• No results found

Chapter 6 Experiments and discussions 1 5 2

Chapter 6 Experiments and discussions 1 5 3 W h e n the rig runs 3 0 minutes after reaching 700°C, the temperatures of s o m e parts in

the rig are listed in Table 6.1. T h e list represents the highest temperatures though the rig keeps running for extra four hours, which means that heat balance can be obtained in thirty minutes after reaching the required temperature. A s the bearings are designed to be able to w o r k at 360°C and the grease at 290°C, the cooling system enables the bearings to w o r k properly. T h e load cell and capacitance sensor are under 70°C, therefore they work in the allowed temperature range.

Table 6.1 Temperature distribution after half-hour running at 700*C

position of measurement original tip of capacitance sensor

middle of cooling pipe top of cooling pipe velocity measuring disk optoelectrical sensor mounting b e a m

roller support b e a m bearing holding block

roller

temperature 55°C 95°C 50°C 70°C 70°C 65°C 110°C 430°C

T w o preliminary experiments were conducted in the following conditions,

1. roller material: mild steel 2. temperature: 700 °C

3. pressure between the roller and the disk: 9 0 M P a

Chapter 6 Experiments and discussions 154 Rolling speed of the first experiment w a s 1.4 m/s and the experiment lasted 150

minutes. The experimental curve for the change of diameter of roller is shown in Figure 6.5, and the friction force is shown in Figure 6.6.

CM

O

u

35

a

CL V)

CM CM CD

CM

co CM

CO

CO

a

O

r

E

01 +d

V)

= n

</>

11

CM O O

c -a

* £ M 2

<

LU CD

O CC

CL O LL CO LU

ct

LU CL X

£ O O

d

Q.

o z

a u

ac

e L o (0

e Kl

PI*

k.

O X

ftp

310

BJ0

o o CM *-o <J5

O

m • - o r--£_ E

> a: •£

4> Q .

CM O O O

vf>

CS CM CM 1 00 o

o o

\D

r—

0)

*

-*d

•V

£ o

d-<A Vi a, • • .. ^

* <« ii w

w Q * -o

-^- - ^ i- *

c e = %

* fc !$:§

x x * ^ o

LU ui i— 2. as

At OP

a

IT) CM

O 00

o

CQ

CD u_ ^

LU L_

Q A

<-> ^ —

^

e E

Oi

£

o 3 C

C

0)

o

i-u 01

•dd OI

E 2 -3

5*-o oi M C M

u

I/) 01

u 9 CXI

in CM CM

Chapter 6 Experiments and discussions 1 5 5

IMTT

<8>

r-K)

d'

1

O) CO

o

Id.

</> u

31

a a. v>

a

CM CM U)

IZ

CM

• ~ 6)

L^

CM

IZ

_

*7

O)

[^ LL.

CD

•o

i

o

E

** 0)

11

k

CM O O

o

* —

it t

Or

i - * (A X

«/

C Lb U _l LU

£ O O

d

in o.

O

f •»

Q.

o

\

Z a

i—

cc s

1—

•p

•S .2

5 S

<K

2 * N. o

§ g

•X 00 o

IBIS 2 iB IB

© * > i * — *

_l

z

-J

o

H-O X o:

o

u_

CO

1—

z

LU Of LU CL LU

o

CM o

<

o u \t> -D T O - <T«

w •£ ..

^ Z </i .5 fe Z ^ 1- ^f

~ > 2 * =

> 0£ .£ 1-

0-CM O O ^

o

0"v 1 CM CM 00 O

d

o ; o •

VO

iii

E

-U 'w

.a »: •• -1

-V) d * •o

** +* h %

x x .* £ 3

U U h- I a c

"o

Q

^5

o

c

< «=

• ^*A

!>

c

£ E

Ch ^-/ 4>

£

O

£

a» TT CM O

o >

c o o 1=

.a >?

^ _ _ _ «. A

3 C O tri

a

01

*—.

u

oi

O

u e _o

"•5

_u

"E

b-OI la

3

Because the roller is m a d e of normal mild steel ( A S 250), the wear is severe. C h a n g e of diameter is about 60pm in 150 minutes. It is found that the curve is not quite good. Due to three pauses of the rig in the experimental process, the heating balance was

Chapter 6 Experiments a n d discussions 1 5 6 interrupted and the long tips of the capacitance sensor and support b e a m of the roller

shrank. T h e shrinking length of the long tips of the sensor must be longer than that of the support b e a m since their volume and heat capacity are smaller and they transfer heat quickly. Therefore, pauses caused the transient readout of displacement to be smaller.

W e found this kind of situation in B C , D E , and F G sections, which correspond to the time from 35 to 45 minutes, 8 0 to 9 0 minutes, and 140 to 145 minutes. Because all the pauses only took about 1 or 2 minute stop of the heating device, the heating balance resumed in 10 minutes.

Heating device cutoff circuit automatically stops the heating device when the motor stops or the protective m e s h covering the test rig is open. This circuit causes the heating interrupted by the pauses in the experiment, which results in a transient heating balance w h e n the experiment continues. It also takes time to resume the required temperature in pre-adjustment process, for example, in step 9 of the experimental process described in Section 6.1.

It is also found that wear of the roller is very uneven in the direction of the length of the roller, as s h o w n in Figure 6.7. Only about 6 0 percent of the surface in the front end of

the roller wore in the 150-minute experiment This is because the high temperature causes the disk to warp upwards, as s h o w n in Figure 6.8. A n y w a y , the expansion due to the high temperature cannot be released in the inner part of the disk, but expansion in the outer part of the disk can be free if there is no restriction from the screws. Though long screws, loose connection and big mounting holes are considered in the design of the rig to allow for expansion of the disk, it seems that the screws still restrict the expansion.

Chapter 6 Experiments and discussions 1 5 7

Figure 6.7 U n e v e n w e a r of the roller

roller position w h e n warpage

original position of the roller

warpage of the disk aftex-heating^i:

P

r

• f

original shape of the disk screws

heat

Figure 6.8 Contact condition of the roller a n d the disk after heating

The warpage also influences the flatness of the disk at high temperature, and results in

vibration of the rig and more wear of the roller. Uneven wear makes it difficult to measure the diameter of the roller. So the mounting angle of the capacitance sensor should be adjusted properly. The angle of the tapered roller is designed to be 6.59".

Chapter 6 Experiments and discussions 158 After the severe wear the angle at the front end of the roller is 15.4°. Thus, the distorted angle of the disk in the experiment is 4.3°. T h e warpage caused the contact zone concentrating o n the front end of the roller which m a k e the diameter of the roller decrease quickly in the early period of the experimental process, as s h o w n in Figure 6.5.

No water is applied in the experiment. The average of friction coefficient is about 0.06, which almost remains the same during the experimental process. T h e surface of roller is

still smooth after the experiment. But from 5 0 to 100 minutes, s h o w n as B C section in Figure 6.6, the friction force measured is smaller than the whole trend. It could be due to the fact that the linear bearings linking the support b e a m and holding frame were not smooth enough in this period. The friction force in the bearings decreased the contact force to the load cell.

The instability of the measurement also comes from the diameter runout of the roller and unflatness of the disk due to thermal distortion. T h e runout of the roller is s h o w n in

Figure 6.9, which is measured by the capacitance sensor w h e n the roller is mounted in the rig. T h e runout results from the following reasons:

1. roundness of the tapered surface of the roller;

2. concentricity of the tapered surface and the bearing surfaces of the roller;

3. roundness and concentricity of the bearing mounting holes;

4. concentricity of the inner and outer rings of bearings, clearance of bearings

The major factors should be the first two reasons. The maximum runout is about 0.1 m m , which directly influences the measurement of the diameter of the roller. Although

Chapter 6 Experiments and discussions 1 5 9 the digital filter is employed in the software, it is still difficult to eliminate the influence

from such a large runout

270

Figure 6.9 Diameter runout of the roller

Then the tapered angle of the roller w a s modified to be 15.4°, which is equal to the angle after wear in the 150-minute experiment to improve the measurement of diameter of the roller. T h e linear bearings were readjusted, and low viscosity grease w a s applied instead of the high viscosity one, and cleaning the rolling ball races to allow free running of the bearings. Then another 180-minute experiment w a s conducted uninterrupted. T h e rolling speed in this experiment w a s 2.0 m/s. Experimental curves for the change of diameter of roller and friction coefficient are s h o w n in Figure 6.10 and Figure 6.11.

Chapter 6 Experiments and discussions 1 6 0

to CM

a>

o o L.

U

<*i

c

V)

CM CM CO

CM CO

CM CO

CO

QJ

•a

a

X Is E

a>

«/;

11

1 1

d.

m

'

.

h

o o

5 >

1 k t f

3- =

* V,

2 * = TJ - * in

X

€>

C

u.

Ul -1

u

Z

O O

o o

CL

r ^

p°-r...

*

z

e u

O o.

o

i— cc