WO1999061886A1

WO1999061886A1 - Characterisation of thixotropic material specimen

Info

Publication number: WO1999061886A1
Application number: PCT/GB1999/001453
Authority: WO
Inventors: Plato Kapranos; David Harvie Kirkwood
Original assignee: University of Sheffield
Current assignee: University of Sheffield
Priority date: 1998-05-26
Filing date: 1999-05-26
Publication date: 1999-12-02
Anticipated expiration: 2000-11-26
Also published as: EP1082598A1; GB9811237D0; GB2337822A; GB2337822B

Abstract

Thixotropic specimen slugs (22) are isothermally compressed at high and constant speed to induce a shear rate in excess of 10 s-1, the load required to maintain the compression being monitored at a rate exceeding 2 KHz so that a peak force required at commencement of the compression to break down the microstructure of the specimen, before shear thinning is experienced, can be detected and employed to characterise the specimen.

Description

CHARACTERISATION OF THIXOTROPIC MATERIAL SPECIMEN

The present invention is in the field of thixoforming of metals and alloys, and relates to a method of, and apparatus for, characterising specimens of such metals and alloys to enable their suitability for thixoforming to be determined, and to compare different materials and materials prepared under different conditions.

Semi-solid metal forming, tMxoforming, thixocasting and thixoforging are all terms applied to a relatively new technology in the forming of complex shapes of metal bodies in a mould or die. A metal or alloy is given a suitable microstructure, usually by stirring during solidification or by remelting of a re-crystallised structure, but in any event, so that it comprises essentially globular, non-dendritic grains. For example, EP- A-0305375 describes one method of preparation of an alloy suitable for thixoforming.

Then, a slug of the material is heated under precisely controlled conditions to around its freezing point where it comprises solid grains in a liquid matrix having a variable fraction solid, depending on the temperature.

Usually the slug is allowed to soak for a period of several minutes during which the temperature and fraction solid of the slug is stabilised, and possibly during which some modification of the structure occurs, for example, further spheroidisation of the globules. After this time the slug is then rapidly injected into a die. Under high shear, the viscosity of the material in this condition rapidly dirninishes and it flows into the die in a controlled and relatively uniform manner, it having the consistency under such shear akin to thick oil. The injection is characterised by an essentially smooth front of the injected material as it penetrates the die so that gas is reliably ejected from the die without entrapment.

This contrasts with the injection of molten metal in normal die casting, where the metal or alloy tends to have a much lower viscosity akin to water, so that it has a tendency to jet and spray into the die possibly trapping air and leading to porosity of the final product. Secondly the thermal shock to the die is considerably larger in normal die- casting; so that die life is dramatically improved by tmxofoπmng. Finally, it is found that the microstructure of the material is not significantly affected by the forging process and so, as long as the microstructure required in the finished product corresponds with a microstructure which lends itself to thixofoπning (ie non-dendritic), that microstructure can be established in the raw material for the slugs which are to be injected. Thus the integrity of the component is improved and usually less postprocessing of a thixoformed product is required.

However, it is found that different materials have different characteristics; and even allegedly the same materials from different supphers, as well as different batches of the same material from the same supplier, may sometimes exhibit differences. This means that each new batch of material must be tried in varying conditions to establish the precisely appropriate condition for tmxoforming.

There is therefore a need to characterise the rheological properties of thixotropic materials so that manufacturers can recognise how materials will behave in thixoforming and adjust their injection conditions, including the heating and soaking steps, accordingly. Also there is a need to be able to compare thixotropic properties of materials for the purposes of quality assurance. No such methods are currently available.

Accordingly it is an object of the present invention to provide such a method, and also an apparatus for effecting such method.

In accordance with a first aspect of the present invention a method of characterisation of a thixotropic specimen comprises the steps of:

heating the specimen to a temperature at which it has a predetermined fraction solid;

soaking the specimen for a period of time;

compressing the specimen with one of a predetermined load or a predetermined compression rate, sufficient to establish an average shear rate in the specimen of at least 10 s^"1, preferably at least 10² s^'1, more preferably at least 10³ s^"1;

detecting the load applied or the compression rate, at a sampling frequency of at least 10³ s^"1, preferably at least 2 x 10³ s^'1; and characterising the specimen on the basis of the change in load or compression rate, as the case may be, detected during compression.

Preferably, the predetermined compression rate or load applied is maintained constant during compression. Preferably, it is the compression rate which is predetermined, and in which event it is the load applied to maintain such compression rate that is detected.

Preferably, said characterisation comprises a peak value of load or negative peak (ie drop) in compression rate on commencement of compression of the specimen.

Such a method has two essential features. Firstly, the usual industrial conditions of thixoforrning are reproduced. Secondly, and equally importantly, the sampling frequency (ie rate at which measurements are made) is sufficiently high to detect the apparent peak of viscosity of a thixotropic material in the moments before its structure "yields" or "breaks down" as substantial shear is applied and at which point its viscosity rapidly reduces to the low level at which it flows. Indeed, it is the appreciation that the shape of the load/(time or compression) curve, particularly the peak (if any), is characteristic of thixotropic behaviour that the invention finds its application. Depending on the curve, which might possibly be reduced to one or two critical values taken therefrom, the suitability or otherwise of a material for tmxoforming, the expected behaviour of the material, the conditions under which it should be heated, soaked and injected, and comparisons with other materials, can all be determined, or at least approximated.

The first aspect of the present invention also provides thixoforming material characterisation apparatus, comprising:

a first plate on which to support a specimen slug of the material to be characterised;

a second plate opposite the first;

a ram to urge the two plates together;

control means, adapted to operate the ram at a predetermined velocity, regardless of load, or a predetermined load, regardless of velocity, to establish an average shear rate in the slug in excess of 10 s^"\ preferably at least 10² s^' more preferably at least 10³ s^"1; and

a sensor to detect one of the load applied to, or compression rate experienced during compression of, the slug by the ram, the sensor being adapted to operate at a sampling frequency of at least 10³ s^'1, preferably 2 x 10³ s^"'.

Preferably, said control means is adapted to maintain the predeterrnined velocity or predetermined load constant during compression. Preferably, it is the velocity which the control means is adapted to maintain at said predetermined rate, and said sensor is a load sensor.

Preferably, the ram moves the first plate and the load sensor is disposed between the second plate and a fixed frame of the apparatus. Preferably, the plates comprise a relatively hard, thermally non-conductive material such as an asbestos, or asbestos-like, composition, for example Sindanyo. This ensures as far as possible isothermal conditions during the compression. The hardness of the material reduces any cushioning effect.

Preferably, the second plate is mounted in an aluminium plug having tapered sides and received in a corresponding bore of the frame, the sensor being mounted between the said plug and the frame.

Preferably, the frame extends from said bore around a slug mounted on said first plate.

Said first plate may be integral with the ram. The tapering of the plug ensures that it lifts substantially fiictionlessly from the frame on compression of the slug by the first plate.

In a second aspect, the present invention comprises a method of characterisation of a thixotropic specimen comprising the step of quantifying the breakdown point at which flow of the specimen commences. In this aspect, the invention also comprises thixoforming material characterisation apparatus, comprising means to quantify the breakdown point at which flow of a specimen to be characterised commences. In this respect, prior to compression of a thixotropic material, and before its viscosity is reduced to a level where injection into a die may be effected with minimum resistance, its structure temporarily resists such compression, and the present invention appreciates that such resistance is an important characteristic which varies from sample to sample and depends on the condition and constitution of the sample and its mode of preparation. Therefore, the detection and quantification of this initial resistance, which looked at another way may be considered as the material's yield point, in specific materials will be of value to manufacturers employing thixoforrning processes. The method and apparatus of the first aspect of the invention are suitable for putting the second aspect into effect.

The invention is further described hereinafter, by way of example, with reference to the accompanying drawings, in which:

Figure 1 is a schematic view of apparatus according to the present invention; Figure 2 is a detailed view of a compression cell of the apparatus of Figure 1; Figure 3 is a plot of load or force applied (in KN) against time (in s);

Figure 4 is a plot of viscosity (in 10⁴ Pa.s) calculated from the values of Figure 3 against calculated average shear rate (in s^'1); and

Figure 5 is a micrograph of the microstructure of an aluminium alloy sample after rapid compression in the apparatus of the present invention.

In Figure 1 of the drawings, a thixofoirning press 10 is shown comprising a frame 12 having a base 14 on which an actuator 16 is disposed. The actuator 16 has a ram 18 which is adapted to rise and fall and which, on its end, has a pedestal 20 on which a slug 22 of the specimen material is supported. In a heating zone 24, the frame supports induction heating coils 26 supplied by cables 28 from a power source 30.

The induction coils 26 when energised with current heat the slug 22. After the heating process (described further below), is complete, the ram is raised to introduce the slug 22 into a compression cell or die 32 held by a die clamp 34 of the frame 12. Figure 2 shows the compression cell in more detail. The pedestal 20 comprises a first plate made of Sindanyo which has low thermal conductivity. The pedestal 20 slides in a close sliding fit in a graphite core 36 mounted in the frame 12. At its end, the graphite core supports a further Sindanyo second plate 38 which is mounted in an aluminium plug 40 having tapered sides 42. The plug 40 is mounted in a correspondingly shaped bore of the graphite core 36, and on its other side mounts a steel cap 44. A load cell 46 is pressed between the cap 44 and a steel backing plate 48 of the frame 12. The plug 40 has tapered sides so that, when the ram 20 raises the slug 22 against the plate 38, the transmission of the load to the steel plate 48 and load cell 46 is essentially instantaneous and frictionless. Returning to Figure 1, a computer 50 controls the apparatus 10 and receives signals from the load cell 46 for subsequent processing.

Slug 22 comprises any metal or alloy which it is desired to characterise in relation to its rheological properties and will typically comprise an alurninium alloy or other metal/alloy. In the present embodiment, it comprises a cylindrical slug 42mm high and

36mm in diameter of an A357 aluminium alloy which had previously been warm extruded. The slug is first heated for about a minute within the heater section 24 to a temperature just above the freezing point of the alloy, whereupon it will have a fraction solid of about 0.5 depending on the precise temperature. The sample is then allowed to soak at constant temperature for about three minutes. Steps had previously been taken to ensure that at the end of heating and soaking radial variation in temperature of the slug of less than 1°C existed. This comprised inserting two thermocouples halfway down the slug, one near the surface and one in the centre. Thixotropic alloys typically employed in thixo forming will have a microstructure substantially as shown in Figure 5.

This shows glor- ales of 60 of solid material in a matrix 62.

Because the velocity of compression is arranged to be between 200 and 2,000mm per second, and if earned out between plates 20, 38 of low thermal conductivity and within the chamber defined by the graphite core 36, it is assumed that isothermal conditions pertain for the period of the test.

When the temperature of the slug is at the desired level, the ram is raised and the slug 22 compressed between the plates 20,38 at the defined rate. The load cell 46 is arranged to provide signals corresponding to the compressive force on the slug 22 and the computer control 50 is arranged to collect such signals at a rate of two kilohertz. In Figure 3 the measured load is plotted against time for the alloy A357 at three different temperatures, namely 572°C, 574°C and 576°C. In this example, the ram was advanced at 500mm per second so that a time of 0.06 seconds represents a compression of the slug from 42mm to 12mm in height with consequent sideways expansion until contact with the core 36 and resultant rapid growth in the load when the progression of the ram 20 is stopped.

The three curves a, b, c in Figure 3 are as follows:

Curve a is at the lowest temperature, 572°C and represents a fraction solid of 0.6. At time zero the load applied rises rapidly to a peak value of approaching 4 KN before equally rapidly decreasing to a minimum of about 0.4 KN at time 0.01 seconds, whereupon the load remains substantially constant over the next 0.025 seconds before the approaching end of the test when the load rises rapidly. Thus the peak lasts for a period of less than 0.01 seconds.

Graph b shows the alloy at a higher temperature, 574°C and therefore a smaller fraction solid of about 0.5. Here the load rises at the same point as for curve a, but reaches a first peak at a value of 1 KN. However, this probably represents a temporary collapse in the structure of the slug which is picked up again almost immediately so that the peak load of about 2 KN is reached and at substantially the same time as the peak for curve a. Again, the load drops off rapidly, and within a corresponding time frame to curve a.

Finally, curve c shows the alloy at a temperature of 576°C and the peak on this curve is hardly noticeable and occurs after a delay of almost 0.01 seconds. The delay probably represents a slump of the slug in view of its much lower fraction solid (of about 0.4) so that there was very little resistance to compression until the end of that delay. However there is still a marked peak, representing the breakdown of the viscosity, but only of about 0.3 KN.

Thus, armed with this information in relation to a test sample of A357, a manufacturer wishing to employ that alloy for thixoforging knows that it is certainly suitable therefor, it having a low viscosity once the peak load has been overcome and, that the appropriate temperature of operating can be selected depending on the manufacturer's requirement. For example at 576°C, the alloy has the lowest viscosity but may not have sufficient integrity to be self-supporting. At 572°C however the initial loading may be higher than desirable increasing the impact on the ram. Thus graph b may represent the ideal condition in which the material is sufficiently stiff at rest to maintain its integrity but the initial load required in order to initiate the thixotropic effect is not substantial.

Turning to Figure 4, this shows the calculated average viscosity as a function of the average shear rate, also calculated. It is apparent in Figure vith increasing temperature of the sample, its viscosity decreases, in both its pea. ÷ue and minimum value during compression.

Claims

1. A method of characterisation of a thixotropic specimen comprising the steps of:

soaking the specimen for a period of time;

compressing the specimen with one of a predetermined load or a predetermined compression rate, sufficient to establish an average shear rate in the specimen of at least 10 s^"1, preferably at least 10² s^"1, more preferably at least 10V;

detecting the load applied or the compression rate, at a sampling frequency of at least 10³ s^"1, preferably at least 2 x 10³ s^"1; and

characterising the specimen on the basis of the change in load or compression rate, as the case may be, detected during compression.

2. A method as claimed in claim 1, in which, the predetermined compression rate or load applied is maintained constant during compression.

3. A method as claimed in claim 1 or 2, in which, it is the compression rate which is predetermined, and in which event it is the load applied to maintain such compression rate that is detected.

4. A method as claimed in claim 1, in which said characterisation comprises a peak value of load or negative peak value of compression rate on commencement of compression of the specimen.

5. A method as claimed in claim 3, in which said characterisation comprises a peak value of load on commencement of compression of the specimen.

6. A method as claimed in claim 5, in which said characterisation comprises the duration of said peak and or the difference between said peak and a minimum value of said load after said peak.

7. A method as claimed in any preceding claim, in which said specimen is compressed at a velocity between 200 and 2,000mm per second, preferably at about 500mm per second.

8. A method as claimed in any preceding claim, in which said specimen is a cylindrical slug having a diameter of between 20 and 50mm, preferably between

30 and 40mm, and a height of between 20 and 60mm, preferably between 30 and 50mm, more preferably about 40mm.

9. A method as claimed in any preceding claim in which said compression of the specimen is conducted isothermally.

10. Thixoforming material characterisation apparatus, comprising:

a second plate opposite the first;

a ram to urge the two plates together;

control means, adapted to operate the ram at a predetermined velocity, regardless of load, or a predetermined load, regardless of velocity, to establish an average shear rate in the slug in excess of 10 s^"1, preferably at least 10² s^"1, more preferably at least 10³ s^"1; and

a sensor to detect either the load applied to, or the compression rate experienced during compression of, the slug by the ram, the sensor being adapted to operate at a sampling frequency of at least 10³ s^"1, preferably 2 x 10³ s^"

1

11. Apparatus as claimed in claim 10, in which said control means is adapted to maintain the predetermined velocity or predetermined load constant during compression.

12. Apparatus as claimed in claim 10 or 11, in which it is the velocity which the control means is adapted to maintain at said predetermined rate, and said sensor is a load sensor.

13. Apparatus as claimed in claim 12, in which the ram moves the first plate and the load sensor is disposed between the second plate and a fixed frame of the apparatus.

14. Apparatus as claimed in any of claims 10 to 13, in which the plates comprise relatively hard, thermally non-conductive material, preferably an asbestos or asbestos-like composition, more preferably Sindanyo.

15. Apparatus as claimed in claim 13, in which the second plate is mounted in an aluminium plug having tapering sides and received in a corresponding bore of the frame, the sensor being mounted between said plug and the frame.

16. Apparatus as claimed in claim 15, in which said frame extends from said bore around a slug mounted on said first plate, the frame around said slug preferably comprising a heat non-conductive material, more preferably graphite.

17. Apparatus as claimed in any of claims 10 to 16, in which said first plate is integral with the ram.

18. A method of characterisation of a thixotropic specimen comprising quantification of the breakdown point at which flow of the specimen commences.

19. Thixoforming material characterisation apparatus, comprising means to quantify the breakdown point at which flow of a specimen to be characterised commences.

20. Apparatus substantially as hereinbefore described with a reference to the accompanying drawings.

21. A method substantially as hereinbefore described with reference to the accompanying drawings.