MCP39F511N datasheet(46/61 Pages) MICROCHIP

46 / 61 page

 2015-2018 Microchip Technology Inc.

DS20005473B-page 46

MCP39F511N

9.3.2

EXAMPLE OF RANGE SELECTION

FOR VALID CALIBRATION

In this example, the user applies a calibration current

of 1A to an uncalibrated system. The indicated value

in the Current RMS register is 2300 with the system's

specific shunt value, PGA gain, etc. The user expects

to see a value of 1000 in the Current RMS register

when 1A current is applied, meaning 1.000A with

1 mA resolution. Other given values are:

• the existing value for Gain Current RMS is 33480

• the existing value for Range is 12

By using Equation 9-1, the calculation for GainNEW

yields:

EQUATION 9-2:

When using the Auto-Calibration

Gain

command, the result is a failed calibration or a NAK

returned form the MCP39F511N, because the

resulting GainNEW is less than 25,000.

The solution is to use the Range register to bring the

measured value closer to the expected value, such

that a new gain value can be calculated within the

limits specified above.

The Range register specifies the number of right-bit

shifts (equivalent to divisions by 2) after the

multiplication with the Gain Current RMS register.

Refer to Section 5.0 “Calculation Engine (CE)

Description” for information on the Range register.

Incrementing the Range register by 1 unit, performing

an additional right-bit shift or dividing in half is included

in the calculation. Increasing the current range from 12

to 13 yields the new measured Current RMS register

value of 2300/2 = 1150. The expected (1000) and

measured (1150) values are much closer now, so the

expected new gain should be within the limits:

EQUATION 9-3:

The resulting new gain is within the limits and the

device successfully calibrates Current RMS and

returns an ACK.

Notice that the range can be set to 14 and the result-

ing

new

gain

will

still

within

limits

(GainNEW = 58226). However, since this gain value is

close to the limit of the 16-bit Gain register, variations

from system to system (component tolerances, etc.)

might create a scenario where the calibration is not

successful on some units and there would be a yield

issue. The best approach is to choose a range value

that places the new gain in the middle of the bounds of

the gain registers described above.

In a second example, when applying 1A, the user

expects an output of 1.0000A with 0.1 mA resolution.

The example is starting with the same initial values:

EQUATION 9-4:

The GainNEW is much larger than the 16-bit limit of

65535, so fewer right-bit shifts must be introduced to

get the measured value closer to the expected value.

The user needs to compute the number of bit shifts

that will give a value lower than 65535. To estimate

this number:

EQUATION 9-5:

2.2 rounds to the closest integer value of 2. The range

value changes to 12 – 2 = 10; there are two less

right-bit shifts.

The new measured value will be 2300 x 22 = 9200.

EQUATION 9-6:

The resulting new gain is within the limits, and the

device successfully calibrates Current RMS and

returns an ACK.

GAIN

NEW

GAIN

OLD

Expected

M easured

---------------------------



33480

1000

2300

------------



14556

25 000





GAIN

NEW

GAIN

OLD

Expected

Measured

---------------------------



33480

1000

1150

------------



29113

25 000



29113

65535



GAIN

NEW

GAIN

OLD

Expected

Measured

---------------------------



33480

10000

2300

---------------



145565

65535



145565

65535

------------------2.2

GAIN

NEW

GAIN

OLD

Expected

Measured

---------------------------



33480

10000

9200

---------------



36391

25 000



36391

65535



MCP39F511N Datasheet(PDF) 46 Page - Microchip Technology

MCP39F511N Datasheet(HTML) 46 Page - Microchip Technology

Html Pages

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61

Datasheet Download

Link URL