ROCIS Averager

When participants use low cost particle monitors the biggest impressions are often the incidents or spikes that cause the highest readings... like burning, or just cooking.  Of course, it's important to recognize and reduce the frequency of these spikes. 

However, we see a tenfold difference in the median particle counts between LCMP (Low Cost Monitoring Program) sites, and in most homes, the spikes do not appear to contribute significantly to that difference.  

How can we help folks make sense of the data? In one week alone, the three Dylos particle monitors generate 60,000 data points for each site! The ROCIS team has developed several tools to support LCMP participants. How do outdoor particle counts influence indoor counts? Does it help to use a kitchen exhaust fan or to use a portable air purifier?

ROCIS Data Averager

An integral piece of the Low Cost Monitoring Project is the ROCIS Data Averager. The ROCIS Data Averager, Version 6.3, is one of the visualization tools that we use in order to share particle data with ROCIS Low Cost Monitoring Project participants. The data produced by the Dylos uploads shows the particle monitor outputs every minute. In one week alone, this can amount to 60,000 individual pieces of data. Such detail can be great for scientific investigations but it is far too much information for most users: it is difficult to envision what is happening in the house with that many data points. The ROCIS Averager now provides participants with the recorded data in 15 minute averages, accompanied by three graphs. This summary is sent to current participants but anyone with Dylos outputs can play with it themselves.

The graphs in particular allow insights into what is happening inside the house (or office) as well as the effect of outdoor particle counts on indoor monitors. Quick rises in the indoor counts (also known as “peaks” or “spikes”) are usually caused by indoor sources. This can be activities as diverse as cooking, vacuuming, or having kids jump on the couch. While such peaks can be alarming at first glance, it may be that their duration is short term. The bar graph on the lower left gives the cumulative times that the monitor spent in each range. If the bar graph is predominantly green, or Dylos “Good” to “Excellent”, then the number of particles in the house are quite low (regardless of whether there are a number of peaks in the line graph). When the ROCIS data is viewed as a whole, the houses with the highest particle counts show about 10 times more particles than the houses with the lowest counts.

Here are some screenshot examples. The screenshots are not active. The figure below serves as a typical first screen. This is similar to what you will see when you run the Averager.

There are three graphs on this page. If you want to see tables of data (which is useful for pinpointing particle generation events), click on the tab on the bottom labelled “TimeSeries”.

The bar graphs on the lower left-hand side show the cumulative total particle counts in the monitor locations in your house. They are color-coded to match the Dylos monitor ratings. The graph on the top right provides a history of monitor output over the recording period (0.5+ um). The graph on the bottom right shows the same data for the Dylos large counts (2.5+ um). Usually sharp spikes in the Dylos large graph correspond to some indoor activity that produces particles, such as cooking.

The example graph above is quite typical of what we see. In this case, the bedroom monitor and living room monitors are about 30-40% in the green area of the graph or Dylos “Good” to “Excellent”. The line graphs on the top right show the outside green line is usually higher than blue and yellow inside lines, except for the times when indoor peaks push house air higher in particle counts. Even when the indoor counts (0.5+ um) are significantly below outdoor counts, they usually track outdoor counts except except for indoor incidents.

Here are two other examples. In the screen grab below, you can see that indoor and outdoor counts are very much the same for Dylos small (or total) particles. This usually occurs when windows are open to the outside. The bar graph on the lower left confirms the similarities.


This next graph shows the improved particle counts when a participant runs their furnace fan 24 hours a day through a high efficiency filter. What stands out is that indoor air particles are far below outdoor particle levels, although they do follow the general outdoor particle trends in this example.


PART 2 – How to use the Averager

Attached to this discussion is a working Excel file containing the Averager. It is the same file as the ABC graph in the example above. Make sure that your Excel (or other program) is set to enable macros when you open this file. Note that the Averager macros currently do not run on Macs, though the spreadsheet results are viewable.

For the working file, the data comes from an anonymous source ABC. However, you can use any Averager file to start the process. You are simply overwriting the data on the previous file.

Park your cursor over that green outlined box showing “ABC” and put in your own three initials. The box to the right has the interval you would like the Averager to use. Fifteen minutes is common but you could choose 60 minutes or any other suitable time period. If you have three monitors, the last box on the right is good. If you have 2 or 4, change it as desired. Then hit “Go!”. The Averager will ask for the location of your upload files. I usually put in “Outside” as the first one, because almost everyone has an “Outside” file. Then you point the Averager to the where it can locate this file. Once it loads that file, another screen comes up asking if you have files you wish to add to the one loaded, for instance if you have saved two or three different weeks of data. Respond “Yes” or “No” as appropriate. Once all the data is loaded for “Outside”, it will go through the same process for the other sites where you have monitored.

When you have loaded the 2-5 files that you specified you had, the Averager will churn away. Before it produces the graphs, it often (but not always) gives you a couple of warnings. Just acknowledge them by clicking “OK”. Then it creates the file with the three graphs and the tables for reference.


Copyright © 2022 All Rights Reserved.