small resistor values for current sensing

[helix1]

I was just wondering why resistor values of 0.1Ohm or even 0.01Ohm might be used in current sensing applications rather than 1Ohm - is there a particular reason one might want to use such small resistances?

[Kalvin]

Let's assume that you have a system which requires power supply voltage of 5V and it can draw current up to 1A. Now, you place the current sensing resistor in series with the power supply and the system, you can determine the voltage drop across the sensing resistor at the given current using the Ohm's law, and what the system supply voltage will be at different currents. Try different resistor values 0.01 ohm, 0.1 ohm, 1 ohm and 10 ohm with the current values 0.01A, 0.1A and 1A. You should find the answer there.

[Neilm]

Let's assume that you have a system which requires power supply voltage of 5V and it can draw current up to 1A. Now, you place the current sensing resistor in series with the power supply and the system, you can determine the voltage drop across the sensing resistor at the given current using the Ohm's law, and what the system supply voltage will be at different currents. Try different resistor values 0.01 ohm, 0.1 ohm, 1 ohm and 10 ohm with the current values 0.01A, 0.1A and 1A. You should find the answer there.

Don't forget to calculate how much power is wasted in those scenarios. It becomes very significant if running on a battery a power supply.

I have designed battery powered  DC-DC switch mode power supplies that had over 40A peak current in the transformer, I definitely didn't want to use a 1ohm in that situation.

[Simon]

It's good practice to make the resistor as small as possible so that the voltage drop is as low as possible so not to disrupt the circuit. This is the whole reason Dave designed the uCurrent. The same principle can be applied to higher currents and is as valuable as it reduces the power dissipated.

[German_EE]

I recently completed a power supply project where the current sense resistor was 0,015 ohms. Why? With a 50 uA meter connected across the shunt full scale was the maximum current out for the power supply. Building the shunt was easy as I have a bag of 0,15 ohm 5W resistors but even with this low value I drop 0,3V at full current.

[Simon]

ebay is full of sub mOhm resistors with panel meters, I bought one to measure the power coming out of two paralleled power supplies.

mine is 750 uOhm and 375 uOhm

[German_EE]

One other thing, meters are expensive so it's a good idea to wire back to back diodes across the meter terminals and include either a current limiting resistor or a 100mA fuse in one lead. That way if your shunt ever goes open circuit (been there, done that) your meter will not become a light bulb.

[Simon]

hm for 100A ?

[SeanB]

2 1N4001 diodes and a 100mA fuse will work for the100A shunt. It will introduce a slight error in the current measurement, but this likely will be under the width of the needle in nost meter movements, as the resistance of the fuse will be in the 2R range, while the meter will be around 500R. The diodes are not going to be a cause of any error, they are only going to start having an effect at around 300mV drop across the resistor, where most movements will be hard against an end stop in any case, 75mV full scale.

[jolshefsky]

For what it's worth, if you're looking for a cheap way to get a low-value resistance with good current handling capability, try a piece of wire. 24ga. copper wire is about 0.025 ohms/ft; 0.5mm dia. copper wire is about 0.068 ohms/m.

[Kleinstein]

With a resonable low voltage drop at the shunt (e.g. < 200 mV for full scale) one can have diodes in parallel zu protect the shunt and if needed the meter circuit that measures the voltage at the shunt. This way a slow blow fuse can be used for protection. The fuse just has to blow before the diodes.

Especially if AC with higher peaks is includes one might need 2 or even 3 diode in series to allow for a higher voltage.

The power loss in the shunt is an incentive to use a rather low value shunt, especially at higher currents. To get a precise reading the shunt should not get hot. So one has to stay way below (e.g. 1/10) the rated power of resistors, which may be at something like 200 C inside the resistor.

Copper wire is not a good resistor material with a voltage meter. It may work with an uncompensated analog meter as this will have a similar temperature coefficient.

[Kevin.D]

I was just wondering why resistor values of 0.1Ohm or even 0.01Ohm might be used in current sensing applications rather than 1Ohm - is there a particular reason one might want to use such small resistances?

Hi
You choose a resistor value appropriate for the range of current values that you are measuring.
Larger values of Resistor give increased sensitivity per unit of current, but weighing against that is that Larger Rsense values mean increased power dissipation  which means a greater temperature variations in Rsense (so larger temp coeff errors) so larger Sense Resistor required. Power dissipation is much less of an issue when measuring lower current's ( because PD= I^2 R) so required Sensitivity is the dominant factor to consider here. Also 'Burden voltage'in current measuring devices may be another consideration which dictates max Rsense value. Burden Voltage (alternatively 'burden resistance' )is the voltage drop per unit of current through it, which is something you don't particularly want in an good ammeter.

[Zero999]

2 1N4001 diodes and a 100mA fuse will work for the100A shunt. It will introduce a slight error in the current measurement, but this likely will be under the width of the needle in nost meter movements, as the resistance of the fuse will be in the 2R range, while the meter will be around 500R. The diodes are not going to be a cause of any error, they are only going to start having an effect at around 300mV drop across the resistor, where most movements will be hard against an end stop in any case, 75mV full scale.

Be careful when choosing a fuse. Ensure it has adequate breaking capacity, especially if it's DC, otherwise it will become an arc welder.

[dadler]

This seller regularly lists NOS current shunt resistors. I picked up about 15 different values, all for much, much cheaper than list price.

http://stores.ebay.com/HFO-Elektronik/Resistors-Potentiometers-/_i.html?_fsub=510193119

For example, I scored 100 160mOhm 2W resistors for $3 USD:

http://www.ebay.com/itm/262125889282

[macboy]

For what it's worth, if you're looking for a cheap way to get a low-value resistance with good current handling capability, try a piece of wire. 24ga. copper wire is about 0.025 ohms/ft; 0.5mm dia. copper wire is about 0.068 ohms/m.

Copper has a relatively very high temperature coefficient so as your shunt heats up under high current, the value changes significantly.