Air Quality Sensor based on PMS5003, does it work?

[Andrew More]

So recently in a thread there was some discussion of various air quality testers based on the PMS5003 sensor which are available on eBay.

I recently got mine and did a bit of testing in the shop.

So far I have tested it with a few cuts on various pieces of machinery, and I'm not sure what to make of the results.

First, the unit appears to be operating correctly, reporting the PM2.5 number, temperature, and relative humidity. From what I can tell with other sensors in the shop, and humidity, and temperature appear to be correct. Second, it does appear to be reporting a PM2.5 number which fluctuates depending on what's going on.

My concern with the accuracy comes from the lack of response to most tools in the shop. I have run the table saw, miter saw, and band saw within inches of the sensor, and seldom get it above about 10-15 ug/m^3, even with no dust collection.

I have managed to get it to spike, going as high as 50-300 ug/m^3, but the only by sitting it 2" away from my rigid belt sander, and running a piece of pine against the belt for 10-20 seconds, with dust collection turned off. Even with that, the numbers quickly start to go down into the 100s ug/m^3, and then 15-30 ug/m^3 after the operation ends.

Is this pretty normal behavior, or is the sensor under reporting the amount of dust in the air?

[David L Morse]

Which monitor do you have? Does it also report the PM10 value? That would be more sensitive to the dust produce by a belt sander. Do you also have the raw particle counts available?

Smoke is a better test of the PM2.5 channel. Light a candle or small piece of paper and then put it out. You should see all channels increase as the smoke particles diffuse.

All small particles diffuse fairly rapidly with even minor air currents. The values you get near your sander drop rapidly after stopping because of this diffusion, not because they are falling out of the air.

[David L Morse]

I just ran a similar test. My sensor was approximately 1ft from a disc sander. My readings before sanding:
427625
After 30s of sanding:
427626
For comparison, here's my reading inside my house while my neighbor was burning leaves in the fall:
427628427627

[Andrew More]

Which monitor do you have?
This one: http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItemVersion&item=143093699783&view=all&tid=1732652097004



Does it also report the PM10 value?
Yes, and PM1.0

Your point about diffusion makes sense, and the air quality readings don't necessarily decrease. Which is to say that they seem to return to a number slightly higher than before the sanding started. So if it was in the range of 4-6, after sanding it might go back to 7-8 after a few minutes.

However, it doesn't seem to be possible to really put the meter into a "bad" reading for any period of time, where bad is indicated by the red and yellow colors on the built-in scale. So I'm wondering if the air quality sensor is really that good, despite producing what should be enough saw dust to cause problems, or if I need to produce more saw dust to cause problems, and what's a bad reading.

[David L Morse]

I was trying to demonstrate that 30s of sanding doesn't really produce enough particulate matter to be significant. The other example I showed was a leaf fire 100 yards downwind from me. My windows were closed. The readings were almost 10x the readings I just got from the sander. I've gotten much higher readings than those when my wife fries bacon without the exhaust fan at the other end of the house.

The EPA air quality numbers certainly don't agree with some of the more extreme requirements you read about on the internet. Spend some time reading the Australian forum. Some of them want less than 5microgram/cubic meter PM2.5.

OSHA requirements are, iirc, in the range of 1000 to 3000 microgram/cubic meter.

So, yes, your sensor is probably correct.

[Larry Frank]

I have one and it seems to work fine. I am using it to get an idea of the air quality with different operations and which ones I need to do better with dust collection. I have attached a picture of the one I have.

427633

[Andrew More]

I was trying to demonstrate that 30s of sanding doesn't really produce enough particulate matter to be significant.

The EPA air quality numbers certainly don't agree with some of the more extreme requirements you read about on the internet. Spend some time reading the Australian forum.


Okay, so part of this is me coming from having read some of Bill Pentz's musings on the subject, where he makes comments like "walking around my shop was sufficient to stir up enough dust to fail air quality standards", and "My doctor also reminded me to not only wear my apron, but also when making fine dust to wear a cap, light jump suit, and bandanna that I take off before going into my home. This keeps the fine dust from being tracked all over." and "Fine airborne dusts are so unhealthy the EPA indoor air limit for fine dust is only 0.1 milligrams per cubic meter. Most small shops are in typical two-car garages with no ventilation and less than 100 cubic meters of air. This means a typical small shop exceeds the EPA air quality limits as soon as 10 milligrams of fine dust gets launched airborne. This is a tiny amount of fine dust. We easily launch more than 10 milligrams of fine dust when we slap a dusty shop apron or hand saw just over seven inches of a typical 3/4" thick board."

From what I can tell this does not appear to be the case.


I have one and it seems to work fine.

Yeah, I think you were the one who recommended it. I don't want to seem ungrateful for the suggestion, it's a great deal compared to a Dylos, yet I find the readings I'm getting don't seem to jive with the claims made by Mr. Pentz. I certainly did not clear the shop before trying these tests. Instead there's a good amount of sawdust from a variety of tools all over the place that the dust collector was unable to get for one reason or another.

[Larry Frank]

I will get some readings today. But when I walk into the shop the readings are very low. As I move around they creep up some and go up a bunch when I sand or cut wood on the saw. I see a big decrease when I turn my Jet Air Filter on.

I suspect that my sampling techniques are more of an issue than the actual readings. I wish I could output the info to a device and be able to graph over time but do not know how or want to spend the time.

A lot depends on what you are trying to do. I just want to get an idea of what I need to work on and the biggest problems.

BobL has a lot of interesting info on the Australian forum.