Ci a ion: Bˇe ’ák, A.; Zach, J.; Misák,
P.; Vanˇe ek, J. Compa ison o Wood
Mois u e Me e s Ope a ing on
Di e en P inciples o Measu emen .
Buildings 2023,13, 531. h ps://
doi.o g/10.3390/buildings13020531
Academic Edi o s: Paulo Cachim and
Jo ge Manuel B anco
Recei ed: 25 Oc obe 2022
Re ised: 31 Janua y 2023
Accep ed: 11 Feb ua y 2023
Published: 15 Feb ua y 2023
Copy igh : © 2023 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license (h ps://
c ea i ecommons.o g/licenses/by/
4.0/).
buildings
A icle
Compa ison o Wood Mois u e Me e s Ope a ing on Di e en
P inciples o Measu emen
Adam Bˇe ’ák * , Jiˇ íZach , Pe Misák and Jan Vanˇe ek
Facul y o Ci il Enginee ing, B no Uni e si y o Technology, Ve e i 331/95, 602 00 B no, Czech Republic
*Co espondence: [email p o ec ed].cz
Abs ac :
This expe imen compa ed comme cially a ailable mois u e me e s ( h ee capaci i e
me e sand one esis ance me e ) and es ed hei p edic i e abili y a di e en mois u e condi ions on
selec ed beech (Fagus syl a ica) and sp uce (Picea abies (L.) Ka s .) wood samples. The measu emen s
we e ca ied ou on he samples a speci ied mois u e in e als anging om 5% o 30% mois u e
con en (MC). The esis ance me e showed a close co ela ion o g a ime ic MC alues; he in luence
o he measu ing di ec ion o MC below 17% was ound when highe MCs in he ans e se di ec ion
o bo h species we e a chie ed. The di e ence was 4.6 imes highe o so wood and 1.6 imes highe
o ha dwood. Di e ences be ween adial and angen ial ans e se measu ing we e no obse ed.
The close co ela ion coe icien o MC measu emen s was also ound o capaci i e me hods. The
e ec o he di ec ion was ound o all he es ed me e s when highe MC alues in he longi udinal
measu emen s we e ound. This e ec was especially signi ican a an MC o wood highe han
20 w .%. Fo wo capaci i e me hods, he e ec o annual ing de lec ion only in he sp uce samples
was ound whe e highe MC alues in he angen ial di ec ion we e obse ed.
Keywo ds:
wood; g a ime ic me hod; mois u e con en (MC); capaci i e; esis ance; mois u e me e ;
sp uce; beech
1. In oduc ion
In mos buildings, he mois u e con en (MC) o he imbe in he s uc u e may
inc ease a some poin , which may be caused by wa e leaching in o he wood s uc u e, a
echnical aul in he echnical equipmen o he building, o unexpec ed na u al phenomena.
F om a echnological poin o iew, i is necessa y o unde s and he anspo a ion and
sou ces o mois u e ac ing on he s uc u e [
1
]. An inc ease in MC o he hyg oscopic limi
leads o changes in he mechanical physical p ope ies o he wood associa ed wi h swelling
and sh inking and inc eases he isk o wood being a ec ed by mould and wood-bo ing
ungi (MC abo e 20%) [
2
]. Usually, he MC o he imbe inco po a ed in he s uc u e is
be ween 8% and 18%; MC exceeding his le el can only appea in he e en o a ailu e
o he building’s echnical equipmen o ailu e o comply wi h design p inciples [
3
]. The
equilib ium MC o wood depends on he exposu e condi ions o he imbe , and he ela i e
humidi y a ec s i s le el h oughou he yea in cycles which main ain he mois u e balance
o he wood. Changes in he en i onmen and MC o he wood lead o olume changes in
he wood up o he sa u a ion limi o he cell walls [4].
In mos cases, he design o a imbe s uc u e does no include conside a ion o all
c i ical de ails o embedded imbe elemen s. The quali y and p ecision o he con ac o ’s
wo kmanship play a e y impo an ole in e ms o he du abili y o a s uc u e wi h
imbe elemen s. In iew o hese ac s, he du abili y o a imbe s uc u e could be
ensu ed by long- e m moni o ing o he embedded imbe in he s uc u e using mois u e
and empe a u e senso s on selec ed exposed imbe elemen s [5].
Buildings 2023,13, 531. h ps://doi.o g/10.3390/buildings13020531 h ps://www.mdpi.com/jou nal/buildings
Buildings 2023,13, 531 2 o 14
1.1. The De e mina ion o Wood MC
The MC o wood can be de e mined by des uc i e o non-des uc i e me hods, which
a e u he di ided in o di ec and indi ec measu emen me hods. The di ec measu emen
me hod de e mines he ac ual wa e con en o he wood. The indi ec me hod de e mines
he wa e con en o he wood using ano he a iable whose alue depends on he wa e
con en o he wood [
5
,
6
]. The basic di ec me hod is he g a ime ic me hod ( he o en
d ying me hod) which is conside ed o be he mos accu a e me hod o de e mining he
MC o ma e ials acco ding o s anda dized p ocesses (see EN ISO 12570) [
7
]. In comme cial
p ac ice, he woodwo king indus y demands he immedia e de e mina ion o he MC
o he ma e ial wi h he maximum possible accu acy [
8
]. Indi ec me hods based on
elec ical o dielec ic p ope ies ha e been de eloped ha allow mois u e esul s o be
ob ained e y quickly wi h su icien accu acy. Indi ec me hods a e used o in si u
mois u e measu emen s. The mos common measu emen me hods (in he case o wood-
based ma e ials) a ailable a e he capaci ance, elec ical esis ance, mic owa e ield [
9
],
spec ome ic, and adiome ic me hods [5].
1.2. The Resis ance Me hod o Wood MC De e mina ion
The esis ance me hod uses he elec ical esis ance/conduc i i y o he wood [
10
].
S amm (1927) s a es ha esis ance dec eases wi h inc easing MC because wa e has a much
highe conduc i i y han wood [
11
]. The esis ance is measu ed by applying a ol age
be ween wo inden ed pins (poles o elec odes) o an elec ic ci cui ha a e in con ac
wi h he wood and measu ing he cu en ha lows be ween hem [
8
]. The megohmme e
scale ypically indica es he MC o he ma e ial di ec ly based on an empi ical calib a ion
o esis ance e sus MC o a pa icula elec ode con igu a ion in a gi en wood species
a a speci ic empe a u e [
11
]. Resis ance me e s measu e he highes mois u e ac oss he
exposed ends o he pins. Spacing be ween pins is no c i ical as he esis ance is p ima ily
he su ace esis ance be ween he con ac s and he wood and is almos independen o he
leng h o he pa h he cu en akes h ough he wood [8].
Accu acy: This me hod p o ides accep able accu acy o measu ing wood MC om
6% o he body o he g ain sa u a ion. In his ange, he e is an app oxima ely linea
ela ionship be ween he loga i hm o he MC and he loga i hm o he esis ance. Below
6% wood MC, he elec ical esis ance eaches alues oo high o measu e easily [
5
]; abo e
he ib e sa u a ion poin o wood, he accu acy o his me hod dec eases. Ve mass (1975)
lis s se e al di e en ac o s, which can be di ided in o wo ca ego ies, ha a ec he
di ec cu en esis ance da a o wood. The i s is he p ope ies o he wood i sel , e.g.,
mois u e and i s dis ibu ion in hec oss-sec ion, g ain o ien a ion, empe a u e ( esis ance
o wood dec eases wi h inc easing empe a u e), bulk densi y, and chemical cons i uen s
including majo cons i uen s, ex ac i es, and ash con en . The second g oup o a iable
ac o s is e e ed o as expe imen al a iables. These a iables include elec ode ypes and
con igu a ions, con ac p essu e, sample shape and size, elec oly ic e ec s a he elec odes,
and he magni ude and du a ion o he applied ol age [12].
1.3. The Capaci i e Me hod o Wood MC De e mina ion
Capaci i e mois u e me e s use changes in he dielec ic p ope ies o he wood o
de e mine he MC o he wood as a unc ion o he change in capaci ance (o dielec ic
cons an ) [
13
]. Ma e ials, including wood, gene ally ha e a dielec ic cons an
E
(F/m) in
he ange o 2 o 8, which emains (almos ) unchanged, while he wa e con ained in he
ma e ial has a high dielec ic cons an , i.e.,
E
= 80. The amoun o wa e c ea es a a iable
cons an [
14
]. The dielec ic pe mi i i y o d y wood is a ound 2 o i s anhyd ous s a e [
15
].
The ield ha o ms be ween he capaci o pla es and he ma e ial ha ac s as a dielec ic [
16
]
is he measu ed ield. The dielec ic cons an inc eases wi h inc easing wood MC, and
his e ec becomes g ea e as he equency o he elec ic ield dec eases. The mois u e
me e does no pene a e he s uc u e o he sample, and he measu emen de e mines
he a e age MC on he su ace o he wood (<35 mm) wi hou de e mining he mois u e
Buildings 2023,13, 531 3 o 14
g adien [
5
]. Dielec ic me e s measu e he MC o he zone o he wood ha is pene a ed
by he elec ic ield and gene ally indica e he mean MC o his zone. Gene ally, dielec ic
mois u e measu emen is non-in asi e and gauges low MC below he measu emen limi
o esis ance hyg ome e s wi h as e and mo e con enien measu emen o la ge quan i ies
(p oduc ion lines) [
8
]. Fo he measu emen , pe manen , i m con ac wi h he su ace o
he measu ed ma e ial, which is ee o de ec s ( o , chips, esins, e c.), mus be ensu ed [
17
].
Ha ley and Ma chan (1995) s a e ha dielec ic me e s a e gene ally no as accu a e as
esis ance mois u e me e s [8].
Accu acy: An accep able accu acy o wood MC is p o ided om 2% up o he ib e
sa u a ion poin [
18
]. Fo sén and Ta ainen (2000) claim ha he capaci y ange o powe -
loss- ype mois u e me e s is 0% o 25% MC [
17
]. The e ec o mois u e on he dielec ic
p ope ies o wood can be di ided in o a ange below an MC o 5% ( he monomolecula
mois u e o wood) when he dielec ic cons an inc eases slowly and abo e an MC o 5%
when he dielec ic cons an inc eases apidly as a unc ion o MC [
19
]. The bulk densi y
o wood, ma e ial empe a u e, g ain o ien a ion, and al e na ing cu en equency a ec
he dielec ic p ope ies. Un o una ely, he eadings depend on he bulk densi y o he
specimen [
17
]. The dielec ic cons an inc eases wi h inc easing wood bulk densi y and
inc easing wood empe a u e, excep a a e y high MC whe e he e ec becomes e a ic [
8
].
1.4. Objec i es o he Pape
The aim o his s udy was o compa e comme cially a ailable mois u e me e s (based
on capaci i e and esis ance p inciples) in e ms o he mois u e ange o hei possible
usage, and assess hei measu emen accu acy. Addi ionally, he possible dependency o he
measu emen s on he wood pa ame e s o g ain di ec ion (longi udinal and ans e se) and
bulk densi y (so wood and ha dwood) was de e mined. While he mois u e measu emen
o wood used in s uc u es is mos ly accessible ans e se o g ain o ien a ion, his s udy
assessed he possible in luence o he annual ings’ de lec ions (below 45
◦
as he adial
plane o abo e 45
◦
as he angen ial plane). This could be e y impo an knowledge
because he imbe s (bo h adially and angen ially sawn) do show he a iable o he
annual ings’ de lec ion h oughou he c oss-sec ion. Finally, s a is ical analyses we e
pe o med o all he mois u e me e s.
2. Ma e ials and Me hods
2.1. Ma e ials
Two ypes o wood species we e chosen, beech wood (Fagus syl a ica) and sp uce
wood (Picea abies (L.) Ka s .). Ele en samples o each wood species we e p epa ed om a
wood p ism; he samples we e cu a a dis ance o 300 mm om he p ism edge o ul il
he s anda d equi emen . Samples wi hou isible wood de ec s (kno s and c acks) we e
selec ed o es ing. The samples we e cu o he equi ed dimensions o 80
×
50
×
100 mm.
The absolu e mean d y bulk densi y o he samples was 425 kg/m
3
o sp uce wood and
650 kg/m
3
o beech wood. To conside he di e ence in he ings’ de lec ions, he samples
we e spli in o wo g oups: (i) wi h a de lec ion in he ange o up o 45
◦
as adial plane ‘R’
and (ii) wi h a de lec ion in he ange abo e o 45
◦
as angen ial plane ‘T’. Figu e 1shows a
wood sample o he mois u e measu emen .
2.2. Me hods
2.2.1. Mois ening o Wood Tes Samples
The samples we e placed in a d ying o en a 103
±
2
◦
C o de e mine he d y weigh
o he samples. A e eaching a cons an weigh , he samples we e cooled in a desicca o o
oom empe a u e, and each sample was weighed o he nea es 0.01 g (d y weigh ). Based
on he de e mined d y weigh , all he samples we e hen mois ened by wa e sp aying
o he equi ed ini ial 5% MC. To ensu e a uni o m mois u e dis ibu ion h oughou he
c oss-sec ion, he samples we e placed sepa a ely in polye hylene ilm o 72 h. A highe
MCs, he ime equi ed o mois u e s abilisa ion o equilib ium MC was equi alen ly
Buildings 2023,13, 531 4 o 14
inc eased. The mois u e inc emen in e al o he samples was se a 3 w .% up o a alue
o 30 w .% o ensu e he wood MC eached he ib e sa u a ion le el. The e o e, he MC
alues o wood samples o 5%, 8%, 11%, 14%, 17%, 20%, 23%, 26% and 30% we e chosen
o mois u e measu emen by all he es ed me e s.
Buildings 2023, 13, x FOR PEER REVIEW 4 o 15
Figu e 1. Schema o wood samples o mois u e measu emen , i.e., a sample wi h a ing de lec ion
o 50° a a ans e se c oss-sec ion as a angen ially o ien ed sample
2.2. Me hods
2.2.1. Mois ening o Wood Tes Samples
The samples we e placed in a d ying o en a 103 ± 2 °C o de e mine he d y weigh
o he samples. A e eaching a cons an weigh , he samples we e cooled in a desicca o
o oom empe a u e, and each sample was weighed o he nea es 0.01 g (d y weigh ).
Based on he de e mined d y weigh , all he samples we e hen mois ened by wa e sp ay-
ing o he equi ed ini ial 5% MC. To ensu e a uni o m mois u e dis ibu ion h oughou
he c oss-sec ion, he samples we e placed sepa a ely in polye hylene ilm o 72 h. A
highe MCs, he ime equi ed o mois u e s abilisa ion o equilib ium MC was equi a-
len ly inc eased. The mois u e inc emen in e al o he samples was se a 3 w .% up o a
alue o 30 w .% o ensu e he wood MC eached he ib e sa u a ion le el. The e o e, he
MC alues o wood samples o 5%, 8%, 11%, 14%, 17%, 20%, 23%, 26% and 30% we e
chosen o mois u e measu emen by all he es ed me e s.
2.2.2. Mois u e Measu emen s
A e e i ying he achie ed MC o samples using he g a ime ic me hod, measu e-
men s using he selec ed mois u e me e s ope a ing wi h di e en measu emen me hods
(capaci ance and esis ance) we e ca ied ou . The mois u e measu emen s we e ca ied
ou a a empe a u e o 20 ± 2 °C and a ela i e humidi y o 45 ± 5%. The achie ed alues
o MC o e e y me hod we e ounded o he nea es 0.1 w .%.
Wood, as an aniso opic ma e ial, has di e en p ope ies depending on he cou se
o he ib es. To conside he wood aniso opy, measu emen s we e made o samples in
bo h he longi udinal di ec ion (i.e., me e s a ached on bo h he ans e se su aces) and
ans e se di ec ion (i.e., me e s a ached on bo h he longi udinal su aces) a 20 °C. The
mois u e was measu ed six imes o he longi udinal di ec ion and nine imes o he
ans e se di ec ion o each amoun o MC o he wood samples. The measu emen p o-
cess ul illed he equi emen s o s anda ds EN 13183-2 (2002) [20] o esis ance me hods
and EN 13183-3 (2002) [21] o capaci ance me hods. All he MC measu emen s o he es
samples we e pe o med in he middle o he wid h o each es ed su ace.
2.2.3. Capaci i e Mois u e Me e
Th ee capaci i e mois u e me e s, GMI 15 (G eisinge GmbH, Regens au , Ge many),
GMK 100 (G eisinge GmbH, Regens au , Ge many), and Tes o 616 (Tes o, Lenzki ch,
Ge many) we e chosen as ep esen a i es o he capaci i e me hod.
The GMI 15 is decla ed by he manu ac u e o be an ins umen o indica i e mois-
u e measu emen . The ins umen is a speci ied dimensionless uni ( aw da a) ha is
Figu e 1.
Schema o wood samples o mois u e measu emen , i.e., a sample wi h a ing de lec ion o
50◦a a ans e se c oss-sec ion as a angen ially o ien ed sample.
2.2.2. Mois u e Measu emen s
A e e i ying he achie ed MC o samples using he g a ime ic me hod, measu e-
men s using he selec ed mois u e me e s ope a ing wi h di e en measu emen me hods
(capaci ance and esis ance) we e ca ied ou . The mois u e measu emen s we e ca ied
ou a a empe a u e o 20
±
2
◦
C and a ela i e humidi y o 45
±
5%. The achie ed alues
o MC o e e y me hod we e ounded o he nea es 0.1 w .%.
Wood, as an aniso opic ma e ial, has di e en p ope ies depending on he cou se
o he ib es. To conside he wood aniso opy, measu emen s we e made o samples
in bo h he longi udinal di ec ion (i.e., me e s a ached on bo h he ans e se su aces)
and ans e se di ec ion (i.e., me e s a ached on bo h he longi udinal su aces) a 20
◦
C.
The mois u e was measu ed six imes o he longi udinal di ec ion and nine imes o he
ans e se di ec ion o each amoun o MC o he wood samples. The measu emen p ocess
ul illed he equi emen s o s anda ds EN 13183-2 (2002) [
20
] o esis ance me hods and
EN 13183-3 (2002) [
21
] o capaci ance me hods. All he MC measu emen s o he es
samples we e pe o med in he middle o he wid h o each es ed su ace.
2.2.3. Capaci i e Mois u e Me e
Th ee capaci i e mois u e me e s, GMI 15 (G eisinge GmbH, Regens au , Ge many),
GMK 100 (G eisinge GmbH, Regens au , Ge many), and Tes o 616 (Tes o, Lenzki ch,
Ge many) we e chosen as ep esen a i es o he capaci i e me hod.
The GMI 15 is decla ed by he manu ac u e o be an ins umen o indica i e mois-
u e measu emen . The ins umen is a speci ied dimensionless uni ( aw da a) ha is
con e ed o a mois u e alue acco ding o he manu ac u e ’s enclosed ins uc ions. The
ecommended measu emen ange o MC is 0% o 30% by weigh . The measu emen s we e
ca ied ou acco ding o he manu ac u e ’s ecommenda ions wi h a se measu emen
dep h o 30 mm.
The GMK 100 conside s he wood bulk densi y, which is gi en in he ange o 50 kg/m
3
.
The esul ing alue is displayed on he ins umen as a pe cen age o he absolu e wood
MC by weigh . The manu ac u e s a es a measu emen ange o 0% o 100% wi h he
accu acy o he measu emen dec easing om 0.1% o 1% o mois u e alues abo e 19.9%.
The ins umen enables measu emen s a dep hs o 10 mm and 25 mm; o his expe imen
a measu emen dep h o 25 mm was chosen. The manu ac u e o he mois u e me e
p o ides a cha ac e is ic able wi h a speci ied ypical olume weigh o he ype o wood
o be measu ed, which was en e ed in o he measu ing ins umen be o e he mois u e
Buildings 2023,13, 531 5 o 14
measu emen . The ac ual mean d y bulk densi y o he beech sample was 650 kg/m
3
,
esul ing in MC alues being duplica ed by se ing he bulk densi y on he ins umen o
650 kg/m
3
(650) and 700 kg/m
3
(700). Sp uce wood measu emen s we e ca ied ou wi h
he ins umen se o a bulk densi y o 450 kg/m3.
The Tes o 616 measu es up o a maximum dep h o 50 mm, unlike o he capaci i e
mois u e me e s. The ins umen allows o he choice o ma e ial and wood species
o he de e mina ion o he MC by weigh . The ange o he measu ing de ice s a ed
by he manu ac u e is 0% o 50%; he accu acy o he measu emen is no s a ed. The
de ice enables p e-se ing o measu ing so woods (sp uce, pine, la ch, and o he s) and
ha dwoods (beech, oak, maple, and o he s). An elec ode p essu e o 1 o 3 kg o he
mois u e me e on he es ed su ace ma e ial is s ic ly ecommended.
2.2.4. Resis ance Mois u e Me e
The Tes o 606-2 (Tes o, Lenzki ch, Ge many) esis ance mois u e me e measu es he
mois u e-dependen elec ical esis ance using a pai o s op spikes ha a e d i en in o
he wood s uc u e. Measu emen s we e made on all ele en es samples. The dep h o
he spikes inse ed in o he wood was up o 8 mm, which was he maximum leng h o he
spike elec odes used. The ins umen con ains a senso o de e mining he empe a u e
and ela i e humidi y o he en i onmen . The measu emen accu acy s a ed by he
manu ac u e is 1%. The ins umen con ains wo op ional g oups o wood species: he i s
g oup o wood species (beech, i , la ch, and o he s) and he second g oup (oak, pine, maple,
and o he s). The i s g oup was chosen o he expe imen wi h a s a ed measu emen
ange o 8.8% o 54.8%. The mois u e me e con ains ma e ial cu es ha allow he MC o
he ma e ial o be displayed di ec ly in weigh pe cen ages ela i e o he d y weigh .
2.2.5. S a is ical E alua ion o he Da a
The esul s o he mois u e es s by he desc ibed me hods we e compa ed s a is ically
wi h he g a ime ic me hod, which can be conside ed e y accu a e. This s a is ical
analysis was pe o med sepa a ely on he es esul s o one es me hod, one wood species,
one mois u e le el, and one es di ec ion. The i s s ep o he s a is ical analysis was he
assessmen and elimina ion o ou lying es esul s, which was pe o med by means o
1.5 mul iples o he qua ile ange. The es esul s hus adjus ed we e always sub ac ed
om he esul s o he g a ime ic me hod. Hence, 154 s a is ical se s we e conside ed.
Subsequen ly, a se o s a is ical es s was pe o med. Fi s , i was necessa y o assess
he no mali y o he da a, speci ically wi h he Shapi o–Wilk es . The no mali y o he
di e ences om he g a ime ic me hod indica es ha he es esul s we e only a ec ed
by andom ac o s and can be assumed o ha e been una ec ed by sys ema ic in luences.
Fu he mo e, he hypo hesis was es ed using a one-sample - es o he nulli y o he
mean alue o he unde lying se s om whose ealisa ions he s a is ical se s o igina ed.
The posi i e esul o bo h hese es s indica es ha he esul s o he obse ed mois u e
de e mina ion me hod a e no signi ican ly di e en s a is ically om he g a ime ic
me hod and a e he e o e in luenced only by andom ac o s.
3. Resul s
3.1. Wood Mois u e Measu emen
3.1.1. Longi udinal Di ec ion
Figu es 2and 3show a g aphical compa ison o he mean di e ences o he mois u e
eadings om he g a ime ic me hod o each mois u e condi ion, including he sample
s anda d de ia ions. The lowes di e ences we e ob ained wi h he GMK 100 capaci ance
mois u e me e , in bo h cases measu ing beech samples (ha dwood wi h a se bulk densi y
o 650 kg/m
3
) and sp uce samples (so wood wi h a se bulk densi y o 450 kg/m
3
). Fo his
mois u e me e , he lowes di e ences we e measu ed o e he en i e ange o measu ed
mois u e condi ions, i.e., om 5% o 30%. Con e sely, he Tes o 616 capaci ance mois u e
me e showed a signi ican inc ease in measu ed mois u e alues a e eaching 26% MC

Buildings 2023,13, 531 6 o 14
o bo h es ed species o sp uce and beech. In he case o he GMI 15 capaci i e mois u e
me e , he limi se by he manu ac u e o MC calcula ed om ‘ aw’ da a was exceeded
o so wood a an MC o 23 w .% and o ha dwood a an MC o 17 w .%.
Buildings 2023, 13, x FOR PEER REVIEW 6 o 15
de e mina ion me hod a e no signi ican ly di e en s a is ically om he g a ime ic
me hod and a e he e o e in luenced only by andom ac o s.
3. Resul s
3.1. Wood Mois u e Measu emen
3.1.1. Longi udinal Di ec ion
Figu es 2 and 3 show a g aphical compa ison o he mean di e ences o he mois u e
eadings om he g a ime ic me hod o each mois u e condi ion, including he sample
s anda d de ia ions. The lowes di e ences we e ob ained wi h he GMK 100 capaci ance
mois u e me e , in bo h cases measu ing beech samples (ha dwood wi h a se bulk densi y
o 650 kg/m
3
) and sp uce samples (so wood wi h a se bulk densi y o 450 kg/m
3
). Fo his
mois u e me e , he lowes di e ences we e measu ed o e he en i e ange o measu ed
mois u e condi ions, i.e., om 5% o 30%. Con e sely, he Tes o 616 capaci ance mois u e
me e showed a signi ican inc ease in measu ed mois u e alues a e eaching 26% MC
o bo h es ed species o sp uce and beech. In he case o he GMI 15 capaci i e mois u e
me e , he limi se by he manu ac u e o MC calcula ed om ‘ aw’ da a was exceeded
o so wood a an MC o 23 w .% and o ha dwood a an MC o 17 w .%.
Gi en he manu ac u e ’s decla ed measu emen limi (an MC ange o 8.8% o
54.8%) o he Tes o 606-2 as he esis ance me hod, measu emen a an MC le el o 5%
was ca ied ou . I was con i med ha wood MC a a 5% mois u e le el in he longi udinal
di ec ion o bo h so wood and ha dwood canno be measu ed.
Figu e 2. Compa ison o a e age alues o di e ences om he g a ime ic me hod in he longi u-
dinal di ec ion o beech wood using selec ed mois u e me e s GMI 15, GMK 100, Tes o 616, and
Tes o 606-2 in di e en mois u e condi ions and GMI 100 a di e en ins umen se ings o 650
kg/m
3
(650) and 700 kg/m
3
(700).
Figu e 2.
Compa ison o a e age alues o di e ences om he g a ime ic me hod in he longi udi-
nal di ec ion o beech wood using selec ed mois u e me e s GMI 15, GMK 100, Tes o 616, and Tes o
606-2 in di e en mois u e condi ions and GMI 100 a di e en ins umen se ings o 650 kg/m
3
(650) and 700 kg/m3(700).
Buildings 2023, 13, x FOR PEER REVIEW 7 o 15
Figu e 3. Compa ison o a e age alues o di e ences om he g a ime ic me hod in he longi u-
dinal di ec ion o sp uce wood using selec ed mois u e me e s GMI 15, GMK 100, Tes o 616 and
Tes o 606-2 in di e en mois u e condi ions.
3.1.2. T ans e se Di ec ion
Figu es 4 and 5 show a g aphical compa ison o he a e age alues o he di e ence
o he measu ed mois u e alues o o each o he mois u e me e s using he g a ime ic
me hod o each mois u e condi ion, including he sample s anda d de ia ions. The low-
es di e ences o he whole ange o mois u e condi ions om 5% o 30% we e achie ed
using he GMK 100 mois u e me e o beech samples (ha dwood wi h a se bulk densi y
o 650 kg/m
3
). In he case o he GMI 15 capaci ance mois u e me e , as opposed o he
longi udinal di ec ion, i was possible o use he manu ac u e ’s decla ed ecalcula ion o
he mois u e indica ion o e he en i e ange o mois u e condi ions up o 30%.
In con as , he highes di e ences we e ob ained using he Tes o 606-2 esis ance
mois u e me e o beech, while o sp uce he highes di e ences we e ound using he
GMI 15 mois u e me e . Wi h ega d o he manu ac u e ’s decla ed measu emen limi s
(wi h a ange o 8.8% o 54.8%), he measu emen showed ha when using a Tes o 606-2
esis ance mois u e me e below he manu ac u e ’s speci ied ange, he MCs o he so -
wood and ha dwood a a le el o 5 w .% we e measu ed.
Figu e 4. Compa ison o a e age alues o di e ences om he g a ime ic me hod in he ans-
e se di ec ion o beech wood using selec ed mois u e me e s GMI 15, GMK 100, Tes o 616 and
Figu e 3.
Compa ison o a e age alues o di e ences om he g a ime ic me hod in he longi udi-
nal di ec ion o sp uce wood using selec ed mois u e me e s GMI 15, GMK 100, Tes o 616 and Tes o
606-2 in di e en mois u e condi ions.
Gi en he manu ac u e ’s decla ed measu emen limi (an MC ange o 8.8% o 54.8%)
o he Tes o 606-2 as he esis ance me hod, measu emen a an MC le el o 5% was ca ied
ou . I was con i med ha wood MC a a 5% mois u e le el in he longi udinal di ec ion o
bo h so wood and ha dwood canno be measu ed.
3.1.2. T ans e se Di ec ion
Figu es 4and 5show a g aphical compa ison o he a e age alues o he di e ence
o he measu ed mois u e alues o o each o he mois u e me e s using he g a ime ic
me hod o each mois u e condi ion, including he sample s anda d de ia ions. The lowes
di e ences o he whole ange o mois u e condi ions om 5% o 30% we e achie ed
Buildings 2023,13, 531 7 o 14
using he GMK 100 mois u e me e o beech samples (ha dwood wi h a se bulk densi y
o 650 kg/m
3
). In he case o he GMI 15 capaci ance mois u e me e , as opposed o he
longi udinal di ec ion, i was possible o use he manu ac u e ’s decla ed ecalcula ion o
he mois u e indica ion o e he en i e ange o mois u e condi ions up o 30%.
Buildings 2023, 13, x FOR PEER REVIEW 7 o 15
Figu e 3. Compa ison o a e age alues o di e ences om he g a ime ic me hod in he longi u-
dinal di ec ion o sp uce wood using selec ed mois u e me e s GMI 15, GMK 100, Tes o 616 and
Tes o 606-2 in di e en mois u e condi ions.
3.1.2. T ans e se Di ec ion
Figu es 4 and 5 show a g aphical compa ison o he a e age alues o he di e ence
o he measu ed mois u e alues o o each o he mois u e me e s using he g a ime ic
me hod o each mois u e condi ion, including he sample s anda d de ia ions. The low-
es di e ences o he whole ange o mois u e condi ions om 5% o 30% we e achie ed
using he GMK 100 mois u e me e o beech samples (ha dwood wi h a se bulk densi y
o 650 kg/m
3
). In he case o he GMI 15 capaci ance mois u e me e , as opposed o he
longi udinal di ec ion, i was possible o use he manu ac u e ’s decla ed ecalcula ion o
he mois u e indica ion o e he en i e ange o mois u e condi ions up o 30%.
In con as , he highes di e ences we e ob ained using he Tes o 606-2 esis ance
mois u e me e o beech, while o sp uce he highes di e ences we e ound using he
GMI 15 mois u e me e . Wi h ega d o he manu ac u e ’s decla ed measu emen limi s
(wi h a ange o 8.8% o 54.8%), he measu emen showed ha when using a Tes o 606-2
esis ance mois u e me e below he manu ac u e ’s speci ied ange, he MCs o he so -
wood and ha dwood a a le el o 5 w .% we e measu ed.
Figu e 4. Compa ison o a e age alues o di e ences om he g a ime ic me hod in he ans-
e se di ec ion o beech wood using selec ed mois u e me e s GMI 15, GMK 100, Tes o 616 and
Figu e 4.
Compa ison o a e age alues o di e ences om he g a ime ic me hod in he ans e se
di ec ion o beech wood using selec ed mois u e me e s GMI 15, GMK 100, Tes o 616 and Tes o 606-2
a di e en mois u e condi ions and GMI 100 a di e en ins umen se ings o 650 kg/m
3
(650) and
700 kg/m3(700).
Buildings 2023, 13, x FOR PEER REVIEW 8 o 15
Tes o 606-2 a di e en mois u e condi ions and GMI 100 a di e en ins umen se ings o 650
kg/m3 (650) and 700 kg/m3 (700).
Figu e 5. Compa ison o a e age alues o di e ences om he g a ime ic me hod in he longi u-
dinal di ec ion o sp uce wood using selec ed mois u e me e s GMI 15, GMK 100, Tes o 616 and
Tes o 606-2 in di e en mois u e condi ions.
3.2. S a is ical E alua ion o he Da a
Th ough he qua ile ange, 381 es esul s ou o a o al o 9285 measu emen s we e
excluded by means o ou lie s. The analyses showed ha 90.9% o he da a se s ha e a
no mal p obabili y dis ibu ion. Howe e , in o de o conclude ha he es esul s o he
me hods conside ed a e only a ec ed by andom ac o s compa ed o he g a ime ic
me hod, i is also necessa y o show ha he mean alues o he esul s a e no signi ican ly
di e en om ze o. The analysis o he esul s o he mean alue es s showed ha only
10.4% o he esul se s did no de ia e sys ema ically om he g a ime ic me hod. This
implies ha almos 90% o he measu emen s show s a is ically signi ican biases. Table 1
shows he esul s o he mean alue es s in he o m o he pe cen ages o un ejec ed es s
o a gi en measu emen me hod.
Table 1. Tes esul s o mean alues in he o m o pe cen ages o a gi en measu emen me hod.
Me hod o Measu emen (%)
GMI 15 (capaci i e) 10.0
GMK 100 (capaci i e) 16.7
GMK 100–650 (capaci i e) 11.1
GMK 100–700 (capaci i e) 5.6
TESTO 606-2 ( esis ance) 8.8
TESTO 616 (capaci i e) 11.1
Howe e , signi ican biases om he g a ime ic me hod do no necessa ily indica e
ha he measu emen me hods unde conside a ion a e inco ec . As can be seen in Fig-
u es 2
–
5, in mos cases he bias is s a is ically signi ican , bu echnically, i is an a e age
bias o a ew pe cen . Simila biases a e epo ed by he manu ac u e o he measu ing
ins umen s.
The nex s ep o he s a is ical analysis was o e alua e he mul i a ia e analysis o
a iance (ANOVA). The esul s a e p esen ed in Table 2 as p- alues. The ollowing ac o s
we e selec ed as in luencing he esul ing de ia ion om he g a ime ic me hod: meas-
u emen me hod, ype o wood species, he di ec ion o measu emen , and mois u e le el.
Figu e 5.
Compa ison o a e age alues o di e ences om he g a ime ic me hod in he longi udi-
nal di ec ion o sp uce wood using selec ed mois u e me e s GMI 15, GMK 100, Tes o 616 and Tes o
606-2 in di e en mois u e condi ions.
In con as , he highes di e ences we e ob ained using he Tes o 606-2 esis ance
mois u e me e o beech, while o sp uce he highes di e ences we e ound using
he GMI 15 mois u e me e . Wi h ega d o he manu ac u e ’s decla ed measu emen
limi s (wi h a ange o 8.8% o 54.8%), he measu emen showed ha when using a Tes o
606-2
esis ance mois u e me e below he manu ac u e ’s speci ied ange, he MCs o he
so wood and ha dwood a a le el o 5 w .% we e measu ed.
3.2. S a is ical E alua ion o he Da a
Th ough he qua ile ange, 381 es esul s ou o a o al o 9285 measu emen s we e
excluded by means o ou lie s. The analyses showed ha 90.9% o he da a se s ha e a
Buildings 2023,13, 531 8 o 14
no mal p obabili y dis ibu ion. Howe e , in o de o conclude ha he es esul s o
he me hods conside ed a e only a ec ed by andom ac o s compa ed o he g a ime ic
me hod, i is also necessa y o show ha he mean alues o he esul s a e no signi ican ly
di e en om ze o. The analysis o he esul s o he mean alue es s showed ha only
10.4% o he esul se s did no de ia e sys ema ically om he g a ime ic me hod. This
implies ha almos 90% o he measu emen s show s a is ically signi ican biases. Table 1
shows he esul s o he mean alue es s in he o m o he pe cen ages o un ejec ed es s
o a gi en measu emen me hod.
Table 1. Tes esul s o mean alues in he o m o pe cen ages o a gi en measu emen me hod.
Me hod o Measu emen (%)
GMI 15 (capaci i e) 10.0
GMK 100 (capaci i e) 16.7
GMK 100–650 (capaci i e) 11.1
GMK 100–700 (capaci i e) 5.6
TESTO 606-2 ( esis ance) 8.8
TESTO 616 (capaci i e) 11.1
Howe e , signi ican biases om he g a ime ic me hod do no necessa ily indi-
ca e ha he measu emen me hods unde conside a ion a e inco ec . As can be seen
in Figu es 2–5, in mos cases he bias is s a is ically signi ican , bu echnically, i is an
a e age bias o a ew pe cen . Simila biases a e epo ed by he manu ac u e o he
measu ing ins umen s.
The nex s ep o he s a is ical analysis was o e alua e he mul i a ia e analysis o
a iance (ANOVA). The esul s a e p esen ed in Table 2as p- alues. The ollowing ac o s
we e selec ed as in luencing he esul ing de ia ion om he g a ime ic me hod: mea-
su emen me hod, ype o wood species, he di ec ion o measu emen , and mois u e le el.
Table 2. Resul s o mul i a ia e ANOVA. Fac o in e ac ions a e ma ked wi h an abb . ‘ s’.
Fac o p-Value [-]
Me hod 4.91 ×10−149
Wood species (so wood/ha dwood) 1.17 ×10−240
Di ec ion (longi udinal/ ans e se) 6.78 ×10−84
Mois u e le el 1.47 ×10−149
Me hod s. Wood species 2.96 ×10−32
Di ec ion s. Wood species 2.25 ×10−1
Me hod s. Di ec ion 4.57 ×10−199
Me hod s. Mois u e le el 6.17 ×10−281
Mois u e le el s. Wood species 7.46 ×10−2
Mois u e le el s. Di ec ion 4.94 ×10−277
The p- alue in his es is a measu e o he s a is ical signi icance o he esul s. The
p- alue indica es he p obabili y o obse ing a es s a is ic as ex eme o mo e ex eme
han he one calcula ed, assuming ha he null hypo hesis is ue. In he con ex o ANOVA,
he null hypo hesis is ha he e is no di e ence be ween he popula ion means o he
g oups de ined by he ac o le els. The p- alue o each ac o and in e ac ion e m es s
he hypo hesis ha he ac o has no e ec on he esponse a iable. I he p- alue is less
han a chosen signi icance le el (0.05), i sugges s ha he e is s ong e idence agains he
null hypo hesis and ha he ac o does ha e a signi ican e ec on he esponse a iable.
The smalle he p- alue, he s onge he e idence agains he null hypo hesis, and he
mo e likely i is ha he ac o has a signi ican e ec . This means ha a change in he
di ec ion o measu emen combined wi h a change in he ype o wood species es ed does
no signi ican ly a ec he esul s ob ained. The same conclusion applies o he change in
mois u e le el and he ype o wood species.
Buildings 2023,13, 531 9 o 14
4. Discussion
4.1. Compa ison o Tes ed Mois u e Me e s
As can be seen om he measu ed alues, each o he measu ing de ices was bu dened
wi h a ce ain e o , which was di e en acco ding o he ype o wood and he measu ed
MC le el.
The limi o he esis ance me hod is de e mined by he MC o he wood in he ange
below 5%, i could no be de e mined due o high esis ance alues o measu ing [
5
].
Acco ding o EN 13 183-2, he limi a ion is de e mined by amois u e ange be ween 7% and
30% [
21
]. Fo he es ed Tes o 606-2 (using he esis ance me hod), a an MC le el o 5%,
only measu emen in he ans e se di ec ion o bo h he so wood and ha dwood was
possible. Howe e , e en in his di ec ion and a his MC le el, a comple e numbe o alues
(max nine alues) was no eached o h ee samples. Fo so wood samples, only 93% o
he measu emen was de ec ed, and o ha dwood samples, only 85% o he measu emen
was de ec ed. This ins umen showed a close co ela ion coe icien (when compa ing he
measu ed alues wi h he alues de e mined by he g a ime ic me hod), eaching alues
o 96% o 100%. As can be seen in he ollowing Figu e 6, in he whole mois u e ange
om 5% o 30%, he esis ance ins umen indica es highe alues compa ed o g a ime ic
alues. The same end was ound by Li e al. (2018) when, a an MC o 10% o 20%, only
he bound wa e was p esen in he wood, and he au ho obse ed only a small dec ease
in esis ance, while abo e 20%, when ee wa e pene a es in e cellula spaces in he wood,
he dec ease in esis ance was mo e signi ican . Highe MC can lead o an imp o emen in
he elec ical con ac b ough abou by he conduc i i y o wa e [22].
Buildings 2023, 13, x FOR PEER REVIEW 10 o 15
17% o 30%, no signi ican di e ences we e ound. The e ec o he measu ing di ec ion
has been e alua ed by di e en au ho s; o ins ance Ha ley [8] s a es ha he elec ical
esis ance o wood is abou hal as la ge o a cu en lowing pa allel o he g ain as i is
o a cu en lowing pe pendicula o he g ain. This means up o a 2% highe MC eading
o pa allel cu en low han o pe pendicula , con a y o o he au ho s [17,22], who
ound no di e ence be ween he measu ing di ec ions ha we e pa allel and pe pendic-
ula o he g ain.
Figu e 6. Dependence o he a e age measu ed alues o MC using he Tes o 606-2 mois u e me e
on he MC o di e en ypes o wood (S—sp uce, B—beech) and in di e en di ec ions in he MC
ange o 5% o 30%.
The capaci i e me hods (using h ee di e en ins umen s) showed e y simila be-
ha iou o MC measu emen s, as can be seen in Figu es 7–9. Ve y simila esul s o he
g a ime ic me hod we e ob ained using he Tes o 616 and GMK 100 mois u e me e s in
he mois u e ange o up o 23%. Abo e 23% MC, signi ican di e ences in measu ed al-
ues compa ed o g a ime ic alues we e ound. The inc ease in measu ed MC alues
abo e 23% ( o Tes o 616 and GMK 100 in he longi udinal di ec ion) can be a ibu ed o
he con ibu ion o con ac capaci ance. The exis ence o ee wa e in wood could lead o
i s in il a ion a he wood/elec ode in e ace. Consequen ly, ee wa e ills he ai gap
be ween he elec odes and wood [23], and con ibu ions o in e ace capaci ance inc ease
wi h inc easing MC [17,24]. Bo h ins umen s showed a close co ela ion coe icien when
compa ing he measu ed alues wi h he alues de e mined by he g a ime ic me hod,
eaching alues o 96% o 98% o Tes o 616 and 98% o 100% o GMK 100. GMI 15
showed highe di e ences in coun ed MC o bo h o he es ed wood species compa ed
o he g a ime ic alues, and a high co ela ion coe icien o 94%–100% was eached.
Figu e 7. Dependence o he a e age measu ed alues o MC using he Tes o 616 mois u e me e on
he MC o di e en ypes o wood (S—sp uce, B—beech) and in di e en di ec ions in he MC ange
o 5% o 30%.
Figu e 6.
Dependence o he a e age measu ed alues o MC using he Tes o 606-2 mois u e me e
on he MC o di e en ypes o wood (S—sp uce, B—beech) and in di e en di ec ions in he MC
ange o 5% o 30%.
Fu he , he achie ed esul s in he ange o 5% o 17% o MC showed a di e ence
when highe alues we e ob ained o measu ing in ans e se di ec ions, whe eas om
17% o 30%, no signi ican di e ences we e ound. The e ec o he measu ing di ec ion
has been e alua ed by di e en au ho s; o ins ance Ha ley [
8
] s a es ha he elec ical
esis ance o wood is abou hal as la ge o a cu en lowing pa allel o he g ain as i is o
a cu en lowing pe pendicula o he g ain. This means up o a 2% highe MC eading o
pa allel cu en low han o pe pendicula , con a y o o he au ho s [
17
,
22
], who ound
no di e ence be ween he measu ing di ec ions ha we e pa allel and pe pendicula o
he g ain.
The capaci i e me hods (using h ee di e en ins umen s) showed e y simila be-
ha iou o MC measu emen s, as can be seen in Figu es 7–9. Ve y simila esul s o he
g a ime ic me hod we e ob ained using he Tes o 616 and GMK 100 mois u e me e s
in he mois u e ange o up o 23%. Abo e 23% MC, signi ican di e ences in measu ed
alues compa ed o g a ime ic alues we e ound. The inc ease in measu ed MC alues
abo e 23% ( o Tes o 616 and GMK 100 in he longi udinal di ec ion) can be a ibu ed o