Same Apparent Strain And Gage Factor Change example essay topic

ABSTRACT For accurately measuring thermal strains, particularly on large structures where welded strain gages cannot be used, a reversible bonded strain gage was developed. Basically it is a special poly imide strain gage which is same on both the base side and cover side so that it can be used both ways. It can be used to measure strains at temperatures under 250 oC (482 oF) of a structure made of aluminum alloys or composites (to which its difficult to weld a strain gage). These gages can be can be peeled after taking required apparent strain measurements in a furnace and can be attached reverse side up at a required point on a structure.

To measure mechanical stresses on structures at high temperatures it is essential to measure apparent thermal stresses accurately in the first place. In practice, several strain gages in a pack are used to obtain calibration data. The apparent strain and gage factor change of all the gages in the pack are assumed to be same which is not so in practice, in spite of great efforts to reduce scatter of apparent strain. Since reversible strain gages can be reattached to the test structure after taking apparent strain readings, the error caused due to apparent strain scatter (by using different strain gages) can be reduced to great extent. In this paper the thermal characteristics of the reversible strain gage - repeatability of apparent strain, gage-factor change, creep, drift and the output for a given mechanical strain - were investigated. INTRODUCTION There are several problems associated with elevated temperature measurements, static or dynamic, the basic one being that alloys useful as strain gages at these temperatures are also excellent temperature sensors.

Firstly, installation of the strain gage is a problem and secondly the apparent strain and change in gage factor makes it very difficult to measure the actual strain. In aerospace industry we come across a lot of situations when very accurate strain measurements at high temperatures are required, but in spite of a lot of improvement in new high temperature strain gages most of them are welded types. Hence, they cannot be used on materials like aluminum alloys or composites. In this paper, a reversible bonded strain gage is described for use at temperatures under 250 o (482 oF) that can be applied to a structure made of materials commonly used in aerospace industry like aluminum alloys and various composites.

Aircraft wings are often subjected to high temperature and high acoustic noise level and the application of reversible strain gages to accurately measure the stresses is the main motivation behind choosing this paper for review. These gages have an additional advantage that the adhesives used to fix the gage cures at room temperature unlike most adhesives for high temperature strain measurements which needs curing heat cycles (like 1 hr at 180 o). As a result of this heating of a structure during curing can be avoided. BACKGROUND Gage Factor Change The electrical properties of the active strain elements that are most critical to strain gage performance are gage factor (G) and temperature coefficient of resistance (TCR) [1], defined as (1) (2) where T is temperature, the strain and R is resistance. This is particularly true when large temperature gradients are superimposed over the surfaces of engine components. Since static strain gages are not only subjected to applied mechanical strains, but are also subjected to large thermal strains as well, the strain measurement must be capable of distinguishing the relative contributions of the two sources of strain.

Specifically, the total contribution to the measured strain (fractional resistance change) for a given strain application is the sum of the actual mechanical strain at a given temperature and the temperature induced strain. The latter contribution is due to the TCR of the semiconductor and the differences in thermal coefficient of expansion (TCE) between the gage and the substrate. Thus, the relative contributions to the total strain can be represented by (3), (4) and (5). (3) Where, (4) (5) and where's and g are the coefficients of thermal expansion for the substrate and gage, R the electrical resistance, T the temperature and is the strain. From the equations above, it is evident that the TCR should be as small as possible to avoid the need for temperature compensation.

By minimizing the thermal component of static strain (apparent strain) and maximizing the gage factor, it should be possible to maximize the sensitivity of the sensor and permit reliable strain measurements at high temperatures. Thus, as temperature increases the gage factor also change and in order to measure the actual mechanical strain correctly, we need to know the gage factor change when compared to the gage factor under calibration condition which is generally the room temperature. Apparent Strain Apparent strain is any change in gage resistance that is not caused by the strain on the force element. Apparent strain is the result of the interaction of the thermal coefficient of the strain gage and the difference in expansion between the gage and the test specimen.

The variation in the apparent strain of various strain-gage materials as a function of operating temperature is shown in Figure 1. In addition to the temperature effects, apparent strain also can change because of aging and instability of the metal and the bonding agent. Figure 1 Apparent Strain Variation with Temperature Compensation for apparent strain is necessary if the temperature varies while the strain is being measured. In most applications, the amount of error depends on the material used, the accuracy required, and the amount of the temperature variation.

If the operating temperature of the gage and the apparent strain characteristics are known, compensation is possible. It is desirable that the strain-gage measurement system be stable and not drift with time. In calibrated instruments, the passage of time always causes some drift and loss of calibration. For accurately measuring the mechanical strain, apparent strain should be accurately measured and scatter band of apparent strain in a package of strain gages should be small. Thus the relationship between measured strain, strain due to mechanical stresses, gage factor change and apparent strain is m (T) = (Gt / Go) F (T) where, m (T) = measured strain. strain based on mechanical stresses. Go = gage factor at temperature T. Gt = gage factor at room temperature.

F (T) = apparent strain. Knowing apparent strain F (T) and gage factor change we can find out the actual mechanical strain In industrial practice, several strain gages in a pack are used to obtain the calibration data. But to get the actual mechanical strain at each temperature from the gathered data it is assumed that all the strain gages in the pack have the same apparent strain and gage factor change as those of the gage picked out for calibration. But this assumption holds well only when the scatter band of apparent strain is small. Although many advances has been made to reduce the apparent strain scatter but at high temperatures it is still significant even for gages made of advance or karma foil which has a low temperature coefficient of resistance change. The scatter band of apparent strain is roughly proportional to the gage temperature change and its ratio is about 1 /oC for gages made of karma foils.

Thus for structures where strains are about 1000 under service conditions and temperature is about 250 o, the error caused by the estimation of apparent strain is about 250 which is unacceptable. For same combination of strain gage and adhesive the gage factor change with temperature is not much that is they are compensated. Thus it is possible to estimate and calibrate gage-factor change with good accuracy. Thus if the same strain gage is used to measure strains then the apparent strain scatter is quite small. Moreover with proper choice of adhesives its gage factor change can also be taken care of. Since a reversible bonded strain gage is a reusable strain gage therefore apparent strain scatter band is very small compared to conventional strain gages.

This is because in the former one we are using the same strain gage to measure the apparent strain of a structure where as for the later different strain gages are used to measure apparent strains leading to error due to apparent strain scatter. For high temperature tests large plastic deformations often occur so it is desirable to measure the thermal strain very accurately in the first test. DESIGN AND EXPERIMENTAL ASPECTS The requirements for high temperature test measurements are 1. The apparent strain of each gage must be accurately measured. 2. A room temperature curing adhesive to prevent heating of test structures and ensure little variation of gage factor with different gages.

3. The thermal strain must be measured with accuracy on the first structural test. To reduce apparent strain scatter a reusable strain gage is used. In considering the problems associated with surface preparation and cleaning it is made sure that the reusable strain gage should not be adhered with the same surface again, but must be adhered reverse side up onto the test structure. The reversible gage can be used to measure strain with the cover side down and then it can be flipped over to the base side to take strain readings again. Since the same gage is used apparent strain is likely to be same.

Thus the apparent strain scatter is likely to be very less. It is also ensured that there is no gage factor change between the reverse side up gage and the same gage adhered ordinarily. Also a measurable range of 1000 after peeling and re adhering the gage was desired. To meet requirement (2) an adhesive compatible with the strain gage that can be used in the temperature range of 200-250 oC and curable at room temperature was sought for. Initially to find out the gages which can be peeled easily and re-adhered to the reverse surface experiments were conducted with 10 different ordinarily available strain gages and 3 different room temperature curing adhesives.

1. Strain Gages Dimension: about 10 x 5 mm. Base material: bakelite, polyester, epoxy, poly imide. 2.

Adhesives Polyester A: mixture ratio 1 percent. Polyester B: mixture ratio 1 percent. Cya no- acrylate Six different test specimen materials were used steel, stainless steel, aluminum alloy, icon el X, titanium alloy and glass / poly imide composite. Roughness and cleanliness conditions were as recommended for ordinary strain gage installation method and two to three heat cycles up to 250 oC were conducted before carefully peeling with knives. Test results showed the following combinations were easy to peel: 1. Strain gage: poly imide base (one not so thin) 2.

Adhesive: polyesters A, B and cyan o-3. Tool: hacksaw-blade knife. Initially 33 ordinary strain gages were peeled and re adhered reverse side up position and heated in a furnace to test for their thermal characteristic data but it was found that 30% of them didn't read here. This happened because ordinary strain gage leads are put on the gage tabs and connected by soldering or spot welding but this leads to two projections on the cover side so when it was adhered cover side down the projections were obstructing proper adhesion of the gage. As a result a new gage was developed. To overcome this problem gage lead was made thinner and was connected by spot welding as flatly as possible.

Also gage tabs were made thinner to keep the connected part away from the sensing element of the gage. In spite of doing all these things the problem with proper adhesion with the reversed side up persisted. But finally after many trials the optimal dimension for the reversible strain gage was set at 16 x 6 mm, with a thickness of 50 - 70 m and a gage lead thickness of 20 -40 m. This configuration was arrived considering that transmitting strain from the structure to the gage requires minimum length outside the sensing element to be about 10 times more than the thickness. In the actual design this length was set to 1.0 - 1.5 mm and after that the gage tab length was determined so that the gage could be adhered even in the worst case, at least 1.0 - 1.5 mm outside of the sensing element, ensuring proper adhesion. A picture of the reversible strain gage is given in figure 2.

Figure 2: Reversible strain gage Figure 3: Ordinary bonded strain gage To test the repeatability of apparent strain the aluminum alloy, steel and stainless-steel specimens were used and 100 reversible gages were tested. At first the gages were adhered cover-side down position onto the specimens and were taken two or three times at temperatures up to 250 oC in a furnace. Then the gages were peeled and were re adhered with the reverse side down position and were taken up to temperatures of 250 oC twice. Polyesters A and B were used and it was ensured that curing time and clamping pressure for the peeled gages were made same as that to the first adhesion.

Otherwise test results would be affected. Cure time was 48 hrs. Data range: 22 oC (room temperature) to 50 oC Results are shown in figure 4. Open legends - cover-side down position (1, 2, 3) Closed legends - base side down position (4, 5) Figure 4: Apparent strain Conditions to be satisfied by the gages to ensure good repeatability are: 1. There should be no great difference in apparent strains of first and second cycles. 2.

If condition is not satisfied a third heat cycle is run and if the apparent stain data obtained agrees with the second heat cycle then this gage is used. 3. In the second heat cycle zero shift must be small. 4. The apparent strain must not be large. Out of 100, 6 got damaged while peeling 11 failed to satisfy the conditions mentioned above.

Thus data for remaining 83 reversible strain gages were obtained. OBSERVATIONS AND RESULTS It was observed that the apparent strains in the third and fourth heat cycles were in good agreement with the fourth and fifth heat cycles (first and second heat cycles of the re adhered gages) respectively. Results for scatter band of apparent strain in 2nd (cover-side down) and 4th (base-side down) heat cycles are shown in figure 5. Figure 5: Scatter band of apparent strain of 20 reversible strain gages. It was observed that the scatter band grows with increase in temperature and the ratio is about 0.7 oC (which is less than that of karma foils) Open legends - cover-side down position Closed legends - base side down position Figure 6: Difference of apparent strain for 50 reversible strain gages The differences of apparent up to 225 oC have the same tendency and the maximum difference is less than 50 The differences go up to a maximum of 55 at 250 oC. Open legends - cover-side down position Closed legends - base side down position Figure 7: Difference of apparent strain for 32 reversible strain gages Figure 7 shows the apparent strain data for gages which were heated till 250 oC and not 260 oC in the first cycle for stabilization.

It is also seen that the maximum apparent strain difference is larger than those in fig 4. Gage factor change is shown in figure 8. In the cover down position the gage output reduction was almost the same as that in the base side up position hence it is easy to compensate for gage factor change by simple calculation. Open legends - cover-side down position Closed legends - base side down position Figure 8: Difference of apparent strain for reversible strain gages Creep data is shown in figure 9, Specimen is Icone l X... the creep and drift under 250 oC would be less than these at 250 oC. Open legends - cover-side down position Closed legends - base side down position Figure 9: Creep at 250 oC for 10 reversible strain gages. Open legends - cover-side down position Closed legends - base side down position Figure 11: Strain output of 10 peeled and re adhered reversible gages Conclusions 1.

Reversible strain gages can be applied to a large structure to be tested at high temperature. 2. Error caused by apparent is comparatively much less. 3.

Repeatability of apparent strain is fairly constant. REFERENCES Primary Reference (paper that is reviewed) Paper presented by Koichi e.g. awa at the Fourth SESA International Congress on Experimental Mechanics held in Boston, MA on May 25-30, 1980. Original manuscript submitted: August 26th, 1980. Final version received: July 21, 1981.

Proceedings of the Society for Experimental Stress Analysis (SESA) Experimental Mechanics Vol - IX, 1982, Pg 161. Central Library IIT Madras, Call No. DI 12/60, A cnt No. 11997 Other

Bibliography

1. Otto J. Gregory and Qing Luo, A self-compensated ceramic strain gage for use at elevated temperature, Sensors and Surface Technology Partnership, Department of Chemical Engineering, University of Rhode Island, Kingston, RI 02881, USA Received 27 July 2000;
revised 27 September 2000;
accepted 1 October 2000.
Available online 13 February 2001.
Sensors and Actuators A: Physical, Volume 88, Issue 3, 5 January 2001, Pages 234-240, web 2.