Recent Posts

Pages: [1] 2 3 ... 10
1
Support / Are all LAeq values obtained by converted LA10 values?
« Last post by roger on July 30, 2021, 02:46:24 PM »
No, the noise index that is calculated depends on the calculation method that you use.
NoiseMap has three different types of noise source, namely Roads, Railways and Open Site sources.  Each type of source is calculated by its own calculation method as follows.

For ordinary roads, NoiseMap uses 'Calculation of Road Traffic Noise' (CRTN), which calculates levels in terms of the index LA10, so if you need LAeq values when using CRTN, you do have to convert the index. NoiseMap has a built-in conversion method that you can use.

For noise from Open Sites, NoiseMap uses the BS5228/ISO9613 methodologies, which calculate levels in terms of the LAeq index.  These methodologies consider 'Haul Routes', which are like roads, but you have to set up noise source details for the vehicles and machines using the Route; and fixed sources which are stationary at a given point.

For Railway Noise, the sources are Rail Tracks and the calculations are made either in terms of the LAeq index or the LAmax index.

A noise model can contain any or all of these sources.  However, when you run a calculation, you first choose the calculation method, which then only evaluates the sources relevant to the chosen method.   If you have additional types of sources, you have to run additional calculations using the relevant calculation method.

When you have calculated for the required noise sources, you can use the Results Processor to compare or combine the various source levels.
2
Hints and Tips / My model has got extremely large. How can I reclaim unused space?
« Last post by roger on July 19, 2021, 07:01:24 PM »
I made a lot of intermediate calculations which took up a huge amount of disk space.  Since I did not need them, I deleted these calculations.  I ran the database optimisation but this did not recover any space. How can I reduce the model to a sensible size before I take the project forward?

There is a neat possible solution but you can’t export a complete scenario tree.

You can export a single scenario using file>export archive. Select NoiseMap archive (.nma)
  • This will export the plant, activities, geography, receivers (but not the results) for a single scenario, but this might be enough.
  • Then start a new model and ‘import archive’.  This will recreate the scenario but without the dross. 
  • Don’t forget to commit changes, preferably to a new scenario.
  • I tried this with your test model and it drops to about one-tenth of the size.
  • Of course, I haven’t checked whether there is important information missing, other than the scenario tree.
But at first glance it seems fine.  And it’s very quick to do.
3
At the moment I have historical road models with assumptions on topography and earthworks etc., but as detailed design progresses I am needing to refresh the scheme contours and it would be neater if I could have a consolidated road and rail model.

a.   There are various ways that you can combine separate road and rail models.  One way is to start with the most complex model first (probably the rail model).Having worked out the main differences between the road and rail models (I would expect the receivers and topography to be similar in both), go to the road model and  export the items that you need to add to the rail model.  For example, the road segments should be exported as shapefiles, as should any additional barriers and topography.  However, the traffic flows will have to be exported as a CSV file.  Do this first.
b.   Now export the road segments. Go to View>Export Shapefile and untick everything except road segments.
c.   Open a copy of the old model to use as the combined model.
d.   Now go to the new combined model and go to View>import Shapefile.  Under Profile, click the dropdown and select NoiseMap exported roads.  Click OK
e.   At this stage, I would answer No to ‘do you want to load the database covering the DXF/Shapefile Area.
f.   Now go to View>Convert Layer and select object to convert into as Roads.  Click Next
g.   You  could try setting the import tolerances to 0, although sometimes this causes an issue so set to 0.1
h.   Turn the road visibility to On.
i.   Check that the traffic flows have been loaded (in the Traffic Flow Manager) If happy, commit changes.
j.   Add the other objects likewise – only export from the roads model the items that change in the combined model.
k.   All parapets go down through the ground.  If you don’t want them to screen, you can delete them in the roads scenario.
4
I am concerned about this because the road segments cross the rail tracks, so I need them to be in a single model.
a.   You can include all the various types of noise source (road, rail, construction, etc)  in a single model but you have to calculate each source separately.  Any road segments are ignored in a rail calculation, and any rail track segments are ignored in a road calculation. All the other topographical elements (buildings, ground topography, barriers, etc) are common to all calculations.  A particular issue for road and rail is that they use different calculation indexes, but a standardised conversion lets you get from LA10 to LAeq.
b.   Once you have done the calculations, you can use the Results Processor to add the contributions together either for individual receivers or for contours.
c.      The Results Processor Guide explains how to combine the results from separate calculations.
5
Support / What is Local height with reference to Railway Track Segments?
« Last post by roger on May 28, 2021, 07:01:27 PM »
Track Segment Properties contain a parameter called 'Local Height'.
It is primarily intended as a 'helper' parameter to enable the height of a track segment to be calculated from the NoiseMap ground model contours, but it can also have another use, which is to define when a section of track is above ground level, on a bridge or viaduct.

The ground level profile is defined only by the ground contours.  Whether the track is on level ground, on an embankment or in a cutting, there should be ground contours (in fact ground profiles, as they need not be at a constant level) along the edges of the 'permanent way'.  However, there is likely to be a small distance between the edge of the closest ground profile and the edge of the adjacent track segments.  If 'Local Height' is set to zero, NoiseMap assumes that the ground slopes uniformly from the ground profile to the track segment, as these may not be at exactly the same height.

However, suppose the track segments are on an elevated structure?  In this case, the ground will pass at a lower level, below the track: it will not slope up to meet it.  In this situation, you would set the Local Height at a suitable value.  For example, if the structure is 10 m above the local ground level, set the 'Local Height' parameter to 10.  NoiseMap will then assume that the ground is 10 m below the track and that in essence there is open space beneath it.


1.   To allow NoiseMap to calculate the track height if you don’t have an accurate height profile of the track –  if you click on ‘select method to calculate height', NoiseMap will find the heightof the ground from the ground model and use the local height and to calculate the height of the track segment.
2.   Local Height also defines how the ground rises or falls to the edge of the track.
a.    Thus if 'Local Height' is zero, NoiseMap assumes that the ground rises or falls from the last ground contour until it meets the track with the track at ground level. This avoids having to have a contour that exactly follows the edge of the track.
b.   If Local Height is positive, then the track is assumed to be on a structure with a gap underneath it.
c.   If it is negative, then the track is in a cutting and the ground drops from the last contour until it meets the edge of the track.  The track is not underground.


If you set the height above local ground to a huge value and leave it at that value when the calculation is run,  NoiseMap does not change either the height of the ground or the  track.  It uses the track level already defined in the Object Position coordinates, and also assumes that the ground profile runs smoothly between the contours on each side of the track. In other words, it does not check that the height above local ground is actually correct at the stage the calculation is run.

When you do click on ‘calculate height’ and NoiseMap resets the height of the start and end of the track, it doesn’t reset the height above ground parameter.  If you are using the local height, we would advise that it is best to set this to zero where the tracks are at local ground level.

6
Support / What tolerance should I use when converting railway tracks?
« Last post by roger on September 13, 2020, 12:33:27 PM »
Q. When I set the tolerance to zero, NoiseMap crashes (sometimes it just hangs forever).
A. Your input data has already divided the railway tracks into 10 m segments.  Experience shows that NoiseMap will keep to the original 10 m tracks if you set the tolerance to a low value, say 0.1m.  I suspect that if you set the tolerance to zero, then due to the precision with which decimal values can be stored in in binary values in the computer, there will be some values that can't be represented with zero tolerance. As a result, the curve-fitting algorithm will try forever to get an exact fit, which is unachievable.  Hence the program appears to hang. 
The answer is to set tolerances to a small value, say 0.1 m, and NoiseMap will stick to the original segmentation.
7
Support / Creating PDFs
« Last post by admin on August 04, 2020, 02:40:22 PM »
I've seen your advice to use a 'pdf printer' for making portable high-resolution maps.  I have tried a 'freebie' pdf printer but NoiseMap will only create maps at A4 size.  What should I do?

NoiseMap creates maps at an exact scale on any printer.  To do this it needs to know the exact size of the 'paper' on which it is printing, even if it is virtual paper.  Because any printer might be connected to the computer, Windows requires the printer driver to supply the hardware characteristics of the printer back to the user software (in this case NoiseMap).  Unfortunately some poorly-written printer drivers don't do this properly, and it is even more complex with 'virtual printers' where paper size and other characteristics are simulated in software. Whilst good PDF software will allow you to change things like paper size, if they don't supply this change back to NoiseMap, there is no way that NoiseMap knows what size of paper has been chosen and it will probably default back to A4 or US letter size.

The solution is to use one of the printer drivers that we recommend, such as Adobe Acrobat or Win2PDF, or to try out various others for yourself.  Do be careful with 'freebies'.  These may carry malware or 'Adware' which could turn out to be very expensive in the long run.
8
Announcements / NoiseMap 5.2.10 released 29 October 2018
« Last post by roger on April 03, 2019, 03:03:11 PM »
NoiseMap 5.2.10 was released in October 2018 and contains a number of major improvements.

ISO9613-2 Barrier Calculation
The headline feature in this release is a new ISO9613-2 barrier calculation option.  This option is part of the SiteNoise Calculation module, which is based on BS5228 with enhancements.

This adds to the existing barrier calculation options in the SiteNoise module, so now you can choose from:
• ISO9613-2 barrier calculation
• BS5228 simple barrier calculations,
• BS5228 octave band spectrum barrier calculations,
• CRTN barrier calculations.

The ISO calculation requires octave band noise source data, as the attenuation is evaluated for each octave. The ISO procedure has a ‘meteorological correction’ which has the effect of reducing the path difference used to calculate the barrier attenuation.  This correction is used to take into account the scattering of sound into the barrier’s shadow zone, caused by atmospheric turbulence under downwind propagation. This means that the barrier attenuation will be less than calculated by some other procedures.

Many new scripting options have been added – allowing you to import and export Categories and Combinations, and to export full analysis of calculation details for receiver points.

[The intermediate versions between 5.2.6 and 5.2.10 were not public releases.]
9
Hints and Tips / Modelling low traffic flows
« Last post by roger on April 03, 2019, 02:49:06 PM »
Calculation of Road Traffic Noise (CRTN) states that is unreliable in situations where flows are less than 50 vehicles per hour.  How should traffic flows on a waste disposal site be modelled when they are this low?

CRTN is based on the LA10 value, i.e. the noise level  exceeded for 10 % of the time. When flows are very low, vehicle noise may not be present at significant levels for 10 % of the time, in which case the LA10 is more a function of the ambient level than of the traffic noise.  This is probably the case at 50 veh/h which is less than one vehicle per minute. 

Here, you would be advised to use the ‘haul road’ calculation of BS5228, given in the ‘SiteNoise’ module of NoiseMap, which provides the LAeq of the source.  This calculates the contribution of sound energy from each moving point source, so is applicable to any flow rate and speed, as long as you have the sound power level or LAmax at 10 metres for the applicable vehicle speed. 

This will give an appropriate calculation of the LAeq values.  As previously noted, LA10 is probably not the most appropriate index to use in these situations. 
10
Support / Sloping or overhanging barriers in Rail Noise Calculations
« Last post by roger on December 03, 2018, 12:32:19 PM »
CRN and TNPM don’t provide for sloping or overhanging barriers.

NoiseMap did implement an overhanging capability for use in Hong Kong, where the overhanging barriers were also absorbent.

We are not aware of much in the way of research on this matter (other than measurements we took in Hong Kong), although admittedly we have not searched for recent work.  If users are aware of any guidance, then we will be glad to consider it.

For these reasons, the implementation in NoiseMap is very simple:  we assume that an overhanging barrier behaves as a simple barrier with the diffracting edge at the lip of the overhang.  We have assumed absorptive properties because we assume that there is no reverberant field within the enclosure space.  This is because the behaviour of sound within the space between the train and the semi-enclosure is not accurately defined. It will depend on many factors like separation distances.

Is the overhanging barrier tool is the best one to model the sloping barrier situation?
We think that if the diffracting edge is above the top of the train envelope, this would be the best way to do it. If the barrier slopes down toward the train, then reflections off the barrier will be directed downwards, so the assumption of absorbency would be reasonable.

If the sloping barrier is still at a distance from the nearside track, you could model it as a vertical barrier closer to the track and then you can choose whether or not to use absorbency.

This is an area that would benefit from further research.
Pages: [1] 2 3 ... 10