Campbell Scientific CR10 CR10 Measurement and Control - Page 176

3.

SECTION

13.

CR1O

MEASUREMENTS

essentially the same as

the

input source

resistance of configuration

A.

Moving

Rl

out

to

the thermistor as shown in Figure 13.3-7C

optimizes the signal settling

time

because

it

becomes

a

function of

R1

and

C*

only.

Columns

4 andT

list

the signal voltages as a

function of

temperature

using

a

2000

mV

excitation for configurations A and

C,

respectively.

Although configuration

A

has

a

higher output signal (2500 mV input range),

it

does

not

yield any higher resolution than

configuration

C

which

uses

the

t250

mV

input range.

NOTE:

Since

ft'attenuates

the signal

in

configuration

B

and C, one might

consider

eliminating

it

altogether.

However,

its

inclusion "flattens" the non-linearity

of

the

thermistor,

allowing

more accurate curve

fitting over

a

broader

temperature

range.

Where

possible,

run

excitation

leads and

signal leads in separate shields

to

minimize

transients.

Avoid

PVC-|nsulated conductors to

minimize

the

effect

of

dielectric absorption

on

input settling time.

Use

the

CR10

to

measure

the

input settling

error associated with a given configuration.

For example,

assume

long leads are

required but

the

lead capacitance, Cw,

is

unknown.

Configure

Rf

on

a length of

cable

similar to

the measurement.

Leave

the

sensor end open as shown

in

Figure 13.3-8

and measure

the

result using

the

same

instruction parameters

to

be used with the

sensor.

The

measured deviation

from

0V

the

input settling error.

6.

Most Campbell Scientific sensors are

configured with

a

small

bridge resistor,

R1,

(typically

1

kohm)

to

minimize

the

source

resistance.

lf

the

lead length of a Campbell

Scientific

sensor

is

extended by connecting

to

the

pigtails directly, the effect

of

the

lead

resistance,

R;,

on

the

signal must be

considered.

Figure 13.3-9 shows a

Campbell Scientific Model 107 sensor with

500

feet

of extension lead connected

directly to

the

pigtails.

Normally the signal

voltage is

proportionalto

Ry'(R.+R6+R), but

when

the

pigtails are extended,

the

signal

is

proponional

to

(R1+R1)/(R"+R6+R1+R1).

R1

is

much smaller than the other

terms

in

the

denominator and

can

be

discarded. The

effect

on

the

signalcan

be analyzed by

taking

the

ratio

of

the

signal with extended

leads,

V.'to

the

normal signal, V":

VsrNs

=

(Rt+R)/R1

Plugging

in

values

of

Rr=1k

and

Rt=.012k

(500'at

23 ohms/1000',

Table

13.3-2) gives

an

approximale

17"

error

in

the signal with

extended

leads.

Converting

the

error

to

oC

gives approximately

a

0.33=oC error at

OoC,

0.53"C

error

at

20"C,

and a

0.66"C

error

at

40"C.

The error

can

be

avoided by

maintaining

the

pigtails

on

the

CR10 end of

the extended leads because

R1

does not

add to

the

bridge completion resistor,

Rr,

and its influence

on

the

thermistor

resistance is negligible.

30.0

27.8

23.4

15.2

10.6

6.1

1

1

1

1

1

1

2.2

6.6

15.9

32.8

42.4

51.8

TABLE 13.3-7. Source Resistances and Signal

Levels

lor

YSI

#44032

Thermistor Configurations

Shown in Figure

13.3-7

(2V

Excitation)

----A----

-----B..--

"""-C----

Ro

V"(mV)

Ro@P

Ro

V"(mV)

(kohms)

(kohms)

(kohms)

TRs

(kohms)

40

884.6

-20

271.2

0

94.98

+25

30.00

+40

16.15

+60

7,60

29.0

27

22.8

15.0

10.5

6.1

66

200

480

1000

1300

1596

13-10

Campbell Scientific CR10 CR10 Measurement and Control - Page 176

kohms

Page 176 highlights