PM abnormally high values issue. Apparently solved
26 views
Skip to first unread message
Alberto Villa
Apr 30, 2017, 2:28:21 PM4/30/17
to Safecast Air Quality Discussion
Hi to all,
As previously mentioned, about ten days ago I noticed a huge increase in my PM values (5-6 times greater than before), without any dramatic environmental event that could have justified it. Probably I have solved the issue, and in this case I hope the story will be of some help. I am thinking for instance at Tara's very high values.
The attached images summarize what happened.
The problem started just after I opened the air kit lid to check the display (I wanted to see the number next to "WIFI ok", wondering what parameter it is referred to). The first day I imagined it could have been due to some accumulated dust moved around in the air while playing with the case. So I expected a decay of the readings, back to the previous orders of magnitude.
But ten days later the captured values remained very high, even if the weather was beautiful with clean fresh air and a continuous breeze. So, the 27th I decided to take the unit back indoor and clean it carefully, outside surface and air filters first, then inside. Beginning with a vacuum cleaner, and then with a detergent-wetted cloth for the refinements.
I restarted the device indoors, and kept it there for a while. Great was the satisfaction seeing that the uploaded PM values were again the low ones I registered during my very first test, in March. After 24 hours the air kit was again outdoors, same place as before. And now it seems to go on as expected...
Of course I am not 100% sure that the cause of the issue and its solution were both due to my actions. Can't exclude mere coincidences, of course...:-) For instance: the cleaning and restart activities of the device occurred when the Website was down for maintenance. I wonder if during the maintenance some changes were made in the processing of the collected data.
Hoping to have been helpful, best regards to everybody!
Alberto
As previously mentioned, about ten days ago I noticed a huge increase in my PM values (5-6 times greater than before), without any dramatic environmental event that could have justified it. Probably I have solved the issue, and in this case I hope the story will be of some help. I am thinking for instance at Tara's very high values.
The attached images summarize what happened.
The problem started just after I opened the air kit lid to check the display (I wanted to see the number next to "WIFI ok", wondering what parameter it is referred to). The first day I imagined it could have been due to some accumulated dust moved around in the air while playing with the case. So I expected a decay of the readings, back to the previous orders of magnitude.
But ten days later the captured values remained very high, even if the weather was beautiful with clean fresh air and a continuous breeze. So, the 27th I decided to take the unit back indoor and clean it carefully, outside surface and air filters first, then inside. Beginning with a vacuum cleaner, and then with a detergent-wetted cloth for the refinements.
I restarted the device indoors, and kept it there for a while. Great was the satisfaction seeing that the uploaded PM values were again the low ones I registered during my very first test, in March. After 24 hours the air kit was again outdoors, same place as before. And now it seems to go on as expected...
Of course I am not 100% sure that the cause of the issue and its solution were both due to my actions. Can't exclude mere coincidences, of course...:-) For instance: the cleaning and restart activities of the device occurred when the Website was down for maintenance. I wonder if during the maintenance some changes were made in the processing of the collected data.
Hoping to have been helpful, best regards to everybody!
Alberto
ta...@kithub.cc
May 1, 2017, 11:48:51 PM5/1/17
to Safecast Air Quality Discussion
Alberto this is helpful -- I saw a spike too. I thought it was the increased temperature, Santa Ana winds, being near the largest urban park in the US and right next to the 5 freeway all causing the readings to go up. I will follow your lead and clean it out and see if it changes.
Thank you!
Alberto Villa
May 2, 2017, 11:24:57 AM5/2/17
to Safecast Air Quality Discussion
You're welcome, Tara!
By the way:
1) I can confirm that the 50011 readings should be correct now. In fact the relatively low levels of PM I am measuring in these days after cleaning and restarting the device are not only coherent with the climate, but also with the readings of official monitoring stations in my city that I found yesterday on their website. This increases the confidence in a happily working air kit after the treatment.
2) When drawing away the dust from the air intake, I have kept the vacuum cleaner tube at about 1,5 cm from the sensor's intake hole. A precaution to avoid possible damages to the (unknown) inner parts of the sensor. In any case, the steel mesh of my tea filter cup made it impossible to get closer than that. A good protection also against too aggressive vacuum cleaners!
It could be interesting to ask the manufacturer of the air sensor an opinion about this problem. What happens if the sensor is "overloaded" by dust? Is the vacuum cleaner treatment OK, or potentially traumatic? Is there an alternative procedure?
Good luck with the air kit washing, but remember to use a mild spin cycle...:-)
Best regards,
Alberto
By the way:
1) I can confirm that the 50011 readings should be correct now. In fact the relatively low levels of PM I am measuring in these days after cleaning and restarting the device are not only coherent with the climate, but also with the readings of official monitoring stations in my city that I found yesterday on their website. This increases the confidence in a happily working air kit after the treatment.
2) When drawing away the dust from the air intake, I have kept the vacuum cleaner tube at about 1,5 cm from the sensor's intake hole. A precaution to avoid possible damages to the (unknown) inner parts of the sensor. In any case, the steel mesh of my tea filter cup made it impossible to get closer than that. A good protection also against too aggressive vacuum cleaners!
It could be interesting to ask the manufacturer of the air sensor an opinion about this problem. What happens if the sensor is "overloaded" by dust? Is the vacuum cleaner treatment OK, or potentially traumatic? Is there an alternative procedure?
Good luck with the air kit washing, but remember to use a mild spin cycle...:-)
Best regards,
Alberto
Alberto Villa
May 11, 2017, 1:44:24 PM5/11/17
to Safecast Air Quality Discussion
Hi to all,
As shown in the attached pdf, the readings continue to look fine during the last dozen of days after vacuum-cleaning the air sensor.
Only a single anomaly emerged: a huge peak in PM10 of 617.7 ug/m3 (see first chart).
Looking at the PM1, PM2.5 and PM10 outputs before and after that sampling time (2017-05-09 09:19:45 UTC), it seems like if the sensor miscalculated the PM10 by a factor of ten or so. The second chart shows the parameters after changing this value into 61.77 ug/m3.
The third chart is a zoomed view of a couple of blackout periods. I think a problem of my ADSL provider...
Regards,
Alberto
As shown in the attached pdf, the readings continue to look fine during the last dozen of days after vacuum-cleaning the air sensor.
Only a single anomaly emerged: a huge peak in PM10 of 617.7 ug/m3 (see first chart).
Looking at the PM1, PM2.5 and PM10 outputs before and after that sampling time (2017-05-09 09:19:45 UTC), it seems like if the sensor miscalculated the PM10 by a factor of ten or so. The second chart shows the parameters after changing this value into 61.77 ug/m3.
The third chart is a zoomed view of a couple of blackout periods. I think a problem of my ADSL provider...
Regards,
Alberto
Reply all
Reply to author
Forward
0 new messages