A preliminary air quality analysis of my shop

[cgallery]

I just noticed you posted the form (i must have been snoozing) and want in on a 1/5 micron unit, but I won't have access to a fax till I get back to work on Tuesday. Can we email this in or phone it in?

Michael

In that case E-Mail all the information on the form, except for CC information, to support@dylosproducts.com

Then call them at 877/351-2730 and read them your credit card #.

WHen you're done don't forget to E-Mail groupbuy@cgallery.com so we can make sure you're added to our list that we will share w/ Dylos to make sure nobody falls through the cracks.

[mpauly]

I just noticed you posted the form (i must have been snoozing) and want in on a 1/5 micron unit, but I won't have access to a fax till I get back to work on Tuesday. Can we email this in or phone it in?

Michael

[cgallery]

I would bet that the unit is not very efficient at picking up the smallest bin size (.5 micron). The only way to test that is to compare it side by side with an industrial airborne particle counter certified down to at least 0.5 micron (most go down to at least 0.3 micron).

The developer was silultaneously using his models along with a $4500 six-channel particle counter. His background is in designing cleanroom monitoring equipment (that runs into the tens of thousands of dollars).

In addition, the actual size of the particle that is measured varies greatly depending on shape, angularity, etc. These units are calibrated using perfect spheres which are electrostatically sorted by size for accuracy.

Electrostatically? Wouldn't that pull the particle in from the field towards the edge and make it impossible to count?

I admit that I have absolutely no idea how the technology works.

All that said, what you're really looking for in this case is relative cleanliness, so it should work just fine, with a few caveats.

First, the number of particles it counts will depend greatly on where the unit is placed. I'd put it about the same height and position of where you'll be working, since that's what you're really concerned about.

Second, this unit has a very low sample rate. I have a small hand held unit at work (semiconductor cleanroom) that draws 0.1 CFM, while this unit is 0.01 CFM. A lot of you may have heard the term "Class 1 cleanroom". This refers to the number of particles >0.5 micron in a cubic foot of air. FED STD 209 (now obsolete, but still used) has a lot of statistical work behind it, and to say an area is class 1 with confidence you have to draw a cubic foot of air. For cleanroom certification work, a larger unit that draws 1.0 CFM or more is used to get the job done quicker. For the same confidence level, you would need to run this unit for 100 minutes, assuming it meets other requirements which I won't detail here. For your usage, I wouldn't run the thing for a few minutes and use that number. I'd recommend turning it on for the full 100 minutes. This will help eliminate variability in the test due to randomness. Bottom line, longer test times are better.

The amount of airborne dusts varies greatly in a shop environment. There really is no such thing as eliminating variability. Walk around a little and your numbers can spike by 5x. Sweep the floor and within seconds you'll increase your counts by a factor of at least 10x (often quite a bit more).

Third, the federal standard assumes uni-directional flow (sometimes called laminar flow, although this is technically incorrect), whereas airflow in most workshops is typically random.

All this said, I think this tool can be pretty useful. If one area rates as a "10" for example, and another is "8.5", and both measurements are done in the same manner as stated above, you can say that they are roughly the same cleanliness. If one is 10 and another is 100, you can probably safely say one is significantly cleaner than another.

The best usage would be to determine if measures you take to improve quality of air are effective. For example, before and after dust collection devices are installed. Shop air cleaners can be very cost effective.

Alternatively, copy exactly Rod's shop, I think you won't go wrong.

I have the 1/5 model and will purchase the .5/2.5 as well. I am quite confident that the #'s reported are quite accurate as far as particle counter tolerances allow. He talked a little about the differences and variability in even the multi-thousand dollar units. It seems like he really has a handle on the technology and is delivering a product that will work very well in the shop.

[gugie]

Caveats on usage of this device.

"Based upon those measurements, he says sanding pine produces more >1-micron particles than .5-micron particles."


I would bet that the unit is not very efficient at picking up the smallest bin size (.5 micron). The only way to test that is to compare it side by side with an industrial airborne particle counter certified down to at least 0.5 micron (most go down to at least 0.3 micron). In addition, the actual size of the particle that is measured varies greatly depending on shape, angularity, etc. These units are calibrated using perfect spheres which are electrostatically sorted by size for accuracy.

All that said, what you're really looking for in this case is relative cleanliness, so it should work just fine, with a few caveats.

First, the number of particles it counts will depend greatly on where the unit is placed. I'd put it about the same height and position of where you'll be working, since that's what you're really concerned about.

Second, this unit has a very low sample rate. I have a small hand held unit at work (semiconductor cleanroom) that draws 0.1 CFM, while this unit is 0.01 CFM. A lot of you may have heard the term "Class 1 cleanroom". This refers to the number of particles >0.5 micron in a cubic foot of air. FED STD 209 (now obsolete, but still used) has a lot of statistical work behind it, and to say an area is class 1 with confidence you have to draw a cubic foot of air. For cleanroom certification work, a larger unit that draws 1.0 CFM or more is used to get the job done quicker. For the same confidence level, you would need to run this unit for 100 minutes, assuming it meets other requirements which I won't detail here. For your usage, I wouldn't run the thing for a few minutes and use that number. I'd recommend turning it on for the full 100 minutes. This will help eliminate variability in the test due to randomness. Bottom line, longer test times are better.

Third, the federal standard assumes uni-directional flow (sometimes called laminar flow, although this is technically incorrect), whereas airflow in most workshops is typically random.

All this said, I think this tool can be pretty useful. If one area rates as a "10" for example, and another is "8.5", and both measurements are done in the same manner as stated above, you can say that they are roughly the same cleanliness. If one is 10 and another is 100, you can probably safely say one is significantly cleaner than another.

The best usage would be to determine if measures you take to improve quality of air are effective. For example, before and after dust collection devices are installed. Shop air cleaners can be very cost effective.

Alternatively, copy exactly Rod's shop, I think you won't go wrong.

[cgallery]

We're now up to 38 orders. If we hit fifty by the end of business on Monday we will be getting a 38% instead of 35% discount.

To order, use this form:
http://www.cgallery.com/jpthien/dc1100.pdf

[cgallery]

Here is the order for if you want to get in on the group buy. The forms must be faxed to Dylos by end of business Monday.

http://www.cgallery.com/jpthien/dc1100.pdf

[gearbuilder]

I would be interested in one of these meters.

Jamie

[garymuto]

I might be interested too

[cgallery]

Furthermore, I find it interesting that last night I was sanding my face frame with the meter within 2-feet of me and the 1-micron and 5-micron #'s barely budged (maybe they went up by 1(00) each). The sander was a PC ROS connected to my Ridgid vac and my cyclone lid.

[cgallery]

Just spoke w/ Roger at Dylos. First, he did some pine and MDF testing this last weekend with his PC finish sander and some fine sandpaper. He used both a .5-micron and 1-micron meter, as well as a 6-channel particle counter. He is E-Mailing the results but I will paraphrase our discussion.

With the 1-micron model, sanding pine, he took an initial reading of 480(00), and an after reading of 4700(00). A 10x increase in 1-micron dust.

With the .3-micron model, sanding pine, he took an initial reading of 2000(00), and an after reading of 9630(00). This is approx. a 5x increase.

Based upon those measurements, he says sanding pine produces more >1-micron particles than .5-micron particles.

Next, he did some MDF.

With the 1-micron meter, his before reading was 520(00), and his after was 5450(00). Again a 10x increase.

With the .5-micron, his before was 2170(00), and his after was 10140(00). This again is a 5x increase.

His take was sanding MDF isn't that much different than sanding pine. More "coarse" (his term) than fine particles.

On the larger (5-micron) scales, he started at 85(00) both times and ended-up at 1580(00) (for pine) and 1960(00) for MDF.

To perform this test he setup a fan (w/ a filter) to blow-over the area he was sanding. Down-wind of this he setup the meters.

Due to the nature of his test, he doesn't have any settling data. And quite honestly I understand that he may have some resistance to getting into the business of measuring how quickly .5-micron dust settles compared to 1-micron. He did say that .5-micron dust settles more slowly that 1-micron, and that we could google "settling times" for more data. Until get get .5-micron units that will have to do.

Kim (his wife) who will handle the processing of the orders wasn't in today. I proposed to Roger that we all fax our orders in individually and that they charge the cards individually and ship them for us. He said that sounded fine, but he will defer to Kim (I will put all this in an E-Mail to her).

He also said he would defer to Kim on shipping units to Canada via USPS, but that this sounds reasonable, as well.

He said it would be fine if some of you want serial ports and some don't.

So there it is in a nutshell. I will finish the details up w/ Kim hopefully tomorrow and have a PDF order form posted Tuesday night or Wednesday. You will be able to download/fax the form and the order cutoff will probably be set to sometime next Monday (21st) or Tuesday (22nd).

It will take 5-10 days after they receive the order (depending on configuration like whether you want a serial port) for them to begin shipping the units.

**********

Roger just E-Mailed me his test results, here it is:

Hi Phil,

Here are my raw test results.

Test Conditions-

A fan is blowing filtered air towards the test setup
A Finishing sander is set up 2 feet from the fan
A 6 channel particle counter is set up 6 inches downstream of the sander
220grit sandpaper is used

The following are the concentration of particles per cubic foot
Size Channel Before Pine After MDF
> 0.3 micron 207618 963456 217774 1014391
>0.5 micron 85672 701476 93127 769685
>1.0 micron 48356 477381 52858 545445
>2.0 micron 34459 368715 37769 436662
>5.0 micron 8719 158895 8331 196115
>10.0 micron 1756 73679 2314 79891

The sander was run for 15 seconds with light pressure - the data for Pine and MDF was taken during the sanding.

What I noticed most about these numbers is that the distribution is very much skewed towards larger particles. Also, the size distribution seemed very close between the pine and the MDF.

[Garasaki]

I think your observations are really interesting.

I hope that a number of you guys pick these up, and post what you find while playing with them.

I am a tinkerer myself, but I am a cheapskate tinkerer - I can't justify 100 dollars on a product like this...

[gugie]

Guys, if you're thinking of getting one of these, I'd recommend getting the less sensitive model and save a few bucks. There is a good correlation between large and small particles (# small># large), and since the 1 micron unit picks up plenty of data, you'll get a good idea of how clean the room is. The reason to go to a more sensitive model is that as the room gets cleaner, it takes longer to get the data you need to get decent confidence.

[mpauly]

I'm in for 1, either the 1 or .5 micron one, depending on the results.

Michael

[Carlos]

I'm in for one.

[cgallery]

Interesting deal if we can get a group discount, but I'll bet there's only about 10 of us who are tinkerers and measurers.

Do you have a HEPA filter on the Shop vac? It would be interesting if you let everything settle and then just turned on the shop vac and see what happens... that would tell you if the shop vac is pumping/recirculating dust through its filter.

We're up to 17 commitments between woodnet and sawmillcreek (in six hours) so I think we will be able to hit 25.