Calc resistors for current sense?

D

David Knaack

Guest
Greetings,

I want to use a MAX472 to measure the current flow to and from an car
battery, excluding the starter motor. I don't think I'll see current
peaks of more than 120A, and probably rarely over 50A. About 12-15vdc.

The MAX471/472 datasheet gives some typical configurations
(http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1108/ln/en link to
datasheet near top right corner), but only up to 10A.

I'm not confident that I understand how to calculate the values of the
three resistors (Current-sense, gain and output). I think that I'll
need about 0.5 milliohm for the current sense, but I don't really
understand the other ones.

I wonder if someone could point me in the right direction here. I'd
like to know how to calculate the right values since I'll probably need
to adjust them later (after I have a better idea of what the actual
current will be).

Thanks
DK
 
David Knaack wrote:
Greetings,

I want to use a MAX472 to measure the current flow to and from an car
battery, excluding the starter motor. I don't think I'll see current
peaks of more than 120A, and probably rarely over 50A. About 12-15vdc.

The MAX471/472 datasheet gives some typical configurations
(http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1108/ln/en link to
datasheet near top right corner), but only up to 10A.

I'm not confident that I understand how to calculate the values of the
three resistors (Current-sense, gain and output). I think that I'll
need about 0.5 milliohm for the current sense, but I don't really
understand the other ones.

I wonder if someone could point me in the right direction here. I'd
like to know how to calculate the right values since I'll probably need
to adjust them later (after I have a better idea of what the actual
current will be).

Thanks
DK
Click to expand...
How much voltage can you afford to lose at peak amps (and how much
power are you willing to waste)? This sets the upper limit for the
sense resistor. Remember that the lower its resistance the lower the
voltage loss and power waste, but the lower the precision of the
measurement.

Once you have selected R sense, you pick the combination of gain and
output resistors to amplify the small voltage drop across R sense.
The total gain is the ratio of these resistors, but the errors and
power consumption depend on their values. You can make the gain
different for currents in opposite directions by choosing two
different values for RG1 and RG2, since only one is involved at a time
(one for each current direction). The chip forces one of these
resistors to have the same voltage drop as the sense resistor. It
also forces both that gain resistor and the output resistor to share
the same current. So if the gain resistor and output resistor are
equal, the net gain of the current sense voltage is 1. If the output
resistor is 10 times as large as the gain resistor, the net gain is
10, etc. The upper limits for the gain resistor involve the voltage
caused by the 2.5ua offset current of the amplifier. The lower limit
involves the 1.5ma Iout limit. The output voltage must always stay at
least 1.5 volts lower than the voltage at the amplifier end of the
gain resistor.

Table 1 on page 9 suggests gain resistors between 200 and 50 ohms. I
would probably try to stay in that range. The output resistor can be
any value that does not drop more than about 8 or 9 volts when it is
carrying a current that will cause a voltage drop across the gain
resistor that has a drop that matches that of the sense resistor.

--
John Popelish
 
To measure that much current, you should seriously look at Hall sensing the
magnetic field generated by the current flowing in the wire.

That is lossless.


"John Popelish" <jpopelish@rica.net> wrote in message
news:4096EB2A.910A35AB@rica.net...
David Knaack wrote:

Greetings,

I want to use a MAX472 to measure the current flow to and from an car
battery, excluding the starter motor. I don't think I'll see current
peaks of more than 120A, and probably rarely over 50A. About 12-15vdc.

The MAX471/472 datasheet gives some typical configurations
(http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1108/ln/en link to
datasheet near top right corner), but only up to 10A.

I'm not confident that I understand how to calculate the values of the
three resistors (Current-sense, gain and output). I think that I'll
need about 0.5 milliohm for the current sense, but I don't really
understand the other ones.

I wonder if someone could point me in the right direction here. I'd
like to know how to calculate the right values since I'll probably need
to adjust them later (after I have a better idea of what the actual
current will be).

Thanks
DK

How much voltage can you afford to lose at peak amps (and how much
power are you willing to waste)? This sets the upper limit for the
sense resistor. Remember that the lower its resistance the lower the
voltage loss and power waste, but the lower the precision of the
measurement.

Once you have selected R sense, you pick the combination of gain and
output resistors to amplify the small voltage drop across R sense.
The total gain is the ratio of these resistors, but the errors and
power consumption depend on their values. You can make the gain
different for currents in opposite directions by choosing two
different values for RG1 and RG2, since only one is involved at a time
(one for each current direction). The chip forces one of these
resistors to have the same voltage drop as the sense resistor. It
also forces both that gain resistor and the output resistor to share
the same current. So if the gain resistor and output resistor are
equal, the net gain of the current sense voltage is 1. If the output
resistor is 10 times as large as the gain resistor, the net gain is
10, etc. The upper limits for the gain resistor involve the voltage
caused by the 2.5ua offset current of the amplifier. The lower limit
involves the 1.5ma Iout limit. The output voltage must always stay at
least 1.5 volts lower than the voltage at the amplifier end of the
gain resistor.

Table 1 on page 9 suggests gain resistors between 200 and 50 ohms. I
would probably try to stay in that range. The output resistor can be
any value that does not drop more than about 8 or 9 volts when it is
carrying a current that will cause a voltage drop across the gain
resistor that has a drop that matches that of the sense resistor.

--
John Popelish
Click to expand...
 
In theory, you could use the battery lead itself as the sense resistor,
albeit it'd need to be calibrated.

Good Luck!
Rich


"mook johnson" <mook@mook.net> wrote in message
news:lrClc.61155$Dn1.38046@fe2.texas.rr.com...
To measure that much current, you should seriously look at Hall sensing
the
magnetic field generated by the current flowing in the wire.

That is lossless.


"John Popelish" <jpopelish@rica.net> wrote in message
news:4096EB2A.910A35AB@rica.net...
David Knaack wrote:

Greetings,

I want to use a MAX472 to measure the current flow to and from an car
battery, excluding the starter motor. I don't think I'll see current
peaks of more than 120A, and probably rarely over 50A. About
12-15vdc.

The MAX471/472 datasheet gives some typical configurations
(http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1108/ln/en link to
datasheet near top right corner), but only up to 10A.

I'm not confident that I understand how to calculate the values of the
three resistors (Current-sense, gain and output). I think that I'll
need about 0.5 milliohm for the current sense, but I don't really
understand the other ones.

I wonder if someone could point me in the right direction here. I'd
like to know how to calculate the right values since I'll probably
need
to adjust them later (after I have a better idea of what the actual
current will be).

Thanks
DK

How much voltage can you afford to lose at peak amps (and how much
power are you willing to waste)? This sets the upper limit for the
sense resistor. Remember that the lower its resistance the lower the
voltage loss and power waste, but the lower the precision of the
measurement.

Once you have selected R sense, you pick the combination of gain and
output resistors to amplify the small voltage drop across R sense.
The total gain is the ratio of these resistors, but the errors and
power consumption depend on their values. You can make the gain
different for currents in opposite directions by choosing two
different values for RG1 and RG2, since only one is involved at a time
(one for each current direction). The chip forces one of these
resistors to have the same voltage drop as the sense resistor. It
also forces both that gain resistor and the output resistor to share
the same current. So if the gain resistor and output resistor are
equal, the net gain of the current sense voltage is 1. If the output
resistor is 10 times as large as the gain resistor, the net gain is
10, etc. The upper limits for the gain resistor involve the voltage
caused by the 2.5ua offset current of the amplifier. The lower limit
involves the 1.5ma Iout limit. The output voltage must always stay at
least 1.5 volts lower than the voltage at the amplifier end of the
gain resistor.

Table 1 on page 9 suggests gain resistors between 200 and 50 ohms. I
would probably try to stay in that range. The output resistor can be
any value that does not drop more than about 8 or 9 volts when it is
carrying a current that will cause a voltage drop across the gain
resistor that has a drop that matches that of the sense resistor.

--
John Popelish
Click to expand...
 
Rich Grise wrote:

In theory, you could use the battery lead itself as the sense resistor,
albeit it'd need to be calibrated.

Good Luck!
Rich


"mook johnson" <mook@mook.net> wrote in message
news:lrClc.61155$Dn1.38046@fe2.texas.rr.com...

To measure that much current, you should seriously look at Hall sensing

the

magnetic field generated by the current flowing in the wire.

That is lossless.


"John Popelish" <jpopelish@rica.net> wrote in message
news:4096EB2A.910A35AB@rica.net...

David Knaack wrote:

Greetings,

I want to use a MAX472 to measure the current flow to and from an car
battery, excluding the starter motor. I don't think I'll see current
peaks of more than 120A, and probably rarely over 50A. About

12-15vdc.

The MAX471/472 datasheet gives some typical configurations
(http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1108/ln/en link to
datasheet near top right corner), but only up to 10A.

I'm not confident that I understand how to calculate the values of the
three resistors (Current-sense, gain and output). I think that I'll
need about 0.5 milliohm for the current sense, but I don't really
understand the other ones.

I wonder if someone could point me in the right direction here. I'd
like to know how to calculate the right values since I'll probably

need

to adjust them later (after I have a better idea of what the actual
current will be).

Thanks
DK

How much voltage can you afford to lose at peak amps (and how much
power are you willing to waste)? This sets the upper limit for the
sense resistor. Remember that the lower its resistance the lower the
voltage loss and power waste, but the lower the precision of the
measurement.

Once you have selected R sense, you pick the combination of gain and
output resistors to amplify the small voltage drop across R sense.
The total gain is the ratio of these resistors, but the errors and
power consumption depend on their values. You can make the gain
different for currents in opposite directions by choosing two
different values for RG1 and RG2, since only one is involved at a time
(one for each current direction). The chip forces one of these
resistors to have the same voltage drop as the sense resistor. It
also forces both that gain resistor and the output resistor to share
the same current. So if the gain resistor and output resistor are
equal, the net gain of the current sense voltage is 1. If the output
resistor is 10 times as large as the gain resistor, the net gain is
10, etc. The upper limits for the gain resistor involve the voltage
caused by the 2.5ua offset current of the amplifier. The lower limit
involves the 1.5ma Iout limit. The output voltage must always stay at
least 1.5 volts lower than the voltage at the amplifier end of the
gain resistor.

Table 1 on page 9 suggests gain resistors between 200 and 50 ohms. I
would probably try to stay in that range. The output resistor can be
any value that does not drop more than about 8 or 9 volts when it is
carrying a current that will cause a voltage drop across the gain
resistor that has a drop that matches that of the sense resistor.

--
John Popelish
Click to expand...
Automotive ammeters don't need much precision, and a length of wire is
not too bad in the accuracy department. If you're not going to go to
the trouble of temperature compensating over the -40 to +85C temperature
range it's going to see why bother calibrating it at room temperature?

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top