By C. H. KAIN and E. W . D O W SO N (Members) two test-pieces o f reasonably uniform properties.
W hen how ever the tongue of the block was space between the prim ary crystals prevented the effective gravitational feeding o f the middle exactly uniform properties. The present Paper describes the steps taken in the foundry with
It seems clear that directional solidification and adequate feeding alone will no t produce three test-pieces having identical properties, since in these tests all three bars were cut from solid m etal below the feed head and pipe.
The growing tendency o f inspecting authori
ties to dem and three tests (usually tensile, bend
possible from the centre line. T he results are possible surface being available to produce this condition. I f the bars cut as show n in Fig. 3 freezing, even though these provide direct feeding influences. previous publications,2 3 how ever, dem onstrate th a t such a m ethod is n o t satisfactory.
N on-ferrous and certain C ontinental steel 62
foundrym en seem to have given m ore attention to the production of satisfactory test-blocks than steel-foundry investigators in this country.
Fig. 4 is an illustration o f a test-block which is som etim es used in this country for high- shrinkage alloys, such as m anganese bronze, and is used in C ontinental foundries even for metals like stainless steels. It is claim ed4 th at this block gives very good results.
A fter consideration, certain modifications were m ade to this design, although the basic principle was retained. It was decided to use three bars o r tongues instead o f fo u r and to em ploy a circular head with the diam eter of the inscribing circle. In order to avoid the danger o f cleavage planes, it was decided to investigate the possibility o f a rounded section and a block was evolved as shown in Fig. 5.
Ta b l e IV .— Results given by Test-pieces Machined from.
Block shown in F ig. 5.
H eat No.
P osi
tion.
Y .P . M.S. E. R .A.
Tons per sq. in.
Tons per sq. in.
Per cent.
Per cent.
A .1272 lSquare 2 0 .6 4 3 3 .3 2 25 3 3 .5
2 2 0 .4 8 3 3 .5 2 31 4 4 .9
3 3 0 .3 6 3 3 .4 0 30 4 3 .3
4 Centre 2 0 .8 4 2 8 .4 0 10 1 2 .0
In addition to the avoidance of cleavage planes, the rounded section has the advantage of requiring slightly less m etal and being som e
what easier to machine. The first results are shown in T able IV.
T he inferiority o f the bar cut from the centre dem onstrates clearly the advisability o f adher
ing to the first two rules enunciated. Also, the slight superiority o f the rounded bars over the
Ta b l e V.— Results obtained from the Design of Test-block shown in F ig. 6.
H eat N o.
P osi
tion.
Y .P . M.S. E. R .A.
Tons per sq. in.
Tons per sq. in.
Per cent.
Per cent.
A. 1285 1 19.32 3 1 .0 4 33 4 6 .0
,« 2 19 .3 2 3 1 .0 8 34 4 2 .0
, , 3 1 9 .0 8 3 0 .9 2 34 4 7 .2
, , 4 centre 18.68 2 9 .2 4 15 1 6 .8
Fi g. 6 .— Fi n a l De s i g n o f Te s t- Bl o c k Ev o l v e d.
square section justifies the adoption o f this section.
A final design was then adopted as shown in Fig. 6, to conform with all three rules. The bars freeze under identical conditions; they have
4 M , [
Fi g. 5.— Fi r s t Ty p e o f Ve r t ic a l l y Di s p o s e d Te s t-Bl o c k. 63
a u n iform ly large surface fo r prim ary freezing
A lthough the im pact test is not as yet fully understood and has not been accepted by steel founders for general use, com parative tests were m ade as show n in T able IX.
In every case exam ined, the clover-leaf type o f bar yielded slightly higher and very m uch more uniform results.
Summary
(1) The freezing phenom ena of test-blocks designed for the provision o f test-pieces for steel castings have been considered.
(2) Rules governing the design of an ideal test-block are suggested and a section resembling a clover-leaf recom m ended.
(3) D ata are given in support o f this recom m endation.
Acknowledgments
T h e authors wish to record their thanks to the directors of Lake & Elliot, Lim ited, for permission to publish these results.
R E F E R E N C E S . 1 P ro c. I. B rit. F ., vol. X X I I I , p . 507.
2 „ ,, ,, vol. X X V I, p. 501.
3 „ „ ,, vol. X X V III, p . 661.
* “ F o u n d ry T ra d e J o u rn a l,” vol. 62 (8940), p. 7.
W R I T T E N D I S C U S S I O N
M r. F . W. Ro w e, M.Sc., w ro te :—This design o f test-block is a m atter to which the writer has devoted a certain am ount of tim e, but w ith
out the successful results which appear to have attended the authors’ efforts. I t is thought that answers to the following queries would be of general interest: —
(1) Is the clover-leaf type of test-block equally satisfactory with a 9-in. length o f test-bar as with a 6-in. length?
(2) Is the com parative gross weight with head available as against the ordinary type of tongue block usually used, providing the same length of test-bar?
(3) A t first sight it looks as though the clover- leaf test-block m ight be m ore expensive to cut up than the ordinary one. H ow is the block cut? A re the head and runner burnt off and then the three test-pieces milled out, or how?
On the usual tongue block the w riter cuts his test-pieces from the block w ith a hacksaw.
(4) In the m ajority o f cases even now only two test-pieces are needed, either tensile and bend or tensile and im pact, bu t a tongue block about the same proportions as are shown in the P aper is called for, because the “ C ” portion is, as the Paper proved, usually rather poor.
(5) Could another P aper be w ritten on the best design of test-block where only two test- bars are needed?
A U T H O R ’S R E P L Y
The Au t h o r s answered Mr. R ow e’s queries
as follow s: —
(1) The clover-leaf test-block has been found to be perfectly satisfactory for test-bars as long as 15 in.
(2) The clover-leaf type of bar is a few pounds lighter than the equivalent tongue type of block m ade for three bars.
(3) The question of cutting off the bars is an aw kward one, and it is usual to slit down the three tongues with an oxygen cutter before heat- treating. T he bar can then easily be cut oil with a hacksaw.
(4) Even where only two test-pieces are re
quired, it is the practice always to provide three in case a spare is required or in case o f dispute
(5) There can be no doubt that the tim e will come when the im pact test will be much more generally used than it is to-day, and it is really advisable to be prepared to produce three good test-bars, when required. N o great difficulty is thought to arise in the production o f three test- bars of reasonably uniform properties.