Fig. 5. Selecting the glue. (a) Typical air void defect into the glue at the interface with a transparent precision tubing. (b) Typical bodies used to test simultaneously fourdifferent glues during fatigue tests.effects on the strain measurements:
Bragg{
T}
Bragg{
T}
=
b{
T}
T
⇒
ε
1
a{
ε}
Bragg{
ε}
Bragg{
ε}
−
b{
ε}
b{
T}
Bragg{
T}
Bragg{
T}
However, even if the four FBGs are very close one to another(Fig. 6(a)), their measurements do not correspond exactly to thesame physical area. This small distance between two gratingstherefore introduces some time and space discrepancies in themeasurements. This is all the more true in transient states, espe-cially with steep temperature gradients, considering the fact thatthe time response and the sensitivity to temperature of a FBGdedicated to strain measurements (glued to the plug) are also dif-ferent from the “free from stress” temperature compensation FBG.These discrepancies still exist in steady states, however in someextent, one can consider that they are reproducible from cycle tocycle, therefore enabling additional elaborated numerical compen-sation techniques to be implemented to take into account thesephenomena.The plug shown in Fig. 6(a) has been designed keeping in mindthat all the four out-coming optical fibers must be kept alignedwith the plug axis (radial direction) in order to be easily routed(without any break) to the roll center and then to the externaldata acquisition station as detailed in the following. The FBG usedfor temperature compensation and the FBG measuring the radialstrain εmrrare inserted one close to another into two separate holesaligned with the plug axis. The two other optical fibers dedicated toεmand εm45measurements are inserted into the plug according to amore complicated scheme in order to get the appropriate FBG ori-entations in the surface vicinity. The curvature radii are limited toRmin= 10 mm to avoid mechanical fatigue breaks as well as signif-icant optical signal power losses, two reasons which may preventany measurement. Technically, such a complicated path cannot bedrilled inside the plug body besides the fact that the optical fiberswould be very difficult to insert. That is the reason why two slots onthe sides of the plug are made with specific shapes at the bottom,then a thin hole is drilled from the slot with the right orientationto roll surface. Optical fibers are inserted at the bottom of the slots,and then inserted and glued in the hole with the same procedure asthe two radial ones. A picture of manufactured plugs is presentedin Fig. 6(b).Two successive optical fibers are spaced 1.25 mm along the axialdirection. It is considered that the stress field is sufficiently homo-geneous along the axial direction at this scale, so that one canconsider all optical fibers at the same axial position. A first test of theplug before insertion inside the roll body is performed under pressand a picture is presented in Fig. 7(a). FBGs responses present verygood linearity as shown in Fig. 7(b), where F is the vertical force (inNewton) applied by the press and εTrepresents the deformation ofthe FBG free from stress dedicated to temperature cross-sensitivitycompensation (no significant sensitivity under pressure at roomtemperature can be noticed in comparison with the other FBGs).
Fig. 5 เลือกกาว (ก) โดยทั่วไปอากาศบกพร่องโมฆะเป็นกาวที่ต่อกับท่อความแม่นยำโปร่งใส (ข) ร่างกายโดยทั่วไปใช้ในการทดสอบ fourdifferent ผลิตภัณฑ์สำหรับยึดติดกันระหว่าง tests.effects ล้ากับการวัดต้องใช้:Bragg {T }Bragg {T }=b {T }T⇒Ε 1{การΕ}Bragg {Ε}Bragg {Ε}−b {Ε}b {T }Bragg {T }Bragg {T }However, even if the four FBGs are very close one to another(Fig. 6(a)), their measurements do not correspond exactly to thesame physical area. This small distance between two gratingstherefore introduces some time and space discrepancies in themeasurements. This is all the more true in transient states, espe-cially with steep temperature gradients, considering the fact thatthe time response and the sensitivity to temperature of a FBGdedicated to strain measurements (glued to the plug) are also dif-ferent from the “free from stress” temperature compensation FBG.These discrepancies still exist in steady states, however in someextent, one can consider that they are reproducible from cycle tocycle, therefore enabling additional elaborated numerical compen-sation techniques to be implemented to take into account thesephenomena.The plug shown in Fig. 6(a) has been designed keeping in mindthat all the four out-coming optical fibers must be kept alignedwith the plug axis (radial direction) in order to be easily routed(without any break) to the roll center and then to the externaldata acquisition station as detailed in the following. The FBG usedfor temperature compensation and the FBG measuring the radialstrain εmrrare inserted one close to another into two separate holesaligned with the plug axis. The two other optical fibers dedicated toεmand εm45measurements are inserted into the plug according to amore complicated scheme in order to get the appropriate FBG ori-entations in the surface vicinity. The curvature radii are limited toRmin= 10 mm to avoid mechanical fatigue breaks as well as signif-icant optical signal power losses, two reasons which may preventany measurement. Technically, such a complicated path cannot bedrilled inside the plug body besides the fact that the optical fiberswould be very difficult to insert. That is the reason why two slots onthe sides of the plug are made with specific shapes at the bottom,then a thin hole is drilled from the slot with the right orientationto roll surface. Optical fibers are inserted at the bottom of the slots,and then inserted and glued in the hole with the same procedure asthe two radial ones. A picture of manufactured plugs is presentedin Fig. 6(b).Two successive optical fibers are spaced 1.25 mm along the axialdirection. It is considered that the stress field is sufficiently homo-geneous along the axial direction at this scale, so that one canconsider all optical fibers at the same axial position. A first test of theplug before insertion inside the roll body is performed under pressand a picture is presented in Fig. 7(a). FBGs responses present verygood linearity as shown in Fig. 7(b), where F is the vertical force (inNewton) applied by the press and εTrepresents the deformation ofthe FBG free from stress dedicated to temperature cross-sensitivitycompensation (no significant sensitivity under pressure at roomtemperature can be noticed in comparison with the other FBGs).
การแปล กรุณารอสักครู่..
