DWR Performance

[Stephen Seeber]

Richard:

My testing follows the guidance provided by  ASTM D737.  The air velocity measurement is accomplished using a series of Dwyer Variable Orifice Flow Meters. With these, I can measures flows from .43 CFM/ft2 to 700 CFM/ft2.  I rely on the flow meter valves to adjust test pressure.  Test pressure across the test item is measured with both  a Fieldpiece SDMN5 two channel digital manometer and a  Dwyer magnehelic gauge.  The test fixture follows the basic design published on BPL by LanceM.  However, my duct is 5 duct diameters in length to reduce air flow turbulence.  I measure the static pressure using pitot tubes installed at three locations in the duct, which allows me to assess the flow turbulence.    Initial calibration was conducted by producing calibration plates.  These plates are a series of disks, each with a unique hole size drilled at the center of the plate.  The plates were sent off to Frazier and Frazier sent back flow measurements for each from which I can produce a calibration curve.   I have also sent actual fabric or garment samples to both Frazier and Touch Stone labs for testing so I can compare my test results with those from the labs.

Of course, looking at the table I posted above, you can see that for test results that you posted for which I had samples sent to the labs, the test lab results differ, generally dramatically, from your results.  The lab results are pretty close to mine.  I am sure you would agree that periodic calibration tests of test instrumentation is an important requirement for any test lab.  We know that over time, test instruments  can deviate from their original specifications for a variety of reasons.  As a result, professional test labs will provide calibration certificates to their customers and some customers will insist that proof of calibration be provided along with their test results.    In view of the wide variation I have documented between your results, my results and test lab results, I respectfully request that you describe how you have verified the accuracy of your instrumentation since their purchase.  If you have a dependable calibration procedure, yet we continue to be faced with disparate test results, I suggest we work together to explain these discrepancies so that members of the community can have some confidence in the data that we both publish.

Regarding the Alpine Start, as I mentioned in my post above, I am awaiting receipt of Alpine Start jackets manufactured in several different years.  I will test these to see if fabric performance has deviated over time.  If necessary, I will send the garment that I tested to Touch Stone Labs and get their results.  I will then post the results.

 

[Richard Nisley]

Jimmy,

I was referencing a specific 2014 knit windshirt. What happened with this garment was stretched, to any degree, is as you stated. I don’t believe this specific windshirt’s performance analysis is applicable to any rain jacket design that I am aware of.

[California Packrafting]

Indeed, I must say, we have a right to complain…we pay for it!

[Tuukka U]

I find Stephen’s measurement of 13.2 CFM for the Alpine Start strange. The Massdrop windshirt is specified at 12 CFM and EE spec the Membrane 10 taffeta at 10 CFM, and these two have a very similar “Darth Vader feel” to me. My 2018 Alpine Start, however, Vaders nowhere near the impermeability of these fabrics. I have to breath through 5-7 layers of the Alpine Start to have a similar resistance.

This is, of course, only one rough data point, but the magnitude of difference is big enough that there must be something wrong here.

 

[Stephen Seeber]

I have mentioned twice previously in this thread that I am waiting to receive a collection of Alpine Starts from different vintages.  I will retest and post the results. This may happen next week.

[Stephen Seeber]

I have received three Alpine Starts. According to the individual who supplied the jackets, the 2015 was well used, the 2016  was purchased on Ebay but appears little used and the 2017 jacket was worn several times.  I tested the air permeability of each.  The 2015 averaged 10.3 CMF/Ft2, the 2016 averaged 10.2 CMF/Ft2, the 2017 averaged 7.9 CFM/Ft2.  In order to check my instrument calibration, I measured the flow using a 1/8″ orifice plate.  The flow through this plate was measured by Frazier Instruments to produce a flow of 10.8 CFM/Ft2.  The flow I measured through this plate on my instrument at the conclusion of the test of the three jackets was 10 CFM/Ft2.  Close enough for my purposes.   These values are considerably below the values attributed to the manufacturer in 2014.  I have been provided with the name of the person at Schoeller USA who provided the data.  Where that data came from, I have no idea.  Having checked four of these garments spanning 2015 to 2018, I think the data supplied by the Schoeller representative is inaccurate.  For fans of the Vader test, I compared with a Houdini Air, which I have measured to be 14.3 CFM/Ft2.  As best I can tell, they are pretty similar.

[Stephen Seeber]

A little addendum:  In 2014, Richard Nisely measured 65.9 for the Alpine Start.  It is certainly possible that Schoeller changed the fabric after 2014.  However, my test results show fairly consistent results for this product between 2015 and 2018.  I have asked Richard Nisely in this thread to document his calibration procedure in view of  substantial discrepancies with accredited testing lab results I have paid to obtain versus his postings, as listed in the table  included in this thread.  He has chosen not to disclose his calibration procedures.

[Richard Nisley]

Stephen,

It is now 2020; you are are continuing to whine about a garment that myself and MANY others tested in 2014.

My test occurred in 2014 using a 2014 commercial machine calibrated by the manufacturer in 2014.

The following are the identification tags on the jacket I tested that roughly concurred with all the similar high air permeability assessments including the manufacturer’s (your assessment being an outlier either do to equipment, reasoning or both).

Maybe Black Diamond can help you with your reasoning if you provide them this information plus the same information for the other year’s versions.

 

Ideally forum members will independently contact Black Diamond and post the fabric characteristics of this particular garment.

[Stephen Seeber]

This subject was brought up by Tuukka U based on the numbers that you published in 2014.  If people are relying on your test results to set their expectations, they should expect that your results reflect reality to some degree.  The examples I have cited in the table above are so divergent from paid lab test results that they cannot be relied upon to reflect reality.   You appear to be unable to accept that perhaps your instrumentation is not as accurate as you have claimed.  You also appear to be unwilling to subject your test instruments to periodic calibration to prove their accuracy.  Here are some words of wisdom that you will recognize from Isaac Asimov: “I believe in evidence. I believe in observation, measurement, and reasoning, confirmed by independent observers.”   I am asking you  to provide confirmation by independent observers, just like I have provided for my test procedure.

[Richard Nisley]

There are lots of different air permeability testers that have been created in different countries. Those that have been proven accurate by independent standard groups have been certified and listed in the standards. Most of those standard’s provide conversion algorithms from each tester type to the others. Stephen doesn’t have a certified machine listed in any standard. He decided to build a modified copy of a Frazier machine and have one of his associates help him calibrate it. Paying for calibration is a mandatory requirement for a DIY design only. He wants to use his custom calibration markers as a smokescreen to claim that only he can determine air permeability and no other machine is viable without similar calibrations points.

That is BS; each standard based air permeability tester is uniquely calibrated by the manufacture during production. Most commonly the user can then measure a range of his or her relevant material types and then archive the material. If there a significant change in a reading, the units typically have to be be sent back to the manufacture for repair. Standards based units cost big dollars to purchase and big dollars to repair but they rarely go out of calibration.

Stephen claims that his test value CFM for the BD Alpine Start from other years are the only ones with validity. The 2014 tests done by many individuals and organizations doesn’t mix well with his CoolAid flavor (sour grapes). After drinking his CoolAid, he then chants over and over and over “only my unique machine calibration procedure is valid and without that there is no truth.”

Nothing is served by him which doesn’t go well with his CoolAid. The freshest side dish he could have offered with it is called the BD Distance. In 2/14 I tested and posted the CFM and HH values to the forum at 3.6 CFM and 246mm HH. At that time there was no information for either value searchable by Google.

Much later he tested this same garment. He was apparently unaware that I had tested and posted the results about a year earlier. His test values were 1.9CFM and 251mm HH. His results totally contradicted the chant that goes along with his cool aid. The chant is that only he can determine truth.

Using Stephen’s range of possibility CFM values that can be measured (699.57), my test value yielded the air permeability position at a point at .005 of that range and his test value at .003. of that range. The Federal government lists inter lab variances of 5 % as being equivalent.

Using my range of possible HH values before potential fabric damage (3500mm), my test value yielded the HH position .070. Stephen’s point was at .072.

It is obvious to me after 6 years of posting test results to BPL, that it routine procedure for venders to use the same trademark name and change fabric characteristics. I posted detailed multi year change analysis for both the Squamish and Houdini. In the case of the Squamish, in addition to fabric weave changes, they also changed the coverage ratio of an inside PU coating they used for multiple years.

In my opinion Stephen frequently provides something of value combined with a dose of BS. I know enough to to glean the value part and point out the BS part. That results in a lot of time responding to his attacks. Hence, this missive.

[Roger Caffin]

each standard based air permeability tester is uniquely calibrated by the manufacture during production.
Uniquely: just so. Hydrostatic Head testers and Air Permiability testers are not the same as a set of calibrated masses and a beam balance (for example). Keeping one of these testers calibrated to an international standard is both hard work and expensive. As noted elsewhere, a thermal insulation test system for air mats or for quilts is even more tricky, as evidenced by the very high cost of testing.

routine procedure for venders to use the same trademark name and change fabric characteristics.
Looking behind the curtain a bit, what this is about is the reality of the commercial world.

You buy a roll of fabric from a mill somewhere in Asia. Will next year’s production be identical to this one? Almost certainly not.

You then get the roll of fabric coated by a processing plant somewhere in Asia. Next year you get a roll of slightly different fabric coated. Will the two lots of coating be identical? Almost certainly not.

So, will this year’s fabric be identical to last years? Almost certainly not, doubled. How do you explain this to the Public? You can’t, so you don’t. Most of them won’t notice anyhow.

Cheers

[Jerry Adams]

and it varies from different places on the same roll

still, the testing is useful.  approx results are good

[Roger Caffin]

and it varies from different places on the same roll
True, all true.

approx results are good
When you find one fabric which tests somewhere around 3,000 mm HH and another one which tests around 1,000 mm, yep, it helps.

Cheers

[Stephen Seeber]

Hi Roger:

Can you look at this: http://www.frazierinstrument.com/products/fap/fap-faq.html#replacementcalibrationchart
Air Permeability Test Plates & Test Plate Calibration:
Because of the increasingly stringent needs of our customers for a means of checking  the calibration of their Frazier Differential Pressure Air Permeability instruments as demanded by laboratory calibration authorities, Frazier Company has agreed to produce a series of test plates for checking the calibration of our instruments.
The calibration of the Frazier Differential Pressure Air Permeability Machines is intrinsic in the Calibrated Venturi Nozzles that it uses to measure air flow.  Once each of the nozzles is calibrated, that calibration is permanent and is not affected by use of the instrument.  Only actual physical abuse of the nozzles themselves such as dirt, scratches, nicks, corrosion, etc. will affect their calibration.
The instrument itself can be caused to malfunction through abuse or age.  However, these malfunctions are easily reversible.  Some conditions that can cause the instrument to malfunction include oil in the system, air leaks in the system, aged or improper oil, improper leveling of the instrument, obstructions in the system and malfunction of the electrical components.  All of these problems are easily repairable by the owner.
The standard test plate provided with each instrument should be used, on a regular basis, to check the system to ascertain if the instrument is operating properly.  It cannot be used to adjust the instrument or affect the calibration which is intrinsic to the nozzles.  The test plates function by simulating a relatively unchanging test sample.  The nozzles themselves can only be calibrated or recalibrated by Frazier Company at our production facility.
As a general guide, under good condition, a reading obtained when a test plate is used should be within plus/minus 3% of the test plate calibration.  However, a variation in reading within plus/minus 2% is preferred.  A variation in readings of up to about plus/minus 1.5% due to changes in atmospheric density is normal in addition to the above.  A reading discrepancy of more than plus/minus 5% is usually unacceptable.  Variations of plus/minus 5% can occur when the nozzle is not calibrated in the instrument in which it is to be used.  This tendency to larger error is especially true when a nozzle is used in a non-standard instrument or a test plate is used to check non Frazier Instruments in which case larger variations can be expected.
The standard test plate can also be used to check the calibration of the 4 mm nozzle to ascertain if it has been subject to abuse or has become dirty.
Because the standard test plate only can be used to check the calibration of the 4 mm nozzles, and because of the increasingly more demanding needs of our customers, Frazier Instrument Company has developed and is selling a special series of test plates that can check the calibration of the other eight nozzles that are standard with our instruments.  These test plates also cannot be used to adjust the instrument or calibrate it or calibrate the nozzles.
We recommend that Test Plates be recalibrated yearly by Frazier Company due to possible changes through handling over time.
Yearly calibration is also a partial requirement for ISO certification.
I know nothing about Richard’s machine and if it has a calibration plate.  The calibration plate is the route I have gone for mine.  I have four flow meters, a digital manometer and a magnehelic gauge.  I can also have a leak in the tubing through which all the air flows. Any of these can fail. With a calibration plate, I can know that everything is working if the number comes out right when using the plate.  If not, then I have to isolate any issues.  I would expect that similar issues can occur with Richard’s machine.  I do not know if he purchased a new or used machine, but in any case, a new or used instrument must be calibrated.

I agree with your point about fabric variation.  I see it in  the readings I get from individual garments at different locations.  I have an Arc Squamish 2019 and a 2017.  The fabrics are totally different.  I measure void space, the gaps between yarns at weave intersections.  The 2019 has 2.9%.  The 2017 has 6.9% and these differences show up as huge  changes in both permeability and MVTR.

However, I don’t think this can explain the differences in coffee filter performance, where Richard claims 70 CFM/Ft2 and the lab test results were 26.9.  There is some variance in filters (I have tested many), but not of this kind of magnitude.  I am also not sure this would explain the variance for Richard’s Alpine Start vs manufacturer’s data and my measurements.  In this thread, I relate the results for product years 2015-2018 and they are pretty consistent.

I appreciate your insights on this issue.

[Roger Caffin]

For Stephen and Richard

The problem with some of these ‘Standards’ is that usually there are many of them. When I was working on moisture breathability or vapour transmission, I found that different fabric vendors each used a different Standard, each selecting the one which gave them the highest position in the hierarchy. Understandable, but not always really helpful.

What is the position regarding air permeability? Are there multiple possible Standards? I would imagine so. I welcome a discussion of the differences between them, and of how different test equipment works. I am sure a lot of BPL readers would be interested. I will be.

I am sure the discussion can be conducted in a polite and respectful manner – like at a science conference. OK, maybe slightly better than at some science conferences?

Cheers

[Stephen Seeber]

I would be happy to work with Richard on this.   Perhaps on its own thread.

[Roger Caffin]

This could be good.

Cheers

[Ken Larson]

Stephen I second Roger’s remarks.

my2cents worth!

[Richard Nisley]

When I asked what international standards have certified his DIY machine accuracy, none was provided. Instead Stephen said, “My testing follows the guidance provided by ASTM D737.”

OK, lets use ASTM D737 to illustrate how to solve the relative accuracy question. Stephen ignored almost identical test results on the same 2019 windshirt (plain continuous filament yarns). The ASTM D737 standard states that two laboratories testing this type of material would expect up to a 4CFM difference to be about the best that should be expected.

Stephen then focused on a knit material from different manufacturing years to claim that only his CFM reading should be viewed as accurate. Even if the same custom knit material from the SP 2014 was tested by two labs, it could easily vary ~50CFM or more according to ASTM D737.

The 2014 difference between the Schoeller CFM measurement and the Black Diamond CFM measurement can also be explained by the ASTM D737 guidance.

[Roger Caffin]

Richard and Stephen

Clearly there is a difference of opinion here, and it would be to everyone’s benefit to get some public resolution. There is a way to do this. During my career in CSIRO Div of Textile Physics, we used to run what were called ‘Round Trials’ for certain fabric materials. This meant that samples of a number of different materials were sent to the participants (mostly industrial fabric processors, plus some international test labs) for testing, and the results were returned to our Division, collated, analysed and the results published.

I suggest we get a range of fabrics and have each of you test them, with multiple tests on each fabric. (Testing commercial garments at the start might be a bit too expensive I think.) Then you send the results to an independent third party for analysis and publication. I am willing to stick my hand up for the analysis and publication; I don’t have any air permeability equipment you see.

We would have to find a range of different fabrics of course, but I am sure that could be done. The many MYOG enthusiasts here at BPL might be able to provide samples from their own cupboards. We would need to work out a minimum sample size to fit in the test equipment of course. Posting A4-sized fabric samples as letters would be cheap.

What do you think?

Cheers
Roger Caffin

[Author]

Posts