Short Introduction to Power PROFET™

Short Introduction to Power PROFET™
S h o r t I n t r o d u c t i o n t o P o w e r P R O F E T TM
What the designer should know
Application Note
Rev. 1.0, 2015-01-20
Automotive Power
Application Note
Power PROFETTM
Table of Contents
Table of Contents
1
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3
3.1
Inductive Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Free Wheeling Diode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
4
4.1
4.2
4.2.1
4.3
Short Circuit Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Relay and Fuse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Semiconductor Switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Current Trip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Repetitive Short Circuit Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10
10
10
10
11
5
5.1
5.2
5.2.1
5.3
Current Sense . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Current Sense Signal in Nominal Current Range . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Current Sense two-points calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Calibration example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Calibrated dkILIS variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12
12
13
14
16
6
6.1
6.2
6.3
Typical Application Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Loss of VS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Loss of GND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Inverse Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
18
18
19
19
7
7.1
7.1.1
7.1.2
7.1.3
7.2
7.3
7.4
7.4.1
7.4.2
7.4.3
7.5
Specific Application Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Low Battery Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Requirements from Tier1 / OEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Undervoltage Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Countermeasures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Load Jump . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Parallel Operation of Power PROFETTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Power Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DC operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PWM operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fault conditions during PWM operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Capacitor at IN pin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
21
21
21
21
21
22
23
24
24
25
26
27
8
Revision History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Application Note
2
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Abstract
1
Abstract
Note: The following information is given as a hint for the implementation of the device only and shall not be
regarded as a description or warranty of a certain functionality, condition or quality of the device.
This Application Note is intended to provide useful information to designers using Power PROFETTM high side
power switch in the automotive environment. With those information, the designer is able to optimize the system
by better understanding the constraints for short circuit protection and repetitive switching of inductive loads.
Table 1
Terms in Use
Abbreviation
Meaning
αIS0
Temperature coefficient for IIS0(cal) (see below)
CIN
Input Capacitor
Csense
Capacitor for Sense Signal Filtering
dkILIS
Current Sense Differential Ratio
dkILIS(max)
Current Sense Differential Ratio, maximum value
dkILIS(min)
Current Sense Differential Ratio, minimum value
dkILIS(typ)
Current Sense Differential Ratio, typical value
dkILIS(cal)
Current Sense Slope Ratio measured after 2-points calibration
Δ(dkILIS)
Current Sense Differential Ratio spread over temperature and repetitive pulse
operation after 2-points calibration
DC
Duty Cycle of the PWM signal
EOFF
Energy Associated to the switch off phase
EON
Energy Associated to the switch on phase
kSW
Correction Factor
ICL(0)
Current Trip Detection Level
IGND
Ground Current
IIS
Sense Current
IIS(cal)1
Sense Current at the first calibration point
IIS(cal)2
Sense Current at the second calibration point
IIS0(cal)
Current Sense Offset (after calibration)
IIS0
Current Sense Offset
IISn,min
Minimum Value of the Sense Current at a generic load condition
IL(cal)1
Load Current at the first calibration point
IL(cal)2
Load Current at the second calibration point
ILmin,det
Lowest Detectable Load Current
IL,min
Minimum Load Current
IL,max
Maximum Load Current
ILn
Generic Value for Load Current
IIS(FAULT)
Sense Signal Current in Fault Condition
IL(NOM)
Nominal Load Current
L
Inductivity
PDC,Logic
Logic Circuitry Power Dissipation Contribution
PDC,Power
DMOS Power Dissipation Contribution
Application Note
3
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Abstract
Table 1
Terms in Use
Abbreviation
Meaning
PDC,TOT
Total Power Dissipation under DC operation
PSW
Power Dissipation due to Switching Losses
RDS(ON)
On-State Resistance
RIN
Input Resistor
RLOAD
Load Equivalent Resistor
RIS
Sense Resistor
Rsense
Protection Resistor between Sense Pin and Microcontroller
RθJA
Thermal Resistance Junction to Ambient
RθJA(2s2p)
Therma Resistance Junction to Ambien (FR4 2s2p board
RVS
Integrated Resistor
tSW(ON)
DMOS Switch ON time
tSW(OFF)
DMOS Switch OFF time
TA
Ambient Temperature
Tcal
Temperature at which the calibration is performed
TPWM
PWM signal period
TJ
Junction Temperature
Tx
Temperature at the operating point
tOFF(TRIP)
Overload Shutdown Delay Time
VDS(CL)
Drain-Source-Clamping Voltage
VIS
Voltage at the IS pin
VIN
Input Voltage
VIN(H)
High Level Input Voltage
VIN(L)
Low Level Input Voltage
VOUT
Output Voltage
VS
Supply Voltage
Z
Zener / Clamping Diode
Application Note
4
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Introduction
2
Introduction
The BTS50015-1TAA, also named Power PROFETTM, is a single channel High Side Switch with very low RDS(ON)
(typical 1.5mΩ). It integrates also protective functions, diagnosis and Infineon Reversave capability. The target of
this Application Note is to give the user some deeper information about the main features of the device as well as
to provide some advices on how to use the Power PROFETTM in the real application.
Table 2 shows the main parameters extracted from the datasheet.
Table 2
Product main features
Parameter
Symbol
Value
Operating voltage
VS(OP)
8V...18V
Extended supply voltage
VS(DYN)
3.2V...28V
Maximum on resistance at TJ= 150°C RDS(ON)
3mΩ
Minimum nominal load current
33A
IL(nom)
Typical current sense differential ratio dkILIS
51500
Minimum detectable load current
ILmin,det
9.2A
Minimum short circuit current
threshold
ICL0_min
135A
Minimum stand-by current for the
whole device with load at TA = TJ =
85°C
IS(OFF)
18μA
Maximum reverse battery voltage at
TA = 25°C for 2 min
-VS(REV)
16V
Application Note
5
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Inductive Loads
3
Inductive Loads
Each application and each circuitry consists of some inductive load, either intentionally as motor, solenoid or valve,
or as parasitic inductivity due to the wire harness. Whatever is the origin of the inductive component, there will be
a certain amount of energy stored in the inductance, which is proportional to both the inductance value itself and
the square of the current.
When switching OFF inductive loads with high side switches, the voltage at OUT pin drops below ground potential
because the inductance does not allow an instantaneous current value variation (the current can not go to zero
suddenly) and keeps driving the current in the same direction (see Figure 1).
Voltage spike at VS due to the
inductance at battery side
Voltage at OUT goes
below ground
VS
VS
VS
ON
0V
OFF
OFF
OUT
OUT
OUT
The inductance tries to
keep the current
a)
Figure 1
b)
c)
Effect of the inductance on the application
The OUT pin voltage will remain below ground as long as the energy stored in the inductance has not been
completely dissipated (Figure 1 b).
Moreover any inductive component on the battery side brings some positive spikes on the VS (see Figure 1 c).
To avoid the destruction of the device during this events due to too high voltages, an overvoltage protection is
implemented. This overvoltage protection clamps the voltage (see Figure 2) between VS and OUT at VDS(CL).
Voltage between drain
and source is clamped
IL
VDS
Current goes to zero after the
inductor is completely demagnetized
Figure 2
Voltage clamping across the DMOS
Application Note
6
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Inductive Loads
Nevertheless it is not recommended to operate the device repetitively under this condition. Therefore when driving
an inductive load or any other load with additional inductive component (e.g. due to wire harness or to the load
itself) which can cause the device to be clamped, a free wheeling diode should be always placed.
3.1
Free Wheeling Diode
The need for implementing a free wheeling diode depends on the type of load driven by the Power PROFETTM.
Motors, solenoids, etc. are some typical examples of inductive loads. However some loads can show also
inductive behavior due to parasitic inductances. The former require definitely a free wheeling diode. For the latter
it is not so trivial to determine whether the free wheeling diode is really mandatory.
Some basic steps are recommended in order to decide whether the free wheeling diode should be considered in
the application:
•
•
•
•
•
•
connect the load to the Power PROFETTM with a cable with impedance as close as possible to the real wire
harness present in the application
connect the battery to the Power PROFETTM with a cable reproducing the impedance of the wire harness
actually present in the application
set the current at the highest value expected in the application
set the temperature at the lowest value within the application conditions
switch on/off the device through the IN pin
checking with the oscilloscope the voltage across the DMOS (VDS, voltage between VS and OUT)
The VDS shoult not exceed, in any circumstance, the minimum voltage of the parameter 6.1.11 of the datasheet,
(VDS(CL), min = 28V).
Figure 3 shows that in some cases (a and c) the condition is clear and well defined (resistive or inductive). In reality
also “pure” resistive load may have inductive components due to the inductance of the wires (Figure b).
VDS
VDS
28V
VDS(CL)
VDS(CL)
VDS
28V
VDS(CL)
Max VDS always lower
than 28V
28V
VDS is clamped
t
IL
t
IL
t
Pure resistive load
Figure 3
t
IL
t
t
Resistive load with wire
inductance
Inductive load
Different effect of the load inductive component
The easiest recommendation about the free wheeling diode is to select a Schottky diode, thanks to its very low
forward voltage and fast switching. The choice of the proper Schottky diode should consider also:
Application Note
7
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Inductive Loads
•
•
maximum DC reverse voltage. The Schottky diode, whose cathode is connected to OUT and anode to ground,
is normally subject to an inverse polarization and should withstand the maximum battery voltage the load can
experience in the application.
maximum peak forward surge current. Since the diode is conducting only during the switch off of the inductive
load, this parameter plays a relevant role for the choice, more than the average current
Fugure 4 shows how the OUT pin voltage is limited while dropping below ground when a free wheeling diode is
used (b), differently from what would happen without any diode (a).
VIN
VIN
t
t
VOUT
VOUT
VS
VS
t
t
VS -V DS( CL)
VS -VDS(CL)
IL
IL
t
t
Without free wheeling diode
Figure 4
With free wheeling diode
Effect of the free wheeling diode
After selecting the proper diode and then embedding it in the application, it is absolutely important to ensure the
effectiveness of the solution. Any not optimized connection with the load or in the ground path might significanlty
jeopardize the effectiveness of the diode.
For example, the high impedance connection (e.g. wire too long) depicted in the Figure 5 (upper schematic)
generates a delay in the switch on of the diode, which means in turn that the VDS could increase. As consequence
the goal of the free wheeling diode might be not achieved.
Application Note
8
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Inductive Loads
BTS50015 -1TAA
Free Wheeling Diode
Inductive
Load
Not optimized free wheeling path
Inductive
Load
Recommended free wheeling path
BTS50015 -1TAA
Free Wheeling Diode
Figure 5
Recommended free wheeling diode connection
Application Note
9
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Short Circuit Protection
4
Short Circuit Protection
There is always a risk for a short circuit in application. Such short circuit can happen anywhere, between each
possible pin combination. Some short circuits are more severe than others. Power PROFETTM has implemented
protection features to prevent the device from damage in case of short circuit.
In this chapter the pros and cons of the different protection concepts are discussed and how to optimize the shortcircuit protection within the application.
4.1
Relay and Fuse
Since many years, semiconductors are replacing more and more relays in automotive applications . Especially at
the body controller (BCM - Body Control Module) it is almost standard to switch loads like bulbs or small motors
by semiconductor switches instead of relays. The semiconductors do not only replace the switch function in the
application, but they also cover the protection functionality to support the removal of the fuses.
Nevertheless, there are still a lot of applications where the relay is used as a switch. The reason could be either
technically, when no adequate semiconductor switch is available, or commercially, because relays are often still
cheaper than semiconductors. When relays are used, the protection against overload or short-circuit is done by
fuses. The advantage is the price because fuses are quite cheap. But there are some disadvantages:
•
•
•
•
If fuses are blown, they must be replaced, therefore the fuse box must be always mounted where it can be
easily accessible.
Relay + fuse requires more space than semiconductor switch.
Contact resistance between fuse holder and fuse is increasing over lifetime, especially when fuse is blown and
must be replaced, resulting in increased voltage drop and power losses over the fuse.
Due to the long reaction time to overload (e.g. double nominal current), the complete system must be
developed to withstand currents higher than nominal current for a certain time. This leads to thicker wires than
needed under normal conditions, resulting in higher costs and weight.
4.2
Semiconductor Switch
The Smart High Side Switches like the Power PROFETTM embed several kind of protections. The protection
themselves and the approach to pursue these protection can vary from device to device. Some products are
features with reverse polarity protection, some not, some devices are protected against GND disconnection, etc.
A protection against overcurrent or short circuit events is instead a must which should be integrated in any Smart
Switch.
The overcurrent protection approach differs in accordance with the target application, technologies, etc. This
results in different methods to assure an effective reaction under these circumstances.
An extensive dissertation of the possible overcurrent protections is beyond the aim of this Application Note. The
next chapters will focus on the description of the protections embedded in the Power PROFETTM.
4.2.1
Current Trip
The protection against overcurrent conditions or short circuit is achieved through the so called "current trip"
approach. Basically a current threshold is internally set (ICL(0) ,defined as in the datasheet). The current through
the DMOS is monitored and as soon as it exceeds the minimum value of ICL(0) , the device switches off.
The device does not switch on again (it is latched) unless the IN pin is toggled.
Considering this protection approach, it is important to take care every time a load with inrush current is driven.
The user should make sure that the inrush current is, under any condition, below the minimum value of ICL(0).
The benefit of such a protection is definitely the safety (the device switches off and not automatically switches on
again) and the limited thermal stress for the silicon.
Application Note
10
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Short Circuit Protection
4.3
Repetitive Short Circuit Operation
In many applications, OEM or Tier1 require a certain number of short circuit cycles. The test conditions, definitions
and classifications are well described in the AEC-Q100-012.
There are two ways to specify the short-circuit robustness: Either by giving a extrapolated failure ppm-rate for a
certain number of cycles or to grade the device according to the cycle capability. Power PROFETTM uses the
second approach, they are in Grade D, i.e. a pre-defined number of parts was tested in short-circuit mode for
100kcycles and no device was failing this test. The test was performed at ambient temperature of -40oC. When
using a different ambient temperature or different conditions than specified in the datasheet, the maximum
achievable short-circuit cycles before getting a failure might be lower or bigger than specified. For example, a
higher ambient temperature (e.g. 85oC) will lead to a reduced number of short-circuit cycles.
Application Note
11
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Current Sense
5
Current Sense
5.1
Current Sense Signal in Nominal Current Range
Figure 6 shows the current sense as function of the load current in the Power DMOS. The upper and lower solid
curves represent respectively the dkILIS(min) and the dkILIS(max) (minimum and maximum slope of the current sense
ratio), whereas the dotted curve between them is the dkILIS(typ) (typical slope of the current sense ratio).The lower
curve should theoretically cross the y-axis at a negative value, although in reality the sense circuitry can´t provide
a negative current. The crossing point between this curve and the x-axis indicates the lowest detectable current
ILmin,det.
I Lmin ,det = I Ln – ( I ISn ,min ⋅ dk KILIS ( max ) )
(1)
where the IISn,min is the minimum value of sense current at a certain load current ILn, as indicated in parameters
from 6.1.43 to 6.1.46 in the datasheet. For BTS50015-1TAA, ILmin,det is around 9A.
In general, once a sense current is measured, the estimated load current can be in the following range:
( I IS – I IS0 ) ⋅ dk ILIS ( min ) < I L < ( I IS + I IS0 ) ⋅ dk ILIS ( max )
(2)
3.5
dkILIS (min)
3
dkILIS (typ)
2.5
dkILIS (max)
IIS (mA)
2
1.5
1
0.5
IIS0(max)
0
0
ILmin,det
20
IL1
40
IL2
60
80
IL3
100
120
IL4
140
160
IL (A)
Figure 6
Current Sense for Nominal and Overload Condition
.
Application Note
12
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Current Sense
5.2
Current Sense two-points calibration
For some application, the specified current sense range might be too inaccurate. To achieve a better accuracy,
Power PROFETTM allows a calibration with two calibration points. The general idea of 2-points calibration, shown
in Figure 7, can be described as follow:
•
•
•
•
two different loads (IL(cal)1 and IL(cal)2) between ILmin,det and the minimum overcurrent threshold must be selected
the sense currents IS(cal)1 and IS(cal)2 are measured at the two load conditions above mentioned
These points (IL(cal)1 , IL(cal)2, IS(cal)1 and IS(cal)2 ) define the typical slope of the calibrated current sense curve IIS/IL
(Equation 3) and the crossing point at the y-axis, for simplicity called offset from now on (Equation 4)
I S ( cal )2 – I S ( cal )1
1
------------------------ = --------------------------------------dk ILIS ( cal )
I L ( cal )2 – I L ( cal )1
(3)
I L ( cal )1
I L ( cal )2
I IS0(cal) = I S ( cal )1 – ------------------------ = I S ( cal )2 – -----------------------dk ILIS ( cal )
dk ILIS ( cal )
(4)
in order to take into account the variation with the temperature and the possible repetitive pulse operation, the
parameters Δ(dkILIS(cal)) and αIS0 (respectively 6.1.47 and 6.1.54 in the datasheet) should be used to correct
the slope and the offset. The bluish area in Figure 7 is the result of this correction and represents the accuracy
achieved after 2-points calibration.
The load current, after the 2-points calibration, can be calculated as follow:
Δ ( dk ILIS ( cal ) )⎞ ⎛
I IS0 ( cal )
I L = dk ILIS ( cal ) ⋅ ⎛ 1 + --------------------------------- ⋅ I IS – ----------------------------------------------⎞
⎝
⎠ ⎝
100
1 + α ( T – T )⎠
IS0
x
(5)
cal
where dkILIS(cal) is the current sense slope ratio measured after two-points calibration (defined in Equation 3),
IIS0(cal) is the current sense offset (calculated after two points calibration, see Equation 4), Tx is the operating
temperature, and Tcal is temperature at which the calibration is performed (e.g. 25°C). The Equation 5 actually
provides two values for load current, considering that Δ(dkILIS(cal)) can be both positive and negative (see
parameter 6.1.47 in Table 6 of the datasheet):
Application Note
I IS0 ( cal )
8 -⎞ ⋅ min I – ---------------------------------------------I L, min = dk ILIS ( cal ) ⋅ ⎛ 1 – -------IS
⎝
⎠
1 + α IS0 ( T x – T cal )
100
(6)
I IS0 ( cal )
8 -⎞ ⋅ max I – ---------------------------------------------I L, max = dk ILIS ( cal ) ⋅ ⎛ 1 + -------IS
⎝
100⎠
1 + α IS0 ( T x – T cal )
(7)
13
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Current Sense
.:
1
dk ILIS(cal)
IIS
8%
8%
IIS(cal)2
IIS
IIS(cal)1
IIS0(cal)
IL
IL(cal)2
IL(cal)1
IL
Calibration points
Figure 7
Improved Current Sense Accuracy after Two-Points Calibration
5.2.1
Calibration example
In order to further explain the calibration procedure, here below an example will be described.
When the device is operating in the real application, the load current can be estimated using equation 3 of the
datasheet.
Assuming that the Power PROFET is driving a load and the measured current sense is 580μA, the load current
can be estimated as follows:
I L, min = ( I IS – I IS0 ) ⋅ dk ILIS, min = ( 580 – 150 )μA ⋅ 43700 = 18.8A
(8)
I L, max = ( I IS + I IS0 ) ⋅ dk ILIS, max = ( 580 + 150 )μA ⋅ 58200 = 42.5A
(9)
In case the achieved accuracy does not match the application requirements, a two-points calibration should be
performed.
As first step, two points in the range between 9A (ILmin,det ) and 135A (ICL(0), minimum overcurrent threshold) should
be selected . In Figure 7, IL(cal)1 = 20A and IL(cal)2 = 40A are selected. Assuming that the measured sense currents
are IIS(cal)1 = 350μA and IIS(cal)2 = 750μA, it is possible to calculate the slope and the offset:
Application Note
14
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Current Sense
I S ( cal )2 – I S ( cal )1
1
750 – 350 )μA- = 20 ⋅ 10 –6 ⇒ dkILIS ( cal ) = 50000
------------------------ = --------------------------------------- = (-----------------------------------dk ILIS ( cal )
( 40 – 20 )A
I L ( cal )2 – I L ( cal )1
(10)
I L ( cal )1
I L ( cal )2
20A- = – 50μA
I IS0(cal) = I S ( cal )1 – ------------------------ = I S ( cal )2 – ------------------------ = 350μA – -------------dk ILIS ( cal )
dk ILIS ( cal )
50000
(11)
Therefore the typical curve after two points calibration starts from the point IIS0(cal) = -50μA on the y-axis and has
a slope of 1/50000. It is important to remind that a basic assumption is that the calibration is performed at 25°C
ambient temperature. Both the slope and the offset must be corrected with the temperature, whereas the slope
should be corrected also considering the repetitive pulse operation. Assuming then that the application works
between -20°C and 85°C, the correction for the offset can be applied as follow:
I IS0 ( cal )
– 50μA
I IS0(cal)max = --------------------------------------------------- = ---------------------------------------------------------- = – 40.7μA
–3
1 + α IS0 ( T 85C – T cal )
1 + 3.8 ⋅ 10 ⋅ ( 85 – 25 )
(12)
I IS0 ( cal )
– 50μA
I IS0(cal)min = ------------------------------------------------------ = --------------------------------------------------------------- = – 60.3μA
–3
1 + α IS0 ( T – 20C – T cal )
1 + 3.8 ⋅ 10 ⋅ ( – 20 – 25 )
(13)
The correction for the slope is already provided by the parameter 6.1.47 of the datasheet.
When the device is working in the final application, for a given value of the sense current (in this example, IIS =
580μA) , it is possible, using Equation 5, to estimate the load current:
8 -⎞ ⋅ ( 580μA – ( – 60.3μA ) ) = 34.6A
I L, max = 50000 ⋅ ⎛ 1 + -------⎝
100⎠
(14)
8
I L, min = 50000 ⋅ ⎛⎝ 1 – ---------⎞⎠ ⋅ ( 580μA – ( – 40.7μA ) ) = 28.6A
100
(15)
Comparing Equation 6 and Equation 7 with Equation 13 and Equation 14, the contribution of the calibration to
the accuracy is significant. Figure 8 shows that also graphically.
Application Note
15
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Current Sense
Calibration points
(40A, 750µA)
IS
(20A, 350µA)
8%
8%
IIS=580µA
IIS0(cal)=-50µA
18.8A
28.6A
34.6A
IL
42.5A
Accuracy after 2-points calibration
Accuracy without calibration
Figure 8
Example of Two-Points Calibration
5.3
Calibrated dkILIS variation
As described in the previous chapters, the slope and the offset after two-points calibration can be corrected using
the parameters 6.1.47 and 6.1.54 of the datasheet. These parameters are defined by a characterization over a
limited number of samples and not subject to production tests. However the results can be considered with a good
level of confidence.
Moreover, focusing especially on the slope variation (Δ(dkILIS(cal))), it is possible to do some considerations in order
to have further improvements in terms of accuracy.
As stated in the datasheet, the parameter dkILIS(cal) variation are due to two different contributions:
•
•
temperature
repetitive pulse operation
The Figure 9 shows the behavior of dkILIS(cal) , highlighting the variations caused by the above mentioned
parameters.
The red curve is the variation due to the temperature (reference point should be 25°C), whereas the blue curve
represents the additional variation of dkILIS(cal) due to aging.
Looking at Figure 9 it is evident how the ±8% stated in the datasheet is significantly far from this characterization
curve. For that reason, although the parameter is given only as typical value, it provides a reliable approach to
achieve a good accuracy after two points calibration.
Application Note
16
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Current Sense
Additionally, the Figure 9 shows also that, narrowing the temperature range (e.g. from -20°C up to 85°C, as the
example of the previous chapter, described by the dotted window in the graph), the variation is further reduced
and better accuracy might be achieved.
106.0%
typ
influence of aging
105.0%
influence of the
temperature
-20°C --> 85°C temperature window
104.0%
dKILIS Variation [%]
103.0%
102.0%
101.0%
100.0%
99.0%
98.0%
97.0%
96.0%
-50
0
50
100
150
200
TJ [°C]
Figure 9
dkILIS variation over temperature and aging
Application Note
17
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Typical Application Aspects
6
Typical Application Aspects
Power PROFETTM is compliant with regard to ISO7637-1 and ISO7637-2. The device itself is passing the ISO
pulses at standard levels without problem or damage, both in ON- and OFF-mode.
Special attention must be taken to ISO pulse 1 in ON mode. Power PROFETTM tends to switch-OFF during ISO
pulse 1. Because Power PROFETTM is designed to switch-OFF in latch mode, measures must be implemented to
trigger the restart of the device. Such measure can be a capacitor between IN and GND which ensures a small
delay in voltage ramp at VS and IN after ISO pulse, resulting in an auto-restart of the device. Recommended
values for such capacitor are typically in the range from 150nF to 220nF (see also the application diagram in the
datasheet).
Furthermore, suppressor diodes are recommended, either between VS and GND or in parallel to the load. The
usage is depending on load configuration and application requirements.
6.1
Loss of VS
In every application it could happen, that the battery connection of a module is disconnected even if the application
is still running. If this occurs at systems using Power PROFETTM, special measures must be implemented to avoid
damages to the semiconductor due to high negative voltage at the output.
When disconnecting the supply voltage of a Power PROFETTM with charged inductive load, the output voltage will
become negative (in reference to GND potential). The datasheet specifies a minimum voltage at the output pin of
-64V (parameter 4.1.21 on page 10).
In case the output voltage exceeds the maximum ratings and is less than -64V, the technology limitation is
exceeded, resulting in a high current flowing into ground (Figure 10). It is depending on the load and temperature,
which magnitude and duration this ground current will be. One countermeasure against an output voltage
exceeding the maximum ratings is the usage of an external zenerdiode, a varistor or any VS clamping power
switch at the output (solution B in Figure 11) or between VS and GND (solution A in Figure 11). It is important to
highlight that only one of them is necessary, either A or B, not both.
Vbb
Z (AZ) GND
IN
IS
VS
7 5V
VIN
Lo gic/G atedriver
RIN
Z (AZ) IS
2* Z (E SD)
RVS
OUT
GND
RIS
Figure 10
Possible Internal and External Voltage Values for Loss of Supply without additional protection
measures
Application Note
18
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Typical Application Aspects
Vbb
IN
A
VS
7 5V
Z (AZ) GND
Z (AZ) IS
Logic/G atedriver
RIN
VIN
Z(E SD-L)
Z(ES D-H)
RVS
Either (A) or (B) is
recommended , not both
OUT
IS
GND
B
RIS
Figure 11
External Protection Components for Loss of Supply
6.2
Loss of GND
When the device GND is lost, the Power PROFETTM automatically switches off. This is not a latched switched off,
i.e. the device should switch on again as soon as the GND is connected again. Nevertheless, it is mandatory to
add a input capacitor between IN and GND to ensure a proper reset of the device (see also Chapter 7.1.2).
There are applications where a current-controlled switch might be required. For such cases, Power PROFETTM
could be used by switching the GND path. However, it must be considered that the device was not developed and
specified for switching via GND pin. Therefore deviations from the specification are possible.
6.3
Inverse Operation
Inverse Operation or Inverse Mode means, the Output voltage is higher than the voltage at the supply pin. Such
situation can happen for a transient time for certain application conditions, e.g. when a motor load acts as a
generator.
Power PROFETTM is using a feature called Inversave capability to limit the power dissipation in inverse operation.
Usually the current will flow through the body diode, leading to a high power dissipation as result of the inverse
current and the voltage drop over the body diode (typically 1V, depending on current and temperature). For Power
PROFETTM, the DMOS channel will be active even in inverse mode as long as the device is switched on (VIN >
VIN(H)). If the input is OFF (VIN < VIN(L)), the device can be switched on by the input voltage even if the output voltage
is higher than the supply voltage. As long as VIN is smaller than VIN(L), the inverse current will flow via body diode,
causing an increased power dissipation and increasing junction temperature. It has to be considered that under
inverse conditions the protection functions are not active and the device could be destroyed.
Application Note
19
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Typical Application Aspects
VOUT
(a) Inverse spike during ON -mode
for short times (< tp,INV ,noFAULT)
VOUT
VS
(c) Inverse spike during ON -mode with short
circuit after leaving Inverse mode
VOUT
VS
VS
t
t
IIS
(b) Inverse spike during ON -mode
for times > tp,INV ,noFAULT
t
> t p, INV ,noFAULT
< t p, INV ,noFAULT
IIS
tOFF (trip )
IIS
IIS (fault )
IIS (fault )
tsIS (ON)_J
t p, noINV, FAULT
tpIS (FAULT )
t
t
t
Internal Fault -flag set
Figure 12
Inverse Behavior - Timing Diagram
Parameter 4.2.7 in the datasheet functional range specifies a maximum slew rate at the output voltage. When the
voltage rises faster than specified, there is a risk that the DMOS will not be activated in inverse mode. This is due
to the limited speed of the comparator which monitors the voltage drop over the DMOS.
For VOUT > VS, the internal fault flag is set (Figure 12). If the duration of the inverse mode is shorter than tp,INV,noFault,
the output switches off after tOFF(TRIP) and with a certain delay time (tpIS(FAULT)), the sense signals the failure (case
(a) in Figure 12). If the inverse mode duration is longer than tp,INV,noFault, the internal fault flag is reset and no failure
will be observed at the sense output (case (b)). Nevertheless, if the device enters any failure mode (overload,
overtemperature) after leaving the inverse operation, the failure is visible at the current sense output (IIS(fault)).
Application Note
20
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Specific Application Requirements
7
Specific Application Requirements
7.1
Low Battery Voltage
The low battery voltage condition might be due to a weak discharged battery or to cranking during engine start.
The weak battery is more a permanent state, whereas cranking is a dynamic effect.
7.1.1
Requirements from Tier1 / OEM
Figure 13
LV124: Starting pulse requirement, UT_min = 3.2V for t4 = 19ms
7.1.2
Undervoltage Behavior
BTS50015-1TAA is specified for a minimum dynamic undervoltage capability of 3.2V. Regarding voltage
requirement, LV124 is fulfilled. However, it only can be guaranteed that the device will operate for max 6ms at a
battery voltage of 3.2V. For cranking duration >6ms, the device will operate as source follower.
It is recommended to use an input capacity to ensure a proper reset of the device in case of short negative
transients (see datasheet, chapter 8 “Application Information”). But not only for short negative spikes, also for
rising VIN and VS synchron, such capacity is required. Due to this capacity, the input voltage is slightly delayed to
VS, which allows a safe switch on.
The capacity (see also Chapter 7.5) range is in range from 150nF and 220nF. But those values are only valid for
VIN=5V and VS=13.5V. When using a setup with VIN and VS connected, the input capacitor value must be increased
and evaluated in the specific application.
7.1.3
Countermeasures
In order to ensure a proper operation at 3.2V or lower voltages for longer times (19ms as required in LV124 or
even longer), the supply voltage must be stabilized by using electrolytic capacitors between VS and GND. Those
capacitors must be in the μF range and are therefore expensive, i.e. an assessment is mandatory whether the
application absolutely requires the operation at very low supply voltages.
Application Note
21
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Specific Application Requirements
7.2
Load Jump
In some applications it may happen that the load current is changing during the ON-mode. The so called “Load
Jump” can be observed, for example, when Power PROFETTM is used as supply switch and the connected loads
are activated / de-activated independently.
V BAT
Semiconductor or
Electromechanical Switch
VS
Power
PROFETTM
RIN
uC
OUT
Load 1
Rs ens e
Load 2
RIS
GND
Cs ens e
Figure 14
Application Example with Power PROFETTM as Supply Switch
When changing from high- to low-ohmic load and back to high-ohmic load, when an inductive component is
present at primary side, an undesired fault condition could be notified at IS (Figure 15). This unwanted fault
notification is mainly due to huge spikes at VS and internal device capacities, which are bringing up the VOUT as
well.
Ch1 = VIN [V]
Ch2 = VIS [V] with RIS=1kΩ
Ch3 = VOUT [V]
Ch4 = IL (50A/Div)
TJ = 25°C
VS = 14.5V
2nd
1st jump
jump
Figure 15
Load Jump without External Measures
Some external components, which are anyway recommended for the typical application, might help to avoid the
unwanted IIS(FAULT) signal during load jump (Figure 16):
•
•
220nF capacitor between VS and GND
Zener diode between VS and GND
Both external components are needed anyway: the capacitor for better EMC robustness and the diode for loss of
battery protection. A standard zenerdiode is too weak and cannot withstand the energy under worst case
conditions. For evaluation, suppressor diode 1N6283A was used.
With those measures, the supply voltage is buffered and the failure was no longer observed, for the whole
temperature and supply voltage range(Figure 17). It is however recommended to verify under actual application
conditions.
Application Note
22
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Specific Application Requirements
V BAT
Semiconductor or
Electromechanical Switch
VS
uC
OUT
Power
PROFETTM
RIN
Load 1
Rs ens e
60V
RIS
GND
Load 2
220nF
Cs ens e
Figure 16
Recommended Application Schematic for Improved Load Jump Behavior
Ch1 = VIN [V]
Ch2 = VIS [V] with RIS=1kΩ
Ch3 = VOUT [V]
Ch4 = IL (50A/Div)
TJ = 25°C
VS = 18V
L = 10µH
Figure 17
Load Jump with External Measures
7.3
Parallel Operation of Power PROFETTM
The Power PROFETTM achieves a very low RDS(ON) value.
However, in some applications, the load current could be
higher than the specified nominal current for a single device.
For such cases, two or more Power PROFETTM can be
switched in parallel.
This allows to reduce the power dissipation in ON mode by
sharing the load current between two or more devices.
A proper current sharing is facilitated by the temperature
dependance of the resistance (in this case RDS(ON)).
It is important to remind that the system (the n devices in
parallel) will have the lowest ICL(0) among the thresholds of
the devices in parallel.
Figure 18
Application Note
23
2 x BTS50015-1TAA in parallel
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Specific Application Requirements
7.4
Power Considerations
The idea of maximum dissipated power is linked to different operation of the device (continuous or PWM), different
load to drive (resistive, inductive,etc.), and so on.
7.4.1
DC operation
Assuming that the Power PROFET is driving continuously (IN pin always high, no PWM operation) a resistive load,
the power dissipation is the sum of two contributions:
(16)
P DC, Tot = P DC, Power + P DC, Logic
These contributions can be defined as follow:
2
P DC, Power = R DS ( ON ) ⋅ I L
(17)
P DC, Logic = V S ⋅ I GND + ( V S – V IS ) ⋅ I IS + V IN ⋅ I IN ≅ V S ⋅ I GND + ( V S – V IS ) ⋅ I IS
(18)
The maximum power the device can dissipate is linked to the thermal resistance and the ambient temperature, as
shown by the following formula:
TJ – TA
P DC, Tot ≤ -----------------R thJA
(19)
The thermal resistance, for a defined package, is mainly dependant on the PCB features on which the device is
soldered. Assuming the device soldered on a board compliant with Jedec JESD51-2, -5, -7, with natural
convection, the thermal resistance junction to ambient can be considered as indicated in Table 4 of the datasheet,
that is RthJA(2s2p) = 20K/W.
The maximum junction temperature is considered 150°C. Therefore, in case the application has a maximum
ambient temperature of 85°C and under the assumption of using a PCB compliant with the above mentioned
standard, the maximum power the Power PROFET can dissipate is:
150 – 85 )K- = 3.25W
P DC, Tot ( max ) = (----------------------------K
20 ----W
Application Note
24
(20)
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Specific Application Requirements
7.4.2
PWM operation
Although the PWM operation is not a common requirement of most of the target applications of the Power
PROFET (lighting application are the most demanding from this point of view), the Power PROFET can be driven
with a relatively low frequency PWM signal.
Whenever a PWM signal is used to drive the Power PROFET, the choice of the PWM frequency must take into
account:
1. switching on and off time of the device
2. switching power losses
The choice of the frequency should be a compromise between the application requirements and the above
mentioned parameters.
With regard to the timing, it is evident that the period can´t be shorter than the time necessary to switch on and off
the DMOS and therefore the parameters from 6.1.17 to 6.1.20 of the datasheet should be considered.
The switching power losses basically occur during the switch on or switch off process, when both the voltage and
the current have not reached the steady state condition yet. The estimation of the switching losses is not as easy
as the calculation of the DC Power losses, described in previous chapter. However some assumptions on the
current-voltage transition may help to do some effective approximations. Very often, a linear or a piecewise linear
approximation are good way to describe most of the voltage-current switching responses (see Figure 19).
Power
Power
VDS
VDS
Load current
Figure 19
Load current
Approximation of the drain-source voltage and load current during switch off, with piecewise
linear approach (left) or simply linear (right)
It is worth, for a better description, introducing a reference variable, exclusively dependant on the load:
P matching
2
VS IL
VS
= ------- ⋅ ---- = ------------------------2 2
4 ⋅ R LOAD
(21)
The switching power losses can be calculated as:
Application Note
25
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Specific Application Requirements
E ON + E OFF
P SW = ---------------------------T PWM
(22)
where EON and EOFF are respectively the energies associated to the switch on and switch off phase, whereas the
TPWM is the PWM switching period (inverse of the PWM switching frequency). The switch on and off energies can
be calculated as follow:
E ON = P matching ⋅ k SW ⋅ t SW ( ON )
(23)
E OFF = P matching ⋅ k SW ⋅ t SW ( OFF )
(24)
where kSW is a correction factor (0.67 for linear approximation, 0.47 for linear piecewise) and tSW(ON) and tSW(OFF)
are the switch on and off time.
For a deeper analysis of the above shown calculations, please refer to ANPS061E.
The total power dissipation can be then estimated as:
(25)
P DISS = P DC, tot + P SW
where PDC,tot is, in this case:
2
P DC, tot = [ V S ⋅ I GND + ( V S – V IS ) ⋅ I IS + R DS ( ON ) ⋅ I L ] ⋅ DC
(26)
where DC is the duty cycle of the PWM signal.
7.4.3
Fault conditions during PWM operation
In case a fault condition occurs while the device is operating in PWM mode, it is important to consider the limitation
given by the parameter 4.1.9.
The meaning of this parameter is that, in case of overcurrent or overtemperature, the device should not be
switched on again before one second.
Application Note
26
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Specific Application Requirements
7.5
Capacitor at IN pin
In some application the VS and IN pins might be connected to the same voltage or it could happen that the IN pin
is kept high while the battery is disconnected or also that the battery voltage, for any reason, is temporary lower
than the voltage at the IN pin.
In these conditions, to assure a proper start up of the device, it is recommended to add an external capacitor at
the IN pin in order to slow down a little bit the voltage ramp at this pin (see Figure 20).
The value of this capacitor depends to some extent on the battery voltage (in the first case) and generally on the
ambient temperature (at cold temperature, a higher value might be required). However a general advice is to
connect a capacitor in the range between 150nF and 220nF.
V BAT
VS
IN
Power
PROFETTM
CIN
OUT
Load
GND
Figure 20
Recommended capacitor to connect to IN pin
Application Note
27
Rev. 1.0, 2015-01-20
Application Note
Power PROFETTM
Revision History
8
Revision History
Revision
Date
Changes
1.0
2015-01-20
Document release
Application Note
28
Rev. 1.0, 2015-01-20
Edition 2015-01-20
Published by
Infineon Technologies AG
81726 Munich, Germany
© 2015 Infineon Technologies AG
All Rights Reserved.
Legal Disclaimer
The information given in this document shall in no event be regarded as a guarantee of conditions or
characteristics. With respect to any examples or hints given herein, any typical values stated herein and/or any
information regarding the application of the device, Infineon Technologies hereby disclaims any and all warranties
and liabilities of any kind, including without limitation, warranties of non-infringement of intellectual property rights
of any third party.
Information
For further information on technology, delivery terms and conditions and prices, please contact the nearest
Infineon Technologies Office (www.infineon.com).
Warnings
Due to technical requirements, components may contain dangerous substances. For information on the types in
question, please contact the nearest Infineon Technologies Office.
Infineon Technologies components may be used in life-support devices or systems only with the express written
approval of Infineon Technologies, if a failure of such components can reasonably be expected to cause the failure
of that life-support device or system or to affect the safety or effectiveness of that device or system. Life support
devices or systems are intended to be implanted in the human body or to support and/or maintain and sustain
and/or protect human life. If they fail, it is reasonable to assume that the health of the user or other persons may
be endangered.
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising