ground loop pipework and the building.  The pipes may pass close to, or touch the foundations, or vibration may even pass through the ground.  Access to these pipes is not possible.  There are similar potential problems with working on pipework in inaccessible parts of the building.

There is therefore a practical limit to what further reductions can be achieved

 

Richard A Collman BSc (Jt. Hons), CEng, MIOA, Tech IOSH is Managing Director of Acoustical Control Engineers Ltd and Acoustical Control Consultants Ltd having joined the company in the 1980s and has specialised in the measurement and assessment of sound from industrial and commercial plant for over 35 years. He pioneered the use of digital instrumentation for short duration consecutive logging techniques. As an expert on sound from refrigeration and air conditioning plant he represented the Institute of Refrigeration on BSI committees responsible for various acoustic standards.

Mike Hewett MIOA is Principal Acoustician with Acoustical Control Consultants Ltd having joined the company in 2021 bringing more than 30 years’ experience of Acoustic consultancy. Mike’s particular expertise is in the assessment, prediction and control of noise and vibration from structures, plant and equipment. He is a former examiner for the Noise Control Engineering module of the IOA Diploma and a former Secretary and Chair Noise and Vibration Engineering specialist group.

Controlling Noise from ground source heat pumps

Acoustical Control Engineers Limited is 50 years old in 2022.  As part of the celebrations the company is preparing a series of articles describing some of the technical challenges encountered in finding practical solutions to noise problems. 

Ground source heat pumps (GSHP) are becoming an increasingly common response to the requirements for sustainable and efficient sources of heat, but they are not without problems, noise being one of these. In this article, Richard Collman and Mike Hewett describe some of the issues with finding solutions to noise problems caused by GSHP installations.

Background

Over the past couple of years, Acoustical Control Engineers and Consultants (ACEC) has been working at numerous high quality multi-occupancy retirement living premises to control the noise within some apartments due to GSHP installations. Most of the affected apartments are directly adjacent to the plant rooms but some are elsewhere in the buildings. This work has involved analysis of the problems; identification of the most appropriate solutions; and then design, manufacture, delivery and installation of acoustic engineering solutions and specification of other work as appropriate.

Being high quality premises the residual sound level within the apartments is generally very low, providing little masking sound, and the residents’ expectations are understandably relatively high.  The elderly residents may also have reduced higher frequency hearing sensitivity (presbycusis) potentially making them more sensitive to the lower frequency plant noise. Sound from the plant exhibits both tonal and amplitude modulation characteristics and can also excite modal responses (standing waves) within rooms. The buildings are relatively acoustically live lightweight concrete construction, with the plant mounted on what are in effect suspended (beam and block) concrete floors.

In order to avoid or minimise any disruption to the residents, it has been necessary to carry out the installation work while the plant has remained operational i.e. without altering pipework, plant location, etc.

The GSHP system distributes hot water directly through the building’s radiators and indirectly provides the building’s domestic hot water via a heat exchanger. 

The system comprises:

• Two GSHPs
• Two ground loop pumps (one per GSHP) to pump the fluid (brine) through the ground loop and GSHP
• Various pumps to circulate the water through the GSHPs, storage cylinders, and the building, heat exchangers, pipework, storage cylinders, valves and other fittings

Much of the equipment and pipework is supported in the plant rooms by a support frame constructed from an MFMA (Metal Framing Manufacturers Association) standard strut channel system (e.g. Unistrut). At the different sites this was found to be connected to plant room walls floors and ceilings with a variety of connection methods.

Prior to ACEC’s involvement the client had previously:

• Fitted inertia bases to GSHPs at some sites;
• Fitted resilient pads between the support frames and building surfaces to which they were attached;
• Installed false ceilings in some plantrooms; and
• Relocated some plant items to a new external plant room building.

However, for various reasons, this appeared to have provided little if any improvement.

Preliminary investigations

To start with, several of the sites were visited to gain an initial understanding of the situation. The visits included some preliminary testing such as:

• Discussion with residents to establish the nature and scale of the problem and the most significant times of day/night;
• Measurement and subjective observation in affected apartments simultaneously with controlled plant operation to evaluate the level and character of problematic sound, identify the relative significance of different items of plant and assess likely sound transmission paths;
• Indicative airborne sound insulation testing between the plantroom and adjacent apartments; and
• Indicative impact energy transmission tests to the pipework, using a mallet, to assess the relative strength of some structure-borne energy transmission paths.

This initial research established the following:

• The predominant acoustic energy transmission path between the plant and apartments was structure-borne, with the building providing adequate airborne sound insulation, probably without the need for the false ceilings that had been fitted in some cases;
• The most significant sources of plant noise within the apartments were the ground loop pumps not the GSHPs themselves;
• The use of inertia bases for the GSHPs was inappropriate. The sole purpose of an ‘inertia’ base (i.e. mass of concrete attached to the source) is to reduce the amplitude of vibration of the attached plant, which, in this case, is not particularly significant. However, the vibration isolators on which the inertia base was mounted were appropriate and a more suitable solution for the GSHPs would be to mount on the vibration isolators using support frames;
• It appeared that significant energy was being transmitted into the structure via pipework connections to the plantroom floor, walls and, in some cases, ceilings as well as directly from the plant supports into the floor;
• The energy in the pipework through the remainder of the buildings was generally relatively insignificant;
• The building construction was relatively live and provided an efficient structural energy transmission path, exacerbated by the beam and block spanning over void ground floor construction, which did not allow the ground underneath to provide damping to vibration transmitted from the plant above;
• There were strong modal responses in some rooms, typically in one or more of the 100 Hz, 250 Hz and 315 Hz one third octave bands;
• On some occasions there were significant interactions between the two GSHP systems adding a beat frequency to the sound;
• Effective vibration isolation of the relevant parts of the GSHP system should be able to provide significant attenuation to the sound within the affected apartments without the need for any additional airborne sound insulation; and
• The strut channel frames used to support the plant and fittings could provide a useful mounting system for appropriate vibration isolation in most cases. However, it was not sufficiently stiff for such an application with the ground loop pumps without significant alteration or replacement with stiffer structural support frames.

In addition to several sites with existing noise problems, three new sites were under construction.  This gave the opportunity to incorporate appropriate mitigation during the design and installation phase of the GSHP systems. However, this could not delay construction, so decisions had to be made based on the information available at that time.

Remedies

At the existing sites, the plant and pipework were decoupled from the building structure.  The resilient pads that had been fitted were unsuitable both because they would be expected to provide little deflection (and associated vibration isolation) under the loads and in the locations where they had been applied and because the frames were bolted directly to the structure through the resilient pads, bridging any isolation they may otherwise provide.  It was determined that the necessary isolation could best be achieved with steel spring isolators.

Given the very limited space, difficulty of supporting the existing strut frame sections, and large number of connection points that required steel spring vibration isolation, it was beneficial to develop some low cost vibration isolators that could relatively easily be installed, provide good vibration isolation (high deflection spring, effective ‘noise’ pad, and height adjustment) without incorporating unnecessary materials or taking up more space than necessary, which was particularly important as many parts of the plant rooms were very congested.

Results

The initial results were mixed; with reductions large enough to satisfy the residents at some sites and significant remaining levels and little improvement at others.  There were several reasons for this:

• It was difficult to identify all structural bridging points due to the often too crowded plant rooms preventing access to view all areas;
• For the same reason it was very difficult to access all of the identified bridging points particularly with the need to keep system running in the occupied buildings; and
• The ground loop pipes exited the plant room through the floor and had been ‘concreted in’ creating a rigid connection to the floor. This rigid connection was generally in a part of the floor close to an external wall providing a short route for connection to the rest of the building.

At several sites it became apparent that the issue of ground loop pipe coupling to the floor was significant and we advised that the concrete around the pipes should be broken out (by others). This work was also undertaken while the systems remained in operation and was successful in achieving further reductions of varying significance at different premises. This also allowed yet further reductions to be achieved by adjustments and refinements to the other isolating elements.

 

Graph 1 shows the time varying levels of overall sound and the three most significant one third octave bands measured in an apartment at one of the premises during various stages of the work.  The first series was measured once the initial vibration isolation work had been completed.  Although this had achieved some improvement, the overall level remained very high, with significant amplitude modulation, particularly in the 500 Hz band. This was when the ground loop pipe concreting was identified as a significant transmission path. The pipes were therefore cleared, following which, the temporal modulation virtually ceased leaving the overall level steadily around 35dBA, due primarily to the 100 Hz band. Further adjustments were made to the vibration isolation systems which reduced the 100 Hz level by around 7dB. The corresponding overall dBA level fell but was then more affected by residual sound so did not show as great a reduction. Finally, one further fixing was found and removed providing a further improvement. It should be noted that the final 100 Hz band shows some temporal amplitude modulation, although the level is sufficiently low that this is not significant.

Other sites

At another site, there remained a problem after similar work had been completed. Initial investigations had already included checking that pipework outside the plantroom was not a significant transmission path. It was then found that a steel column within the plantroom may have been compromising the attenuation system’s performance.  The concrete was broken out from around the column and the level and character of sound in the apartment became demonstrably suitable. However, the resident remained dissatisfied. It was identified that sound from the plant was most noticeable close to one wall, so the wall lining was removed and replaced with one that was decoupled from the structure, providing a further slight improvement for the resident.

At a different site, although the sound level was low, its tonal nature meant it was still slightly intrusive, although the resident was happy with what had been achieved. Further testing showed that the GSHPs were generating standing waves within the plantroom, resulting in relatively high levels of tonal sound that was then breaking through the ceiling/floor slab into the apartment above.  Rather than trying to increase the slab’s sound reduction index, a better solution was to install some strategically placed absorption within the plantroom to reduce its modal response.  This achieved a further improvement as predicted.

Some of the residents are still disturbed by sound which has been reduced to well below standard criteria (BS 8233, NANR45 etc).  This may in part be due to the very low residual sound levels in the flats but also to a degree of hyper-sensitivity and sensitisation in the residents.

As the isolation/attenuation schemes were installed, adjusted and refined the sound reductions achieved were significant and most of the residents felt that the results were satisfactory. However, inevitably at some of the sites, plant noise was still present to some degree. Options for further reductions, within the constraints of the sites, are limited as there are potential transmission paths and issues which cannot be practicably addressed. For example; there is a possibility that there are further transmission routes between the underground ground loop pipework and the building.  The pipes may pass close to, or touch the foundations, or vibration may even pass through the ground.  Access to these pipes is not possible.  There are similar potential problems with working on pipework in inaccessible parts of the building.

There is therefore a practical limit to what further reductions can be achieved

 

Richard A Collman BSc (Jt. Hons), CEng, MIOA, Tech IOSH is Managing Director of Acoustical Control Engineers Ltd and Acoustical Control Consultants Ltd having joined the company in the 1980s and has specialised in the measurement and assessment of sound from industrial and commercial plant for over 35 years. He pioneered the use of digital instrumentation for short duration consecutive logging techniques. As an expert on sound from refrigeration and air conditioning plant he represented the Institute of Refrigeration on BSI committees responsible for various acoustic standards.

Mike Hewett MIOA is Principal Acoustician with Acoustical Control Consultants Ltd having joined the company in 2021 bringing more than 30 years’ experience of Acoustic consultancy. Mike’s particular expertise is in the assessment, prediction and control of noise and vibration from structures, plant and equipment. He is a former examiner for the Noise Control Engineering module of the IOA Diploma and a former Secretary and Chair Noise and Vibration Engineering specialist group.