Isolation Amplifier

Technical Data

HCPL-7840

Features Description ments, providing for smoother

• 15 kV/µs Common-Mode The HCPL-7840 isolation amplifier control (less “torque ripple”) in
Rejection at VCM = 1000 V family was designed for current various types of motor control
• Compact, Auto-Insertable sensing in electronic motor drives. applications.
Standard 8-pin DIP Package In a typical implementation, motor
• 0.00025 V/V/°C Gain Drift vs. currents flow through an external
resistor and the resulting analog Functional Diagram
Temperature
voltage drop is sensed by the
• 0.3 mV Input Offset Voltage
HCPL-7840. A differential output IDD1 IDD2
• 100 kHz Bandwidth voltage is created on the other VDD1 1 8 VDD2
• 0.004% Nonlinearity side of the HCPL-7840 optical VIN+ 2 + 7 VOUT+
+
• Worldwide Safety Approval: isolation barrier. This differential
UL 1577 (3750 Vrms/1 min.) output voltage is proportional to VIN– 3 – – 6 VOUT–
and CSA (pending), VDE the motor current and can be con-
GND1 4 5 GND2
0884 (Option #060 only) verted to a single-ended signal by SHIELD
• Advanced Sigma-Delta (Σ−∆) using an op-amp as shown in the
A/D Converter Technology recommended application circuit.
• Fully Differential Circuit Since common-mode voltage
Topology swings of several hundred volts in
tens of nanoseconds are common The product can also be used for
• 0.8 µm CMOS IC Technology
in modern switching inverter general analog signal isolation
motor drives, the HCPL-7840 applications requiring high
Applications was designed to ignore very high accuracy, stability, and linearity
• Motor Phase and Rail common-mode transient slew under similarly severe noise con-
Current Sensing rates (of at least 10 kV/µs). ditions. For general applications,
• Inverter Current Sensing we recommend the HCPL-7840
• Switched Mode Power The high CMR capability of the (gain tolerance of ± 5%). The
Supply Signal Isolation HCPL-7840 isolation amplifier HCPL-7840 utilizes sigma delta
• General Purpose Current provides the precision and (∑-∆ ) analog-to-digital converter
Sensing and Monitoring stability needed to accurately technology, chopper stabilized
monitor motor current in high amplifiers, and a fully differential
• General Purpose Analog
noise motor control environ- circuit topology fabricated using
Signal Isolation

CAUTION: It is advised that normal static precautions be taken in handling and assembly of this component to
prevent damage and/or degradation which may be induced by ESD.
 2

Agilent’s 0.8 µm CMOS IC stability over time and regulatory safety standards. (A
process. Together, these features temperature. This performance gull-wing surface mount option
deliver unequaled isolation-mode is delivered in a compact, auto- #300 is also available).
noise rejection, as well as excellent insertable, industry standard 8-pin
offset and gain accuracy and DIP package that meets worldwide

Ordering Information
Specify Part Number followed by Option Number (if desired).

HCPL-7840 = ± 5% Gain Tol.; Mean Gain = 8.00

Option: #YYY

No Option = Standard DIP package, 50 per tube

060 = VDE 0884 Option

300 = Surface Mount Option

500 = Tape/Reel Packaging Option, 1k min. per reel

Package Outline Drawings

Standard DIP Package

9.65 ± 0.25
(0.380 ± 0.010)

8 7 6 5

A 7840 DATE CODE

YYWW

1 2 3 4 7.62 ± 0.25
(0.300 ± 0.010)
1.78 (0.070) MAX.
1.19 (0.047) MAX.
6.35 ± 0.25
(0.250 ± 0.010)

4.70 (0.185) MAX.

0.51 (0.020) MIN.

2.92 (0.115) MIN.

0.20 (0.008)
1.080 ± 0.320 0.65 (0.025) MAX. 5° TYP. 0.33 (0.013)
(0.043 ± 0.013)
2.54 ± 0.25
(0.100 ± 0.010)

DIMENSIONS IN MILLIMETERS AND (INCHES).

Note: Initial or continued variation in the color of the HCPL-7840’s white mold compound is normal and does not affect device
performance or reliability.
 3

Gull Wing Surface Mount Option 300

PAD LOCATION (FOR REFERENCE ONLY)

9.65 ± 0.25 1.016 (0.040)
(0.380 ± 0.010) 1.194 (0.047)

8 7 6 5
4.826 TYP.
(0.190)
A 7840 6.350 ± 0.25
YYWW (0.250 ± 0.010) 9.398 (0.370)
9.960 (0.390)

1 2 3 4

0.381 (0.015)
1.194 (0.047) 0.635 (0.025)
1.778 (0.070)

1.780 9.65 ± 0.25

(0.070) (0.380 ± 0.010)
1.19 MAX.
(0.047) 7.62 ± 0.25
MAX. (0.300 ± 0.010)
0.20 (0.008)
4.19 MAX. 0.33 (0.013)
(0.165)

1.080 ± 0.320 0.635 ± 0.25

(0.043 ± 0.013) (0.025 ± 0.010)
0.635 ± 0.130
2.54 (0.025 ± 0.005) 12° NOM.
(0.100)
BSC
DIMENSIONS IN MILLIMETERS (INCHES). LEAD COPLANARITY
TOLERANCES (UNLESS OTHERWISE SPECIFIED): xx.xx = 0.01 MAXIMUM: 0.102 (0.004)
xx.xxx = 0.005

Maximum Solder Reflow Thermal Profile

260
240
∆T = 145°C, 1°C/SEC
220
∆T = 115°C, 0.3°C/SEC
200
TEMPERATURE – °C

180
160
140
120
100
80
∆T = 100°C, 1.5°C/SEC
60
40
20
0
0 1 2 3 4 5 6 7 8 9 10 11 12
TIME – MINUTES

(NOTE: USE OF NON-CHLORINE ACTIVATED FLUXES IS RECOMMENDED.)

 4

Regulatory Information
The HCPL-7840 is pending approval by the following organizations:

VDE UL CSA
Approval under VDE 0884/06.92 Approval under UL 1577, com- Approved under CSA Component
with VIORM = 891 VPEAK expected ponent recognition program up Acceptance Notice #5, File CA
prior to product release. to VISO = 3750 Vrms expected 88324 expected prior to product
prior to product release. release.

VDE 0884 Insulation Characteristics*

Description Symbol Characteristic Unit
Installation classification per DIN VDE 0110/1.89, Table 1
for rated mains voltage ≤ 300 Vrms I-IV
for rated mains voltage ≤ 450 Vrms I-III
for rated mains voltage ≤ 600 Vrms I-II
Climatic Classification 55/100/21
Pollution Degree (DIN VDE 0110/1.89) 2
Maximum Working Insulation Voltage VIORM 891 VPEAK
Input to Output Test Voltage, Method b**
VIORM x 1.875 = VPR, 100% Production Test with VPR 1670 VPEAK
tm = 1 sec, Partial discharge < 5 pC
Input to Output Test Voltage, Method a**
VIORM x 1.5 = VPR, Type and Sample Test, VPR 1336 VPEAK
tm = 60 sec, Partial discharge < 5 pC
Highest Allowable Overvoltage VIOTM 6000 VPEAK
(Transient Overvoltage tini = 10 sec)
Safety-limiting values—maximum values
allowed in the event of a failure.
Case Temperature TS 175 °C
Input Current*** IS,INPUT 400 mA
Output Power *** PS,OUTPUT 600 mW
Insulation Resistance at TS, VIO = 500 V RS >10 9 Ω
*Insulation characteristics are guaranteed only within the safety maximum ratings which must be ensured by protective circuits within
the application. Surface Mount Classification is Class A in accordance with CECC00802.
**Refer to the optocoupler section of the Isolation and Control Components Designer’s Catalog, under Product Safety Regulations
section, (VDE 0884) for a detailed description of Method a and Method b partial discharge test profiles.
***Refer to the following figure for dependence of PS and IS on ambient temperature.
OUTPUT POWER – PS, INPUT CURRENT – IS

800
PS (mW)
700
IS (mA)
600

500

400

300

200

100

0
0 25 50 75 100 125 150 175 200
TA – CASE TEMPERATURE – °C
 5

Insulation and Safety Related Specifications

Parameter Symbol Value Unit Conditions
Minimum External Air Gap L(101) 7.4 mm Measured from input terminals to output
(Clearance) terminals, shortest distance through air.
Minimum External Tracking L(102) 8.0 mm Measured from input terminals to output
(Creepage) terminals, shortest distance path along body.
Minimum Internal Plastic Gap 0.5 mm Through insulation distance conductor to
(Internal Clearance) conductor, usually the straight line distance
thickness between the emitter and detector.
Tracking Resistance CTI >175 Volts DIN IEC 112/VDE 0303 Part 1
(Comparative Tracking Index)
Isolation Group III a Material Group (DIN VDE 0110, 1/89,
Table 1)

Absolute Maximum Ratings

Parameter Symbol Min. Max. Unit Note
Storage Temperature TS -55 125 °C
Operating Temperature TA - 40 100
Supply Voltage VDD1, VDD2 0 5.5 V
Steady-State Input Voltage VIN+, VIN- -2.0
VDD1 +0.5
2 Second Transient Input Voltage -6.0
Output Voltage VOUT -0.5 VDD2 +0.5
Solder Reflow Temperature Profile See Package Outline Drawings Section

Recommended Operating Conditions

Parameter Symbol Min. Max. Unit Note
Ambient Operating Temperature TA -40 85 °C
Supply Voltage VDD1, VDD2 4.5 5.5 V
Input Voltage (accurate and linear) VIN+, VIN- -200 200 mV 1
Input Voltage (functional) VIN+, VIN- -2 2 V
 6

DC Electrical Specifications
Unless otherwise noted, all typicals and figures are at the nominal operating conditions of VIN+ = 0,
VIN- = 0 V, VDD1 = VDD2 = 5 V and TA = 25°C; all Min./Max. specifications are within the Recommended
Operating Conditions.

Parameter Symbol Min. Typ. Max. Unit Test Conditions Fig. Note
-2.0 0.3 2.0 TA = 25°C
Input Offset Voltage VOS mV 1,2
-3.0 3.0
Magnitude of Input
Offset Change vs. |∆VOS /∆TA| 3.0 10.0 µV/°C 3 2
Temperature
Gain (± 5% Tol.) G 7.60 8.00 8.40 V/V -200 mV < VIN+ < 200 mV 3
Magnitude of VOUT 4,5,6
Gain Change vs. |∆G/∆TA| 0.00025 V/V/°C 4
Temperature
VOUT 200 mV Nonlinearity NL200 0.0037 0.35 % -200 mV < VIN+ < 200 mV 5
Magnitude of VOUT 200 mV
Nonlinearity Change |dNL200 /dT| 0.0002 % / °C 7,8
vs. Temperature
VOUT 100 mV Nonlinearity NL100 0.0027 0.2 % -100 mV < VIN+ < 100 mV 6
Maximum Input
Voltage before |VIN+ |MAX 308.0 mV 9
VOUT Clipping
Input Supply Current IDD1 10.86 15.5 VIN+ = 400 mV 7
mA 10
Output Supply Current IDD2 11.56 15.5 VIN+ = -400 mV 8
Input Current IIN+ -0.5 5.0 µA 9
Magnitude of Input
Bias Current vs.
|dIIN/dT| +0.45 nA/°C 11
Temperature
Coefficient
Output Low Voltage VOL 1.29 V
10
Output High Voltage VOH 3.80 V
Output Common-Mode
V OCM 2.2 2.545 2.8 V
Voltage
Output Short-Circuit
|IOSC | 18.6 mA 11
Current
Equivalent Input
RIN 500 kΩ
Impedance
VOUT Output
ROUT 15 Ω
Resistance
Input DC Common-
CMRR IN 76.1 dB 12
Mode Rejection Ratio
 7

AC Electrical Specifications
Unless otherwise noted, all typicals and figures are at the nominal operating conditions of VIN+ = 0,
VIN- = 0 V, VDD1 = VDD2 = 5 V and TA = 25°C; all Min./Max. specifications are within the Recommended
Operating Conditions.
Parameter Symbol Min. Typ. Max. Unit Test Conditions Fig. Note
VOUT Bandwidth (-3 dB) BW 50 100 kHz VIN+ = 200 mVpk-pk 12,13
sine wave.
VOUT Noise NOUT 31.5 mVrms VIN+ = 0.0 V 13
VIN to VOUT Signal Delay tPD10 2.03 3.3 mVrms Measured at output of 14,15
(50 – 10%) MC34081 on Figure 15.
VIN to VOUT Signal Delay tPD50 3.47 5.6 µs VIN+ = 0 mV to 150 mV step.
(50 – 50%)
VIN to VOUT Signal Delay tPD90 4.99 9.9
(50 – 90%)
VOUT Rise/Fall Time tR/F 2.96 6.6
(10 – 90%)
Common Mode Transient CMTI 10.0 15.0 kV/µs VCM = 1 kV, TA = 25°C 16 14
Immunity
Power Supply Rejection PSR 170 mVrms With recommended 15
application circuit.

Package Characteristics
Parameter Symbol Min. Typ. Max. Unit Test Conditions Fig. Note
Input-Output Momentary RH < 50%, t = 1 min.,
VISO 3750 Vrms 16,17
Withstand Voltage TA = 25°C
Resistance (Input-Output) RI-O >10 9 Ω VI-O = 500 VDC 18
Capacitance (Input- CI-O 1.2 pF F = 1 MHz 18
Output)
 8

Notes: 7. The input supply current decreases 10 kHz) and increases with increasing
General Note: Typical values represent as the differential input voltage frequency.
the mean value of all characterization (V IN+–VIN- ) decreases. 14. CMTI (Common Mode Transient
units at the nominal operating conditions. 8. The maximum specified output supply Immunity or CMR, Common Mode
Typical drift specifications are determined current occurs when the differential Rejection) is tested by applying an
by calculating the rate of change of the input voltage (VIN+–V IN-) = -200 mV, exponentially rising/falling voltage
specified parameter versus the drift pa- the maximum recommended operat- step on pin 4 (GND1) with respect to
rameter (at nominal operating conditions) ing input voltage. However, the out- pin 5 (GND2). The rise time of the
for each characterization unit, and then put supply current will continue to test waveform is set to approximately
averaging the individual unit rates. The rise for differential input voltages up 50 ns. The amplitude of the step is
corresponding drift figures are normalized to approximately -300 mV, beyond adjusted until the differential output
to the nominal operating conditions and which the output supply current (VOUT+ –VOUT-) exhibits more than a
show how much drift occurs as the par- remains constant. 200 mV deviation from the average
ticular drift parameter is varied from its 9. Because of the switched-capacitor output voltage for more than 1µs.
nominal value, with all other parameters nature of the input sigma-delta con- The HCPL-7840 will continue to func-
held at their nominal operating values. verter, time-averaged values are shown. tion if more than 10 kV/µs common
Note that the typical drift specifications 10. When the differential input signal mode slopes are applied, as long as
in the tables below may differ from the exceeds approximately 308 mV, the the breakdown voltage limitations
slopes of the mean curves shown in the outputs will limit at the typical values are observed.
corresponding figures. shown. 15. Data sheet value is the differential
1. Agilent recommends operation with 11. Short circuit current is the amount of amplitude of the transient at the
VIN- = 0 V (tied to GND1). Limiting output current generated when either output of the HCPL-7840 when a
VIN+ to 100 mV will improve DC output is shorted to V DD2 or ground. 1 Vpk-pk, 1 MHz square wave with
nonlinearity and nonlinearity drift. If 12. CMRR is defined as the ratio of the 40 ns rise and fall times is applied to
VIN- is brought above V DD1 – 2 V, an differential signal gain (signal applied both VDD1 and VDD2.
internal test mode may be activated. differentially between pins 2 and 3) 16. In accordance with UL 1577, each
This test mode is for testing LED to the common-mode gain (input pins optocoupler is proof tested by
coupling and is not intended for tied together and the signal applied applying an insulation test voltage
customer use. to both inputs at the same time), ≥ 4200 Vrms for 1 second (leakage
2. This is the Absolute Value of Input expressed in dB. detection current limit, II-O ≤ 5 µA).
Offset Change vs. Temperature. 13. Output noise comes from two primary This test is performed before the
3. Gain is defined as the slope of the sources: chopper noise and sigma- 100% production test for partial
best-fit line of differential output delta quantization noise. Chopper discharge (method b) shown in
voltage (VOUT+–V OUT- ) vs. differential noise results from chopper stabilization VDE 0884 Insulation Characteristic
input voltage (VIN+–VIN- ) over the of the output op-amps. It occurs at a Table.
specified input range. specific frequency (typically 400 kHz 17. The Input-Output Momentary With-
4. This is the Absolute Value of Gain at room temperature), and is not stand Voltage is a dielectric voltage
Change vs. Temperature. attenuated by the internal output filter. rating that should not be interpreted
5. Nonlinearity is defined as half of the A filter circuit can be easily added to as an input-output continuous voltage
peak-to-peak output deviation from the external post-amplifier to reduce rating. For the continuous voltage
the best-fit gain line, expressed as a the total rms output noise. The rating refer to the VDE 0884 insula-
percentage of the full-scale differential internal output filter does eliminate tion characteristics table and your
output voltage. most, but not all, of the sigma-delta equipment level safety specification.
6. NL100 is the nonlinearity specified over quantization noise. The magnitude of 18. This is a two-terminal measurement:
an input voltage range of ± 100 mV. the output quantization noise is very pins 1–4 are shorted together and
small at lower frequencies (below pins 5–8 are shorted together.
 9

VDD1 VDD2 +15 V

0.1 µF

1 8
0.1 µF

2 7 10 K
+
HCPL-7840 VOUT
0.1 µF 10 K –
3 6 AD624CD
GAIN = 100

4 5 0.47 0.47 0.1 µF

µF µF

-15 V

Figure 1. Input Offset Voltage Test Circuit.

0.8 0.39 8.035

VOS – INPUT OFFSET VOLTAGE – mV
VOS – INPUT OFFSET VOLTAGE – mV

vs. VDD1
0.7 0.38
vs. VDD2 8.03

G – GAIN – V/V
0.6 0.37
8.025
0.5 0.36
8.02
0.4 0.35

8.015
0.3 0.34

0.2 0.33 8.01

-55 -25 5 35 65 95 125 4.5 4.75 5.0 5.25 5.5 -55 -35 -15 5 25 45 65 85 105 125
TA – TEMPERATURE – °C VDD – SUPPLY VOLTAGE – V TA – TEMPERATURE – °C

Figure 2. Input Offset Voltage vs. Figure 3. Input Offset vs. Supply. Figure 4. Gain vs. Temperature.
Temperature.

VDD1 VDD2 +15 V +15 V

0.1 µF 0.1 µF

1 8
0.1 µF 0.1 µF

404 2 7 10 K
VIN
+ +
HCPL-7840 VOUT
13.2 6 10 K – –
3 AD624CD AD624CD
GAIN = 4 GAIN = 10
0.01 µF

4 5 0.47 0.47 0.1 µF 0.1 µF

µF µF

-15 V -15 V

10 K

0.47
µF

Figure 5. Gain and Nonlinearity Test Circuit.

 10

8.032 0.03 0.005

0.025

NL – NONLINEARITY – %
NL – NONLINEARITY – %
8.03
G – GAIN – V/V

0.02 0.004

8.028 0.015

vs. VDD1 0.01 0.003

8.026 vs. VDD1
vs. VDD2
0.005
vs. VDD2
8.024 0 0.002
4.5 4.75 5.0 5.25 5.5 -55 -25 5 35 65 95 125 4.5 4.75 5.0 5.25 5.5
VDD – SUPPLY VOLTAGE – V TA – TEMPERATURE – °C VDD – SUPPLY VOLTAGE – V

Figure 6. Gain vs. Supply. Figure 7. Nonlinearity vs. Temperature. Figure 8. Nonlinearity vs. Supply.

4.2 13 0
IDD – SUPPLY CURRENT – mA
VO – OUTPUT VOLTAGE – V

IIN – INPUT CURRENT – µA

-1
3.4
10
-2
2.6
-3
7
1.8
IDD1 -4
VOP
IDD2
VOR
1.0 4 -5
-0.5 -0.3 -0.1 0.1 0.3 0.5 -0.5 -0.3 -0.1 0.1 0.3 0.5 -0.6 -0.4 -0.2 0 0.2 0.4 0.6
VIN – INPUT VOLTAGE – V VIN – INPUT VOLTAGE – V VIN – INPUT VOLTAGE – V

Figure 9. Output Voltage vs. Input Figure 10. Supply Current vs. Input Figure 11. Input Current vs. Input
Voltage. Voltage. Voltage.

1 50 5.5
PD – PROPAGATION DELAY – µS

0
0 4.7 Tpd 10
-50 Tpd 50
Tpd 90
-1 3.9 Trise
-100
PHASE
GAIN

-2 -150
3.1

-200
-3 2.3
-250

-4 -300 1.5
10 1000 10000 10 1000 10000 -55 -25 5 35 65 95 125
FREQUENCY FREQUENCY TA – TEMPERATURE – °C

Figure 12. Gain vs. Frequency. Figure 13. Phase vs. Frequency. Figure 14. Propagation Delay vs.
Temperature.
 11

10 K

VDD1 VDD2 +15 V

0.1 µF

1 8
0.1 µF
0.1 µF

2 7 2K
VIN
–
HCPL-7840 VOUT
6 2K +
0.01 µF 3 MC34081

4 5 0.1 µF
10 K

-15 V

VIN IMPEDANCE LESS THAN 10 Ω.

Figure 15. Propagation Delay Test Circuits.

10 K

VDD2 150 pF
78L05
IN OUT +15 V
0.1 µF
0.1 0.1 1 8
µF µF 0.1 µF

2 7 2K
–
9V HCPL-7840 VOUT
6 2K +
3 MC34081

4 5 0.1 µF
10 K

150
PULSE GEN. pF -15 V

+ –
VCM

Figure 16. CMTI Test Circuits.

 12

Application Information an RC anti-aliasing filter (R2 and some sort of simple isolated
Power Supplies and Bypassing C2). Although the application supply can be used, such as a line
The recommended supply con- circuit is relatively simple, a powered transformer or a high-
nections are shown in Figure 17. few recommendations should frequency DC-DC converter.
A floating power supply (which in be followed to ensure optimal
many applications could be the performance. An inexpensive 78L05 three-
same supply that is used to drive terminal regulator can also be
the high-side power transistor) is The power supply for the used to reduce the floating supply
regulated to 5 V using a simple HCPL -7840 is most often obtained voltage to 5 V. To help attenuate
zener diode (D1); the value of from the same supply used to high-frequency power supply
resistor R4 should be chosen to power the power transistor gate noise or ripple, a resistor or
supply sufficient current from drive circuit. If a dedicated inductor can be used in series
the existing floating supply. The supply is required, in many with the input of the regulator
voltage from the current sensing cases it is possible to add an to form a low-pass filter with
resistor (Rsense) is applied to the additional winding on an exist- the regulator’s input bypass
input of the HCPL-7840 through ing transformer. Otherwise, capacitor.

HV+ +
FLOATING
GATE DRIVE POWER
• • • SUPPLY
CIRCUIT
–

D1 C1
5.1 V 0.1 µF
R2

39 Ω
C2 HCPL-7840
0.01 µF

MOTOR
+ R1 –
• • •
RSENSE

• • •

HV–

Figure 17. Recommended Supply and Sense Resistor Connections.

 13

As shown in Figure 18, 0.1 µF bypass capacitor (C2) is also the input signal. The input filter
bypass capacitors (C1, C2) recommended at the input due to also performs an important
should be located as close as the switched-capacitor also forms reliability function—it reduces
possible to the pins of the HCPL- part of the anti-aliasing filter, transient spikes from ESD events
7840. The bypass capacitors are which is recommended to prevent flowing through the current
required because of the high- high-frequency noise from sensing resistor.
speed digital nature of the signals aliasing down to lower
inside the HCPL-7840. A 0.01 µF frequencies and interfering with

POSITIVE
FLOATING
C5
SUPPLY
HV+ 150 pF

GATE DRIVE R3
• • •
CIRCUIT
10.0 K
U1 +5 V
78L05 +15 V
IN OUT C8
0.1 µF
C1 C2 1 8
C4
0.1 0.1
µF µF 0.1 µF
R5 2 7 R1

68 –
C3 2.00 K
U2 U3 VOUT
0.01 6 R2 + MC34081
µF 3
2.00 K
MOTOR C7
+ – 4 5
• • • C6 R4
RSENSE 0.1 µF
150 pF 10.0 K
HCPL-7840
-15 V
• • •

HV–

Figure 18: Recommended Application Circuit.

PC Board Layout
The design of the printed circuit possible distance between the PC board does not pass directly
board (PCB) should follow good input and output sides of the below or extend much wider than
layout practices, such as keeping circuit and ensuring that any the body of the HCPL-7840.
bypass capacitors close to the ground or power plane on the
supply pins, keeping output
signals away from input signals,
the use of ground and power C2 C4
R5
planes, etc. In addition, the
layout of the PCB can also affect
the isolation transient immunity
(CMTI) of the HCPL-7840, due C3

primarily to stray capacitive

coupling between the input and
the output circuits. To obtain TO VDD1
optimal CMTI performance, the TO VDD2
layout of the PC board should TO RSENSE+ VOUT+
TO RSENSE– VOUT–
minimize any stray coupling by
maintaining the maximum
Figure 19. Example Printed Circuit Board Layout.
 14

MOTOR OUTPUT POWER – HORSEPOWER

Current Sensing Resistors 40
the resistor-to-ambient ther-mal
The current sensing resistor 440 V resistance increases. This effect
35 380 V
should have low resistance (to can be minimized by reducing the
220 V
30 thermal resistance of the current
minimize power dissipation), low 120 V
inductance (to minimize di/dt 25 sensing resistor or by using a
induced voltage spikes which 20
resistor with a lower tempco.
could adversely affect operation), Lowering the thermal resistance
15
and reasonable tolerance (to can be accomplished by
10 repositioning the current sensing
maintain overall circuit accuracy).
Choosing a particular value for 5 resistor on the PC board, by
the resistor is usually a compro- 0 using larger PC board traces to
0 5 10 15 20 25 30 35
mise between minimizing power carry away more heat, or by
MOTOR PHASE CURRENT – A (rms)
dissipation and maximizing accu- using a heat sink.
racy. Smaller sense resistance
Figure 20. Motor Output Horsepower For a two-terminal current
decreases power dissipation, vs. Motor Phase Current and Supply
while larger sense resistance Voltage. sensing resistor, as the value
can improve circuit accuracy by of resistance decreases, the re-
utilizing the full input range of sistance of the leads become a
the HCPL -7840. can also be easily calculated by significant percentage of the total
multiplying the sense resistance resistance. This has two primary
The first step in selecting a sense times the square of the maximum effects on resistor accuracy.
resistor is determining how much RMS current, which is about 1 W First, the effective resistance of
current the resistor will be sensing. in the previous example. If the the sense resistor can become
The graph in Figure 20 shows the power dissipation in the sense dependent on factors such as
RMS current in each phase of a resistor is too high, the resistance how long the leads are, how
three-phase induction motor as a can be decreased below the they are bent, how far they are
function of average motor output maximum value to decrease inserted into the board, and how
power (in horsepower, hp) and power dissipation. The minimum far solder wicks up the leads
motor drive supply voltage. The value of the sense resistor is during assembly (these issues
maximum value of the sense re- limited by precision and accuracy will be discussed in more detail
sistor is determined by the current requirements of shortly). Second, the leads are
being measured and the maxi- the design. As the resistance typically made from a material,
mum recommended input voltage value is reduced, the output such as copper, which has a
of the isolation amplifier. The voltage across the resistor is also much higher tempco than the
maximum sense resistance can be reduced, which means that the material from which the resistive
calculated by taking the maxi- offset and noise, which are fixed, element itself is made, resulting
mum recommended input voltage become a larger percentage in a higher tempco overall.
and dividing by the peak current of the signal amplitude. The
that the sense resistor should see selected value of the sense Both of these effects are
during normal operation. For resistor will fall somewhere eliminated when a four-terminal
example, if a motor will have a between the minimum and current sensing resistor is used.
maximum RMS current of 10 A maximum values, depending on A four-terminal resistor has two
and can experience up to 50% the particular requirements of additional terminals that are
overloads during normal op- a specific design. Kelvin-connected directly across
eration, then the peak current is the resistive element itself; these
21.1 A (=10 x 1.414 x 1.5). When sensing currents large two terminals are used to monitor
Assuming a maximum input enough to cause significant the voltage across the resistive
voltage of 200 mV, the maximum heating of the sense resistor, the element while the other two
value of sense resistance in this temperature coefficient (tempco) terminals are used to carry the
case would be about 10 mΩ. of the resistor can introduce load current. Because of the
nonlinearity due to the signal Kelvin connection, any voltage
The maximum average power dependent temperature rise of the drops across the leads carrying
dissipation in the sense resistor resistor. The effect increases as
 15

the load current should have no resistor resistor, while VIN- (pin amplifier. Generally, op-amps
impact on the measured voltage. 3) is shorted to GND1 (pin 4), with bipolar input stages exhibit
with the power-supply return better offset performance than
When laying out a PC board for path functioning as the sense line op-amps with JFET or MOSFET
the current sensing resistors, a to the negative terminal of the input stages.
couple of points should be kept current sense resistor. This
in mind. The Kelvin connections allows a single pair of wires or In addition, the op-amp should
to the resistor should be brought PC board traces to connect the also have enough bandwidth
together under the body of the HCPL-7840 circuit to the sense and slew rate so that it does not
resistor and then run very close resistor. By referencing the input adversely affect the response
to each other to the input of the circuit to the negative side of the speed of the overall circuit. The
HCPL-7840; this minimizes the sense resistor, any load current post-amplifier circuit includes a
loop area of the connection and induced noise transients on the pair of capacitors (C5 and C6)
reduces the possibility of stray resistor are seen as a common- that form a single-pole low-pass
magnetic fields from interfering mode signal and will not interfere filter; these capacitors allow the
with the measured signal. If with the current-sense signal. bandwidth of the post-amp to
the sense resistor is not located This is important because the be adjusted independently of the
on the same PC board as the large load currents flowing gain and are useful for reducing
HCPL-7840 circuit, a tightly through the motor drive, along the output noise from the isola-
twisted pair of wires can with the parasitic inductances tion amplifier. Many different op-
accomplish the same thing. inherent in the wiring of the amps could be used in the circuit,
circuit, can generate both noise including: MC34082A (Motorola),
Also, multiple layers of the PC spikes and offsets that are TLO32A, TLO52A, and TLC277
board can be used to increase relatively large compared to the (Texas Instruments), LF412A
current carrying capacity. small voltages that are being (National Semiconductor).
Numerous plated-through vias measured across the current
should surround each non-Kelvin sensing resistor. The gain-setting resistors in the
terminal of the sense resistor post-amp should have a tolerance
to help distribute the current If the same power supply is used of 1% or better to ensure adequate
between the layers of the PC both for the gate drive circuit CMRR and adequate gain toler-
board. The PC board should use and for the current sensing ance for the overall circuit.
2 or 4 oz. copper for the layers, circuit, it is very important that Resistor networks can be used
resulting in a current carrying the connection from GND1 of the that have much better ratio
capacity in excess of 20 A. HCPL-7840 to the sense resistor tolerances than can be achieved
Making the current carrying be the only return path for supply using discrete resistors. A
traces on the PC board fairly current to the gate drive power resistor network also reduces
large can also improve the sense supply in order to eliminate the total number of components
resistor’s power dissipation potential ground loop problems. for the circuit as well as the
capability by acting as a heat The only direct connection required board space.
sink. Liberal use of vias where between the HCPL-7840 circuit
the load current enters and exits and the gate drive circuit should Please refer to Agilent
the PC board is also be the positive power supply line. Applications Note 1078 for
recommended. additional information on using
Output Side Isolation Amplifiers.
Sense Resistor Connections The op-amp used in the external
The recommended method for post-amplifier circuit should be of
connecting the HCPL-7840 to the sufficiently high precision
current sensing resistor is shown so that it does not contribute
in Figure 18. VIN+ (pin 2 of the a significant amount of offset
HPCL-7840) is connected to the or offset drift relative to the
positive terminal of the sense contribution from the isolation
 16

FREQUENTLY ASKED QUESTIONS ABOUT THE HCPL-7840

1. THE BASICS
1.1: Why should I use the HCPL-7840 for sensing current when Hall-effect sensors are available
which don’t need an isolated supply voltage?
Available in an auto-insertable, 8-pin DIP package, the HCPL-7840 is smaller than and has better linearity,
offset vs. temperature and Common Mode Rejection (CMR) performance than most Hall-effect sensors.
Additionally, often the required input-side power supply can be derived from the same supply that powers
the gate-drive optocoupler.

2. SENSE RESISTOR AND INPUT FILTER

2.1: Where do I get 10 mΩ resistors? I have never seen one that low.
Although less common than values above 10 Ω, there are quite a few manufacturers of resistors suitable for
measuring currents up to 50 A when combined with the HCPL- 7840.
Example product information may be found at Dale’s web site (http://www.vishay.com/vishay/dale) and
Isotek’s web site (http://www.isotekcorp.com).

2.2: Should I connect both inputs across the sense resistor instead of grounding VIN- directly
to pin 4?
This is not necessary, but it will work. If you do, be sure to use an RC filter on both pin 2 (VIN+) and pin 3
(VIN-) to limit the input voltage at both pads.

2.3: Do I really need an RC filter on the input? What is it for? Are other values of R and C okay?
The input anti-aliasing filter (R=39 Ω, C=0.01 µF) shown in the typical application circuit is recommended
for filtering fast switching voltage transients from the input signal. (This helps to attenuate higher signal
frequencies which could otherwise alias with the input sampling rate and cause higher input offset voltage.)
Some issues to keep in mind using different filter resistors or capacitors are:
1. (Filter resistor:) Input bias current for pins 2 and 3: This is on the order of 500 nA. If you are using a
single filter resistor in series with pin 2 but not pin 3 the IxR drop across this resistor will add to the offset
error of the device. As long as this IR drop is small compared to the input offset voltage there should not be
a problem. If larger-valued resistors are used in series, it is better to put half of the resistance in series with
pin 2 and half the resistance in series with pin 3. In this case, the offset voltage is due mainly to resistor
mismatch (typically less than 1% of the resistance design value) multiplied by the input bias.
2. (Filter resistor:) The equivalent input resistance for -7840 is around 500 kΩ. It is therefore best to
ensure that the filter resistance is not a significant percentage of this value; otherwise the offset voltage will
be increased through the resistor divider effect. [As an example, if Rfilt = 5.5 kΩ, then VOS = (Vin * 1%) =
2 mV for a maximum 200 mV input and VOS will vary with respect with Vin.]
3. The input bandwidth is changed as a result of this different R-C filter configuration. In fact this is one of
the main reasons for changing the input-filter R-C time constant.
4. (Filter capacitance:) The input capacitance of the -78XX is approximately 1.5 pF. For proper operation
the switching input-side sampling capacitors must be charged from a relatively fixed (low impedance)
voltage source. Therefore, if a filter capacitor is used it is best for this capacitor to be a few orders of
magnitude greater than the CINPUT (A value of at least 100 pF works well.)
 17

2.4: How do I ensure that the HCPL-7840 is not destroyed as a result of short circuit conditions
which cause voltage drops across the sense resistor that exceed the ratings of the HCPL-7840’s
inputs?
Select the sense resistor so that it will have less than 5 V drop when short circuits occur. The only other
requirement is to shut down the drive before the sense resistor is damaged or its solder joints melt. This
ensures that the input of the HCPL-7840 can not be damaged by sense resistors going open-circuit.

3. ISOLATION AND INSULATION

3.1: How many volts will the HCPL-7840 withstand?
The momentary (1 minute) withstand voltage is 3750 V rms per UL 1577 and CSA Component Acceptance
Notice #5.

4. ACCURACY
4.1: Can the signal to noise ratio be improved?
Yes. Some noise energy exists beyond the 100 kHz bandwidth of the HCPL-7800(A). Additional filtering
using different filter R,C values in the post-amplifier application circuit can be used to improve the signal to
noise ratio. For example, by using values of R3 = R4 = 10 kΩ, C5 = C6 = 470 pF in the application circuit
the rms output noise will be cut roughly by a factor of 2. In applications needing only a few kHz bandwidth
even better noise performance can be obtained. The noise spectral density is roughly 500 nV/√ Hz below
20 kHz (input referred).

4.2: I need 1% tolerance on gain. Does HP sell a more precise version?

The HCPL-7800A is gain-trimmed and matched to within ± 1% tolerance (at room temperature.)

4.3: Does the gain change if the internal LED light output degrades with time?
No. The LED is used only to transmit a digital pattern. HP has accounted for LED degradation in the design
of the product to ensure long life.

5. POWER SUPPLIES AND START-UP

5.1: What are the output voltages before the input side power supply is turned on?
VO+ is close to 1.29 V and VO- is close to 3.80 V. This is equivalent to the output response at the condition
that LED is completely off.

5.2: How long does the HCPL-7840 take to begin working properly after power-up?
Within 1 ms after VDD1 and VDD2 powered the device starts to work. But it takes longer time for output to
settle down completely. In case of the offset measurement while both inputs are tied to ground there is
initially VOS adjustment (about 60 ms). The output completely settles down in 100 ms after device powering
up.
 18

6. MISCELLANEOUS
6.1: How does the HCPL-7840 measure negative signals with only a +5 V supply?
The inputs have a series resistor for protection against large negative inputs. Normal signals are no more
than 200 mV in amplitude. Such signals do not forward bias any junctions sufficiently to interfere with
accurate operation of the switched capacitor input circuit.
www.semiconductor.agilent.com
Data subject to change.
Copyright © 1999 Agilent Technologies
Obsoletes 5965-8976E
5967-5626E (11/99)