PUBHS47 Comparison ventilation methods final author version.pdf (485.39 kB)
Comparison of two tracer gas dilution methods for the determination of clothing ventilation and of vapour resistance
journal contribution
posted on 2009-12-04, 12:47 authored by George HavenithGeorge Havenith, Ping Zhang, Kent Hatcher, Hein A.M. DaanenTwo methods for Clothing microclimate ventilation determination (Crockford (CR), 1972 and
Lotens & Havenith (LH), 1988) are compared for reproducibility, validity, and usability. Both
methods showed good sensitivity and reproducibility (average coefficient of variation 1.5-
2.3% for the method, up to 7% for method and dressing/movement effects combined) and
produce values close to calibration values in forced ventilation tests (r=0.988). Weak points
for CR are limits in time constant of the equipment, causing an upper limit of measurable
ventilation (around 800 l/min), and the measurement of clothing microclimate volume,
showing large errors. Alternatives (whole body scanner or manual circumference
measurements) were shown to produce good results. For LH, the distribution of the tracer
gas over the whole skin surface becomes a problem factor at very high ventilations (above
1000 l/min).
When the measurement is used to determine water vapour resistance, the choice of tracer
gas (O2, Ar, CO2, SF6) affects results.Clothing microclimate ventilation is an important parameter in climatic stress and in
contaminated environments. The two main methods for its determination (Crockford (CR),
1972 and Lotens & Havenith (LH), 1988) are, after further development, compared in terms
of reproducibility, validity, and usability. Both methods are shown to have a good sensitivity
and reproducibility (average Coefficient of Variation 1.5-2.3% for the method, up to 7% for
method and dressing/movement effects combined) and produce values very close to
calibration values in forced ventilation tests (r=0.988). Weak points for CR are the limits in
the time constant of the measurement apparatus, causing an upper limit of ventilation that
can be reliably measured (around 800 l/min), and the for this method required measurement
of clothing microclimate volume. The original ‘vacuum oversuit’ method was cumbersome
and prone to large errors. Alternatives (whole body scanner or manual circumference
measurements) were shown to produce good results. For LH, the distribution of the tracer
gas over the whole skin surface becomes a problem factor at very high ventilations (above
1000 l/min).
As all methods use tracer gasses (O2, Ar, CO2, SF6) with diffusivities smaller than that of
water vapour, this potentially creates a problem in the calculation of vapour resistance from
the ventilation values in the region where the emphasis of vapour transfer moves from
diffusion to convection. In most real life situations, where body and air movement are
present, a correction is not required however as the error remains below 10%. Clothing microclimate ventilation is an important parameter in climatic stress and in
contaminated environments. The two main methods for its determination (Crockford (CR),
1972 and Lotens & Havenith (LH), 1988) are, after further development, compared in terms
of reproducibility, validity, and usability. Both methods are shown to have a good sensitivity
and reproducibility (average Coefficient of Variation 1.5-2.3% for the method, up to 7% for
method and dressing/movement effects combined) and produce values very close to
calibration values in forced ventilation tests (r=0.988). Weak points for CR are the limits in
the time constant of the measurement apparatus, causing an upper limit of ventilation that
can be reliably measured (around 800 l/min), and the for this method required measurement
of clothing microclimate volume. The original ‘vacuum oversuit’ method was cumbersome
and prone to large errors. Alternatives (whole body scanner or manual circumference
measurements) were shown to produce good results. For LH, the distribution of the tracer
gas over the whole skin surface becomes a problem factor at very high ventilations (above
1000 l/min).
As all methods use tracer gasses (O2, Ar, CO2, SF6) with diffusivities smaller than that of
water vapour, this potentially creates a problem in the calculation of vapour resistance from
the ventilation values in the region where the emphasis of vapour transfer moves from
diffusion to convection. In most real life situations, where body and air movement are
present, a correction is not required however as the error remains below 10%. Clothing microclimate ventilation is an important parameter in climatic stress and in
contaminated environments. The two main methods for its determination (Crockford (CR),
1972 and Lotens & Havenith (LH), 1988) are, after further development, compared in terms
of reproducibility, validity, and usability. Both methods are shown to have a good sensitivity
and reproducibility (average Coefficient of Variation 1.5-2.3% for the method, up to 7% for
method and dressing/movement effects combined) and produce values very close to
calibration values in forced ventilation tests (r=0.988). Weak points for CR are the limits in
the time constant of the measurement apparatus, causing an upper limit of ventilation that
can be reliably measured (around 800 l/min), and the for this method required measurement
of clothing microclimate volume. The original ‘vacuum oversuit’ method was cumbersome
and prone to large errors. Alternatives (whole body scanner or manual circumference
measurements) were shown to produce good results. For LH, the distribution of the tracer
gas over the whole skin surface becomes a problem factor at very high ventilations (above
1000 l/min).
As all methods use tracer gasses (O2, Ar, CO2, SF6) with diffusivities smaller than that of
water vapour, this potentially creates a problem in the calculation of vapour resistance from
the ventilation values in the region where the emphasis of vapour transfer moves from
diffusion to convection. In most real life situations, where body and air movement are
present, a correction is not required however as the error remains below 10%. Clothing microclimate ventilation is an important parameter in climatic stress and in
contaminated environments. The two main methods for its determination (Crockford (CR),
1972 and Lotens & Havenith (LH), 1988) are, after further development, compared in terms
of reproducibility, validity, and usability. Both methods are shown to have a good sensitivity
and reproducibility (average Coefficient of Variation 1.5-2.3% for the method, up to 7% for
method and dressing/movement effects combined) and produce values very close to
calibration values in forced ventilation tests (r=0.988). Weak points for CR are the limits in
the time constant of the measurement apparatus, causing an upper limit of ventilation that
can be reliably measured (around 800 l/min), and the for this method required measurement
of clothing microclimate volume. The original ‘vacuum oversuit’ method was cumbersome
and prone to large errors. Alternatives (whole body scanner or manual circumference
measurements) were shown to produce good results. For LH, the distribution of the tracer
gas over the whole skin surface becomes a problem factor at very high ventilations (above
1000 l/min).
As all methods use tracer gasses (O2, Ar, CO2, SF6) with diffusivities smaller than that of
water vapour, this potentially creates a problem in the calculation of vapour resistance from
the ventilation values in the region where the emphasis of vapour transfer moves from
diffusion to convection. In most real life situations, where body and air movement are
present, a correction is not required however as the error remains below 10%.
History
School
- Design
Citation
HAVENITH, G. ... et al, 2010. Comparison of two tracer gas dilution methods for the determination of clothing ventilation and of vapour resistance. Ergonomics, 53 (4), pp.548–558.Publisher
© Taylor & FrancisVersion
- AM (Accepted Manuscript)
Publication date
2010Notes
This article was accepted for publication in the journal, Ergonomics [© Taylor & Francis]. The definitive version is available from: http://dx.doi.org/10.1080/00140130903528152ISSN
0014-0139;1366-5847Language
- en