QUALITY CONTROLLED HOT DIP GALVANIZING
AND
PHOSPHATE ETCHING

by
Kenneth Stephen F r a z ie r
T*fitri

A THESIS
Submitted to the School of Graduate Studies of M ichigan
State C o lleg e of A gricu ltu re and Applied S cien ce
in p a rtia l fu lfillm en t of the req u irem en ts
for the d egree of
MECHANICAL ENGINEER

D epartm ent of E ngineering
1953

ProQuest Number: 10008306

All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a com plete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.

uest
ProQuest 10008306
Published by ProQuest LLC (2016). Copyright of the Dissertation is held by the Author.
All rights reserved.
This work is protected against unauthorized copying under Title 17, United States Code
Microform Edition © ProQuest LLC.
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 4 8 1 0 6 - 1346

ABSTRACT OF THESIS
ON
QUALITY CONTROLLED HOT DIP GALVANIZING
AND
PHOSPHATE ETCHING

This th e s is co v ers the re su lts of r e se a r c h in determ ining the
value of a ccu ra te controls in the hot dip galvanizing of ferrou s
pr od u cts.
S p ecifica lly it deals with equipm ent design, cleaning procedur
rin sin g , flux coating, con trolled galvanizing, phosphate treatm en t
and econom y.
It com p ares the r e s u lts of te s ts on norm al and fo rced alloy
growth coatin gs, of uniform and uneven coatin gs, of ductile and
b r ittle coa tin g s, of plain and phosphate treated coatings and of
coatings over d ifferen t com position s t e e ls .
It c o v er s the value of the accu rate con trol of a ll p h ases of the
p r o c e s s in the reduction of d ro ss form ation and r e je c ts in the
in te r e st of econom y.
It in d icates the proper hanging and handling of work to insure
quality galvanizing and m inim um labor req u irem en ts.
It show s a method of determ ining proportionate d ro ss fo rm a ­
tion from the work and the galvanizing kettle and c o r r e c t produc­
tion for m in im izin g the amount of d ro ss form ed per ton of m ateria
pr o c e sse d .
It co v er s s a lt spray and hum idity te s ts on quality con trolled
coatings c le a r ly showing the value of phosphate treatm en t.
A r e fe r e n c e table of com parative c h a r a c te r istic s of a ll types
of zinc application is included to p rop erly c la s s ify hot dip galvan­
izing as a p ro tectiv e coating.

F O RE WORD

A fter many y e a r s of in te r e st and r e s e a r c h in the p ro ­
tectio n of s te e l by hot dip galvanizing with in vestiga tion s of
galvanizing plants in the United States and England, the
tech n ica l staff of the D etroit S teel Products Company d e­
sign ed , constructed and put into operation a fu lly controlled
and m echanized hot dip galvanizing plant for the p rotective
trea tm en t of s te e l window fra m es.
This is the f ir s t plant so designed in th is country and
has afforded a la rg e field for r e s e a r c h work in quality
co n tro l.
The v e r y fact that hot dip galvanizing in its cru d est
fo rm p rovid es con sid erab le p rotective value, has clouded
the n e c e s s ity for con trols to obtain uniform ity and econom y.
A s the nam e im p lie s, galvanizing (hot dip galvanizing)
is the e le c tr o -c h e m ic a l form ation of a z in c -ir o n alloy
c r y s ta llin e coating on b a se m etal form ed during su b m er­
sion of this m etal in m olten zinc, coated with fr e e zinc as
the work is withdrawn.
The nam e galvanizing com es from experim ents r e ­
corded in 1786 by Luigi Galvani, Italian P h y sio lo g ist,
coverin g h is work in showing the d ifferen ce of e le c tr ic a l
potential in d issim ila r m eta ls.
Zinc is higher in the electro m o tiv e s c a le than iron
and p ro tects the iron from co rr o sio n by becom ing the
anode of an e le c tr ic a l couple when in the p r e se n c e of an
e le c tr o ly te .
The iron w ill not corrode when coupled with
zinc so that sm a ll voids in a zinc protection do not d e s ­
troy its p ro tectiv e value as is true with many other m etal
c o a tin g s.

K. S. FRAZIER
Chief R esea rch Engineer
D etroit S teel P roducts Company

QUALITY CONTROLLED HOT DIP GALVANIZING
AND
PHOSPHATE ETCHING

Weight and C om position of Coating
It is g en era lly recogn ized by au th orities that the life of a zinc
coating over s t e e l under given atm osph eric conditions is d irectly
proportional to the th ick n ess of this coat.
E xceptions to this ru le of thumb have, how ever, co n sid era b le
bearing on coating life and it is th ese exceptions which we w ish to
con sid er in th is t h e s is .
In the handling of th ese item s it w ill be shown that rigid con­
tr o ls in the Hot Dip Galvanizing p r o c e s s are not m erely good h o u se­
keeping p r a c tic e s but a re profitable to both the galvanizer and the
cu stom er of galvanized work.
Hot Dip G alvanizing is w ell into its second century of u se and
its proven valu e of protection over s t e e l is esta b lish ed .

It is g en er­

a lly conceded, how ever, that sin ce reason ab le cleaning of the b ase
m etal and the dipping of this m etal into m olten zinc w ill gen era lly
produce a good p ro tectiv e coating, c lo s e controls of the p r o c e s s have
been v ery slow in finding their way into industry.
In the m odified hot dip p r o c e ss gen erally known a s G alvannealing, advances in control have been much m ore rapid.

Galvannealing,

gen era lly a sso c ia te d with sh eets and strip w here form ing is required
after zinc coating, has forced the hand of industry in developing con ­
tr o ls of coating th ick n ess, zinc iron alloy growth and annealing.

The

lim its of coating th ick n ess in th is p r o c e ss d ivorce it from im m ed iate
co n sid era tion w here m axim um w eight “quality coating** is d esira b le
1

for long life p rotection p a rticu la rly w here a d v erse atm osph eric co n ­
ditions e x ist.
Since the z in c -ir o n a llo y growth of hot dip galvanizing form s
w hile the b a se m etal tem p eratu re com es up to the tem p eratu re of
the zinc bath it follow s that heavy m aterials such as stru ctu ral s t e e l
w ill accep t n orm ally a heavier coating than light weight sectio n s
such as used in s t e e l windows.

Although the findings shown h ere

w ill apply in many in sta n ces to the h eavier m a teria l a ll te s ts w ere
made with the lighter sectio n s and w ere pointed d ire ctly to that
fie ld .
In order to cla rify this point a sim p le t e s t was made.

Coupons

of 16 gauge (approxim ately l / l 6 ”), l / 8 ” and 3 / l 6 ” in th ick n ess w ere
p rep ared .

A fter thorough cleaning they w ere placed on the sam e

rack and low ered into m olten zinc at 850° F . for a period of one
m inute.

T h ese coupons w ere then stripped in accordance with

A m erican S ociety of T esting M aterials (A.S.T.M .) P roced u re A -90
to determ ine the w eight of accepted coating.
T hree sep arate t e s t s , each containing th ree coupons, w ere run
of each w eight of p la te.

The th ree te s ts showed the follow ing

c lo s e ly p a r a lle l averaged resu lts:

the 16 gauge coupons picked up

2.85 ounces of zinc per square foot of plate or 1.42 ounces of zinc
per square foot of su rface;

the l / 8 ” coupons picked up 3.28 ounces

of zinc per square foot of plate or 1.64 ounces of zinc per square
foot of su rfa ce, a gain of .22 ounces of zinc per square foot of su r ­
face;

the 3 /1 6 ” coupons picked up slig h tly m ore coating than the

l / 8 ” coupons.
Subsequent te s t s indicated that th is growth of z in c -ir o n alloy

2

was p ecu liar to thin sh eets and that beyond 3 / l 6 ” th ick n ess this
change in a llo y growth was not m a teria l.
It is com m on p ra ctice in industry to rec o rd the w eight of zinc
coatin gs in rela tio n to weight of black s t e e l being galvanized.

This

m ust not be confused with the above findings.
A 16 gauge s te e l sh eet w ill have approxim ately 100% m ore s u r ­
fa ce per pound of black s te e l than a l / 8 ” sh eet w hile the zinc co a t­
ing per square foot of su rface w ill only be approxim ately 13% le s s .
F r o m this we find that the w eight of zinc coating accepted on 16
gauge s t e e l w ill be approxim ately 75% m ore per pound of s t e e l than
on a l / 8 Ms t e e l sh eet.
C arrying th is trend of thought into the field of light irregu lar
sectio n s a testin g program was s e t up using 18 sa m p les each of four
differen t se c tio n s of different w eights or a total of 72 t e s t sa m p le s.
Units of this group w ere galvanized at th ree different tem p era ­
tu res of m olten zin c, nam ely 840° F ., 850° F ., and 860° F . and with
d ifferent p eriod s of su b m ersion approxim ately 1 minute, 1 1 /2 m in ­
u tes, 2 m inutes and 2 1 /2 m in u tes.

A ll accepted coatings w ere

photom icrographed and stripped to determ ine w eight of coating.
S everal things w ere apparent fro m the m icrophotographs.
1.

Within th is range of tem p eratu res and su b m ersion tim e,

the coatings w ere w ell proportioned betw een z in c -ir o n alloy growth
and fr e e zinc and w ere r e la tiv e ly uniform .
2.

E x c e s s iv e tram p p a r tic le s of alloy did not show in the fr e e

zinc layer excepting w here a rough s t e e l su rface cau sed p alisading
of the c r y sta llin e z in c -ir o n a llo y growth.
3.

The in c r e a se in alloy growth was co n sisten t at a d e c r e a s ­
3

ing ra te with in c r e a se in tim e of su bm ersion in the m olten zin c.
4.

The h eavier sectio n s produced a greater growth of z in c -

iron alloy with the sam e tim e of su b m ersion in m olten zin c.
5.

The variation in tem perature betw een 840° F . and 860° F .

had v ery little influence on the w eight of accepted coating.
The w eight of coating as determ ined by stripping, substantiated
points 3, 4 and 5 and added cred en ce to ea rlier t e s t s .
F ig . 1 shows graphically the re su lts of th ese t e s t s .

The

cu rv es in th is graph w ere constructed using the average r e s u lts of
a ll t e s t p ie c e s involved and a re a ccu rate only for the conditions of
th is te s t or for com parative r e s u lts betw een d ifferent w eights of
b ars in the light c la ssific a tio n .
F ro m te s ts conducted by H einze Bablik^ and from our own ob­
serv a tio n s we find that the fr e e zinc layer accepted by the work as
it is withdrawn fro m the m olten zinc is r e la tiv e ly uniform under
given conditions r e g a r d le ss of the tim e of su b m ersion .

We can

th erefo re cut this fr e e fro m our chart leaving ju st the a llo y growth
as indicated in the graph.
Having estab lish ed the gen eral c h a r a c te r istic of z in c -ir o n alloy
growth on ligh t sectio n s it appeared advisable to esta b lish this
growth for longer p eriod s of su b m ersion in m olten zinc duplicating
a ll te s ts in order to elim in ate recogn izab le e r r o r s .

1 - A lectu rer at the T ech n ical U n iversity of Vienna and m anager
of a la rg e galvanizing fa cto ry .
4

*

Q uality
\ j 2, 2.0

a.

6

a-1.50

i

Z

3
4

l*3l
-t.lS

86
.83

r

z:
ili
_i

CL

r

<

1. 10-

OWTH

LLOy

VO .90
K .? o
<

Oi
di
(0

L30-.
120

r <
2*0
( t at - ~o
p* 0
.j j— f X
h -O
xff
T ?
vtf4"
,71
5 *

5

.SO

A ppr o x im a teT R E t Et Z I N A L a Y E tR

.40
30
.3 0
.(o
,o o
12

E 2 Z Z 3

T l l i t o}/ uBMtR/ l OH in

MINUTE/'
./

^ O A T ' N ^

LlAHT

A ^ t P T A M d E r
©+

1 RR&6ULAR
f

l<3- 1

yfr<:TIOH/

/
/
Ml
k ./p -

4

l / 8 n by 1” hot rolled low carbon bars w ere used in this te s t.
A ll b ars w ere hung on the hooks of a sin g le rack, thoroughly d e­
g rea sed , p ick led , rin sed , coated with a zinc am m onium chloride
(ZnC 1 23NH^C 1) and low ered into m olten zinc at 8 50° F .

At regular

tim e in terv a ls 2 bars w ere rem oved from the zinc and air cooled .
The tim e in terv a ls w ere 10, 10, 10, 10, 10, 10, 20, 20, 20, 60, 60,
60, 60, 60, 60, 60 and 60 seconds so that the final bars w ere r e ­
moved after 10 m inutes of su b m ersion .
S ection s taken at the center of each of th ese bars w ere photom icrographed and adjacent sectio n s w ere stripped to determ ine
w eight of coating.
In F ig . 2 we have shown a com p osite p ictu re of th is growth
having ca refu lly s e le c te d the width of m icr o se ctio n s used to r e p r e ­
sen t tim e in the zinc bath.

It develops a picture sto ry of the z in c -

iron alloy growth in ten m inutes of su b m ersion .

Since a galvanized

su rfa ce is not en tirely uniform and s in c e each m ic r o se c tio n at 125X
co v ers only a fraction of an inch, any individual sectio n may not be
tru ly rep r e se n ta tiv e of the total coating but a re averaged out in a
p ictu re of th is type.
Independent of this com p osite picture a graphic rep resen tation
at the sam e s c a le as the picture was developed fro m the w eights of
coatings obtained by stripping the sa m p les.
It w ill be noted that the r e s u lts indicated by th e se two independ­
ent m ethods of p ortrayal p a r a lle l v ery c lo s e ly .
F ro m th is inform ation we find a very rapid z in c -ir o n alloy
growth during the f ir s t minute of su b m ersion .

6

^OOD

IhEEEA/IHL BRITTLEHE//
OuMCE/

QUAL1T y

TEM

A

MINUTE: / U B n E E / \ o n

IM M O L T Er l

Z 1 PIC

^ o m p o / i t e P i c t u r e / h o v/i!H<L A l l o y e I e o v t h
T h r . o u 6 h Teh M im ute L y l l e
12 5
Q

COO D

u a l i t y

52
IMEEEA/1M6

BEITTLEISE//

hJ
YJ
k /.fr

M1MUTE/

Zm C Growth

and A ^ c p t a m ^ e

D E T E R r i l N E D

Q u a lity

Mo t D i p

ky

by V / e i ^ h t

/ * T R. J P P 1 PS C

Q ALVAm zm 6

Li 6 h t

/e

D h a p t

l t io n /

Fig. 2

7

This is v e r y im portant sin ce a few seconds either way shows
up ap p reciably in the w eight of accepted coating and coating w eight
would th erefo re be hard to control in this a rea .
This v e r tic a l co lu m n a r-cr y sta llin e growth which is b rittle in
nature slow s down rapidly after a m inute of su b m ersion to a lm o st
no growth at about th ree m inutes probably due to the in crea sin g r e ­
s is ta n c e of fr e e zinc and b ase m etal to contact each other and alloy .
A ccording to D aesen^ this a llo y growth starts as Fe^Zn^Q in
cubic c r y s ta ls of a very thin layer and graduates through FeZn^ to
F eZ n ^3 in the columnar form .
At th ree m inutes of su b m ersion tim e the norm al a lloy growth
is apparently com p lete, uniform and le a v e s a b ase for uniform r e ­
ception of the fr e e zinc.
For the next two m inutes of su b m ersion , from 3 m inutes to
5 m inutes, the a llo y growth is b arely apparent, the quality of the
coating rem a in s good.
Photom icrographs do not indicate e x c e s s iv e tram p p a r tic le s of '
z in c -ir o n a llo y but through th is retarded alloy growth zone th ere is
unquestionably so m e breaking away of the uneven ends of columnar
c r y sta llin e growth.

T h ese broken cr y sta ls of a lloy being heavier

than zinc s e ttle to the bottom of the zinc bath to fo rm d r o ss, p ick ­
ing up zinc in th eir d escen t and showing about 97% zinc to 3 % iron
in the d ro ss fo rm .
This factor is reco g n ized in the old axiom , "“Dr os s begets
d r o s s .”

2 - Consulting M etallu rgical E ngineer
8

From, this covera ge it can rea d ily be seen that any reduction in
su b m ersio n betw een 3 m inutes and 5 m inutes w ill not m a teria lly
red u ce the w eight and quality of coating but w ill red u ce the d ross
form ation and w ill im prove the econom y of operation.
In creasin g the su b m ersion tim e beyond five m inutes shows a
rapidly a cc e le r a tin g new z in c -ir o n alloy growth which tapers off
again at around nine m inutes to p ra ctic a lly no in c r e a se .

In this

ran ge of su b m ersion photom icrographs show a decided in c r e a se of
tram p p a r tic le s of alloy in both quantity and s iz e in the fr e e zinc
layer which rem ain in the coating only by virtu e of having been
trapped by the fr e e zin c.

T h ese p a rticles a re much la rg er than

found with the shorter su b m er sio n s.

The indication is that much

m ore d ro ss is form ed in the zinc bath in this extended su b m ersion
p eriod.
To esta b lish this c r ite r ia , bars of each su b m ersio n period w ere
bent 180° over a soft m andrel.

It was found that the radius of the

bend in crea sed in approxim ate d irect proportion to the thickness of
the a lloy growth.
Exam ining th ese coatings after bending it was found that up to
fiv e m inutes of su b m ersion tim e a coating was produced that r e ­
m ained sm ooth and showed no checking with the bending.

With sub­

m e r sio n tim e s of s ix m inutes through ten m inutes the coating
showed in cr ea se d wrinkling and checking in the bending.

At ten

m inute su b m ersio n , th is fracturing of the coating was quite s e v e r e .
To r e v e a l the extrem e, t e s t bars w ere subm erged in m olten
zinc at 850° F . for fifteen m in u tes.

When th ese bars w ere bent

over the soft m andrel, the coating flaked off in la r g e a r e a s showing

9

ex trem e b r ittle n e ss and poor adhesion.
U sing th is extrem e com parison again, two 3 / 4 ” x 3/16" s te e l
b ars w ere se le c te d ;

one of th ese was subm erged in m olten zinc at

850° F . for 45 secon d s and one for 15 m in u tes.

The coatings w ere

photom icrographed, F ig . 3, and stripped to determ ine w eight of
coating.
F r o m a study of the m icr o se ctio n s at 250 m agn ification s, it was
found that for sam ple one, the alloy growth is the in itial norm al
allo y growth found in good galvanizing.
a llo y a re sm a ll and not w ell defined.

The tram p p a r tic le s of
In bending this sam ple

showed ex ce llen t adhesion of the coat with no checking.
For sa m p le two, the in itia l norm al a lloy growth is w ell defined
at about the sam e le v e l as for sam ple one.

T here is then a forced

a llo y growth which b ecom es v e r y ragged with larg e w ell defined
tram p p a r tic le s at and above its su rfa ce.

T h ese tram p p a rticles

show a co n sid era b le breaking away fro m the b ase a llo y growth and
much of th is has freed its e lf into the bath to form d r o s s.

The in ­

c r e a se d su rfa ce zinc ap p lies b ecau se of the roughness of the a lloy
growth.

In bending, la rg e a rea s of this su rfa ce flaked off partly

due to its ex trem e th ick n ess and p artly due to in cr ea se d b rittlen ess
fro m the proportion of alloy to fr e e zinc.

10

C om parison C hart

T h i c k n e 3S
Weight
S u b m e r s i o n C o a ti n g
T im e
P e r Sq. F t .

C oating
Ave.

%
Alloy

%
F r e e Zinc

Sam ple # 1

45 S e c.

1.7 Oz.

.0027 in.

54

46

S a m p l e #2

1 5 Min.

4.0 Oz.

.0070 in.

71

29

2 50X

45 S e c .

1 5 M in.
Fig.

B r e a k i n g a w ay of b a s e l a y e r on long s u b m e r s i o n
F i g . 3a

11

It is reco g n ized that the v ery nature of applying zinc by the h otdip method due to the alloyin g action of zinc and iron produces a
d en ser coating than in any other form of zinc application.
This d en sity is an added p rotection again st co rr o sio n and g ives
grea ter life s in c e it slow s down the s a c r ific ia l requirem ent of the
zinc to p ro tect the iron.
A s the a llo y growth is allow ed to continue after its norm al
origin al growth the com pact v e r tic a l c r y s ta ls b ecom e m ore t r e e ­
lik e and irreg u la r in creasin g the p orosity of the coat.

A s this

p o ro sity in c r e a s e s , the s a c r ific ia l co rro sio n of the zinc to f ill the
void s b ecom es greater and le s s e n s the effectiv e advantage of this
th ick er coat.

12

F le x ib ility of Coating
A u th orities indicate that for the usual galvanizing method it is
m ainly the th ick n ess of the a llo y layer which governs the fle x ib ility
of the coating.
We have found so m e indication, how ever, of the breaking up of
the adherent a llo y layer on l / 8 ” x 1” open hearth s t e e l strip s after
su b m ersio n for 10 m inutes in m olten zinc at 850° F ., air cooled.
This condition as shown in m icro sectio n , F ig . 3a, red u ces the
adherent quality of the coating and m akes it m ore su b ject to flaking
away to b a se m etal under s t r e s s .
Although this condition is u sually found from rapid cooling, this
was not the c a s e in th e se te s t s and the proportion of su rface thus
affected was com p aratively sm a ll.
We m u st a ssu m e from subsequent te s ts that the condition as
shown would be in crea sed by either longer su b m ersion or fa ster
co olin g .
In ju stic e to the above statem en t a ten m inute su bm ersion with
th e se light sectio n s g ives a r e la tiv e ly heavy alloy growth, the point
being that the heavy allo y which d e c r e a se s fle x ib ility may be further
handicapped by the breaking down of the ad h esive la y er.

13

DROSS
It is recog n ized in the industry that one of the major econom ies
in hot dip galvanizing is obtained by reduction in d ro ss form ation.
It is parad oxical that the con trols required in tim ing, tem p era ­
ture and c le a n lin e ss of work to control d ross form ation autom ati­
ca lly im prove the uniform ity and quality of the galvanized work
pr oduced.
Or in r e v e r s e - controls required to produce a m ore uniform
product of b etter quality tend to red u ce the amount of d ro ss fo r ­
m ation.
The s iz e , shape, method of heating of the k ettle, type of p ro ­
duction and in stallatio n d etails a ll affect the c h a r a c te r istic s or
p e rso n a lity of any se le c te d galvanizing k ettle.
In d isc u ssin g this su bject th erefo re any in stallation m ust be
co n sid ered on its individual m e r its and any data acquired w ill only
apply d ire ctly to the unit from which it was secu red .
This data m ay be used in form at and to som e extent co m ­
p ara tively for u se with other in sta lla tio n s.
A ccep tin g the data from a regular production schedule for a
p eriod of one year we find that by rearran gem en t we can graph a
straigh t lin e of in cr ea sin g production.

A graph of d ro ss form ation

is then su p erim p osed on the sam e chart.
Both of th e se lin es a re extended by extrapolation to zero
production.
A t this point we find a d ro ss production of 5500 pounds per
month which r e p r e se n ts the d isso lu tio n of iron fro m the sid e s of the
k ettle into m olten zinc at 850° F .

14

F ig . 4 is taken fro m a chart prepared by E. D ieg el of Julius
P in tsch Corp.^

This chart indicates the com parative solu b ility of

iron in m olten zinc at d ifferent tem p eratu res within the range of
850° and 1000° F . for a particular s te e l.

This curve w ill rem ain

con stant for the low er tem p eratu re with various s te e ls but w ill vary
con sid erab ly with the higher tem p era tu res.
With the introduction of production into the k ettle, two other
m ajor so u r c e s of d ro ss a re found.

F ir s t the d ross form ed from

the work it s e lf which is gen era lly in proportion to the work put
through.

Secondly, the additional d ro ss form ed fro m the sid es of

the k ettle due to the in cr ea se d heat input through the sid e s of the
k ettle to re p la c e the heat absorbed by the work, the radiated heat
fro m the top of the bath and the washing action of the m olten zinc on
the sid e s of the k ettle during production.
C onsidering this second so u rce fro m the k ettle and refer rin g to
ch art F ig . 4 it can be read ily seen that the d ro ss form ation w ill in ­
c r e a s e by a rapidly a ccelera tin g curve with in crea sed heat tr a n s­
m issio n through the s id e s of the k ettle due to in cr ea se d production
of w ork.

T his curve with e x c e s s iv e production w ill r i s e so rapidly

that it b eco m es incom patible with good econom ics through d ross f o r ­
m ation, handling and kettle life .
We have found that if the m olten zinc is held at 850° F . by
e le c tr ic a l con trol of the heating units a therm ocouple placed c lo s e to
the outer w all of the k ettle w ill r e g iste r 860° F . with no production.
A s production is in crea sed th is tem perature w ill r i s e to about
1100° F .

Since the production galvanizing is a p eriod ic p rocedure

1 - Published in Z eitsch rift D es V erein es D eutscher Ingenieure
May 1, 1915.
15

SQLTJBILOT OP XR03ST IK ZBTC

SOLOBXLIH

25

30

mwm

15

10

840

060

880

900

920

960

900

1000

T M P . OP ZISC - DBG. P
3?ki* carve 'will vary in the higher temperat-ares vith
steels of different analysis.
r

n « . #4

16

th is in c r e a se is fluctuating in nature and the avera ge with definite
production can only be estim ated .

T h is, how ever, giv es us a

reason ab ly c lo s e p ictu re.
Since with no production the tem p eratu re of the air at the out­
sid e of the k ettle is 860° F . and of the m olten zinc is 850° F ., the
tem p eratu re of the in sid e of the kettle w ill a vera g e only a few
d e g r e e s above the zinc tem p eratu re, say 852° F .

At this tem p era ­

ture the additional alloying action of the s te e l kettle over norm al
standing c o r r o sio n can be con sid ered n eg lig ib le.
F ro m th is point we esta b lish an area of in creasin g solu b ility
of iron and zinc with in cr ea sin g kettle w all tem perature due to in ­
c r e a sin g production in keeping with the solu b ility curve in F ig . 4.
Subtracting area of d ro ss for standing kettle and area of d ross
fro m k ettle due to production from total d ross form ation, we have
the area of d ro ss fro m the production its e lf as shown in F ig . 5.
T h ese a r e a s of d ro ss shown as A - for norm al d ro ss from the
k ettle, B - d ro ss fro m the k ettle due to production,

and C - d ro ss

fro m production a re su b ject to individual and combined c o n sid era ­
tion.
The area A m u st be co n sid ered m o stly in the origin al d esign
of equipment and when once esta b lish ed , d ro ss fro m this sou rce w ill
be rea so n a b ly uniform throughout the life of the settin g .
Mr. W. H. Spow ers, Jr.,* has done con sid erab le work in this
field and finds that a cera m ic n on-alloying su rface on the in sid e of
the k ettle elim in ates d ross fro m this area and in reducing contam ­
ination of the bath im p roves the quality of the work.
1 - P r e sid e n t, Spow ers R esea rch L ab ora tories, New York.
17

DROSS BASH) 05 H08TH’ S PBCOTCTI05

16000
14000
13000
12000

MOSS IK P0OKDS

11000

*0000

9000
8000
7000

i V ■ -■

4000
3000
2000
-tt f.'r* V«t7

w
s&
W
>*

.

I’-:

' •

1000

j-rr^

FHRCEBTAGB OP DEOSS SEC■ 5CEPTI.E

to
TOTAL DROSS
A- EZ3 - BOBMAL DROSS PROM ZSPTLB
0 .1 I - DROSS PROM PRODUCT105
B - ® - DBOfcS TECU KETTLE DUS TO PROD.
It.

t

PIS- 5

18

C ertain d ifficu lties in the application and m aintenance of this
cer a m ic lining have held it fro m u n iv ersa l accep tan ce in the indus­
try but fro m the im m ed iate standpoint of theory and p o ssib le future
u s e it would s e e m to be a contribution of major im portance.
U niform heating of the heat tr a n sm issio n area of the k ettle,
s e e P ig . 6, m inim um exposed su rfa ce radiation area of m olten zinc
and proper s t e e l in the kettle it s e lf a ll contribute to low d ross
form ation .
The area B is affected by the sam e con sid eration plus s e v e r a l
m o re.

The keeping of the work away from the sid e s of the k ettle

to m in im ize w ash is m o stly a d esign item .

The control of p ro ­

duction to rem ain below c r itic a l h ea t-rep len ish m en t to the bath
and the holding of the work above the settled d ross area a ll keep the
zinc bath clea n er, tend to red u ce d ro ss form ation, and allow for
better quality work.
The area C of d ro ss form ation fro m the work depends on the
preparation or c le a n lin e ss of the work, the speed of su b m ersion and
withdrawal, the length of tim e of su b m ersion , and the clea n lin ess of
the bath fro m suspended d r o s s.
It is noted that the in sid e su rface of the kettle is co n sisten tly
hotter and th erefo re m ore solu b le in zinc than the su rfa ce of the
work being galvanized.

This is not a d esira b le situation fro m the

standpoint of d ro ss form ation, but would be much w orse if it w ere
not reduced by the follow ing alloyin g action.

An a lloy c r u st form s

on the in sid e su rfa ce of the k ettle which if not w ashed or scraped
away d eters the alloyin g action, w hereas work being galvanized con­
tin u ally p r e se n ts a clean su rfa ce to the m olten zinc for alloying,

19

■LPe- & P Hole

PtEPH<M-.e

3*

RMC-R

h o l t e w
T m p. R « o

2 1HC

C oo pL e.

il? £ J E P rtoLfc
i JSLuRf1*

B oslH&s .

y ° P E r tT 6 -A S

/p e H T
v \

^ X x \X \
/

H o y

D t p

e

-^ j i o

pi

A LV A ti 5 2 I JS 6
f- Lg

k & T i L&

£

20

m o st of which b eco m es a perm anent part of the fin ish ed coating.
It is re a d ily seen fro m the foregoin g d iscu ssio n that autom atic
co n tro ls of tim e and tem p eratu re, scheduled con trols of production
and bath analyzation a ll contribute to both the econom y and quality
of production.
The K ettle
B e c a u se of the com p aratively heavy c o s t of rep lacin g a burned
out or d issip a ted k ettle, k ettle life has alw ays been a su bject for a
great deal of study.
The co m p osition of m etal used, method of w elding, annealing,
con stru ction of the setting and uniform ity of heating a re a ll design
p rob lem s that should be given much thought.
Production sch ed u les and m ethods, how ever, bear greatly on
k ettle life .
F r o m F ig . 5 we find that with no production approxim ately
100% of d ro ss form ed com es from the in sid e su rface of the kettle
and with uniform heating the p ra ctica l life of an idle kettle at m ain­
tained tem p eratu re of m olten zinc could be calculated with c o n sid e r ­
able a ccu ra c y .
A s production is in cr ea se d the total d ross in c r e a s e s , both from
production and fro m the k ettle, reducing the life of the k ettle.

The

p ercen tage of d ro ss fro m the kettle to total d ro ss red u ces in this
settin g to only 63% with fu ll production.

This actually shortens the

k ettle life but in c r e a se s the life in term s of tons of production.
A t a certa in point in the in c r e a se of production the tra n sm issio n
of heat through the k ettle w all in c r e a se s the solu b ility of this s te e l
in zinc to a point w here the in c r e a se in d ross form ation fro m the
21

k ettle in p ercen ta ge to to^.1 d ross r i s e s sharply, reducing again the
k ettle life in te r m s of production tonnage.

This condition is known

in the industry as “beating the p o t.”
We find then that either a low production or a high production
in te r m s of the settin g c h a r a c te r istic s is not econ om ical, and
p a rticu la rly in the high area tends to keep d ro ss p a r tic le s in s u s ­
p en sion thus reducing the quality of the produced coating.
With ca re fu l a n a ly sis of any galvanizing settin g and production
req u ir em en ts, and barring a ccid en ts, the kettle life can be c lo se ly
an ticipated.

ZZ

EFFECT OF CHEMICAL TREATMENT
ON CORROSION RESISTANCE OF GALVANIZED SURFACES
Phosphate Etch A fter G alvanizing
It is reco g n ized fro m the r e su lts of many te s ts that the in itial
co r r o sio n of a new zinc su rface in m o ist atm osphere is com p ara­
tiv e ly rapid.

However th is in itial c o rr o sio n or oxidation form s a

su rfa ce that m a teria lly slo w s down the co rr o sio n p r o c e s s .
C ertain a tm osp h eres have been found in England that do not
allow this oxid ized su rfa ce to form , in which c a s e the d issip atio n of
the en tire zinc p rotection continues at a com p aratively rapid rate.
It m ay be assu m ed that if the in itia l co rr o sio n could be r e t a r ­
ded e ffe c tiv e ly , co n sid era b le p rotection could be saved and the
effectiv e life of the coating lengthened.
Many organizations have worked to this end and their findings
have been rev iew ed .
Even in the field of phosphate etching th ere are many v a r i­
ab les to be co n sid ered in the exact form ulation, p retreatm en t te m ­
perature req u ir em en ts, su b m ersion tim e, resu ltan t appearance and
end u se of the product thus trea ted .
We w ill confine our findings to one s e t of conditions which are
com p arative and in d icative of the value of this p r o c e s s .
For the purpose of our exam inations, b ars 1” x l / 8 ” x 16” w ere
u sed .

A ll bars w ere hot dip galvanized at 850° F . at the sam e su b ­

m e r sio n tim e .
Half of th e s e bars w ere then dipped in a phosphate bath @ 170°
for th ree m in u tes.
O ne-half of the galvanized bars and on e-h alf of the galvanized,
23

ph.osph.ated b ars w ere placed in a s a lt spray chamber in a 20% sa lt
sp ra y at 100° F .

The rem ainder of the bars w ere placed in a

hum idity cham ber at 100% hum idity at 100° F .
A t the end of each tw enty-four hour period one bar each of
galvanized, and galvanized p h osp h ate-treated was drawn fro m each
cham ber.
V isu al in sp ection of th e s e bars was record ed and stripping
showed the rem aining w eight of coating which was a ls o reco rd ed .
The r e s u lts fro m this te stin g w ere definite and showed our
e a r lie r p r e m is e to be c o r r e c t.
1.

See F ig . 7.

The slig h t phosphate etching and application of phosphate

s a lts did not p ercep tib ly red u ce the origin al w eight of coatin g.
2.

The in itia l c o rr o sio n of the zinc coating of the galvanized

b ars was com p aratively rapid.
3.

The in itia l co rr o sio n of the su rfa ce on galvanized,

p h osp h ate-treated bars was retard ed beyond the norm al co rro sio n
after oxide p rotection form ation.
4.

The norm al c o rr o sio n rate after so m e th ree hundred hours

in both cham bers b ecam e uniform and p a r a lle l for both coatin gs.
5.

In retarding the in itia l co rro sio n of zinc a p rotectiv e

th ick n ess of coating approaching a recogn ized co m m ercia l e le c tr o ­
galvanized coating had been saved , com parably extending the a n tic i­
pated life of the p rotection .
T h ese te s ts w ere conducted in the laboratory te s t cham bers
and under the su p erv isio n of qualified technicians of a recogn ized

24

CQHPAHASITB GOHHOSION HAS® OP PBQTECTIVE CO^IHGS
UHDJfc ACCELMUSPS)

T ’E S S T S Q .

2 .0

Ounces of Coating Per

Sq. Ffe. of S u rface

1 .9

1.8
1 .7

1.6
1 .5
1 .4

1.2

1.1

1.0
130
Hours in

240

480

600

720

840

960

Humidity
or S a lt Spray
C abinet
30$ © 1 0 0 ° P.
100$ © 100° P .

HUMIDITY TEST
1 *
Hot Dip
2 Hot D ip
3 4 -

360

G alvan ized and P hosphate T rea ted
G alvan ized Only

SPBAY TEST
Hot D ip
G alvanized and Phosphate T reated
Hot D ip
G alvan ized Only
Fig. 7

su r fa c e p rotectio n organization-^ and checked in the laboratory of a
ga lvan izer
In addition to the co rr o sio n retarding effect, p rop erly co m ­
pounded solu tion and tem p eratu re and tim e controlled phosphate
trea tm en t red u ces the spangle or flo w ers of zinc to a uniform grey
s u r fa c e .

It a ls o provides an ad h esive b ase for d ecora tiv e paint and

for glazing compounds which w ill not adhere to a new untreated zinc
su rfa ce.
Subsequent te s ts of com p lete s te e l window units con trol galvan­
ized and p h osp h a te-treated have shown that in a 20% sa lt spray at
100° F . norm al zinc oxidation or w hite r u st is apparent up to about
200 hours rep resen tin g the fr e e zinc su r fa c e .
F r o m 200 hours to approxim ately 800 hours through the zinc
iron a lloy growth d isco lo r a tio n b ecom es m ore apparent.
At ap p roxim ately 800 hours red r u s t appears showing a co m ­
p lete but spotty breakdown of the coating to b ase m etal.
Up to 1100 hours th ere is s t ill su fficien t zinc on the su rfa ce
to stop any pit ru stin g through its c h a r a c te r istic s a c r if ic ia l action .
F ig . 8 is a m ic r o se c tio n of galvanizing m odified to show the
reducing action of s a lt spray indicating the long life of the treated
zinc p ro tection .

1 - P ark er R ust P roof Company
2 - D etro it S teel Products Company
26

MATERIAL.S AND METHODS

B a se M a teria ls
In order to p ro tect the many econom ic and quality advantages
in con trolled hot dip galvanizing, it b eco m es im p erative that a ll
p h ases of b a se m a te r ia ls and p r o c e s s e s used in the p rep aration of
th is m a teria l for galvanizing be thoroughly in v estig a ted .
It is g en era lly reco g n ized that low -carb on open-hearth s t e e ls
a re m o st su itab le for hot dip galvan izin g.

For con trol of u n iform ­

ity in the accep ta n ce of this coating even low -carb on s t e e ls vary to
a co n sid era b le d eg ree .

Som e of th e se va ria tio n s show e m b r ittle ­

ment of the b a se m eta l in the p r o c e s s w hile others produce a long
rapid growth of n on in ter-ad h eren t colum nar z in c -ir o n a llo y c r y s ta ls
with e x c e s s iv e p orp osity, which due to the iron content darken ra p id ­
ly at the su rfa ce.
F r o m r e s e a r c h by many individuals in this fie ld it can be
lo g ic a lly a ssu m ed that the e x c e s s iv e n ascen t hydrogen intake of
certa in s t e e ls during the su lp h u ric-a cid pickling is p artly r e sp o n ­
sib le for th e s e con d ition s.

The constant r e le a s e of hydrogen fro m

the s t e e l through the form ing z in c -ir o n allo y growth during sub­
m e r sio n in m olten zinc p artly r e lie v e s em b rittlem en t of b ase m etal
but c r e a te s unorthodox z in c -ir o n allo y growth.
Sam ple bars w ere se le c te d for a n a ly sis as follow s:

27

P R O G R E S S IV E CORROSIVE DISSIPATION OF
P H O S P H A T E T R E A T E D ZINC COATING
IN SALT SPRAY

R e c o n s t r u c t e d f r o m T e s t D ata
(See F i g . 7)

IN CR EA SI N G
W H IT E RU ST

0

100

200

I N C R E A S IN G
W H I T E R U ST & D I S C O L O R A T I O N

300

400

500

600

700

800

INCREASING
RED RUST

900

1000

1100

T I M E IN HOURS IN 20% S A LT S P R A Y A T 1 0 0 ° F .

Fig. 8

28

1*

B a rs of good ductility and uniform ly ductile coating.

2.

B a rs showing rapidly darkening su rfa ce.

3.

B a rs showing em b rittlem en t.

The s te e l of th e s e bars was analyzed and m ic r o se c tio n s
photographed.

F ig . 9.

In gen era l the bars producing the b e s t galvanizing without
becom ing b r ittle and showing a uniform su rfa ce of n early pure zinc
w ere of open hearth of .05 to .13 carbon, .040 phosphorus or le s s
and s ilic o n of .02 or l e s s .

B e s s e m e r s t e e l showing higher p h o s­

phorus to .09 showed con sid erab le b r ittle n e ss after galvanizing,
r esu ltin g in broken b ars under norm al s t r e s s .
B ars with high s ilic o n content to .09 showed strong attack by
the zinc and an abnorm al z in c -ir o n a llo y growth with p r a c tic a lly
no fr e e zinc accep tan ce on the su r fa c e .

Due to the iron in the

su rfa ce zinc and extrem e p o ro sity , th is coating darkened rapidly
with oxidation.
This ite m iz e s only a few of the m ajor req u irem en ts in s te e l
for quality galvanizing but d efin itely in d icates the need for contin­
uous and ca re fu l a n a ly sis and testin g in the s e le c tio n of b a se m eta l.
F ab rica tion
Many fa c to r s should be co n sid ered in the fab rica tio n of
m a teria ls for hot dip galvanizing.
Cold upsetting or sharp bending tends to c re a te internal
s t r e s s e s that b r ittliz e in the galvanizing p r o c e s s .
A nnealing of this work after form ing or hot bending and hot
p r e s s u r e upsetting elim in ates th e se in tern al s t r e s s e s and m in i­
m iz e s r e j e c t s .
29

A rc w elding with, coated rods le a v es a flux resid u e su rface
d ifficu lt to clea n and one that w ill not be adherent to zinc if not
clea n ed .

W ire brushing after welding im p roves th is condition.

Light coated or bare w ire in welding le a v es a su rfa ce that zinc w ill
adhere to.
P r e s s u r e butt welding of oily bars burns the oil fro m the w eld
but im p regn ates it into the s t e e l c lo s e to the w eld.
E x ten siv e experim enting shows that m o st ca u stic cleaning
operations w ill not co m p letely rem ove this im pregnated o il and bare
spots w ill occur in the galvanizing.
Cleaning with a su itab le vapor d e g rea ser such as tr i chi or eth yl­
ene r e m o v e s th is im pregnated oil e ffe c tiv e ly .
If thorough d egreasin g of a ll b ars is done b efo re w elding this
im pregnation w ill not occur and the le s s e r cleaning m edium s a re
su fficien t,
Appliqued m a te r ia ls coverin g s u r fa c e s b efo re galvanizing m ust
be c lo s e ly attached and tight fitting to p reven t the holding of pickling
acid s and m o istu r e .

With this condition the zinc in galvanizing

w ill co m p lete ly s e a l the edges of the joined p ie c e s and the c l o s e ­
n ess of the attaching w ill tend to elim in ate w arpage.

Hanging of the Work
It is quite obvious that hanging d ev ices and the manner of the
hanging of work to be hot dip galvanized is im portant.
The d eg ree of th is im portance is borne out in rep eated testin g
and d efin es it s e lf in the r e s u lt s .

30

S T E E L C H A R A C T E R I S T I C AND A P P E A R A N C E
A F T E R HOT D I P GALVANIZING AND P H O S P H A T E E TC H IN G
1
Light
Ductile

2
D ar k
Ductile

3
Light
B rittle

D e s ir able

Carbon

.07

.08

.04

0.05 - 0.13

M anganese

.38

.38

.33

Phosphorus

.010

.010

.065

S u lp h u r

.040

.031

.041

S i l ic o n

.02

.09

.01

Chr o m ium

.02

.02

Copper

.11

.12

Nickel

None

None

N one

M olybdenum

None

N one

None

C o a t i ng
B ase M etal

0.04 M a x.

0.02 M a x .

N one

LIGHT

DARK

LIGHT

DARK
F ig , 9
31

In a ll of the cleaning baths proper drainage of the work tends to
m in im ize the contam ination caused by the liquid fro m one bath
p a ssin g into the next.
This both aids in the econom y and in the effectiv en ess of the
p rocess.
This c o r r e c t drainage b eco m es im perative in the m olten zinc
bath sin c e uniform p rotection depends on proper zinc flow as the
work is withdrawn fro m the bath.
Many exp erim en ts indicate that in an open fram ew ork, no
individual bar should have a slop e of le s s than 30° to the horizontal
for proper flow .
When work is thus hung and slow ly and evenly withdrawn from
the zin c, the flow of zinc takes p la ce uniform ly at the su rface of
the zinc w here the tem p eratu re rem ain s uniform and at about 60° F .
above the m eltin g tem perature of zin c.

The su rfa ce becom es

sta b le fro m air coolin g c lo se ly above this lin e.
If the work is withdrawn quickly, it c a r r ie s e x c e ss zinc w ell
above the bath allow ing ch illin g during the flow and causing runs and
sags in the coating.
If withdrawn e r r a tic a lly , lap lin es and rid g es w ill be apparent.
This in d ica tes c le a r ly the advantages of con trolled m echan­
ization.
F ig . 10 illu s tr a te s the extrem e com p arison of r e s u lts betw een
the hanging of a fra m e so that part of the bars a re h orizontal and
with fra m es that a r e hung so that a ll bars a r e at an angle to the
bath.
If the h orizon tal hanging w ere to be done in a production group,

32

PROPER HANGING

IMPROPER HANGING
F ig . 10
33

it would be im p o ssib le to fe ttle off a ll of the drip or sta la ctite
form ation b efo re settin g or freez in g of the zin c.
The C ontrolled P r o c e s s
A s indicated b efore, the proper cleaning of m a teria ls is as
im portant as any other phase of the work.

The vapor degreasin g

r e fe r r e d to rem o v es a ll oil and g r e a se but does not rem ove dirt,
s c a le or o x id es.
To e ffe c tiv e ly and co n sisten tly do this work, tem perature
so u r c e and con trols m ust be su fficien t to keep the vapor in plentiful
supply at a working tem perature of 190°.

Inadequate heat w ill

render poor cleaning and e x c e s s iv e heat w ill w aste cleaning m a teri­
a ls and c r e a te u n n ecessa ry contam ination of surrounding atm os­
p h eres .
In the pickling p r o c e s s for rem oving sc a le ru st and oxides
many acid s and com binations of acids are u sed .

For economy and

c lo s e con trol sulphuric acid is m o st gen erally accepted.
With production work of a co n sisten t nature, controls of
tem p eratu re and tim e should be used for uniform work.

Since this

bath d e te r io r a te s by lo s s of acid concentration and pick up of iron,
uniform ity of pickling can only be obtained by scheduled an alysis of
the liquid with proper co rrectio n .
In crea se of tem p eratu re of bath and/or in crea sed con cen tra­
tion of acid , within lim its , counters the retarding action of in crea sed
iron content.
F or exam ple, as the iron content in c r e a se s with u sag e from 4%
to 7%, the acid concentration may be in cr ea se d fro m 8 % to 10%.
The tem p eratu re m ay be in cr ea se d from 150° F . to 170° F . or tim e
34

of pickling m ay be in cr ea se d .

Beyond 7% iron content, it b ecom es

eco n o m ica l to dump the bath and rep len ish it with fr e sh liquid.
Shortly b efore dumping, acid p ercen tage may be allowed to run
down to say 4% for econom y in n eutralizing with lim e.
Much above 150° F . the contam ination of atm osph ere from acid
c a r r ie d by r e le a s e d n a scen t hydrogen b ecom es a problem .

Longer

pickling tim e in a con trolled sy ste m red u ces the effectiv en ess of
that s y ste m in production by idling other phases of the p r o c e s s .
It can be rea d ily seen that c lo s e correlation of tim e, acid
concentration, tem p eratu re, atm osphere contamination, balanced
operation and operating co sts a r e vital to both econom y and quality.
R insing
A s the work is rem oved from the pick le solution, acid and iron
s a lts rem ain on the su rfa ce and m ust be thoroughly rem oved.
Hot w ater rem o v es th ese m ore rapidly and thoroughly than
cold and agitation a s s is t s m a teria lly .

H owever, agitation holds

this m a teria l in su sp en sion and req u ires m ore water changes per
hour.
A t th is point the b ase m aterial is so thoroughly clean that
sh ort con tact with the atm osphere w ill produce su p erficia l red ru st.
To a lle v ia te th is, one or two cold water r in s e s w ill reduce the
m a te r ia l tem p eratu re retarding this tendency to r u s t at the sam e
tim e insuring that a ll foreign m a teria ls a re rem oved from the
su r fa c e .
In order to p ro tect this clean su rface until it rea c h e s the
galvanizing k ettle an im m er sio n in a flux bath, usually zinc am m on­
ium ch lo rid e, Z n C ^ B N H ^ C l, is provided for dry galvanizing or

in the c a se of w et galvanizing a flux foam blanket is built up d irectly
on the m olten zinc through which the work m ust p a ss.
In the dry p r o c e s s , the tem p eratu re of the flux bath and its
con cen tration a re r e la tiv e ly c r itic a l to quality galvanizing and here
again rig id con trols p rove valuable.
D rying
M oistu re ca r r ie d fro m th is bath into the zinc would cause
ste a m spitting or e x p lo sio n s.

A drying p r o c e ss before galvanizing

is th erefo re req u ired and it is found that fa st moving air at about
150° F . does th is e ffe c tiv e ly .

A ir below this tem perature lo s e s its

econom y and if above 180° F . tends to d eteriorate the flux coating.
Contam ination of this air fro m fuel ga ses m ust be avoided.
When dry, the work w ill show a thin, shiny, brownish coat of
flux p rotection which w ill oxidize or v o la tiliz e as it contacts the 850°
F . m olten zin c .

We a re then, in a controlled p r o c e s s , a ssu red that

work re a c h e s the zinc clean and that good coverage and proper
adhesion w ill be attained.

C O M P A R A T IV E Z IN C C O A T IN G S

SHERARPIZING
In this p r o c e s s , the a r tic le s , after being thoroughly cleaned by
pick lin g and sand blastin g a re placed in a m etal drum together with
zinc dust, heated to 500 to 600° F • and the drum rotated.
The resu ltan t coating is about 90% zinc and 10% iron with a
m eltin g point of about 1260° F • and although it is highly re sista n t
to c o r r o sio n develops a red d ish or brownish color from the heavy
iron content.
The s y s t e m its e lf lim its its u se to hardware and other
r e la tiv e ly s m a ll ite m s .
The te r m e le c tr o -sh e r a r d iz in g m erely rela tes to the method
of heating the drum.
Coating w eight is a function of tim e but b ecom es uneven in
h eavier co a ts.
ELECTROGALVANIZING
In th is p r o c e s s the a r tic le s to be plated are usually placed in
b ask ets and a c t as the cathode in the reaction .
m e ta llic zin c.

The anode used is

Both acid sulphate and cyanide baths are used and

additional m a te r ia ls a r e added to obtain bright su r fa c e s.
The high d u ctility of the pure zinc su rface applied is its
standing fea tu re.

out­

For th is rea so n it is su c c e s s fu lly used in the

coating of w ire and is now being used in the plating of sh eet s t e e l.
Coating w eight is up to 0.2 oz. per square foot of su rface beyond
which the p r o c e s s is not econ om ical and much thinner coatings are
the u su al p r a c tic e .
37

Z IN C S P R A Y IN G
In th is p r o c e s s zinc w ire or powder is fed at a controlled rate
into the fla m e of any oxy -gas or o xy-a cety len e torch .

The im ­

pingem ent of th is atom ized m etal on a prepared su rfa ce produces a
layer of fla t in terlock ed p a r tic le s which a re m ech an ically bonded to
the su rfa ce being coated.
The w eight of coating is m erely a function of the amount
applied but is com p arativ ely porous.
The coat b ecau se of th is p orosity m akes an ex cellen t paint
b a se and is u sed effectiv ely in repair surfacing of fixed item s such
a s the in sid e su rfa ce of stationary tanks.
GALVANNEALING
A m odified hot dip p r o c e ss w here work is preheated by gas
pickling or drawing through hot lead in lower portion of zinc bath
an d /or fla m e annealed and/or ca rried through furnaces up to 1200°
so that total coating b ecom es z in c -iro n alloy with ex cellen t corrosion
r e s is ta n t p r o p e r tie s.
V ariations:

Z enzim ir P r o c e s s , Sharon P r o c e s s , etc.

HOT DIP
Wet P r o c e s s
C leaned m a teria l is im m er sed through a blanket of flux foam
into m olten zinc at tem p eratu re of approxim ately 850° F ., allowed
to com e to the tem p eratu re of the bath and slow ly withdrawn after
flux blanket is sk im m ed back.
D ry P r o c e s s
C leaned m a teria l is dipped in liquid flux (Zn, 3NH^C1), dried
38

and im m er sed in skim m ed m olten zinc bath at approxim ately 850° F
allow ed to com e to the tem p eratu re of the bath and slo w ly withdrawn
Hot dip galvanizing produces a z in c -ir o n alloy coating with a
c o r r o s io n -r e s is tin g v e r tic a l c r y sta llin e stru ctu re form ed by an
e le c tr o -c h e m ic a l action during su b m ersion .

This a lloy coating is

co v ered with a d u ctile, sm ooth, flat cr y sta llin e n early pure zinc
coating as it is withdrawn fro m the zinc bath.
Surface c r y s ta l form ation (spangle) can be am plified and
b rittlized for appearance with the u se of sm a ll amounts of tin or
antim ony in the bath.

On the other hand th is form ation can be

enlarged, subdued in appearance and refined with the u se of a sm a ll
amount of alum inum in the bath.

COMPARISON CHART

P rocess

U sual
T h ick n ess
P er Sq.
F t. of
Surface

Coating

C ompa r a tiv e
C orrosion
R esista n ce
C harac­
P er Unit
te r is tic s
T hickness

U se

SHERARD- V aries
IZING
with tim e
to 2 OZ. j
.0 0 3 4 ”

10% F e

Good

Sm all parts
Turns
br ownish
or dark

ELECTRO- V aries
with tim e
GALVA­
to 0.2 o z.,
NIZING
.00 03”

P ure Zn

F air

D uctile
Sm all parts
and even w ire,
sh eets

Fair

Por ous
and
uneven

Repair work
and
str uctures

D uctile
with
oxidized
fin ish

Sheets and
str ip

R eason ­
able duc­
tile and
br ight

A ll purpose
for m axi
mum
pr otection

ZINC
SPRAY­
ING

Uneven
to any
weight

Pure Zn

GALVA NN EA LING

to 0.375
oz.
.0 006 ”

ZnF e
A lloy

HOT DIP

to 2 oz.
.0 0 3 4 ”

Good
1/3 to 1 /2
ZnF e A lloy
plus
2 /3 to 1/2
P u re Zn

E xcellen t

40

SUMMARY
To in su re that any galvanizing sy ste m is both econ om ical and
capable of producing a uniform quality product many things m ust be
con sid er ed.
1.

A ll equipm ent m ust be studied and designed to fa cilita te

the handling of the product or products to be galvanized.
2.

C om plete p rov ision for the control of the m ovem ent of

this m a teria l through the sy ste m is e sse n tia l.
3.

A m ple p ro visio n for the com plete cleaning of a ll b ase

m etal of a ll paint, g r e a se , oil, ru st, s c a le and dirt is of prim ary
im p ortan ce.
4.

Adequate flux p rotection by predipping or flux blanket

m u st be m aintained.
5.

C om plete and autom atic tim ing and tem perature con­

tr o ls produce uniform ity.
6.

P r o v isio n for scheduled laboratory a n a ly sis and control

of a ll ch em ica l and elem ents used in the p r o c e ss is im portant.
7.

C ontrolled an a ly sis of b ase m etal to be glavanized is

valu ab le.
8.

C onsidered and controlled fabrication of base m etal

in su res a g a in st r e j e c ts .
9.

A dd itives in coating p erform an ce, such as phosphating,

a re d esira b le on many products.
10.

Good housekeeping and continuous r e se a r c h testin g for

im p rovem ent in the m ech an ical, ch em ical and e le c tr o -c h e m ic a l
p r o c e s s e s w ill im prove quality and r e s u lt in im proved economy.

41