Rds

[kell]

I need a blocking diode and charging current sensing into the positive
terminal of a twelve volt battery. It occured to me that I could use a
mosfet as the blocking diode and bias it on during charging, with the
Rds serving as sense resistor... and the side benefit of eliminating a
wasteful power diode.
How accurate is the Rds listed in mosfet datasheets? They all seem to
have a number with one or two sig figs and no tolerance spec.
And important is the Rds temperature coefficient; supposed positive but
I have no idea of the magnitude.
Max charging current is ten amps in the application. If it works, I
may use the same circuit on a 35 amp charger.

 

[Jim Thompson]

I need a blocking diode and charging current sensing into the positive
terminal of a twelve volt battery. It occured to me that I could use a
mosfet as the blocking diode and bias it on during charging, with the
Rds serving as sense resistor... and the side benefit of eliminating a
wasteful power diode.
How accurate is the Rds listed in mosfet datasheets? They all seem to
have a number with one or two sig figs and no tolerance spec.
And important is the Rds temperature coefficient; supposed positive but
I have no idea of the magnitude.
Max charging current is ten amps in the application. If it works, I
may use the same circuit on a 35 amp charger.

See...

http://www.analog-innovations.com/SED/PerfectDiodeForChargerIsolation.pdf

But I don't think RDS(ON) is going to be a very reliable value.

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice480)460-2350 | |
| E-mail Address at Website Fax480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |

http://www.analog-innovations.com/SED/DefeatHillary.gif

 

[notme]

I need a blocking diode and charging current sensing into the positive
terminal of a twelve volt battery. It occured to me that I could use a
mosfet as the blocking diode and bias it on during charging, with the
Rds serving as sense resistor... and the side benefit of eliminating a
wasteful power diode.
How accurate is the Rds listed in mosfet datasheets? They all seem to
have a number with one or two sig figs and no tolerance spec.
And important is the Rds temperature coefficient; supposed positive but
I have no idea of the magnitude.
Max charging current is ten amps in the application. If it works, I
may use the same circuit on a 35 amp charger.

Rds is a function of a variety of things including process variations,
current and temperature. There is not a single temperature coefficient on
a FET but many which could result in significant variations. I can't see
why you'd want to use one over a resistor thats explicitly designed to do
a good job at this application.

I'm also not clear how you would use a MOSFET as a "blocking diode". If
you are using a diode connected MOSFET with the gate and drain tied
together you are still taking a Vgs drop in the FET which may be more than
a Schottky diode drop.

 

[Noway2]

My project has the same requirement, monitoring the battery charger
current and isolating it from the charger to verify that the battery is
still present. I did use a mosfet to isolate the battery from the
charger and I chose one with a small Rds to limit power disipation as
the charger current can be as high as 10A.

To monitor the current, I used a four terminal resistor of .005R. The
one I used costs a couple of dollars as I chose a high precision
device. The net result is that I can read the current with my 12bit
ADC (0 to 10A) both positive and negative direction with better than
..1Amp accuracy.

I agree that Rds is too variable of a parameter if you wish a
measurement with accuracy. If you are looking for a crude measuremetn
of wheter the charger is charging, it may be sufficient for your needs.

I would also recommend a snubber circuit across the mosfet!.

If you are interested in implementation details, contact me at the
email address listed in the post header and we can discuss.

 

[kell]

Rds is a function of a variety of things including process variations,
current and temperature. There is not a single temperature coefficient on
a FET but many which could result in significant variations. I can't see
why you'd want to use one over a resistor thats explicitly designed to do
a good job at this application.

I'm also not clear how you would use a MOSFET as a "blocking diode". If
you are using a diode connected MOSFET with the gate and drain tied
together you are still taking a Vgs drop in the FET which may be more than
a Schottky diode drop.

I'd use a p-channel mosfet with the source connected to the positive
terminal of the battery. With the mosfet turned off, it would act like
a diode, blocking the battery from discharging. Connecting the gate to
ground (battery negative) during charging would turn the mosfet fully
on and reduce the power dissipated in the mosfet, assuming
I(charge) x Rds
is less than
Vf of the body diode
The mosfet would be conducting in the opposite direction of usual.