l

I

,l

we;
l‘_',. v44 ', A.
k“ ' 3.11531":

7 v

    
  

I

.
,-\
'r, 1'
kg,“
‘Ly

 

p

  
 
 
 

  
       

   

 

   
  

  
  
 

  
  
 
  

.I'f
r.
‘3

4r

  
    

s

\e’:*‘

  

   
  

1
, I

    
 
 
  
  

1..

:1‘5‘. . ‘
«a w 7' ' ”k;
5' 7‘ ' O .71" I 3‘" m'd‘.f':é..‘_fi;. -. ’ ' " .' ? ... \D 3' ; . .g'; ‘ ‘ f .. a . hi. ‘ ‘ .
$.53 5’ 33'2”? W?- ~ .-"~'- 9:.“ 44$, 1% ~7- .‘ - = ”firm
I I .I.. "’ i‘-“‘. "' "I " r ‘ -'. ‘72:. . Zn' ‘7 Ldfi'fi ' -- M" . 31.13.
. L 0 u ”- . . . . k ‘0 ':‘§ 1%“ ’31: '
o {{M$§¢N1§;?I£LEQ;.W“?J.‘ .-
I . _ 3
‘ #7- , '5. I!” 4&‘é‘y ‘ m

L. b. H's";zn§f'.r 4' ‘ , “"f A ’. . '. a t "‘ ‘.'
”flu. n..',‘n:. "I it“ I, - . .' . - ’ " ' . " ‘ r . ' ‘ . v 3-. '- ' 3.‘ '3 - k '..4
..’,.;£#‘: '3”? .. J. ' ’ | ’,J , _ ’ L . . - u , .- , . a {)1 ’3‘ ~\ . ' . . _- -’ b‘req.h‘;$t::.
._ _ > ‘ ' .U.‘
. - .. $37319:
. '.-, j .. ., .~ .' z . 5}; L 55‘Q17" " , ‘ .' his“ I" Q. .431 ‘3‘ ' L“; f 1‘ . 7' _,.; 3 ’l'vfi’fi'
_ .. L' . . "I. ‘ l 4 ..A . 2' . _ .n_ h ' ! ‘ V I: . a ': - , K: .L _' V. -I' “
. . : --- 35‘ . L .5 'a as {L aw . - u. . . ‘. .52 . . LL
. - . , ‘. v -|. L 'k .' . ,l , 2“}. ‘ M M .. ‘. ,...‘ . o i " . ‘ ... ‘| “a; - .‘L~&U"' .. ‘l
| 1 .j ' . I. . - "4113‘ 9" ..., . . ”3 5‘34 j I: z .u': L,- 22'. u ' 5 Jan-'5 053‘“ “1’63;
, J "I. I‘, , .< R f 1. . J‘ '13.?) o “'5.“ 2" - I ‘ , ‘ ‘1‘) ‘1, éfir’
._ . . | . v . ”523%; L “‘3’. a? . f. M ‘
’ '1 J»; ‘ : .

\\

    
  
 

  
  
 

      
  
      
    
   
   
 
      

 

- J: {3.1. “0’ ‘1 1.7;:qu 9“?
L 1;: ~ ~. ' m: . a w "v.1" “
' “332$? 5% i343 9": ,3
V. .{5 ': , ‘7 ., . . ‘3‘ '. ’“ ',
‘ - .
a .
I 0 O
.. §.
.ti :- 7"
"' . 57.
-.'£. ;. .‘
I t - n- o -
" ‘0 I '
.l‘ s o 0 o v ‘
. f‘ . ‘ . . . J .» . .. . .-; .- 1| . 7' , . m.
. . 7_ -,--:g '- * ~ - L 43"?!»v2fi2fi4fififis‘m
. ' ‘H t ‘3 .. ‘l‘c'?3t§:
I. 0 a. o t .. ‘ . ‘.
.. l o ;.: ' . ‘ gn‘Lf'I‘Q‘f” ...
' ' O..“! 4 ‘ ‘ I

9'.
i

‘1'
.g};
N

0
n

o
u‘
".

2:; . ~
‘3‘ \3

s
I

:1
I.“ 1

   

° 5
o 0 g . 9 .
o . . . :‘h'fih ' , L ~- : y . , 4 - . ,- ‘ ' . -' ' ' .
4-H, r . - . - -~ < - ' , , ~ . 2 -‘ ‘ a . A
. o 0 I ’5". {33’ ‘ . ' . ' . V ‘ v ‘ I ' r‘ . ' - ‘ I' ‘ 3 ~ 'a~ . u , b , '. . . . z
‘. o . O , -"l ' ‘ ' I ‘ - l . t'. . ' ' ‘ h,, 7' 1 ~ ., , \ .

. ~"“(u I}
N. 32'? $354, fig}?

   
  

   

  
 
  

 

    
  
     

    
  

    

 
 
    
     
   
        
   

      
      

    
 

   
    
 

 

  
  
    

    
   
   
 

 
 

      
      
  
     
     
      

 
      
      
 
 

  
  
 

   
  
    
  

 

  

. ‘ I ‘ . ' "(ft/.31“? i a. L- , "i. L , - '.
. . g .r 07‘ ‘ ~ m ’ . ' V .' . '

. ... . . , . «am LL. “‘79 , L . - -

. .. - - . . c..w..- June. 5
I - .-. - ‘ M .. J: ' 1.! ‘5' &.v - ... . 1‘ c
. - : .‘ . O -..“;115‘ ,g I . ..., 4 . «u 'Lm“ fitgififly" I
o . '- ”-.u ' i "‘ V . " " ' ‘.‘t.‘*~ -‘.'- ”3%. ‘ -' - .
’ 3" r. f ‘fist‘fié; ,‘WMLJ 4, ‘. - L. V. 59%;: " ‘ " L ' . 1.3.2." u
* ~ '- “ i 'L'. ‘L‘gst-Hé‘zsw J3; - L . -' = _ >63. LL; ': mg.
' -'~». .3 : .-, 9 ‘ . , . .~ ' ,.‘ - "-u. ‘
1 , m M, L w- - — ' 3’“ L 9 ..IL - ~ "3r LL . “3““
‘ ‘ ; L" ... ‘ J" J .. ' _ . '5 u4 .- ‘ . ' $.15“: “4"“: ,2,“ 0“)?4.g.‘o "1° $1.3.“ 4
, ' , - ,. ‘2'“ A ‘ '~ 0. . ' L - ' ' I ' . l “.17 ‘ ' ‘ b

’ 1 .. _ ‘ . ‘. . i, . , ~ . Iflv-afflg‘ggsfig‘f‘ ”é: ~. _ - 1' S .' m&;33%khfir§‘5*m Qt) E’oafiai'ibwfgg ._ u g‘gizy
: I - ' ‘ ' ‘ ' .... ”0 ‘_:¢ r. .'.!5r‘ . ‘.' . . 4". o: :1 5‘ _ ,‘t .. -\ _ ‘I‘ ,L; ‘3 I, .‘u." f ‘ .- . q‘- '1
’ -' - - ; ‘ -~ ' . ”-Nsfiflb‘ 5;; ' ‘ J ‘ ' ' a. ‘ulhxfizgfiué‘ 533:1 “.‘s‘i‘tfiitf .1‘5 'L "V3? 4 Lafis’éfiI w
' 4;. z~ -‘ -‘..°'- 4 » "-f ”$“K” ‘ -. mulf‘fii ‘ ' " -"N~‘"4"v"?"”° . .‘v-n Him???
. i ° . ‘ 5:49:21, 13%}: ‘13:}. ' ‘ ‘f‘. ‘ ‘3 ‘ :t'“"i.:'- ‘ “$5.; 5 “575:3 $.33?
' fi’bfi‘ - ."L’. ’:“~.', . . 1' '~ _- ; h on. ‘1. 43W:
. . ‘L L. ‘_ g’._ , \ ‘ ,,-.!.. r. 1". . Q... ...,I u 9% n 5 II*- N

‘ fi‘ ‘ ‘ " .‘J 3‘ "£5 q, H' * .‘ "‘h‘3 “ 0‘ 4“ ‘t ‘u" “a
m - 'v-‘.:‘ ‘-<‘ ,‘_;~9' LS." «"h‘ ‘5 “Q '.
. % f 6"; r-A" 3",.1i"|:*. ‘ :5.“ Kiwi-1‘" “ “237.5.
‘ ' ° C; ' '{Syfiékfisfi-Skg3'3iifks'é-N'fir'.5"; M}:

' ‘ “-" ’ :‘u w x 2- - :L. .~.L~. ~L~*~f..

‘ ' v!" o 3; . 3“ Eff-$9} “ ' " 3““ 5 ti

3.
I
3..
.‘r
97
x.
.1

 
    
 

£
(4
:J;
4
fi
9
i;
#3 .J'
- .-.“
af-
4“:
x U ‘
(I:
1
vi
( ’
1‘ 3‘"
37 -.
Vt . .
g»
.45.
.r t
w“
J O
‘ '2
‘2
he.

77'.

1'
1r
i‘ih’
if:

 
 
 
  
  

 

     
  
  
 

 

a?

  

Ln" 'L :- :' ..' ‘—."~c"'
.39rޤ-6:S:":-?j'%50 ( , '.'
“’ my: Law: \‘.‘1 ‘ ‘5 "a: ‘ “
a, “11334-“ "s‘ J‘coj‘ . ...”;’~'.luv k; "T ' “.7
'1‘; . «gw w 5‘. 1’15” 4" 1*“ 313‘ ‘ "’4 -’-‘
h'. . -~;~3V_'t« ~" '.’-I’-"‘,a?-" " ' ,
I ..,. q {‘Jy‘ 1 ‘
. ‘0‘.. "'V'A‘w .Hfi‘. S
i -‘d L ‘ ‘ 1 ’ "‘
.m'. "t':.~‘4'=vfi"n 3-"'“~"C"'-:S ' u ‘.
~ L‘»...~J-;;s..“ «$15.2: ~2?’~'s‘r.v hm“. =~
' 919m Fifi-”i311 ‘3'QMM“LY-‘.J 1:
.149 :'- '- ".‘.u-r""
. Etta-Ii?" In. fifiifiifivt’gujfngL-‘l,’ ‘ '
z€L+=N ..I.. an. 36353"? 3;...»

.‘J‘fiot- ~7'J'“ k" - ' 3- ‘ -
-m»eL-~-.LL~:.|L=:L¢~yv 5a rm '-
‘ ' ‘4' °.

    

.'—:-v
“J's
r.
' rr'..~

  
           

2:»
...léé:
1;
c" s‘
-1:
.7 ~ . I"
“::g.: , f (v.1?
‘ iii-Tit"
... q- - ,
4‘).
"a
‘ I
ff»

           
     
   
 

      

   

 

       
  
  

~.

3”:

      
  

A’_r _

‘2
..

        
 

 

2;“W-etg;§;'
'c ' ‘15". “:0"
‘2 f-‘Wdiw .'

t'um“. $.51,"

l

f.
5-"-
;.

."'~"q".'.75‘
{t}: 11'
'."_v-15 ’

1
n

1;.

.. I
9

'a"
Q
..‘N;,

a

fit,
I
:7; .

1’.
1
tiff!
L1.
”-23-”-
.2
’5‘.
r
g:
‘2
’12
'1‘
’t:
:j‘
s!
4‘.

do
..
4..
2‘0.
vv,
.‘o
l.‘
._..
._L,
:4
I3.
.._ "v
:1:-
‘3
,.
'1
‘4.
IF
’ «I
:1
Va
*1: .

1: ‘
J
. 5‘
’f‘
I

:'~0
”1'

. I?
5.7::
.3.
La

...
. '4
” .4
‘
'0'! n

:..

.3‘
.
Lu

_v :‘F 4"
l- a

’ "

‘ '. J.
¢.

1.,
,-
I

.4
v
o

‘ -1.— .

.1 '~

1 Ir,“
.: .53 ' 1.:-

‘9"
{L‘
J
' C" '
9‘?
f
i
'1
sf
. if

a??? »
If!

 

$00.".
- II L.
I.
"
I
f
I
4

.
.0
o
.lv‘
..
1-

I l
”I
J:
J
3) -’
v13.

91
A.
,-
.-
”'1
!
’.'.‘
.JI
.,.

Eris

    

4.
J"".’
o
-‘_
.‘2' 1‘-
to

l:‘ '
5?.“
A}.
'1‘
I'
u .
f

1 I: 'r’;
rt.

3‘ ‘. ‘ ~ '. .
‘1‘;? “‘_§',§~:JY(:I4?O ‘hv1ty

.1
9.91
I. I
.m.
.‘I At.
;‘)’f
3! ,7"
1,
U. ‘
‘IA-'.'.
S.c‘
. ~. '_ ‘ -4
r ' I',‘
'
,’
5,-‘l r

:
'r.
.' 'I'. .
.. .
.r'.

V.’ I L14‘.;
. R, up ., 'I w‘p'. ‘."R4.5"!4."];".‘u’-”'-~.5J‘, :l‘ ‘-
:' “no; 4" " -‘ "7“! 'v int-9L, ' '.' 5." "‘3'?“ ' ‘~3 v -
'.t~r.-t1‘-&.;'Qufi1‘:';-53‘z’fLJiJQ "‘2"? ”..r’" "«‘>"""~“1;.-}“‘~ xviik‘r‘t"

..
.I
.

g.
‘
i

‘ .- “."3.' 5 x.

,n

'1
,f:
I-
’1

‘f
- ’9 .I

' , '.-'t. r“ h

'5.
VI
0
'I
1
I
a"
i.
o

’.“
for
" 1:1
o‘ -‘ .-'.:.-i.‘.'.

“f...
1"
-r

t

‘I

5

f
3!,
“.l I c-' ‘ l
(

O

’1

V

‘

'

'

.

- ,
"(LII ,"I
" I

‘

1‘

f
5!“.
1?"?

’4

v~ . . I ,‘ ' . . ,. .‘ ‘ - II' ‘
:an Mb"-'.-.‘:"s"‘*.:r.‘v-<.w: -:'.‘».~.?~:.~s<.y. Lax!” "friwrn-"r‘nf'mo win."
. . _. V .0 .H‘. V! w , 2-": . ' Ii ": _ ‘ \L . . \I .‘I
‘?-.. - . ' - ‘n -;~’~1 "‘fi‘u’hfl '3-2‘ m u: #:L’1°v-c’.'. 3‘3? ’1’»{'a’:"'W-I‘i'c3 3.‘ ‘3.
' ‘ ‘< - ' II .‘ ‘ILfir‘ I’l’»:4 i‘\“. JIM ‘ ‘ ,' d . "I 4‘ v“
.. .q .1 a ,- x .

.rfi I! w. .- ‘ II. 5 ”4‘.

I
o
n

I" [:1
!l
'.

’ c
If:

t .
LI '
R
Is.

O 0
:— n

J

U 4-1 L
....

h

’ K
Ii
:4)»

-t
"In
I
a
r

"1‘ ‘ vie "0:39.34“ “H. .‘kn -, H 'w
.5: ”I. no.“'.rw.‘.r‘ ,,,‘.u. >0 0“? L "Q‘nhi(°"il- Ih‘ \

’ .-
Too-W»! A

Ofi
2.“
.1
‘.
A
ot"

0v. ”‘1‘
_ ‘ . 7‘
o. o . .‘_‘.' a ~b l”. _ W kg 0 ‘ \u.‘ ‘ 1'. . I ‘ - 3' ‘1 , . ‘ ‘ , 0 .
. O. . :. ...' .. _ . . . II . . _ , ,p Ln 3-. v ‘ .
Q: ‘ . ..‘I ‘42....LLr'q'51-uf~f-;\‘.;.I.- 2.31... ....-. H. I’ ': f ... .O 3" ”Mg: ("I ‘4’.~. \ ’. '3 . ,‘ ..r, I‘ 'L. :‘? .... v ‘ L“ b" ,L‘ , . .“ “q H &,‘ I 0
‘.,' . ‘ ‘. . I ‘ n. ,.' . c pa '- .¢ ' l' ‘n -
4%" ‘h7-‘. “ ' "."L":H‘.'".:".L': @252. ~ 3“." 333'” '.'-.“t"‘.‘ :7'“..‘-§V 1"" '3
h ' ':- ’ , ‘ . , L , . ‘ ’ " n .- '~
“’+ ‘~ "- ‘_' ‘; 4.337.” ..‘; .L ~...'\.I",.f.€" ,«Is, *'-‘~..‘w.,'w Q -s’
_ . ... . , . . _ 'u‘ ' 6"." I}..~I_-
of“ ~"‘o|’\,:., . .‘u u‘. '0‘," H“"“ y. '
L , . .... v: - . ' , . ‘ ‘t . .
‘, 1:1“- ,1 I4:"«‘.’.7.';u'.'¢?", {PC/... - :I «‘4 .‘3Ccu- L‘. .
I‘m-... “fl "1 N 3 :1: '”|\"“."¥1«.~'T~U . ..-, it, |".:.
.; ..‘. J .h 4",. A}‘:“\‘ '1 U ' -"o “Ov..?'p- '
‘0. 0 L. “ "' ' ; " $ ‘ 0“”:V ‘3 ':v '~

: -' . '. '\ '3 - I
, ,, .... , , .. 14 .‘V , . . .l't'?‘ -. [U 'Q- It :
..._ ’ ... .‘ ‘ . u u... ‘. . ‘5 ~. 1" c'"; r 5 IV ... ‘I ._ J 3114‘“. ”*4" u“ '31": ‘.:_- .. ku'fl, ’4‘: ..y.‘ 1.; H.112
~ ”‘?.’;-'*§’:‘-- 'a‘h‘i. l3: V‘C‘l. ‘32?”3' ° '~ ',. "‘- “5" ‘21“.“3 '34 ‘31:» 3'! "‘ -."“‘~5 1‘: ,-
’ .

 

)

&

WW OOE' CWO!“

egg:

 

Some Observations Of The MicroscOpio And Physical

Changes Resulting From Heat-Treatment
And

An Investigation Of The Dilatometrio And Thermal

Changes Resulting from Heat-Treatment
Of
High.8peed Steel
Thesis

Submitted To The Faculty
Of
Hiohigan.8tate College

In.Partia1 Fulfillment
Of The
Requirements For The Degree
or

Master Of'soienoe
Edwin, A? Brophy

June 1933

\ 2”
.1‘5:

I \ -.J

ThEiSrS

 

 

 

ACKNOWLEDGMENTS

The writer wishes to express his gratitude
and thanks to Prof. H. E. Publow for his

unending kindness and many helpful thoughts

and suggestions.

The writer also expresses indebtedness to
Mr. N. I. Stotz for samples and information
of details of practice this heat of steel

underwent.

L.E38 T R A G T.

This thesis, a study of high speed steel, consists
of a short historical sketch, the manufacture from
the raw products going into the electric furnace,
the chemistry of melting and the fabrication through

to the final annealing operation.

A comprehensive metallographic study is shown with
photomicrographs of quenching temperatures from 14000
to 24000 F. A similar study is shown of various
drawing or tempering temperatures; this includes the

recommended hardening practice.

A microsoOpic study shows results obtained using

several etching reagents.

A very complete dilatometric study with curves from

the Chevenard Thermal Analyzer is given.

 

 

. . '
. . ,
. .l , .
u n
x ‘ A
t ‘ -
- o
a - - ‘ .
I
. ‘ _.

-1-

HISTORICAL

Before entering into a detailed study of this analysis as
used today, it might be weal to consider the history of
our present day high speed steel.

Originally all tools were made from a good grade of high
carbon tool steel. Inasmuch as the design of machine tools
was not up to present requirements of feed and speed, these
tools served their purpose in a manner. However as the

speed of shOp machines were increased, a deep need was felt
for greater efficiency and performance in tools. The result
was a tool steel deveIOped by Mushet in England and generally
referred to as Mushet's Air Hardening Steel. This was a high
manganese (austenitic) type of steel, and received its name
from the prOperty of hardening when cooled in air. This air—

hardening property was due to its manganese content.

With the end of the nineteenth century and the beginning of
shop efficiency and big production, it was imperative that
tool costs be cut down; also tools that could stand the abuse
of the improved machine tools. much research brought forth
the epochal discovery at the beginning of this century of our
‘present 18: 4: 1 type of high speed steel by Taylor and White
of Bethlehem Steel Company.

-2-

It is probably the greatest single advancement ever made

in tool steels.

The treatment which they recommended for the standard type

of high speed steel has changed but little since that time.

During the last few years special high speed steels contain-
ing high percentages of carbon, chromium, vanadium, molybdenum,
cobalt, etc; of varying proportions have been developed. The
treatment for these steels will vary somewhat from that recom—

mended for the standard type.

The cost of these steels is prohibitive for general utility‘
purposes, and they only find application where the cost of

tools is secondary.

-3...

MANUFACTURE

This study was made with samples kindly supplied by the
Bracburn Alloy Steel Corporation, and according to their

analysis is as follows:

 

 

 

 

Carbon --------------- .69
Manganese 7—? ——— .18
Phosphorus — — -.014
Sulphur --------------- .011
Silican --------------- .29
Chromium: — -4.09
Vanadium — _=1.07
Tungsten ------------ 17.65

The above is a standard type of what is generally referred
.to as the 18: 4: 1 type of high speed steel and except for
the higher carbon content varies little from the original
analysis as brought out by Taylor and White in their epochal
discovery at the beginning of this century.

Commercial high speed steel is usually made by the electric
furnace process. The Hercult arc-type is most commonly used.
Some of it is made in the electric induction furnaces on a

small scale.

This particular steel was made in a six ton Hercult arc fur-

nace. Figure 1 gives the actual history of the heat of

v...

Figure I.

0'
'2

s._.

Melting Lo

\ c

 

 

.97.: So: #06. S... :0. To. 9 . 1m? .22...

x—JUKL IF?

 

 

 

 

 

......umn. can

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

k. 4 . zinc“. ozm

b. ..p. .2} km .N n». 2325. he—

nmhmoamc 232k qu.~ou.t 00$! “HF Link °m\.é «NE DuluOCO

m1... as; .00 ..Z .023». .<> do .3 Jam .noxu .22 .0

0 0%.? 1.3.: 2.2: ..flao. hev. W cz.n.o_x0uo
o 912‘ awn“ ..-- 3 .. . ......mmm
...»... _ “03.. .m .. l.\ 000 \\\X “waned. ”H.“ 0.... .. V 205-30..

8.. an...” 8 a. .. m .. a imam ,2 2... w

8. 8.... . 2 11%.. w .5ng “new“
.5955 20:150me #1053 min 3.002. .5 .02 ”n...“ \n< .W Qmmmoh- k. .0. bum: manmmnuun

... l<¢0w ZO_PUDDOKE ....(mI “.0 ”Err
.553: ll ...:0... 45.0...
. h.“ .

u... p. N ,3 .ma

0 ¢~ .(> .u...

. 4W.» 33 by» Ed ~\M.>\ W umah<mua2up 0m. . ....m .9...

V5.0 . must... JnNtS \\*0t§w \x‘- momaoz ...0 203.5200 I 1w .2: .uu

Noam ire—vb 391:0 29.20200 «lz‘pc 13% OHN u:0m:0.ua

1. tax... scth<§o N\\wm{\h. zobhOm mo zoEozou 00». \nb {cu oumhmwn.’
- uzokmuz... Oohx lMJUNfuI.
mxmfzum I Oink utmooo<z \K N ha

1 | 930025..

4 3.50: 02.3 000x 4.3.?

>m0»<m0m<.. 3.00 0 0.3 v.9;

muoomhommu 0239‘ 00% m<ammommu 00%;. «.95 bx on.»

.255 .hNN m:... 00.0” Men?» 6.!

.2; .mm: usa<0 23.25 455.522 auoo< m2.» 5.10.25 4:51
m><4u0 . N°¢(IU U_JJ(PU2.ZOZ UOK<IU U.JJ(PM:
.. a: .5... .. “new“. . .... ....u.............. a. and... .u “when. 2:. .2... 3. a . . .... «5. .... ...:

PZMEPE<¢MD MU<ZmDm UFFUMJN

3.2.0.2u

 

-1-

steel from the raw materials to the steel in the ladle

and ingots.

At this point I think it might be well to make brief men-
tion of the chemistry of the electric arc melting process
for steel manufacture. The standard practice, as used in
this case, is to charge the scrap high speed steel, nails,
virgin iron and necessary metallic alloys into the empty
furnace with sufficient lime to form a protective coating
on melting. The current - low‘voltage and high amperage -

is immediately turned on.

This charge was sufficiently melted in two hours to start
adding flourspar to make lime slag less viscous. Where the
charge contains such impurities as phosphorus, sulphur, and
manganese, it is necessary to first run an oxidizing slag,
sometimes called the dephosphorizing period. This period
consists of the removal of un, Si, and P, at low temperat-
ures by the following reactions:

Hn-oFeO = uno rFe
The Inc rises and dissolves in the slag which is later de-
oxidized by the carbon.

Inc «C . CO *Mn
The In returns to the bath and the Co is liberated.

Sio-z Feo = SiOgo-z Fe

The 8103 is fluxed by the lime slag and the Fe returns to
the bath.

-5-

soao + P305 = (0130) 3-9205

The calcium phosphate goes to the slag and is removed when
slag is scraped off before finishing or reducing slag is
formed. If impurities are not great the oxidizing slag is
not removed, but made strongly reducing for finishing by
carbon additions in the form of graphite, or charcoal or
ground coke. During the oxidizing period a small amount of
sulphur is reduced from the metal: I

FeS+ CaO+C-Fe+CaS+CO.
The ideal change for high speed steel should be so prOpor—
tioned as to require little or no oxidizing. Thus with a
carefully selected charge having a low percentage of non,
metallic impurities one slag only is required. This
eliminates the necessity of removing the first or oxidizing
slag and exposes the molten metal to rapid oxidation by air.
Using the one slag throughout the entire process necessitates
careful handling and constant ”doctoring” with lime, spar,
and coke to bring it from a slightly oxidizing condition to
a strongly reducing one. The final slag, to ensure complete
dsoxidation, should slack from the presence of calcium

carbide.

If the bath of metal has been well rabbled the presence of
CaC2 should indicate absence of oxides:

CaC + SFeO- SFe + CaC 4- ZCO
The CaC8 being reacted upon by FeO, or other oxides, cannot

long exist.

-6...

By reference to figure 1 it will be noticed that an addi-
tion of 285 pounds of ferro-tungsten was necessary to
finish with the correct analysis. Large additions of
heavy metallics such as ferro-tungsten while finishing is
not considered the best practice. Ferro-tungsten is heavy
and difficult to get into solution. Steady rabbling at
such a time is necessary, otherwise the finished product
shows tungsten segregation and considerable heterogenity.
Therefore great care should be exercised in the original
calculations to melt down as nearly as possible to the
desired.fina1 analysis. It should be noted that an addi-
tion of 25 pounds of zirconium silicide was added to the

ladle on tapping for degasifying purposes.

-7...

POURING INGOTS AND FABRICATION

This heat of steel was tapped in a slightly cold condition
(see Figure l) and therefore no time was lost in teeming
into ingots which had been heated and thoroughly smoked

with resin fumes.

The ingots cast were 12” square and S” square with the big
end up and a heavy heated clay sinkhead to act as a feeder
to the ingot proper. This concentrates the shrinkage ("pipe")

and permits much less crapping with a sounder and safer ingot.

The ingots on being 'stripped'were - as is the practice at
this mill— charged before they became cold into a furnace
about 16000F. When: thoroughly heated to an approximate
temperature of ZIOOOF, they were forged by the regular mill
practice for a steel of this type, namely, reduction under
hammer to one half of their original cross section or one
fourth of their area. That is the 12' ingots were cogged to
6' square and the 8" ingots were cogged to 4' squares.

The surface of these billets were ground all over while they
were still warm, approximately 400°F. It is the practice at
some mills to anneal billets before grinding. This assists
in locating small surface "checks” and {seams' as well as
adding greater plasticity to the steel, thus reducing the
risk of setting up grinding cracks from induced heat and
strains. It also necessitates somewhat less care in the

grinding operation. The samples used in this investigation.

. qvqo. .a. w‘ .

Fur-u— .....- -..—

'Figure 2.
”x 2500

Figure 3.
'x 600

 

 

 

 

 

 

 

 

 

 

-8...

came from an 8”square ingot which was cogged to a 4" square,

and then finished on a 10” rolling mill to 15/32” by 21/32”..

Figures 2 and 3 show the large carbide grains which have
probably broken down in working and heating from the eutectics
formed during the solidification of the ingot. This particular
specimen was taken from a 1 1/2" square section before annealing
The lower magnification clearly defines the parallel banding
and flow of metal in the direction of rolling. The higher mag-
nification shows definite evidence of a structure beneath the
etched surface, that is, a superimposed structure. The small
emulsified carbides are probably precipitated as such. The
repeated heatings for forging and rolling dissolves much of

the carbide particles and they are again precipitated. But
such action involves chiefly the smallest particles, which

present the greatest surface per unit volume.

According to Grossman and Rain the relative attack upon the
larger carbide particles is very slight, and they assume that
the obliteration of concentrated carbide regions is largely
mechanical. Thus after the final working, it is found that

we have two distinct sets of carbide particles: The small
spherodized carbides precipitated from solution and the larger
ones (usually banded longitudinally to working) from the
mechanical disintegration of the coarse entectics. See figure

4 which shows this condition very clearly.

u .. v.
_ n . .. Km
_ ..." .. B ..
. u m
. 6
_ W3 L8 .x .
. .. .. .F. .
a .
. . .
m ..
m . rs...-

m
u .
a u

_

_
_ a
— .
m .
m
m u
- u
. _
. .
. .
a
. m

 

 

....

C...

W..
9

Ir‘
. .
~-_ 0 ..

(2th

\J

X

CCS

 

 

 

 

 

 

 

      
  

 

MN... ......

.- u" III
E 1k 1 a . I... .I n. D . .~ “ o .
"3-0.”. ,.H:Jm~Fv.n no .flm *Mau..w( .. ....
. . al 4 t . K... I) . . .
v.1 . “49.x: Wm. . ...... o. a,
...}. .r r .... . v ‘ .Vpe
.9 . a 9.)”.( c As . .‘ ....
m 3 .. . - . . ... ..
W. . fiWlfl 0.7“...” Wu! ‘ ~ I 1...“: "JV
2 ~ ... w. v». n . . a .5 O
a 4k wf ”It . la ...! x . .
‘h *"fiwlwu ..q'r U. I **
A “J ..r<(llf t\. .

 

 

 

 

 

 

 

 

 

 

-9-
Figure 5 shows what is probably a richly alloyed band of
carbides. This runs in several bands through the central.
portion of the bar, and is not an uncommon condition in high
speed steel. This specimen was hardened to give the handing

greater prominence when etched.

The reheatings apparently had little or no effect on this
cored condition. The partial breaking down was by mechanical
work and flowed in the direction of the work. Good examples
depicting lack of homogenity of well worked high speed steel
will be found in Figures 4 and 6. One of the characteristics.
of this analysis is the inability of its component parts to
diffuse, even after long anneals and repeated beatings. These
samples show the streaky carbide region wherein the carbide
particles are dense and plentiful. This perhaps indicates an:
inequality of the chemical composition. This difference in
composition could easily be due to the difference in composi-
tion of the first and last (eutectic) grains to be formed.

The differential is seldom equalized by diffusion - the steel

being too sluggish.

It has been suggested that the red-hardness of high speed steel
might be partly due to this atomic immobility or sluggishness

resulting from inhomogenity.

According to mill reports, this particular heat of steel gave

an excellent recovery and showed no indication of trouble.

.. 10...

ANNEALING

High speed steel after finishing on the rolls or hammer shows
cOnsiderable hardness from air cooling. In order to put it
in a state where it can be readily machined, cropped, and

inspected it is necessary to thoroughly anneal.

The Brinell hardness before annealing is quite variable, but ,
is usually well above the 500 number. After annealing it
varies according to the condition of anneal and hardness de-
sired. It is usually in a range between 196 and 248 B.H.N.
An extremely low Brinell hardness is seldom desired as it
seems to be mushy and tears in machining. For general pur-

poses a B.H.H. of 228 is recommended.

Figure 7 and 8 show the microstructure of this steel well
annealed. It will be observed that the carbides are small
and fairly W611 dispersed. In order to get the maximum
machine ability, it is necessary to remove all traces of
austenite and martensite which may be present from the mill

operations. (See Figure 2)
The small precipated carbide particles shown in Figure 5
offer little hardening effect, and therefore lend to machin—

ability.

The generally accepted annealing practice in tool steel mills

Figure 7. , ~ Figure 8.
x 600 ' x 600

Figure 9.
x 600

\.—.d

MC
...-..

1.. a...

A

n
“(‘3

.y‘N f!
t.

\J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-11..

for steel of this type is to anneal in steel or iron pipes
with or without a packing material. This steel was pipe
annealed and packed with 60%-washed river sand and 20% wood
charcoal. As the charcoal becomes ”spent" it is constantly
enriched in order to keep the carbon content between the
limits of 15% and 25%. With such a mixture the character—
istic soft or decarburized skin resulting from annealing is
prevented. Thus, a superior hardening surface is obtained.
'The loaded pipes are charged into a cold or warm furnace and
slowly brought up to approximately 15500 F, It requires 7
or 8 hours to reach this temperature, and it is then held at
this point for about 6 to 8 hours. Gas or electrical current
is then reduced and finally out off. The time of cooling is
about 6 hours. The above times are approximate and vary
with the nature and mass of the charge. When the furnace has
cooled to about 800°F. the pipes can be removed from the
furnace, but the steel is left in the pipes until nearly cool
enough to handle. The speed or rate of cooling below lOOOoF.

is not considered important.

The series of micro structures As shown in_Figures 7,8,9,1O
and 11 give the structure existing after a thorough annealing.
Both cross and longitudinal sections are shown. Figure 7 is
a longitudinal section and shows a well dispersed carbide
condition. This section was taken from a g» round bar. The

worst spot was selected for the photomicrograph.

Figure 11

n“ .

I 600

 

 

I'lsl'l I I- "'

 

 

ll II'I IJ

 

01 515311

,. «iv
353,: I '1

3'5
.2

[I

CCO X

mt‘
:5
“q

I
0..

CCfi X

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-12-

In Figure 8, which is taken in a longitudinal direction,
there is Just a faint trace of handing of the primary car—
bides in the direction of rolling. No matter how much re-
duction is given to the steel this is a very difficult con-
dition to destroy entirely. Further rolling merely means
further elongltion and streaks resembling ghosts lines

result.

Figure 9 is the same %' diameter section shown in Figure 7,
except it was taken from a cross section of the specimen.

The worst spot was again selected for the photomigrograph.
So it demonstrates clearly how well the steel has been worked

and annealed. The carbide distribution is exceptionally good.

Figure 12 was taken at a lower magnification in order to
emphasize the banding or segregation of carbides. This is

a longitudinal section and on close observation it is possible

to discern the crystal (grain) arrangement of the carbides in
the direction of rolling. It is however very slight. It

will be observed, series l3, l4, and 15, that there two distinct
sets of carbide grains, that is, grains of two sizes. The large
grains may be regarded as primary carbides existing from the
original ingot or eutectic structure. The smaller carbide grains
have resulted from eutectic disintegration and precipitation on

reheating for working.

The photomicrographs in Figures 16 and 17 show a longitudinal

and cross section respectively, in the unetched condition

o .r.
. .. /.d4. ....
... .h ...
....
......
$2....
8
.1
I.

' C
-n

_ .
s
. ,. ...
o .
w. . ’. '
I I
"‘ f

 

I.
‘
.... a
. . h ..
.’
.m .. \
.\ .. a.
. 4..., .,. .
4.! A 8.
.0.
In ,.
U m”
....-.
.... ...
..I. .. ..- II
c). ..s
. ‘
w
. k...”

 

 

. --...J

‘
~
_-- 0‘ u- ----a- m

L...

.bf erugii

S"

.I erugii

CCB X

Figure 16.
x 600

Figure.l7.
X 600

 

 

 

\
.‘ ..
9.0.”
.~\.
40"» I;
.. ll s
f

 

 

 

 

 

 

-13...

showing the sonims or slag inclusions. They are fairly small
in size, well distributed and deep—seated and therefore pro-
bably not very harmful. But they do seem a trifle too plenti-
ful for a heat of steel that finished so well in the furnace.

and gave no trouble from beginnhng to end.

Figure 18 shows a graph of the mill anneal from which these
samples were taken. It shows that the furnace was held 8

hours at temperature and then cooled 20° per hour. The anneal

invblved about 40 hours.

Figure 18.

‘1')

I

“I 1’ S .‘.
. AK. 1

H

 

 

/..

F .- -.l iL

2 Q 3 mama : 2 m m a o

 

 

. nQExSvax >\\ \\~\.\ bkkkba

 

 

 

4t

30

25

15

id

f/fi/f IN HUU "5.5..

H/V/VfflL INC CURVE

 

 

 

—14-

SUB-CRITICAL OR WATER ANNEAL

This is resorted to only when great speed is required, such
as test samples from melting furnace to laboratory during

manufacture.

The method the writer has found best is: the sample on be-

ing brough from furnace is placed in a furnace recording

0
1500 F. and allowed to fall and equalize with steel at

14500 to 14750 F. The best results were obtained by hold-
ing sample at this temperature for 20 minutes, but where
greater speed was necessary a 10 to 15 minute soak would
permit drilling. After being thoroughly saturated the sam-
ple is removed from furnace and placed in a spot secluded
from light rays {a pipe closed at one end serves exception—
ally well). 'When color leaves, remove test piece from

pipe and test with file until a soft point has been reached,
a light yellow color shows on filed surface. At this point
quench test sample in cold water, taking out quickly. Test
with file and repeat as above. Cooling should be hastened
after the file "softness“ has been held. Total time required

is less than 20 minutes.

This treatment is not a true anneal. The steel has not gone
through the critical or transformation range. So in reality

it is merely a very high tempering treatment.

-15...

It is advisable to allow the test sample to codl down

considerably before placing in furnace.

This is probably due to the fact that above 6000 to 8000 F
austenite remains stable, and of course if quenched from
14500F. without first allowing a transformation to martensite
at BOOOF. the steel will naturally remain austenitic and

undrillable.

According to Wills good results are obtained by leaving
0 o
specimen in the furnace 5" at 1320 to 1400 F. This gave

him a 364 B.H.N. and was drillable.

-15-

METALLOGRAPHY OF HEAT TREATMENT

The complexity of the chemical analysis of high speed steel
involves an entirely different and radical departuna in
heat-treatment in order to obtain the desired prOperties,

namely, maximum red—hardness.

Red-hardness as its name implies is the ability of the tool
to retain a cutting edge at elevated temperatures without
undergoing considerable erosion or breakage. It must be
hard, yet not sufficiently brittle to cause spalling or
breaking. The maximunm red—hardness has been found to be
directly prOportional to the carbide solution, and this in
turn varies directly with the temperature of quench. Such
elevated temperatures as 23000 to 24000 F. are not uncommon.
With such radical departures in analysis and heat-treatment
it might be thought that this type of steel is different
constitutionally. It is nevertheless fundamentally a hyper-
eutectoid steel. A well annealed sample of this steel
consists of ferrite and carbides, which compares with ferrite

and cementite of hyper-eutectoid steels.

The series of photo-micrographs in figures 7 to 11 inclusive
show a typical annealed structure of this steel, and consists
of a matrix of ferrite (perhaps some sorbite or a modification)
in which innumerable complex carbides of tungsten, chromium

and iron are finely' dispersed. In the final heating for

-17-

working, these excess carbides (not in solution) have been
rejected or precipitated. In later photomicrographs it
will be noticed that the carbides on quenching are not only
in the matrix, but also right within the austenitic grains
retained on quenching. This would be evidence to indicate
that these particular carbides never went into solution

while above the critical range.

It has previously been mentioned, see Figures 13, 14, and
15, that the annealed structure ready for heat-treatment

has two distinct sets of carbide grains within the matrix,
the same material but of different size and from a different
source. The large granules are the remnants of the ingot
eutectics existing after casting and the smaller ones are
reprecipitated on cooling from working temperatures which
are naturally above the critical range. This existing con-
dition, namely, the large excess carbides is the main reason
it is necessary to depart so radically from the regular heat
treating methods.and quench from a temperature much beyond

the transformation range.

In order to get the maximum hardness with this complex type

of analysis, and , as previously mentioned obtain the red-

hardness necessary for tools of this type, it has been found
that maximum solution of carbides of tungsten, chromium and
iron is only obtained when steel is cooled from exceedingly

high temperatures.

' . . K} LII." ' ‘
. ~: .- ' ii
., " ‘,".'.v"0‘i',s

. .1. "-." s '
. lL‘RS‘P-T' i...

   

 

 

 

   

‘ fl '2’ .:_ .
s , ”em .‘1 A
. 3“:le .kawf Wt) 153:}
{“0 {9's 7?” J'yX. ("‘6‘ ”filo-{’1'
. 4'

9:: ’12:.‘0
£25.34» - ifrif‘yékjg ii; -

4-45")? 9' 433$
. .0 “’- “‘
‘ v.

‘I .‘O I
5";

 

 

 
   

 

-18...

It is not difficult at the temperature used for hardening.
22000 to 24000 F., to obtain solution of the secondary

carbides, but the primary carbides seem to remain embedded
in the hardened matrix. Heat apparently has little effect

upon them.

Samples were quenched at varying temperatures starting from
o .
1400 F, which is considerably below the critical range,

and extended to a point where incipient fusion set in. An
attempt was made, by the use of photomicrographs at these
varying temperatures, to show the increase in solution of
these precipitated carbide particles as the quenching
temperature was increased. As has been mentioned, the
quality or hardness of the tools increases with the increased
solution of the carbide grains. There is, however, a limit-
ing point where the brittleness and excessibe grain growth
resulting from overheating more than offsets any gain from
the extra carbide solution. Where scale on tools is a dis-
agreeable condition, it is offtimes necessary to sacrifice
somevsolution (lower temperature) of carbides to retain a

smooth finish to the hardened piece.

Figure 19 shows what may be regarded as the general nature

of structure before hardening.

In series 20, 21, 22 and 23 are shown photomicrographs of

o o 0
samples hardened in oil from 1400 , 1480 , 1510 ,and 16000F.

Figure 20. - ‘ Figure 21.

x 600 . x 600
i
?
'Figure 22. ; ‘».” Figure 23.
x 600 j "'= x 600

“Test?-

~JO

..d
..L

Y .
...3

{\n

a.»

-'

{3213

an

«#J

'31.:

Pr

0. 1
«fl

\4

(Us

 

 

 

 

 

 

‘5'. '

 

‘5?”

'1

I o
.
g:
I '
' 3““
0 .Jo.
-. . .
u|o ,‘
.
.

A.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-19..

respectively. All temperatures are below the critical range,
yet all, with the exception of Figure 20, showed considerable
solution of carbon bearing tungsten before alpha ferrite was

transformed into a solid solution of gamma iron.

The Rockwell hardness of the steel before hardening was 0 16.
After the foregoing quenches they were C 16, C 45, C 50 and
C 52 respectively. These figures show a decided hardening
action. The samples used in all these testswere out before
hardening to a size suitable for microscopic examination.
Thus the time required for heating and cooling was consider-

ably shorter than is usual with tools in practice.

The sample hardened at 16000 F. required about 5" to show

a uniform heating. We can cbmpare this sample Figure 23,
with one quenched from the same temperature namely 16000 F,
but held for a much longer period. Figure 24 is a photo-
micrograph of this sample after being held nearly 15' at

the above temperature. There is perhaps a little increase
in carbide solution , but not nearly as great as one might
expect. The Rockwell hardness was C 52 which would indicate
no appreciable solution of carbides due to the lengthy soak.
The photomicrograph looks as though a slighibreaking up of

carbideshas started.

The series, 25 to 29 inclusive, show photomicrographs from

increasingly higher temperatures with a corresponding increase

shown in carbide solution and in Rockwell hardness. In Fig.26

Figure 24. . Figure 25.

“x 600 ' - x 600
i 7 .Figure 26.
§ 1 soc

 

 

 

v :‘a .‘::I
.l._ . ZIQQ‘OK I .
.....{e- l - -V
"33‘“; ‘9 _ '17.. I
’ ‘ "Q‘s ...; \ ;‘
K. ‘ V. ‘.‘,;; b": ' ..u
4:“: ~-_-.~;\r.,
’.

 

- , t, a ' ‘l A ... .

.' ":- "K -° " ' " -~: L's " ’3 : :kvirJ.’
$.13..." "a '-' .334 ..(g; "-\ 83“}, {Fry} ’ g.
s o -. _.l.'r- |;. ‘Is 5. ‘» ".‘ :“.." ‘:. . _. ”

.\,.A‘90 .... {.gfl ‘lv').’1;“.,\,‘é‘

r “a — "st-H" 3-.‘7;s.-$;3.., g /.‘\.~i,0‘0 ‘ “ a}! ,
-- ' ..." I: ‘30 u" --'. m ' a..." - .\ I '
_ ~ mi..- -.:*.< .-- .2 - amass: 3

 

. 1 mm. L
J”. \ )‘yg'

- ‘ ‘ ... ...“. .
1'.\“.S\ ”3.3..5 to I :‘k' I“'.-

 

 

 

 

 

 

 

 

 

 

 

 

.. 20...

we observe the first slight sign of a polyhedral (austenitic)

structure.

The hardness between 28500 and 24000 F. seems to vary but
little. It would seem as though maximum carbide solution,
in a sample of this size, takes place at a rather low tem-
perature. Owing to the density and sluggishness of high
speed steel, it is always advisable to preheat slowly and
thoroughly at a temperature around lSOOoF. before bringing
up to high heat. If this is not done heat-strains are set
up within the tool and rupture frequently results. The
tests quenched from 18000 F. and above had a preheat at

o
1500 F. The time held at hardening heat is shown in Figure 40.

In Figure 26, sample quenched from 20000 F, the first sign

of an austenitic structure is evident. The samples quenched
from the higher temperatures show this characteristic to a
more marked degree with a much larger growth of the individual

grains.

In Figures 30, 31, and 32 the photomicrographs show the speci-
men at increasing magnifications, after being hardened at

o
2400 F. Figure 32 brings out clearly the austenitic condition.

Figure 34 is the same sample as Figure 33 with the exception

of being reground considerably the second time on account of

the length of time sample was held at the high temperature.

1'.

figure" a7.

..u i300 ',| ‘ ' . ‘

Figure 39.
Vx‘so'o

Figure 38,
x 600

‘D

S srugifi

008 X

. Figure 30. y 3 Figure 31.

"X 600 ‘ E X 1300
§ Figure 32.
3 X 2400

sisal?

I

I

0C8 X

C031 X

 

 

 

 

 

 

 

 

 

J‘t 5v _
. . V3.3... r swish $~ v
A; ‘fh A 1"“! \ .5 ‘fl- \‘w Jan‘s v _
‘4 “a a \a H ..Nh \.. 4 .3 Km -%\\ 0C W) \ ‘c _
1 ‘ c ‘ -. t I t . \
9' V: \ .W\ a. »\u . . $3 . .. V... A.“
. v 1 . D (C w . .
.~ 5‘ ) ‘OQKO ‘\ \x 1 t u \ . HW'... x \Vx c6 W35
u \ ‘ ‘ x \ \ \‘ . 3. ks \ a“ 0 s‘\- _
a ., ... .... 9.9. t .. :
u s .\h .J wall. 5.x * ..Q... _
l‘ 0 \1. . K. \ ...l .. . . 0‘ .Afif Mv
- .‘ II. a
V.\ .9... 2... x. .x L): .. .b a o bka N .
.1 0-5 . \rf.»\0\ A\. N .\ as \ _
C ‘\r.'w . .\ \ .sk‘... . W *Ufibkd fl ‘ § . _
i I u. .I .\ y .
.\ ...r a: d s. . . a}! 3.. . F‘MQ Q
' 3“~ ‘ u ‘ 1" \ ‘ «\K‘ \ VS ‘ .9 «fit i \O‘
'l or a .. \ d 0-H... / . u h a ‘V a
is .5 . ~ I a “a WHVQWW. vvfi .
v 5 I r ‘
\w s: 11! ,\ IAN» Knn.‘ &” \
... ... . r . \
t W. 5. c. I

 

 

 

 

 

 

 

 

 

Figure 33. Figure 34.
x 600 X 600

Figure 35.
X 2400

“fl

(‘1

O
C 3
m

:'-<

w

H

Uz‘

P'"

(1'
U?

srugti

CCFS X

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The only noticeable change is an increase in the width
of the grain boundaries. Figure 35 is the same as Fig.

34 with increased magnification.

Sample in Fig.35 was quenched from the same temperature

as Fig.33 only instead of quenching after‘l& minutes,

when it showed complete saturation of heat,it was held

4 minutes to observe the effect on the solution of carbides.
The effect,exoept for enlarged grain size,is negligible.

In all probability such a tool would show an increase in

- brittleness. The Rockwell hardness remained approximately

the same.

Photomicrograph shown in Fig.38 is a sample heated in a
furnace recording a temperature of 2475° F. It was held

3 minutes and on removal showed excessive scale in the
form of large ”sweat” blisters. The sample was decidedly
overheated and held too long. The Rockwell hardness was
064 which is much below those samples hardened in the reg-
ular manner. The grain growth in this specimen is very
1arge,and it may be possible that excessive decarburisation
at this elevated temperature has brought about the precip-
itation of some ferrite. Figure 39 is a similar sample ens
oept that it was run at a temperature (2500° F.) where in.
cipient fusion has set in. The characteristic eutectic
structure ordinarily found in the solidified steel after

ingot cocling,has now been produced by overheating. It is

 

r 'III e l: - Ill - III! IIII’."I-- ‘

 

 

 

er‘
2:
U.“

A:

Pu

‘I;
f.»

 

 

 

 

 

 

 

 

 

 

 

 

 

-22_‘

quite probable that such a temperature causes decarburizatdon
and ferrite precipation. Austenite is not stable in solid
solution if the minimum carbon percentage is not maintained.
This is a typical structure of badly overheated steel. The
structure that etched dark in Fig.38 as well as Fig. 39 is

in all probability precipitated ferrite resulting from carbon
diffusion. Such a condition as shown in these two samples

is very undesirable; it induces brittleness and weakness.

The samples shown in Figures 36 and 37 show the polyhedral
structure characteristic of austenite which has been retained
in the solid solution after the specimen was cooled in still
air from 23500 F. This indicates that the speed of cooling
is not a factor in the retention of carbides in solution.

The Rockwell hardness is C. 66 and compares very favorably
with those quenched in a more drastic medium. In Figure 737,
which is the same as Figure 36 only at increased magnification,
it looks as though within the grains there is a dark constit-
uent that might possibly be martensitio in nature. It is
exceedingly difficult to etch this in a manner to show the

characteristic needle-like structure of martensite.

In Figure 40 is the detailed treatment given to the varioss

samples with their correSponding Rockwell hardness figures.

. x 2400 -
| ‘I
I .
‘
I -
L
\
l I
‘ V
l .
1 -.
’ -
l. - a
_ _ ,7
‘7 K O
5
' Q
Q t ‘ ‘
i " .
. s y. , ‘
.n' " . '0' f,
“a. . ‘. ‘ ' «3,; .
a" ‘ . -‘ Tfl‘t . > ‘
‘ 4 ‘i‘l ' ”.lff' ‘
' ‘ ‘ 4" " . ‘ -. ,_
K. II I e“ .‘ .
‘ . _ In. 3. "fi ‘ .
-- r' ' ' s "‘7 E": as.
I.’ ' ' "i 3 $- ..
s v .. P - ,.
'r- u
3 . (f' ‘3 w ..v
I f,“ x . d r 4. l“Av'ln _-- . ,
I I I n . ~
. . ..~ \ ., ‘4»,- t _
3 f a. :1 ‘1 .... q q
; ' v ' I " 0} .?~ 3": .
., “b : ' .9” _, ' ,.
q ‘- ’. .1 - .- .1}
V", * . .
. ‘13,. '- '
u ' 1 cl ‘ ‘ ‘
i ‘4 Ila“- _ *p" ‘5 . 4 .
’ “ - vi“ I. ‘1 ‘0 _ .ev- ‘
at c ,2; ' .v " " ”3 1.
. 'o b ‘;’C ’
‘ ‘ 4’ '9.“"v"‘ 2‘ (I ' ‘I v
o ’J E. ‘ I
' Q - I
. i
;- . ‘ '0' w
. ‘l‘ .C ‘
I . 7' '

 

 

 

 

X

‘
.\
.4

CCifi

 

 

 

 

 

 

 

 

 

 

 

Figure 40.

. CV3
9
TI.)
.2
I "'1

-23..

T232-§§-I§38o

Rumba: EZ§:D§§! H188-828§_ E22332};-§
20 ' - 1400°p 10'5" _16 '
21 - 1480 10'0” 44/45
22 - 1510 8'0“ 49/50

223 - 1600 5'0" 50/51
24 - 1600 15'0" 50/52
25 1500 1800 4'15" 57/58
26 " 2000 2'5" .63%

27 fl 2260 2'20n 66/66%
28 ' 2310 2'0" 66%/67
29 . 2350 1'40“ 67
30—32 " 2400 1125- 67
33.35 I 2400 4'0" 66/67
P8 n 2475 2'0*I 63/64
39 n 2500.. 1'18" -
36-37 I 2350 (air) 1'55" 65/66

-24-

In Figure 41 is shown the plotted hardness values at in-

creasing temperatures. It is self—explanatory.

In Figure 42 we have the corresponding values after in-
creasing tempering temperatures. It will be noticed that
with those samples fully hardened the maximum secondary
hardness is obtained between 10000 and 11000 F. This

temperature also gives the maximum red-hardness.

Figure 41.

 

 

C Roe/rum fiflfi’flflfss .

5456026¢ro

3436414650

 

 

 

 

14155171013202122232425

DEGREIS m” HU/YMDTHS
OUf/YCH/NG HflRD/YKSS .

 

 

pt 15.!lu3oblo-ll 1

Figure 42.

O

05‘

45 an;

i

kw

 

(I)

60

9.5 ba‘

74

,

{RD/VSS .

//

46 50

'l

‘ I

l-"(J .‘3 NH '

U

 

 

 

 

 

 

1‘ ‘I
} ,
‘— ‘ ‘0
( n \ o-c/ \
t, I
(I \\ ° “0/. o o . e
o»,- “" // \
>- . / O
t 0'
L0 ‘\r Q
4‘2 .0 r
7“? ,.-- o a— o
\ /-l “\
\
‘o

 

 

 

 

" I * -4 1 1 L l 4 _ .1- - __1___L_ -..1 ___1__.4__J

..‘m; My .k,’ 500 600 I00 600 900 [000 1100 1200 15001400?’

TFMPZ‘V‘K‘IIVQ‘ 6.77/2}ng

 

 

-25-

TEMPERING OR DRAWING

Paradoxical as it may sound the drawing or tempering of

0
high Speed steel at the usual temperature, namely, 1000

to 11000F. results not in a reduced hardness, as is usual
with carbon steel, but in a decided increase in hardness.
This is generally referred to as secondary hardness, and

represents the maximum red-hardness. This condition is

shown graphically in Figure 42.

The sample used for all tempering operations was that
shown in Figure 32 and was quenched in oil from a tempera—

o o
ture of 2400 F. after a 1500 F. preheat.

In the quenched condition it is seen to be typically austenp
itic in structure. The polyhedral grains are well defined
with the primary carbide grains mixed within the grains and
along the grain boundaries. This excess carbide dispersion
and distribution may be quite a factor in the hardness and
heat stability of this type of steel. Its nature would

diminish the atomic mobility and increase sluggishness.

The first few specimens tempered are clear and easily ex-
amined at low magnification, hence the draws between 7000
and SOOOF. were photomicrographed at 600 diameters. Above
this temperature 2500 diameter magnification permitted a

better interpretation of the changes taken place.

. -.- >_,_ __,

.. ~— w...‘ ,

-- .. --.-n- —

‘_Figure’45.‘
x 600“---—’

Figure 44.
x2400

fin.
..r...

'Y r 7-;
9 " a?)

.39 910211

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-26-

The tempering operation was started at 700°F, a point well
below the martensitic transition point. No apparent change
was noticed when etched with the regular 5% Nital reagent.
The Rockwell hardness dropped sharply from C 67 to C 62%.
This would indicate a physical change, probably the removal

of hardening strains.from within the specimen.

All samples were drawn for a period of one hour at tempera—
ture. This should be sufficient time to complete any trans-

formation in a sample of this size.

Figure 43 shows the photomicrograph of sample after a draw
at 760°F. for one hour. This is apparently right at the
point where a start is made in the transition of residual
austenite to the more stable and harder constituent marten—
site. This is well born out by the Rockwell hardness in-
creasing from O 68% to G 63%. No change is apparent in
the photomicrograph, but this is probably due to the
difficulty involved in showing martensite or martensitic

structures with ordinary etching reagents.

Figure 44 shows the same sample after tempering at 84001.

It seems strange that no increase in hardness is shown

with this extra tempering temperature. The Rockwell hard-
ness remains c 63%. The photomicrograph shows no great
change, but with heavy etching a slightly darkened condition

is noticed within the austenitic grain boundaries. An in-

crease in magnification was used to make resolution more

Figure 40.
x 600 .'

 

Q- - O. -

— C~ ‘— a -

5|

 

fiO

AU

.59 erygii

CC5 X

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-87...

certain. This may or may not be a martensitic condition;

it is very difficult to say with any degree of accuracy.

In Figure 45 the sample has been drawn at QOOOF. The Rock-
well hardness is now 0 64, a slight increase. The grain
boundaries do not appear to have changed, they are clear
and well defined. The sample showed a slight tendency to

etch faster.

Figures 46 and 47 show the sample after a 950°F. draw. The
Rockwell hardness showed a small increase to C 66: The
Matrix within the grain boundaries shows a decided transfor-
mation or decomposition. The polyhedral grain boundaries
stubbornly persist, and show very little, if any, effect

from this temperature.

In Figures 48 and 49 the sample has been drawn at 10000F,
The Rockwell hardness remains C 65 which is very close to
the original hardness on quenching. The original poly;
hedral grain boundaries are not as well defined as in former
pictures, but they can be traced with difficulty. The etch
in this case was slightly too deep as it expressed the
nature of the matrix, which somewhat obliterates the defini-

tion of the grain boundaries proper.

Figures 50 and 51 show the sample after a draw at llOOoF.

The Rockwell hardness is now 0 66. The sample is con-

siderably over-etched, it being difficult to find a suitable

 

 

O
' . .
'.
' . Figure 50. .
' . I. . I
. . x 600 ,'
' . I
2 . ;
. l. ,. F I
I I '
. . . I
. i
. " I
I I
- . -.----- o 0 O ‘ 'I _ -
' i
9 .
5 ' 6
' l
s s . ' —
! I
' a
' a. O
i 3 '-
I I "
fir"! ?"\$u:,'. ' I l
‘ '3§>;;. . E . T
always?“ )5... ' : ‘
'1‘ 1 I
e . - -
' i .
I
I
I
. l ., ,
' ' I
I , ' J
- I
--. . . . . __ . .. . .. 3 .— . .

..l
R.

w...

5

.13
We

{JTIY

a a

PM.
.I
h: a

h.
a.

X
a

r.

X

CCPh

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-28-
”etch-time" with Nital to resolve the grain boundaries be-
fore the matrix darkens. However, the grain boundaries are
still faintly defined. It should be mentioned that the
sample was given two separate draws at this temperature to
make certain ample time was given at the tempering tempera-
ture. No microscopic difference was observed with the

second draw.

Figures 52 and 53 show the sample after a lZOOOF. draw.
This is much clearer on focus and etch than the preceeding
Figure, and shows clearly the well defined polyhedral grain
boundaries. The Rockwell hardness of C 62-64 shows a de-
cided drop which would indicate a second phase transforma-
tion has commended. It would probably be some form of
troostite or sorbite stable at this temperature. It points
out clearly the tenacity and permanency of crystal grain
boundaries. The etching solution used for this particular

photomicrograph was less than one percent Nital.

Figures 54 and 55 show the sample after a lSOOOF. draw.
The Rockwell hardness is C 61/63. Under visual micro-
scopic examination the grain boundaries are still in
evidence though not so clearly as the preceeding samples.
Figure 55 requires close examination to detect the grain
boundaries. The Matrix etches dark so quickly that it has
become increasingly difficult to etch in such a manner as

to define the grain boundaries.

I
- ...
b
o . is,
t. D e
h . In: .
0|. I
an M 0*
c‘ .
I
a C
... a a
.5.

I H.0‘Q
..p
en.
|

a
I O
I a w
.... L. 4
J .

s4. .'

”I

are

I
a \

2;.

f»
O

I'r'

 

...
A ca '1
J 44 A.

I“;

a (._

0

91¢;

C5 x

0164.; (tr-"(12:1
(Aha; V

=r

x" '1

 

 

 

 

' :93 4'” '

... . .
.341"

t
. 0
Jose“

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-29-
Figures 56 and 57 show the nature of the steel after a
14000 F. draw. The Rockwell hardness has dropped to
C 52/53. The grain boundaries are no longer discernable,
and the matrix appears to have changed considerably.

There is also some evidence of carbides sperodizing.

By reference to ,Figure 42 will be seen the graph hardness
figures of several specimens quenched and drawn from vary-

ing temperatures. This is self-explanatory.

-30..

COMMERCIAL HARDENING

The generally accepted practice for hardening tools made
'from the regular 18:4:1 type of high speed steel varies
but little from the original treatment as recommended by

Taylor and White.

The treatment should consist of four stages, namely, pre-
heating, high heating, quenching and tempering. All
stages with the possible exception of the quenching are
of the utmost importance to the life of the tool.

Preheating: Owing to the extremely high temperature re-
quired for hardening and the extreme density and sluggish—
ness of this type of analysis, it is of the utmost impor—
tance that a slow and uniform heating be given tools made
from this steel. This slow heating should be adhered to
rigidly until the steel has at least passed through its
critical range, 14800 to 14900 F. Distortion and rupture
are usually the result of excessive strains set up by
heating too quickly through the transformation range.
Owing to the sluggishness of the austenitic transformation
in this steel, it is advisable to hold quite some time at
the preheating temperature to ensure complete transformation.
A slightly low temperature in preheating prevents scaling
of tools. A thorough preheat then permits rapid heating
on transferring to the high heat furnace. A slow heating

and a fairly generous soak during the preheating gives some

-31..

solution of carbides as shown by the Rockwell hardness

values in Figure 40.

High-heat: Quenching in general depends on the size and
nature of the tools to be hardened. Small and intricate
pieces necessarily demand extreme care and somewhat lower
temperatures than larger sections..In general, a tempera-
ture range of 22500 to 24000 F. is usual. The high heat
furnace should be at the desired temperature before
transferring the tools from the preheat furnace. This
ensures fast heating and less scale. When the tool
reaches the temperature of the furnace no great time is
lost before quenching, as if held at this point for any
length of time excessive scaling and grain growth with

its subsequent brittleness takes place. With large
sections there is probably always a differential in
temperature of the core and the outside surface of tool.
This condition cannot be eliminated as the working surface
of the tool must be protected and given first consideration.
The transformation to martensite on cooling takes place
rather slowly, so the differential in temperature referred

to above is not of very great importance.

Quenching: The speed of quenching and the nature of the
quenching medium is of no great importance. It is the
usual practice to quench in a thin soluble oil, but many

tools are cooled in air blast or salt bath. Taylor and

-32-

0
White quenched in molten lead at 1100 F. After tool reaches

temperature of lead it is cooled quickly to or near atmos-
pheric temperature to effect the austenitic—martensitic
transformation. There seems to be little or no gain in

any of the special quenches over that of air or oil.

Tampering: This stage, along with careful heating, is

probably the most important phase of well treated high
0

speed steel tools. The usual draw is 1000 to 11000 F,

where the maximum secondary and red-hardness is obtained.
The length of the draw varies with different shop practices
and different tools from one hour to twenty—four hours or
longer. It is safe to say that no tool made from steel of
this type has ever failed from too long a draw. When heavy
and intricate sections are to be tempered, it is advisable
to warm up slowly to the tempering temperature to prevent
rupture from heat strains. The tool after quenching and

before drawing is naturally in a very strained state.

By referring to Figures 36 and 51 we will see a typical

hardened undrawn and drawn speciment respectively.

-33..
COMPARATIVE ETCHING.

In order to compare the structure with different etching
reagents two samples,one annealed and one hardened,were
selected for examination. The annealed sample was taken
from bar stock which was annealed as shown in Figure 18.
The hardened sample was quenched in oil from 23500 F and
showed a Rockwell hardness of C 67.

All photomicrographs were taken at 2500 diameters.

Figures 58 and 59 are the annealed and hardened specimens
respectively after etching with an alcoholic solution of
picric acid. The annealed sample was etched six minutes,

while the hardened sample required ten minutes.

Figures 60 and.61 are the annealed and hardened specimens

respectively after etching with a solution of hydrogen per—
oxide and sodium hydroxide. The annealed sample was etched
twelve minutes,while the hardened sample required eighteen

minutes.

Figures 62 and 63 are the same as the two previous photo-
micrographs (60 and 61) after repolishing and re-etching.
The tungstides are shown as much larger globules and greater

dispersion through out the entire mass.

Figures 64 and 65 are the annealed and hardened samples re-
spectively after etching with potassium ferricyanide and

potassium hydroxide. The annealed sample was etched five

.n
.. . . .. .
- w. 2 .
.4 .-
. .. _ . .
\ 8 S 1‘ . . . .‘ ..
.... . v. ..t .. q. I
. a .(e \ JD. .. . ..
. « n . ....F . . v
I. a .. . I x. c :5! I
...s. U... I \ . . g
5‘. . 8 ... .... ..fer I . . n. - O
.. I- .“ v)»; . I). ; I. V I
._ 5 .. .. , . 4 5
d . .. rte . .. u a I
o . . .
I. . . _ t . K I
. , z . ..o . .. .... . e
I a,
u . . . . 8
. . 0.... F m
a} \ x . _ 31
- .a. w k .. It 4 . J .....I 9....
I .I ... . . . .n ..
. a . IQ; A N
a . . a \.
. . . \. ..\ . all f . F
. . Sewn. . :93”) . I 3 ...I.
. C)“ S J W. J .I I
. ....t . .c\ o r ...
. ..l. .... ...?vox . . a. u
.o w. .... If
. a a ......i . . .
. . uu. . L ... J In. % . w . ...- . .r
...! 93* . . I . .4 .
. i w I r.
. w .. ...... .... .. .... .f. "v...
..s. r. .. ... ... a .
.fa .I \II . y l . .Il k 1)... A
o . h ..\ .1 fishy
A. ...

..
.
.
t \ .
.. ...
......
i
7
a .
to .
. a! x
w
u. x
h
a
.
..
I O
.
V
\

CCLS K

.83 810311

CICQS X

 

 

 

 

 

 

 

 

 

. .
.- '< "- -
an x ‘ t"
a P Q I .
‘ ' . 5t 7‘ 1“
. L 1'
'Q .1 v J I"
v u -* a ‘ v
-‘ I '
V , .
4- ‘ .
I I ' ' .
. I u'l .— '
‘ ~ F ‘ I
. Q ‘s .
' nu. x_ ‘ '
D ‘. . q I A '
I W b - t ’
' . n ‘ ‘I I
- z. a“ .. - ‘
" —. l '
...O .. I) h " ‘ " -
"1". a, h \ . ‘7 ‘ — . ‘1
. .. . g res
- 4
. . ,f' , - ~... - . 3‘" - "
-. 9 a '
. ._ . 'Q‘ , , .. 0‘ f t J-
. ‘ ’
P! .- ‘5 ‘ 1 " 3."
_‘ ,. ‘ I‘- h
p h 24 I. .
a " I. z
c. I * . .
a , . N Ci: . tr. ‘
_ ' .
' . _ (a. . C ,
» '_.' k ‘ I‘
. a . - ' '
w ‘ ‘
. It
1 ‘5 _ 2 . ‘ . a. - .
. p ' ' r ‘ l
. ‘ ‘Q. . 'i . . J
- ‘ ~ ' --. a .
n. '
A" 5 I v . "
. . J“ o'.
( ‘ . - ‘ “ 'E F 'u “ “r
s ' ’
. I. ..- . - .",. ‘ "‘ F -
.5: , - ’ . f- t'
a ” ‘ "
1 J -0 ' J ‘ .
1 r AI ‘
(V ’ f w. ‘
. .
-'-.I - .0 ‘0 v -'
..
=-
1 'i
a V l
' I
Q
Ax
.
. ., .
. r ' ‘1 ' ‘
‘ P
f .
. .)
a. '4 .~
.‘_ 9 a
O i I . q
s“? ‘

. w'hgui-e' 61’.‘ . ._ ,
. J. ._;x 2400@ _‘ -
V In. "I"

s 3 ~ " ‘

"I
v ‘ -
’ w
l I J
0‘
' .
, t _
‘2 \
9 . 9
. ' )n 4
.. a ’
’ v
.
r
\
‘ .
. 1 ‘4‘ '

7.
m5
7.

a...»
r.»

56.
hr

.IS stugi?

CCQS X

. ‘ I“ Figure 62.
f 7f X 2400

'yiggré 53.;
'x 24bo

cf;
fin...

 

 

 

 

 

 

 

 

 

 

 

 

 

»r1gure 64.
. x 2400

.88 975911

-34..

seconds,while the hardened sample was held fifteen seconds.
The etching solution was used hot. It shows the tungsten seg—

regation.

Figures 66 and 67 show the annealed and hardened samples re-
spectively after etching with g1ycerol,nitric acid and hydro-
chloric acid mixture. The etch shows a very clear and well
defined structure. In the hardened sample there is strong

evidence of a superimposed structural condition..

Figures 66 and 69 show the annealed and hardened specimens
respectively after a very light etch with 4% Nital. The annealed
sample was etched ten seconds and the hardened sample four
minutes. There is again definite evidence of a superimposed

structure in the hardened sample.

.- vs

§

1
v

e
I ,
I

time 661-5 _

J

‘l
O
l

 

..vl
..l
run.“
n...
a e

to
H.\.

V

 

O

..
, 4..

”anal-£354 '_
14240.0

staglfi

«0
Cu

OCQS X

DILATOMETRIC STUDY.

Dilation curves were taken using the Chevenard Industrial
Thermal Analyzert Figure 70 shows a picture of the instrup
ment,while Figure 71 shows the working principle.

A cylindrical test Specimen 2.12 inches long and 0.63’inches
in diameter,with a one-fourth inch hole through the center,
is placed in the silica tube. The closed end of this tube is
square with the horizontal axis. The specimen is so placed
that it rests firmly against the closed end of the silica
tube. A small silica rod rests against the other end of the
specimen and connects through a lever system to a pen arm.

' Any movement of the sample is magnified about 70 times by
the pen. The chart upon which the dilation is recorded is
carried on a revolving drum,the speed of which can be reg-
ulated. lithin the sample is placed a small standard pyros
rod. This rod also bears against the closed end of the silica
tube,and is connected by a mechanism similar to that of the
Otest specimen,to a pen. The movement of the pen is recorded
in degrees of temperature. Thus,we have two curves drawn
simultaneously,one showing changes in length of the specimen,
the other changes in length of the pyros rod; the latter
being calibrated in degrees of temperature. Using the data
obtained from the two curves, another graph may be plotted

using change in length and temperature as co-ordinates.

* Descriptive matter from; Some Observations on the Dilation

of Several Common Steels. Publow,Heath,Batchelor.

 

\

a
a . §
00. .4vt...._l .J in .
. .. .. .$ .. e. A . I.
. . , ...

it

..

 

 

 

 

 

 

i) L

A

Figure 71.

.1? ethglfl

 

kWh N35 ‘43

QQYZM> MTG

 

 

 

 

 

 

W
“Haj/x/x/vxvééefll _
Mycfivv/z/zizéfiwm'

 

 

 
   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-36..

Figure 72 shows a cOpy of the original two curve chart re-

corded by the Chevenard Dilatometer.

In Figure 78 (No.1) the above two curves have been plotted

using temperature and dilation as co-ordinates.
The test specimens were prepared to the standard size.

The test specimens were run in a variety of ways. The two
variables, time and temperature, were used freely. Samples
were heated slowly and cooled slowly, such as Figure 72 and

its corresponding plotted graph No. l in Figure 73. The temp-
erature to which specimen was heated was1970O C (17789 F). The
uneven line on heating is due to varying the resistance while
temperature was rising. The first sign of a transformation
appeared at 825° C (1517° F) and appears to be completed at
900° C (16250 F). The thermal curve, see Figure 72, shows only
a very slight absorption of heat, whereas the dilation curve
shows considerable shrinkage in the specimen. The time of
heating was 72 minutes; cooling was then started. The Ar trans-
formation commenced after 9.6 minutes of cooling, when temp-
erature had reached 725° 0 (13370 F). The transformation was
completed at 675° C (1245° F). The thermal change was of even
less magnitude than that on heating. The dilation change was
not pronounced. A normal cooling curve was again assumed. At
420° 0 (788° F) another retardation in cooling on the thermal
curve, and a decided expansion on the dilation curve is noticed.
This point is spread over a temperature range from 420° C (7780F
to 350° 0 (662° F). The curves again assumed normal cOoling

rates down to atmospheric temperature.

Figure 72.

erugifi

('0
..At

Figure 78,

,8? axial?

 

 

EXPANSION IN INCHES mINCH x101-

 

 

In I I no t
I
.‘\
/ J.
I O
' V r
/ , -‘ A -
‘, - "we [5 ora'r
' .' 1770'! n .‘ «34-,-

 

‘ v——- f I:
n >— . {a ’ f:
N h—' ’ 2 ,‘l a" a. I. ’
/ 0 v . “ '. -'
I f / ,' .‘I f
e. )— 5‘. l / I... U J
.‘ I 1 1 l I I L/’ l J _i .L
300 ‘00 “O ‘0. 1m 0 m m “0 000 100!)
TEMVERATURE UNITS U— ZOO'C .

0*: AIINI,’

LOOL'N'C;

 

 

-37-
The complete series of dilation and thermal curves, with
their corresponding plotted graphs, are shown in Figures
72 to 91. The rate of heating, elevation of temperature
and rate of cooling were varied considerably. This data

is shown in detail separately.

It appears as though the Ac transformation is quite con-
stant and independent of any variable factor, such as

rate of heating. .Its range, however, is spread over at

least 100°C (2120p) with the same specimen run under
similar or varying conditions. It occurs so consistently
at 800° 004720F3) to 900° C(16520F) that we may regard
this as the transformation range on heating. The average
point appears to be approximately 825°C(15170F). A few
tests recorded points below 800°C the lowest was 750°C
(13820F).

o
It should be mentioned that at a point approximattng 700 C
(1292OF) on the thermal curve, there is distinct evidence
of a slight change of contour. It is just possible that
this, rather than being a separate and distinct critical
point, is the beginning of the transformation shown at
a higher temperature. This steel is naturally sluggish to
absorption of heat or any physical change. Therefore, it
seems highly possible that the transformation of alpha iron
to the solid solution of gamma iron is equally sluggish, and

that the transformation shows its first movement as low as

Figure 74.

u .0:

 

a.
TI}

Figure 75.

Figure 76.

(‘0
.1}

t: D
#4

(hi

 

 

 

 

 

 

~03
J
, o
\G E‘-
l
4) o g ._
4!, In L - ‘
7
’7’ -1 ‘ '
f)
/.
’ ” t
r. ’ I
I L
A] ‘ ‘ I
I p ‘
b‘ t- .
I v ‘ I
L L l 1 1'.
L 00 .' I} L' t N

I
_Lnnd
3dr 100‘.

so :
-JUI'
..
VL __ 1.. 1,, 1 1
I :s‘} '-

'[MPrRsqut phII) n - C

_. ._._— _.._ __x .r

 

 

Figure 77.

-c-..

E 1". .

Figure 78.

l Il‘vlel‘t

.8? eiugii

 

e I! 'c
“of!

_

_ _

_

I‘ [LI

 

summations

 

 

 

=

2

m D

m

o

n N d

...2 . :oz_ ... 3202. Z. 2.62.4600

l0

TEMVERATURE UNITS 0(- 200°C c

 

 

--—-‘ ,---.__..-‘_._._____,_#, .

Figure 79.

 

0.02

5o:

 

Figure 80.

we
v43
L.
n1
9.

A:

 

 

EXPANSION IN INCHES "INCH .10" .

 

NO. P N0 0
t— ,A p
/'1 ~’
— . ‘ .\ 1 I,
\ no'c .. I r n)
.. Im'a
. l
n— / re... ’3
e ' ’. ' '
4 e
I

..

d" I’
/ h~ Hf «TING
- »- COOIING

_—

 

 

llllIJILLI

200 400 00. use no. 0 4 It
TEMPERATURE UNITS or zoo'c .

 

 

 

--u--~-----.'«

e--

—-_-.v-v

Figure 81.

.IS erupt?

. C
0
.
.
O a
.
O
O coo-ted
I
0
e O
O ..
O
. .
. O .
. l C
O ... O
. C. C, . Q
C O O
. .. O .
I .
.... C. U C .
0...... I
. . .
O O
. C
OI
ea 0 0
O
U
Q
o
C
C
0
.
O
.
O
'
.
.C '
O O
.
v 0 Q
C O
. C
C
O I
O .
Q
.
.
O Q
.
o
e

mdz

 

Figure 82.

30

‘3)

1d
¥ 0.0

 

 

o" .0:

 

 

 

Figure 83.

7..
LL

3

~$
ax

.... J

 

 

 

EXPANSION IN INCHES INCH ~10"

1'3

 

 

I I 1 J_J__.-,1-J_:

.00 009 m 0 200 .00 .00 600 1000

TEMPERATURE UNITS OF ZOO'C a

 

 

 

-38-
o o o o

700 C(l292 F) and is not completed until 900 0(1652 F) or
thereabouts. This thought is somewhat borne out by the
gradual hardening, see Figure 41, with increasing tempera-
tures. At 1480°F(1804OC) the Rockwell hardness increased
from C 16 to 0 45, but below this temperature no hardening
took place. Of course, the maximum hardness, when based
upon facts from hyper—eutectoid steels, should be obtained

immediately the Ac transformation is completed.

The writer feels the maximum hardness resulting from the
alpha to gamma iron transformation is probably obtained

at 90000(l6520F),but that the super-hardness of this steel
is due in no small measure to increased solution of ex-
cess carbides and the heterogeneity of the carbides of
iron and tungsten retained throughout the matrix on quench-
ing from a very high temperature. This latter condition
might be likened to the interference theory using grain
boundaries instead of crystal lattice planes. An alter-
native thought for this slight thermal change on heating
(700°C)is that it could be the beginning of the solution
of excess carbides. This would account for the lack of

o o
hardening within this narrow range below 800 C(l472 F).

The transformation points on cooling do not show the reg-
ularity of the heating point. Depending on elevation of
temperature or rate of cooling, we may have one or two

Ar critical points.

9.. a-n - 0...:-

— we.-- - ~.--r— .-

Figure 84.

.98 singil

 

I‘D-12-

No.11

 

 

 

0‘.
O
.I-
e
.-
I
e
I
O
.0
0..

 

 

 

 

 

 

Figure 85.

 

 

ExPANSION IN INCHES ... INCH -1o‘.

15

I!

TO

 

 

 

TEMPE RAT URC

UNITS

W ZOO'C c

'9." no u f
./
I
.,/
I“. I,"
/ J
/I .1
e X b“, .‘. f. 5 er" "c
/ ,f‘ 1:00.!" / / I I ya ‘1
. 9'“ /
/ ' /
.. J
.' /
5’. .1 -e “'T I.
A I I 1 I 1 [Al A i l I
20° 4“ I” OH use 0 fl .0 up 6v. mo

..__.p

HE. RT INu
COOLING.

 

 

Figure 86.

.89 erugl?

 

nfioz

 

v v- _-..-_ _ 7

Figure 87.

 

3 .oz

 

Figure 88.

 

 

no I)

 

ION?

l

TEMPERATURE UNITS a

l

LJJ

 

 

u________m
...2 .

. I02. 3. mqu2. 2. Zo_mZ(n_Xu

WWW!”

(00

 

 

-39—, .
o c

When heated at or above 900 C(lBSZ F) and cooled slowly,
see Figures 83, 84, 87 and88, two points appear, namely,
Ar 7OO{BOOOC(1292/l47zoF) and Arg 265/49000(510/9140E1 an
exceptionally wide range of temperature for different
samples is involved. The above temperatures express the
maximum range for all samples. By reference to individual
graphs, the particular point or points can be studied for

each test.

We have one exception to the above statement. Figures 85

and 86 show graphs of the sample which was heated to 84000

(15440F) and cooled at a fast rate, Bzgoper minute. Two

cooling points were observed, Ar 690/765°C(1374/14090F)

and Ara 315/4OOOC(;600?7520F). It is most difficult to

account for this phenomenon when several samples were heated
o

0
between this temperature and 900 C(1652 F) without showing

more than one point on cooling.

The inconsistency of the Ar points leads one to believe that
they might be governed to a certain extent by composition
(solution), or perhaps precipitation of carbides. The true
crystal lattice change is perhaps shadowed by a chemical

phenomenon.

The Ara dilation point is ordinarily of much greater in-
tensity thanthe Arlpoint. This, however, can be altered

by use of the variable factors at hand. If the sample is

—40—
cooled immediately after the Ac transformation is completed,
see Figures 78, 79, 82 and 84, the Arlpoint becomes very
pronounced, and the Ara point is completely eliminated.
This method 0! cooling produces an exceptionally low Rock-

well hardness, as low as C 4 to C 6, see Figures 78 and 79.

Inasmuch as it is generally regarded that the carbon in
this steel exists as a complexcarbide of iron and tungsten,
is it not reasonable to surmise that considerable time and
elevation of temperature is required to break up this con-
.stituent and cause solution of'the carbon in the austenite?
By cooling immediately after the Ac transformation, this
necessary time and temperature required for solution of
carbon is denied, and the resulting or precipitated phase
(austenite—martensite) will be low in carbon and consequent-
ly of a soft nature. This is proven to a certain degree by
the extremely high quenching temperature necessary to ob-
tain maximum hardness. The low carbon content of the aus-
tenite materially raises the temperature of transformation.
Three tests, see Figure 72, 85 and 80 were run at 970°C
(17780F),97OOC, and 975°C(l787OF) respectively. Figures

72 and 80 were cooled at a moderate rate, and show two well

defined points which check with each other very closely.

Figure 85 shows the same sample after a quick cooling. One
0 0
point of considerable intensity is observed at 290 C(554 F)

o o
to 315 C(ZSOO F). This is undoubtedly due to the Ar point

1
being denied time for formation, and the two points are ex-

Figur

89.

.C:P) ¢,'.'I”'C‘.iz"l

.J -- 4—O— v,

 

 

INCH - Io‘ .

EXPANSION IN INCHES

 

 

H015 fi‘C‘

,0
a, \ "
I . C )0 'L

b.
/ ' 1,9u"

eoo'c

“NH!"

. ‘/
IIIHIMEI I,

 

 

TEMPERAIURE UNHS a

H‘ \Hl’lt'.

(.001 ”NC

 

 

 

Figure 90.

..-—-v- «*«s—

Figure 81.

C:

..x
cw

Cu
I

 

 

 

“
I
o 3" ‘
l
T— y. r——
I.)
1
I a pa
3
(n 'N ——
Lu
T.
O .1
:7. 0 ’— 5"
... £7,
1. a f
- f
‘— ,6
0 ¢'-— ,1"
'— I
(f) 1 «
Z /'
,
( n *— /
E1 ("3'
'( ,v
. .'
..J N F— I.
I.
l
.
C
I
up 490

’I
J “J 050 C U‘" ‘
U ‘ I
{.-
I I . ;
I w .
‘o
f I
i
V
I
i‘
J
F C
I .'
I
’
I
I I I | .
I , . . 7 l _ ..A
90° 590 I000 0 2 n a.

'EMPL'RAI URL UNITS .

\s'C

 

__.__.. _.—_ ____,_,_

 

-41-
pressed as one. According to results published by Dr.Albert
Sauveur he gives two Ac points, one at 775°C(l4270F), and
the other at 855°C(15710F). In the Opinion of the writer
these should be regarded as the same point, but the sluggish-
ness of the reaction causes a variation of at least 10000 on
similar tests.

0 o

The writer believes that between 800 and 900 C the trans-
formation from alpha ferrite to a solid solution of gamma
(austenite) iron is completed. MicrosOOpic evidence of an

austeniticcondition after quenching from this point is

lacking, but’such a condition is difficult to show.

The increased hardness resulting from quenching above this
i temperature is thought by the writer to be due entirely to
solution of carbon and carbides and the heterogeneity of

carbides retained in the solid.

The martensite precipitated after a high quench will, of
course, have a much higher carbon (hardening) content if
time and temperature is given for dissociation of carbides
and solution of carbon by the solid solution of austenite.
The hardness of martensite is variable due to its carbon
content. This carbon percentage is of course due to the
percentage of carbon, elemental or as a compound, in solu—

tion in the preceeding phase, namely austenite.

After observing the exceptionally low Rockwell hardness

-42-
obtained by cooling specimen slowly immediately after the
Ac transformation was completed, see Figures 79 and 84,
it was decided to repeat this test with hardened samples
quickly and slowly cooled immediately after completion of
the gamma to alpha iron (Ar) transformation. These tests
are shown in Figures 88, 90 and 91. The first sample run,
graph shown in Figure 88, had a hardness of 059/60 after
hardening at 21000F(1149OC) in air. Specimen was heated
to 820°C(15100F) and cooled slowly until the Ar transfor—
mation was completed and then allowed to cool slowly in
tube of furnace. Rate of cooling was 26.6 deg.Cent. per
minute to the Ar point and from there to atmospheric
temperature was 13.800 per minute. The Rockwell hardness
resulting from this treatment was 010/15. This curve was

not plotted as it is not materially different from many

of the preceeding graphs.

Figs.90 (17 and 18) and 91 show the dilatdétric curve and
plotted graphs of the rehardened specimen. The Rockwell
hardness after treating at 20500F(112000) in air was 053/55.
Both tests were permitted to cool slowly in the silica tube
until the Ar transformation was completed, this point was
at ‘635OC(11750F) in number 17 and 645°C(11930F) in number
18. The rate of codling to this point was 23°C and 23%00
per minute. At this temperature the furnace was removed
from the containing tube and cooling hastened by fanning.
The rate of codling was 59°C and 57.200 per minute. The
Rockwell hardness resulting was 040/42 and 024/25 respec-

tively. With these two samples, treated identically alike,

-43-

there is an appreciable variation in Rockwell hardness.

It is to be noted that quick cooling after the Ar trans-
formation has been completed has caused a considerable
increase in hardness. Inasmuch as there is no evidence
of a critical transformation it is difficult to under-
stand the resulting increase in hardness. Owing to the
differential in the increase in hardness resulting from
identical treatments it is the thought of the writer
that this might be due in a degree to precipitation of

carbides and tungstides.

Figure 92 shows a photomicrograph of a section of the
test specimen after a slow cool immediately following

the Ar transformation. Rockwell hardness 08/10.

13,
86.

14,
87.

Rate of
Heating

19°/min.

43°/min.

18°/min.

420/min.

220/min.

41°/min.

18°/min

Dilatometric Details

Max.
Temp.

970°C

975°C

970°C

910°C

900°C

920°c

920°C

930°C

Ac
' Point

825/900°C

875/900°C

770/815°C

760/800°C

755/810°O

ass/850°C

-915/aso°c

375/90000

Rate of
Cooling

--~gp - c.-

25°/min

22°/min

709/min

26°/min

26°/min

86°/min.

76°/min

17°/min

Ar

Point

Arl
725/7oo°c

Arg
420/3ao°c

Ar1
BBQ/770°C
Arg
490/42000

315/29000

Arl
680/630°C

Arg
320/26500

Ar
sad/635°C

Ar
358/31000

37o/3so°c
375/36500

BSD/800°C

Rockwell
Hardness

o 35/36

C 20/24

0 50/52

0 39/41

o 39/41

11,
84,85.

89.

16,
89.

Rate of
Heating

19°/min

l7°/min

l4°/min

26°/min

50°/min

42°/min

40°/min

Max.
Temp.

Ac Rate of
Point Cooling

 

910°C

890°C

875°C

840°C

840°C

870°C

860°C

BOO/815°C 16°/min

BOO/825°C 14°/min

830/88000 7°/min
loo/min

790/83000

780/815°C 53°/min

750/820°C 26°/min

760/820°C 102°/min

Ar
Point

Ar1
BBQ/630°C

Ara
360/32500

380/32000
780/740°C

745/71500

Ar
768/69000

Arg
400/31500

700/625°C

340/270°C

Rockwell
Hardness

C 39/41

0 4/6

C 8/11

0 28/29

Figure 92.
x 2400

 

 

 

 

 

C 0 N C L U'S I 0 N S

1.The hardening of high speed steel varies with the tem—
perature of quench, reaching a maximum at a point just
preceeding incipient fusion. Going through this point
causes excessive grain growth with a softening and em-
brittling effect. If held for a long period of time

complete diffusion of carbon takes place.

2.Agter passing through the critical range on heating,
800 to 900 C, the increased hardness is due to the in-
creased solution 0f carbon or carbides resulting from
time and temperature given sample. The heterogeneity

of precipitated carbides is shown in photomicrographs
and without doubt is a large factor in the hardness and‘

stability of this steel at elevated temperatures.

3.Elevation of temperature has little or no effect in
breaking up the carbides and eutectics formed at time of

pouring ingot.

4.The irregularity of the Ar critical points of this

steel when cooled from varying temperatures is due entire-
ly to the varying percentages of carbon going into solup
tion with the gamma iron, and the corresponding varying
percentage in the martensite formed on cooling through
this critical range. The hardness of this latter struc-
ture varies directly with the elevation of temperature to

which the specimen was carried.

5.Martensite is of variable carbon composition.

6.Rate of cooling high speed steel after passing through
the Arlpoint increases the hardness slightly. This applies
only to steels cooled immediately after completion of the
Ac transformation and consequently having austenite of low

carbon content.

 

 

 

y.“ (I‘M... ‘
f ‘-

.2 ...
. W :

  
  
        
     
      

 

..‘1

 
 
  
 

I‘-
sl'.. 1’
. ‘. .u '7‘.
' \ - D.
. LR _ '
‘ . .
, . _ fl .' I. 9' .
. ' -- a ' ..-, ' . '
h ,Z'g $"‘ \ ‘3 “ -‘ \A' -. .‘ I” A ‘ ‘
. 'v a .3 _ - e I“ ... .--‘-4 :. , >1
rail!" 'fiilif?$:‘{' r3 ‘2 ' \ 1. fi‘)‘. ‘ (4* b'a{.t-{ 13:. . 11’3” ‘3“. ’0' ‘ (. I " ‘
'." I _ .I . . 5' .-~-u '1 l1“, ... 'I’ ...| i ‘ "_ .‘ I 1‘
. . I. , ,I ’.l ' .t lI‘ ‘ wat‘é‘iJZN- [I 4.”. ‘ ? . 5.1;. b ‘ ‘~h‘ .‘ \'\ 1-: ' 5 ‘ I
I?“ '$>VF’ s \‘ I I ‘ ' I ”I“ ‘ 0" ‘l"._.V. ‘ .
. . MI ... 12; . 1+. .-. ,
‘ 1"; 5“- "Skyh. \‘Q‘?H .‘k I ‘1’ ‘ .Irslf' ‘x-‘VIX \ a}! .3 'i} . XIP‘ ‘* ' I
"""" 3“ " " 1"“): ‘ ' 4 I “a“ r i I J ‘ ‘s -"'\.".". "T .N‘ \ _
A” "~ ‘ i I ’ ‘ Q I r’ ‘ I e v
“Fr/738K 'Eakfll“: '3}? ”(I 71’ .1. 6‘- -" v" " "v .

    
 
   

‘sr,
“ts:
. .925}:
{.3 .3
.. 38;:
23..
i".
g‘

is-

    
 

     
         
  
 
    
   
  

   

“‘3‘" " '
, . .. v '
.10." thw. \ Ii ' 4511.0;
_ 0 K . I ‘ ' I -
Csfi“: I35". .5": “(J ‘ " f \'. ‘
'0 . ‘I ' ‘ I.
"4‘2 M.- .fiéfiig’f. ‘23-"?
="p‘ M .‘I‘.‘-“‘ ’
E’j‘xfi..$bfl *‘3 " .
i. 1.. .‘----:w\x~'.-' «" a" ' ' ‘
\ . .. x, ...M . -
xi f“ 4: ‘VJ " IR- -'" '- ‘ ..H ‘
‘5’ "LR-I‘LF‘LI‘ 71:1; “’ "liq", ‘ '( ‘ I: ‘l I '
‘f 3‘ _C,‘ a: {54". ‘ r1 | 1|... "V ._ . 'n -_l . )l
4:":KI‘JN ' “H l-. . t I'
‘ ‘3. a- ' ~
P i‘ \. ‘ .. -- ‘
c

 
 

‘ 0 ‘d‘ ’-
“ . . .a‘ p. M33.
0'. . 4’ 'I-fi I‘. ‘71.. '€"'t~§1
. -' ' «M vat-N ,
I.- ......Iv ‘ lgg'm?£. .
"t u I ‘. ‘ ' l
.sr ~3’».I*;..»'§’»’-:. - ‘. . "
.vl'hfl » . q' ‘ ‘1 :.; ‘ ... . ..
o I - ‘ ’ III.
; ”‘2‘" is».
My.

     
 
 

    

\"‘

 

'1 . I - . n
' .' .~"‘ I 1 "-9 2". 'r - " l I I- c
. 4: fl... .-j.‘i 2‘9““‘\"5J. S:}-l!t ”...; 08"."- ‘3 '1}.
, «...-h.“ .1 '5. ... ' KII'thwk . ‘. ,l J“ 1'. ‘-
.4;. ~.- “J'.~"~.«..).~'.~ .. rum)“ gin.
- . ..i- mm W . ,
i - I. p I . ' .

.
i . , ,
2b“..' U; '0 "\~ '-I.. ' _ ~‘ 5- . I .“ Q
L '1' $1. 23’! “5‘". u: I. “:11: o .3 :. "3.;N\l;"‘. |§

  

r;
71's
3”

......
f
,.
,.

5.15"" .
959.513 -'
‘ 1‘

I‘—
.~
1‘19..._
-_ 5-H.\/_y

5’.‘

 

         
   
  
 

  
    
     
 

 
   

  

  

         
 
 
  
 

  
 
          

I.“ . _ .
If? ' . a ‘
”If"? . -
I: A". ., .
‘. v ‘1‘ .
ng'i ‘ 'I‘I '
"\, —q 0’
min 7 ,i- " -
..1 I_. v ‘ .
. . :k‘. ;'.._ .,’ '. i
fiat/“4r" 'C'V' . .274." . If" '
24‘ ‘t‘. | "1")".- . .. - .-.' _ w; . . . r A"
«ite‘iiaxviw's. tr“ 2 3-.- r" 4'. «M #4»-
q: ' "J 2:. “WA 2. .‘..5. £3‘,.": ‘. ‘ ' . ~ ‘ ' . 3'") “‘3" |-' ~
‘ 1 ‘9’!!!) “'(0'3'; . ;? 3"." ‘ ~ .~;‘ I!” I’xi (.1. . ;‘. .1 n ) ‘
'. 1.1- "4 .' .h,_,dg "-.'.‘ " '-
‘.’.""\J { fY‘ ' 5:.” f‘"';f:‘ \L’ ‘
f‘) I .' ' a I. \ ‘ . . J "”T ‘ ‘ \. _ ".."n ‘K’"‘ O \
«31"; 1.3 a"? J (f.?.:o; (' -._ .1: 3’4" 33" u ». '
.w" '98“. I Ht 2'2“?“ “'5. 33:; ".-"d
'-..:.‘r.-_I.*c“$l\.'"ifi‘7‘ Lug ‘. . . .~ "11.3.2; .71“ . ‘
. 'L“;;"r2"3r\f2.i-V.¢f 3.99305: " "4';"v\.i7."};..£{ ". 73¢ ’efir'i7.e; _ {:f.
2-'~-’i’...?.’;v?-T°‘~‘I‘v 445%.": 29‘v7!é’;'li)r.\f'- 22M! 1‘79; I" .
-‘-§'- Inf , 4.1;; " ‘.4II--- ".V‘.‘ w . . . ...-,Tnflfiufl" . :- ~
. #‘W‘ \k - U. “ELM." ’. vh‘lfl: r d' J _- 5.: ‘. I." > g .... . U". u‘.’ .‘i' .'
' ' ' ‘ ' ’r“;('£n'.- J A. *2”. I ‘ M ’ ' " " " - ' ‘4 ..‘fi" - .I \H'Wh
".. :5 ' "<6 ‘yafli’gfi‘a’f “03915-237 'J 3'.” "J ‘13 ' . . "' ’ I. ' I”, ‘ Iv'.'s'NJ‘.' J ‘ ..
‘ V. l%‘(."“;p~;""+" "‘9‘6‘3h'fe'r‘ 35‘ 5",! "-';'I"-inuq '15-? @531" ‘1 ' ‘ ' H ", ‘ .fiEWIF-l-IAUI .. .- L ‘
_ “(1,... (3“ 3.141“ ' 3'.“wa . v.:| ¥C?‘:.:~4" if ' “(9. J: ”f . V A ,‘dtak‘vfi A, . ' . _ 5 -4 . .
». '.'~ . . "£-‘.*°I+s"-."'i.i'='.‘"~ {495.5 -. 'I nfi'f‘ own..- ..g ‘. " .. gr. .vy‘fif' '-.~..‘ -
I i'. {3“ I". Livfig’yvs. Jitfv'}! «Jr...» x? "‘g‘ «“33‘ ‘ kg. 9..'€\ ' 'S I V ' l Iv '.{*‘E‘\‘A ~20». .. 3 :' . i ‘- i’
u . _l ' u." :1 ."- t" 'I' “L; .A." ' Z a M‘I'Jf \Q :A.d;ll‘r.’ .1 i: "_ 3",5- _‘.{ ‘1 .- v ,. ‘ 7.11 “k .“ «1". : ‘. ‘I ' '
' ‘ < "' "e“ «W I'«'~ °—i ...-“at?“ A5: a - . MI- $5,153,. _ --
“$1 5' -‘I ...:9' _ .;. ... z‘; :27“. ;; 3"}: .§ k. ' ', 1&1!" 3)" § *antl‘u.’ "Jo I'.'*i;§z.‘7k§. . .' I :27" m‘iw \‘ h. i"| V. . . . . I -. .
Ii ‘1 '1 ‘s 921'?» ‘21"? ea "2:3 3‘ .a' $41M .- . .4 g" a}; "(a ‘J‘Lnflm v. '. .- ,
w: ... “1| 7 ‘9. .8 . ‘\_ -‘ ._.\ 1. ma : .- _. ‘ _ ‘ ._ 9 c . . ' _ .; ‘ I. _ '7 u v... : T ‘
l ' II ' ‘2 3%? ‘W " 4.? 3' 3 1833‘." - “'5: at? ' ‘ flirt-.2 3. {Qflugogn
r \ ... ”A" '\ ‘ {él’afi‘ik’.". 3") V n) . " 't’gi ' t O" .13., a ‘5 .??I Ivy“, . .;l ‘C ‘2 I
‘ .« r . "\|.‘ a‘ I p'-r‘~4’ ’.. ‘)1 | ‘6. ‘3?! :. v-J“‘ IL . ‘w
‘: ...l‘ ' . ‘1' \3' #‘l g u ‘9' '\ Q '2‘ i“; I" '. ‘ u.‘_' . I~ ' . ’-
M h:.’~‘". Lt“\:fi.%d;iéé‘ Qn'!‘ .1 ‘ I. )i-“' — 1*}; . ‘ 1 ' _ ‘2‘") ‘I ‘ -: . ‘ . o. “
xiii-III" 3’ via"? ”533.;"#3845371‘3‘ IWit-I. 5&7.“ '4 {A .I. . 313—.N.,’:.~ "1 I”; ‘ .
-~ "2'5“” ~ ~ ~ , - \',}‘o?"_3f..“-'I’«‘L'ffd ...

’“N . .a . fgalhgfl' .,k,',_ w; ‘ “‘4‘
3m it. :59; fixes-ax .. ‘t

I I ‘

 

-' ft»? ""1 - 7 . ' . ~ ~"- "Ié\!‘r‘\£‘f:q, .‘l' o L“ I 17?, . ‘ti. .‘
. -“"§1‘1‘*“‘** (4.938% .L"“s~e~‘.-'*«r">a'vi:." pfimn 1...: ~2~§a '
‘ :- ‘ 4'le . ’J- "‘1 x ’9". 5‘7““. .. i . a " ... V? '41-" ' «.3 "n.

3- ‘3'

”'TITI'ITIIQITMM MI ljfifllflffl'fiflilfl 111111 Ifijii III?!”