I am using a 3 ohm resistor in series with H bridge driver to detect stall current. When it stalls the current shoots upto 200 mA causing a drop of 0.6 V. I am measuring voltage across the H bridge so, I will sense a voltage of VDD - 0.6, based on which I can decide that the motor is stalled and I can stop driving the motor. My question is will the voltage across the bridge will be very noisy. I am giving the voltage input to ADC of the micro controller directly. Is it manageable in software itself to filter out the noise?
\$\begingroup\$
\$\endgroup\$
11
-
\$\begingroup\$ You may want a smaller sense resistor. You can of course do filtering in software, but be aware that if the analog bandwidth exceeds the nyquist limit of your sampling rate, your resulting representation and any values deduced from it may be misleading in some cases where you may have high frequency components such as switching transients that could alias pathologically. Adding some capacitance may reduce the analog bandwidth to that which your sampling can represent. \$\endgroup\$– Chris StrattonCommented May 28, 2015 at 16:31
-
\$\begingroup\$ @chris I have added a capacitor between VDD and the ADC pin. Value is 100 nF. Why should i go with smaller sense resistor? I will need to detect minute change in voltage ad I go on reducing sense resistor value as voltage across H bridge will not dip more anymore \$\endgroup\$– User323693Commented May 28, 2015 at 16:35
-
\$\begingroup\$ Does Vdd stay constant when 200mA is taken or does Vdd also dip? \$\endgroup\$– Andy akaCommented May 28, 2015 at 16:39
-
\$\begingroup\$ @Andy, VDD doesn't dip(am unsure). It is from alkaline battery directly. Duracell AA, 2800 mAH, 1.5 V, two numbers \$\endgroup\$– User323693Commented May 28, 2015 at 16:42
-
\$\begingroup\$ How many AA batteries in series \$\endgroup\$– Andy akaCommented May 28, 2015 at 16:46
|
Show 6 more comments
2 Answers
\$\begingroup\$
\$\endgroup\$
4
Here is a high-side, differential, current monitor that can run from a single supply voltage as low as 2.5 volts: -
Max voltage supply is 3.6 volts so it's reasonably suited for a 3V system.
It has a gain of 25uA / mV and this means if you use a 10k resistor (as shown above) the voltage gain is 250 and the output is ground referenced. This means that if 0.6 volts is regarded as equating to 200mA drawn by the motor, the shunt resistor (Rsense) need only drop 2.4mV at 200mA i.e. it has a resistance of 12 milli ohms.
This is far better than using a 3 ohm resistor because your will extend battery life.
As usual ensure enough filtering is present to counteract spikes etc. upsetting the digitization process.
answered May 28, 2015 at 17:22
-
\$\begingroup\$ I will try searching for similar part. I have to operate down till 2.2 V, and also has to be cost sensitive. Any thing above 0.1$ for 100,000 is a no no. \$\endgroup\$ Commented May 29, 2015 at 8:31
-
\$\begingroup\$ Would like to take back my comment on price. It can be upto 0.25$, battery life far more important than price. I will search on same lines \$\endgroup\$ Commented May 29, 2015 at 8:44
-
\$\begingroup\$ @Umar did you find a solution to this or did you go with low-side current sensing? \$\endgroup\$– Andy akaCommented Dec 22, 2016 at 18:33
-
\$\begingroup\$ As of the design was concerned, they needed low cost, low size solution. The precision they used to get by measuring voltage across bridge was sufficient. But, for the future design, they are going with costlier but very low power solution, for which I would study this solution again and consider. \$\endgroup\$ Commented Dec 23, 2016 at 1:13
\$\begingroup\$
\$\endgroup\$
2
Would be easier to measure on a low side current sense resistor. Like 0.1R and a simplest opamp amplifier. Most importantly, synchronize your ADC to the switching PWM and sample a before the switching occurs, that's a silent way of current sensing.
-
\$\begingroup\$ That's a good point I can take as an input for my software \$\endgroup\$ Commented May 29, 2015 at 8:26
-
\$\begingroup\$ There is another good point, but it's a bit tricky to explain. Please, google "advanced unipolar PWM". \$\endgroup\$– user76844Commented May 29, 2015 at 8:39