APPLICATICN

S I M f L.ARITV P R I N C I P L E S AND

TURBULENCE RESEARCH TC BED-LOAD MOVEMENT

by
Dr. Ing. A. Shields

T r a n s l a t e d from:
"Anwendung d e r Aehnlichkeitsmechanilr und
d e r Turbulenzforschung a d die Geschiebe
bewegung , " Mitteilungen d e r P r e u s s i s c h e n
Versuchsanstalt f u r Nassexbau nd Schiffbau,

Berlin, 1936
by W. I?.

Ott a n d J . C . van Uchelen

Soil. Conservation Serv<ce

Coczperativc LLzboratory
California Institute of Technology
Pasadeza, C a l 3 o r n i a

n~Ove~%ken2
in 8 P % V & ~aze:

1 ) inadequ.a@sknce:t~Bedgeof the

. '
and 23 the caavoidable maek05 ~a:rnxiaaity
%aH ^ ~ I Q ~ &eqerimLcnt;3,
.

procaeees,

*
eo bad-loa& xckovezmeat in rivers, Mor,;vever, they- -wad.
not cover LPP bed-isad
prob1e:ms e n c o ~ ~ a t a r ein
d hydrauA.ie en gins aria^^ $+*!oreover, the ux~avstsidabls
%

aecGnd aec+Loa caneiders p r i ~ tzle

~ c~s ka~ d~ ~ ~at~ ~$he
s n ebe- C l ,

finds:

in &i"%.ii~rm
fl.osev t&e%e~2511exist only the f.r.i&iou f o ~ c c a

pressure ,Qarcca

re@nl%%ng
f%ozfithe ~ " d ~ iof
~ &be
h t %.water, ~ o n s e y a e n t l y ,the eonaponent of
the weigh5 in the direction

0.E

f k v ~m a s t $e equated to the total slxeax*upon

the perimeter::

with smaL1 d

sin c:--

tzwg @

=S

slope,

-asregarding the fPicdtioaof air, and v~rikfng:

a &.eaf%ag stress

;,an a : ~ e b a ~ @
o-+;er
d the -wetted peri:me&er, As a rule- this

a1 momentum r ~ ~ be
e $applied to the bottom as folPows:
bottom and perpendfeu1a.f zo t11e i s ~ v e ;of
~ the s t r e a m ,

Fig, 2 shows the

subdi~sionalayeas.
Sfnee
-khc &.si&3,g

areas cu9 the iso~selspe-:peneicu%ar, there arc ca

8
shearing fopecs $yaa.sfer~ed
alona
u these asreas,

What rex~-mias.&abe de'kesmia&l age thb2 oubdivisiiandl areas A

--a
fvv~s$igaeisn
of a series
-+-

czpirreg-4a.r
~
~ o u g b r ~ e s *~$h.,$re
s,
pcrir~eteb-a f u ~ i l o ~

se@tieagi.s5
x;iLh

vd
,
-a-

-,f

-.e

05 [ s e e $--age$6 and J-*

di%4s%ors%
sf el:e ercas..seegion can
=&#

b @ad Aw.
isa\*ebs shows -ih& la composite cross1 page 10)a desired a.9;.p.zi.oxbr?~;a;ee

i
l

d:~

*: !

division Ilne a ,

agt;incd

$y ~

- of middle
~
~and arSgL~Ear
~
~

p
*."-- L1- a
- t e of .
a - vn the computed sub&visiana*l vo.e~scf$-Jes,'rhesa e s m p u & a i o n sare ~a>$It,ed..
k e & h 0 ~ 8, M8f Z O V ~ P ,

k e r

eince i t
COW&~$

S J

2~ qtzc.rfisn

B+,%u~,$.

06 the resuits of cir~ul:?!:r.~ ~ h c t h eno$

~ ap~ibicatian
* --a

elq~eri:msnb is p : ~ r n ,ssible
f
$c,r the sub,divisional cross s e ~ fof n in
~

q ~ ~ e ~ itt ~ , advi.sable
~a,
to exnreae the mean part-ve5oeities in relation to
the
cross-.~ec&iona,i
wloci&v, s n c e ;

B"rs%rativcz~aghaess"~
as v ~ c l lsa g;4her f a a c e ~ ~ s' ,$ ~ ~ V ~ ~ . _ t
Vo
C res&
F.

gcaz;raS

cancluaiolln, :2 is ~ + e e e s $ a %-hat
~ r y the q u ~ n t i l yof
- disc%-a&rge
be a raawin3uim va:uc

the latter being an a ~ ~ t ~ - a orequisite,

i.af
F s z * a h ~ % ~ L e dsgfethe processes
SJ-"

a "

involved, vcPocity mmacaur emants a r e necaessrg in order tea eonstruet t h e

isoval patterns.

On the ather hand, since the canstruetion sf complete

i ~ 6 v e pateern
I
requires Baborio~~s
w o r k and i s agtte.cn unnecessary, it i s
more przetics%to d e t e r r ~ i n ea portion sf'kbe pattern, Measurements in MI,
open channel '8.Siterecarried sut in the follo\z~%ng
manner:
A Pitot tube v ~ apEaced
, ~
upon a kzidge in such a n ~ a r n e r%ha%the tube
could be moved horizantafiy across in %he ci~a-mef, A perpendicular velocity
astribution in $%em i d d l e Bane vvas next recorded,
order to deterldnc
ta what wid$& the uelocitiaa of the
zor,e: ecxtsnd~d," h e Pitot tube w a o
%en gradually .r;-.,o~~ed
across the chanael, In this -aanner the region of
validity
%his V~IOC~*~;&.s$ributBon W ~ I
e8kabE8$$ds
Thereupon tsac~:'17e
-farce axqd r n i . 8 ~.scloeiPf w e r 8 graphically dete~x:ained,
ebis region also being dzciaive for bed-load. I T I O V ~ ~ ~ ~ E ~ ~
The value 05 tbe tractive X Q P C ~ ka this region was determined by She

A s Lsng as ekis i%23.~~-mern

*velocityw a s main-

level of the -maximum vz,scitg,

tained at the s~~.rt%acz,

4 t w a g pc~ssfbfe
%O UPC the ~i$iaiter
depth fn t h i ~
C

O % ~ ~ U -

lation, Up to the time the b e d w a s c x e e s ~ i v c 2 ydev:delsged, re-cxaminatisa

*
of the tractive force swsa psssxiika,
&ace no i~4laencew a s exercised by the
e

vta14is, and the km:bBfez~e COU&& d.eveIop freeky,

UI'

nam,

siace P=Til:uza2se',

veluc;:

(see Pn$luanre sf Grain 6&3,pe), A

This rcauPted i n A

.-

va?ues -;thich
-a

-:

E, 48 ia v d i d 8cg-xalp *Perta definite roughness

--

provided

*,
I 1s icx~ewn- could be computed,

d
aza agprax"J"z:~oteiy
4 , O C % s~nalieria,~

one p ~ ~ t cjf&---~ ~against

~ g Isg An- the; g ~ k % c l ~r~'z$bb~.S,fng
al
6f the bed., due
6
8"$e
"
$0
d e w ~ ~ ~ p ~~&~t:d
~ ~ . be
i . ~
dete;:rnrrtea
t ~
~y sgstanxatic dewiationa, The
m a g ~ d t ~ dofc A $OX 2h,e ~;esiaus kinds sf r ~ ~ ~ g h nwith
c s s a quadreABc resistance
given by a q u s & i Q-i-3
~ ~ an6
is
d

&%

&"

"

A
p

- f s

,gs, 4 a~xd5 %ll,u&zz"te tine pi3Miag 02 u agaiaat

far variaus wszar d-.pt::l.,s

Ip this

W P G ccm~tar~t
~
siope,
-9

geckiwfi it jzas been

y i a logarithmic

ft9rm

6& Hz obvious that the friction

S ~ " B Q -I"
~&~
i,Pyc;,
tj%e
*
~j sh,ear

~ - n t&s s ~ h & ~~g e& be

a

--$v ,!?&~a&
.*
~ ~ ~ ~ f dthe
e grad3,s
~ i n gk~:, ;s, bad c ~ fz~;r.~~~.rcr:::~~
g5:~~~j.n
ef.aes, tpbc q,ac:gtion is""".--a

reppase.---Lsa d e c i 3 i ~ ~gzain
e
area {i$h
2

e;x;

~nc
i

[ a / x ~ ),n&icates tb,2 stagp-ation pressure of a crktieal

e
<deei,zta :he resietnnca-cot:ffifici.l%t 31 the grain at a v e l o e i y c .
v~iP1nat g o i~-&oally further definii5im~sf the above magnitudes,

grain shape
veIocieJjl

- -again d e p e ~ d i z gsolely upen

t i

or attenlpt to describe the aspect of b h z flow about the grain, '%Werefer

rather te $&e ~ 1 o rs;e$1 ~
kJf~;~~
(TJ*
c, ~
Da&.,i ~
~: ~
; r 8 ~ ~ u x ~ g ~361)
h e Pvihere
c
-19

sEz~~?SBJ:

type of ro~gkaessvas eonsiderzd { s e e Eaflnence Q-X Grain P O F - ~an6

) , in
which. $he essential :magnHwdes w e r e carefnlly rfieas~ared, There, ati page 8
a
w e see that in a ro3;agh pi2e the ioilswf rig la*$d&:
for velocity pV"eValS8:
s

-2s

%I g

-&

$he ~ ~ - . c a ~
friction
~ e d velsci.Zy

i n e symbol f

e-w

that
d4 ~iqniii....~

f t ~ n c t i o n{and &hers) are kH&ev~&e

dependent 'iijp~ntlke ga BBxn &,ape [see Bn,y&aenc.; sf:G>:zin. shape ),

~yc2~2eR811yvL

"I-"
z.irb

i"'ia.:iduL5.td~e
ccief;.'i.cient

1
-eC.eqc..i3--

02 the grain depends upon both,

a fourth regicn,

\:J;~:~&

gre~kex*
vs*-d)k~
values
'iP

- that i s * a rcgioa

i.17

q-5,

al.1 sftIacae

balds ir8 pap$;$eu]ar for

\yJaF@

e-w,J.&,A

,%-..-

$~.ne,,~s
w i t & sh,ooting f3.o~~
, since
--A".

?:<lade ygitb acpae;zaaDbe

e slopes (
e

"

> i.ZoQf; i , e,, -v~ith

amka,'Jl

depths,

of njovercank sfa m2zrtarL:t,

t - :

id

sav9
', ','*
*,,

E.P& paaBPq

b x i s c ~$0 ~?iasiciiiik~rf:y02 the gsaia a>apes,

" *=,,*
; ~ c c g ~ ~3. LeetaLai,~$.3ii,
c%P. A
I B' because rixjs$ .T
theI
nand
vbr.'..rp

~ K I ~ X ~ Ube-$,
r,~~~;*rxs-;
~ S
"9

L i s also valid for the inyrestigate4 type of broavn coal

8 i n e ~ >-'3

w e can 210: the c o r r r s p o n d i ~ ~;oints

g
in Fig.

bed loadAi,

odorm mi";^,,

6.

The obvious

together w$th Lkc conformity of barite and granite bed laad

--

- - - - .

--=

ere &he b r o v ~ n
exeeg.:the
separatio:~and cornenLnkisn of finer parzs lead to a sanirnewhat l%i@hercritical
i~w,

. A ~ Z G ~ ~

---P-~-e&---.p

gurnishes e-vidsnce that the influence sf "iweigkt upon t h e beginning sf

haa bean ftelly covered,
Influence of Grain Sbape

1
-P
w
--"
.I--

The grain. shape

Tray

many respeces as tbc

i n i i u c ~ c ethe beginning af movement in just as

-~ai.al=aeq
a;ppe:-as-ing in derivszisns of the aqusa"kiongar

$ha b e g i n a l ~ ~a%
g ~xoxren~ent,They we.+e ae follov~s:
P

= in6Buence of grain shape upan p~%r"$rsity

dz

= ixifluznce of grain shape upor. bed-iriciiua coefficient

@.

--

i ~ f l u z n c eof grain shaglie u p ~ n

g r a i n surface

Pa4.Ge:r?rasses

influence of grain s i ~ a p eupon flow c o n d i k i o ~ sin a

l.t is evident from. Fig, 6 $haitthese 131," s a y elczrt opposing i n -

rounded gr~ainsand s h a ~ p - e d g e dg r s i ~ slie sor12ewl1.,at higher than those

grain shape garan:etritaEly,

In. eat:er .loillu&rw& thie point,

@\

for

muat

4,
investigated eeparatzPy, The g ~ r o s i t !@,
~ :an be 2asiI.y dekt~1~~4fLi~$d,
I
q2may 'p"6siblg

be d i t ~ r a+ l ~ n- ebyd ~r?~i:~r:ei~:;
&at is, by- s-irt,usring a few

-.
erains on a beard wli"&-~
ui3iiforzm. +rsdna g;.iled an do IE* thzn b y g r a d ~ ~ a linly
=
f~rbultgia"$ZCI~~tuclining the bsrArd;vhile si-$?&l$ sl~a-icin.~
~3 j$;
- 4@sz~*&~~nding
&e$cas), un.tjl the g2:ains 3-s)iil of;, xfi &f $ manne$f
~ g i $ i @ a~l l ~ p e be
k*

"

used by Cdsey,

ix:,?gi:ther%tikdr~&%
n a p ith he^^ ;lave zk:-,aJa

experit1~1.?n$c
in ;$,?~s

v i t i a aagular grai12.3 9 I ~ Q : X J ~ %
~~F
eq2eti'~"1-tez.7r@9
irl cbp$x? c$aanncls

f"d"i1gzxentsof

ar

sh8-r.m
,*
( t y ~-L,- 3~ IT?
A

slbe to clea.2 up :=hispoint,

-;$ittiv e r y sharp edges) were avail-

To deskzr~mj-ne
definitely %heinfluence of the

grain shape requires that all ather variables

be

* i ~ l ~ 5 : 1granite

ekim-icated, k c ~ a r ~ n g l the
p , axpe5.rncnts
*

- mechanical. m d hydrhulic
\IJ

it11 gzanite gr,~in-t%yes

were

made not only v ~ i t ?ths

-~ s ~ x r ~
unif~>:~n-i
e
g ~ a i ns i z e s {d -. 0,85 mm, d
d

= %,44rnxx) and in

.-

same channel, bat also with

&be

-- 1-23mm,

same slopes, a n d by

using the sams ewperiixeatsl technique as sad by Gasey in his s a r ~ dexperime~:ts~

rzle ~iftzsuj.". fa:: grain sjzeg 3, g5 ";m3yI a2.ltl 2,44 3 1 3 s Eire v e r y cl-.,dractzri ~ t i c;;mid 85-j3 d s p i & b $

F i g s u 9 and lii: ~ > f &

3 pja'kted agdast fie,

!kc
-values for the meny-edged gj:.;iils iie every
time about ten pe:- ;.zent higher than those Isx rounded sends i , e, , an awguls-z
qrsie~p;.o&~~,ces
a p 2 a t e s c ~ I O ~ruesistzt~ee
JY
t&aa z-ounde2 one, Thus, the --a'an
e
velocity, wit;. con star& R 5, i a %ess w i t 3 fra g ~ ~ e n tt ba a . w i t h sand,
il?ustri*';ion ehaivs

ti2ni

)t

& * A h

ff3Pnxaeio!-'aeel-rLc 1

~ ~ s

s U ~c~asgi.::~@,y
&
is

goon ::eache&

Tl~tas,ba

The 3eroistenee

0."
*He

type oi moEfau and Lhe t i m e required for

%t.sitha miovable bed this 3criad indicates $he pcrtiaeng

sea", for bed-

3.4,
Bsad ~mo%~ezx3.,ebfs.%,
[Gee Krey 1 *

t i m e scale %?OF% thzt

e~z;>zripn~nd9
~ 1 : ; ha a t ~ a i g k channc$,
t
Eowevcr,

results ob%&ned*i~itk
e-~i:fe~Trne~t&.Z
c h a ~ a , z hfareish vahsbfe facts {barn-i,%g

P*S;

inus,
considered:

i r o ~t;i ~ erebd-tivc ~ ~ ~ , . g 4 n c$-:l~ce

s s , r,pa.~ii~bHe~
are %.; a be

sne born? of the *&~ataa

suzface

PI PL4

2,

Tbe w a t e r depth

The bed shape

A t the present state 61 seier1t2Ae iaveatig~~tion
it is practical&
iwlpossihte $0 Saifi%3
a?-~&,a,-rc*zy~f~ens--en&s
an6 one a*L.,us"ie oetisfied w i t h
a csaside~aiionof be1? shape w:%..rd vpa$ep dep%&*
3,

It ~ ~ f i c to
e sdesign&%@
b%e bed shape xaerely by the relation oi
heigkk to length, and 0%bed d e ~ f e l s p ~ ~ e irsegulas:i&ies
,ri
5d

thiB

as &he extent 0~~ e dde-:re%oprfie:";:&Ci

d
-3~Hichis

l e s ~ l y&y the relation of the hsEgh4

the w a t e r de;:-&,

&J'a&p$l2e:c

are cafied r i p p i % e ~ ,~

0%the

&e<iaitiwas

$2

LQe b~etow-,
kCCc d

-.sq;.rea sed sLwmeneisn-.

ripples, dx..x~es,axe
att~mp&&~

-*

: ~ ~ e p i1 2~. a~~g&oa;vn
~~ i ~ $&&
$ ; k~~ a ebed

SCOUT

holes to

ergen in

j.oad inducca the

TL

s s a ~ ~ e z$he
* - gs&n matzraai, tnt2
-

2~&%9:jccn
heiabt
e
2

y:

an.d %eae&$
of bed f s ~ ~ ~decpaeese~,
~ ~ i o n.,.r>&e a$ fire sarF,e E-fi2e
3
2 ag ,
iT**

- the r31am3er

scaLe3,

&

0%bed-load

- as s h o ~ lini

r~~ovemze:18
c h a ~ ~ g in
e s the fo"elox:ring sequeaee:

i ~~y q g{ze,-&~
~
~ (f?:eq:ao-&H.br
~
~
~ of , &;2liqnz sl~.aupe)
until -~j&ea
the

s gentle ua&al*&tisn,

oalace v,<d/u
-...
are every Xargc, %ha bed shows

TV
ne raNc b e b ~ ~ e eheight
a
atad waker depth Ekev~lsegza&cL"r;lgdecreases, Tktrs ;
a*
v a r i ~ u smanner^ 0; x % B T $ v c-- ~at~ie9&%
~ c ~ ~s&ortQyafter the beginxing 0%

raovemeni;

- can be rong2iI.g in&cacad by the z ~ a ~ m b k% z * v * * d / u.

k That iss equal 2 c ~ ~ n s 9 dwsu ~ ~ b e x -fo-.

8
similar
$ 09 rklo.i:erfie&,

"ha

grains are $&e ra&.a requisite for

---*----M-w&--.----.---.,-~~-~----~-~-

Fig, 12

.
y
e
-

inetaace,

:%*i

wl~Le
mg,14 ah~xgean sjliclIip
crease of $ra&ve fsl*ce
of the $az.ka l a re>-a;:ed$63

l ~ u ~ $ ~ -of
a ripple
~ ~ o nfGrnli.,t60n

ban'"
-.a-

"-----

& r i db i l B $ t i ~ ~ ) .

coal,

~ I Y Q V J ~

Sne barite,

Wi.&. a in-

was generally ao$$ced"&&13~ith

different l-fianne;:;.e

aa increb-az ir, gg&e

v&c$?ejk';r,

zn& grs&aal%y

-8 % - ?!+

&*

--.

-IT
#,

'
J

d
,""*

"

these raagaitudsa,

"*---

n~

>rp* "

'

and s k a ~ ~
$ac&aps
e
and ally ccsm~?,a?~t$sa
of

,nrs analysi s pro~dsl;:

.7,;th
-.*
still i;rijates

-I1-

pp

I{?l

-*>a]&

also for b-=-, ; : S ~ . ( Q Q coal i s trausgressed.

fFi.i
.~
b ~,f i l gtb,e smbes. hesl~kts
ingjl v ~ B U C : ~ T-'f'O 1!ff,,1 . 6 , _.y)di ag I , E ~ ~..
a

--v e v y

ape e i g z c
*-a

pq--u
gh%

s,dv~2d,3~a~~6.&aa

- 6 ~ -

*'* -?* " ' " Z q

.y.7 e x -

>-& :. &-%a

Grain
diam.e&er $ & ~ p e

an

I L ~ "~ . ~ z

VJ"?"

.m

Fig. 1 .

A
Fig. 2.

F i g . 3 . Ratio of hydraulic r a d i u s t o
depth a s function of depth-width r a t i o .

F i g . 4. Velocity d i s t r i b u t i o n
curves for Barite.
d = 1 . 5 2 r n m , S = 1/500

Fig. 5. Velocity d i s t r i b u t i o n
curves for Barite.
d = 2. .?6mm, S = 1/380

Fig. 6.

Tractive

f o r c e coefficient
(Y,
V4cd
Reynolds n u m b e r of the g r a i n

.
V

Fig. 7.

Y)d

a g a i n s t the

C r i t i c a l t r a c t i v e f o r c e r a g a i n s t weight of g r a i n l a y e r s ,

(y

(the c u r v e s are obtained f r o m F i g . 6 , a s s u m i n g V = 0.012 a n d y = 1 ) .

y)d,

Fig. 8. Cumulative mechanical

a n a l y s i s c u r v e and modulus
M = Af /Ac of a bed m a t e r i a l .

Fig. 10. Resistance coefficient

of the Reynolds number.

Fig. 11.

a s a function

Geometric s i m i l a r i t y of two s t a t e s of flow.

F i g . 15. P l o t of e m p i r i c a l equation f o r
bed-load transportation.

Fig. 16.

Sieve a n a l y s i s ( s q u a r e m e s h ) .