Table o con en s
Supplemen a y da a o he pape : "Nonlinea Beha io o Addi i ely
Manu ac u ed S eel Beams wi h T apped-Powde Dampe s"
Jona han K. Black, 2025-08-07
Con en s
Filename
Desc ip ion
aw_da a.zip
Raw ime his o ies, FRFs, and beam measu emen s
p ocessed.zip
Resul s o linea and nonlinea p ocessing
g aphs.zip
Full se o g aphs e e enced in he pape
Abb e ia ions
•B[x]: These a e he labels o he beam samples. The numbe [x] is
he hickness o he powde pocke in mic ons.
•Mode name abb e ia ions
•D i e poin abb e ia ions, e.g. Z039B: see Fig. 2 in he pape .
•Tes p ocedu e phases: see Table 1 in he pape .
•P1_lin: Phase 1, Linea es ing
•P4_nl: Phase 4, Final nonlinea es ing
Abb e ia ion in pape
Abb e ia ion in da a
Type o mode
Mode Z[#]
bn[#]
So [
non-s i ] di ec ion bending mode [#]
Mode Y[#]
bs[#]
S i di ec ion bending mode [#]
Mode T[#]
[#]
To sion mode [#]
Folde s uc u e
Each zip olde has he ollowing olde s uc u e:
↳Le el 1: Beam label (e.g. B500, B3000)
↳Le el 2: Phase o es p ocedu e (P1_lin o P4_nl)
Fo example, / aw_da a/B500/P4_nl/ con ains he nonlinea es ing-
down da a o he beam wi h a 500µm powde pocke . The esul s o
p ocessing ha da a a e con ained in /p ocessed/B500/P4_nl/, and
he g aphs made om ha p ocessed da a a e con ained in
/g aphs/B500/P4_nl/.
The excep ion o his pa e n is ha he summa y plo s, which combine
da a om mul iple beams, a e con ained in /g aphs/summa y/.
Raw da a: FRFs, ime his o ies,
and beam measu emen s
aw_da a.zip
F equency Response Func ions (FRFs)
Example: / aw_da a/B500/P1_lin/FRF_Z073B.ma
•The FRFs om linea es ing a e s o ed as .ma iles, as expo ed
om Tes Lab
•The p o ided example MATLAB sc ip / aw_da a/pl _FRF.m
shows how o ead in and plo one o hese iles.
No e ha some o he Phase 1 da a we e collec ed wi h an exponen ial window applied o he esponse signal, as no ed in he eadme iles.
This means ha some o he FRFs ha e a i icially highe damping. Howe e , he pu pose o Phase 1 was no o ge accu a e damping alues.
Time his o ies
Example: / aw_da a/B500/P4_nl/meas0_Z090M_med.ma
•The hamme o ce and accele ome e ing-down signals a e s o ed as
.ma iles, as expo ed om Tes Lab.
•The e e ence o ce and he accele a ion measu emen om he
ele an * accele ome e a e s o ed as wo Signal s uc s in he same
ile. De e mine which is which by looking a he uni s in
Signal[#].y_ alues.quan i y.label.
•The p o ided example MATLAB sc ip / aw_da a/pl _Sig.m shows
how o ead in and plo one o hese iles.
•Fo B000 and B100, when Phase 4 was simpli ied, each Signal ile
con ains mul iple hi s. Fo B3000, a di e en me hod was used o
expo da a om Tes Lab, so he o ma is sligh ly di e en . The sc ip
pl _Sig.m wo ks wi h all o hese.
* Fo hi s in he Z-di ec ion, he Z-accele ome e signal is sa ed. Fo Y-di ec ion hi s, he Y-accele ome e signal is sa ed.
Beam measu emen s
/ aw_da a/densi y_measu emen s.xlsx
•This Mic oso Excel sp eadshee con ains he mass and
dimensions measu emen s o he beams, including many ha
we e no included in his s udy.
•See Figu e B.1 o he de ini ions o he dimensions in his shee .
•The Shee i led " aw_da a" con ains he measu ed alues o
mass and he h ee ex e nal dimensions.
•The Shee i led "co ec ions" con ains he o mulas used o
adjus he measu emen s as explained in Appendix B.
•The Shee i led "densi y_20250506" con ains he da a and he
plo used o es ima e he bulk densi ies o he used and un used
powde egions.
Resul s o linea and nonlinea
p ocessing
p ocessed.zip
Summa y boxplo s
Example: /g aphs/summa y/boxplo _bn1. ig
•These boxplo s summa ize all he da a o a gi en mode.
•Each beam has h ee boxplo s: one o each o h ee ampli ude
anges. See he example below.
Model a iance ba g aphs
Example: /g aphs/summa y/ba sd_add_mdln m_bn1. ig
•The ba g aphs compa e he ela i e unce ain y o se e al models
o he da a. See Sec ion 3.4 in he pape o an explana ion.
•The unce ain y o each model is no malized by di iding by he
unce ain y o he linea model.
He e is an example o how o in e p e hese g aphs. This example ba g aph is o B1000, bn1; i
is aken om he igu e ba sd_add_mdln m_bn1. ig.
•The model A has 30% o he linea model unce ain y, meaning accoun ing o nonlinea i y
dec eases unce ain y by 70% in his case.
•The model A, PS has signi ican ly lowe unce ain y han he model A, meaning ha knowing
powde s a e dec eases he unce ain y conside ably.
•The model A, DP has almos he same unce ain y as he model A. The e o e, accoun ing o
nonlinea i y and he d i e poin does no dec ease he unce ain y ela i e o only
conside ing nonlinea i y.
•The model A, PS, DP, HT has 13% o he unce ain y o he linea model. The e o e, he
unce ain y can be educed by 87% by accoun ing o all ou o he ac o s in he
expe imen .
