Will skin effect issues make the 1 milliOhm resistor seem much more than 1 milliOhm?

[treez]

I have an 11 Amp RMS current flowing in a one milliOhm sense resistor (1%).
(its a 1210 size chip resistor)

...this current is amplified by a differential amplifier with a gain of 100, and there are filter capacitors incorporated in the differential amplifier (using OPA335 op amp which has low Vos) so that the amplified output voltage is DC.....(and then this DC output of the opamp is read by an ADC on a microcontroller)

......The 11 Amps current waveshape is of the form of a full wave rectified 50Khz sinusoidal current waveform....so it has a frequency of 100KHz.

.....Do you think skin effect issues will be at large here?......will the 1 MilliOhm resistor act like its a lot more than 1 MilliOhm?

 

[crutschow]

In a film resistor, the current is carried in a thin film of resistive material, so I would expect the skin effect to be quite small at that frequency. For more info see page 6 of this reference.

 

[FvM]

Think! The measured DC current isn't affected by skin effect at all.

Skin effect only affects the AC resistance. I agree with crutschow, that skin effect won't be very strong for film resistors. Some milliohm resistors are however made of thick (e.g. 0.5 to 1 mm) metal bars, and they will show considerable skin effect at 50 or 100 kHz. It would be a problem for precise AC current measurements, not the present problem.

Another point to consider for AC measurements is resistor inductance. For film resistors, it's impact is much more important than skin effect.

 

[treez]

The following shows the schematic and the current waveform in the 1 milliohm sense resistor


Schematic
https://i46.tinypic.com/29n9qh1.jpg

100 KHz current waveform
https://i46.tinypic.com/2vj74gj.jpg



.....do you think ill need a special low inductance 1 milliohm resistor?

i only have to know if the average current flowing is 1A, 2A, 3A, .....10A (10A average corresponds to 11A RMS)

 

[mtwieg]

If you're only interested in the DC component of the current, then parasitic inductance and skin effect won't have an effect on that measurement.

 

[crutschow]

.....do you think ill need a special low inductance 1 milliohm resistor?

i only have to know if the average current flowing is 1A, 2A, 3A, .....10A (10A average corresponds to 11A RMS)

If you only have to differentiate between values that are 1A apart (10% worst-case accuracy) then the parasitic effects of the resistor should not be a significant factor.

 

[FvM]

Fortunately, the 100 µs low-pass will filter the 100 kHz ripple sufficiently. Otherwise the rectifier circuit (which is useless, as I think) would possibly adulterate the DC current measurement. In this case, skin effect and inductance further detoriate the result.

 

[treez]

Sorry, i forgot to say that the opamp rectifier circuit is necessary to act as peak detect and hold for when there is PWM dimming going on.

 

[treez]

I only want to amplify the average value of the waveform in post #4.......will the opamp be able to do this accurately ?


...its just that i worry about the effect of GBW product..................

OPA335 GBW = 2MHz so with a gain of 100, i'll have about 20kHz bandwidth, the amplifier isn't fast enough to even see the waveform.........................is this a problem?

....i thought the capacitors C1 and C2 in the schem in #4 post would filter the current sense resistor waveform, and result in the average vaue of the current sense resistor voltage being accurately amplified?.......................am i wrong in this?

- - - Updated - - -

I only want to amplify the average value of the waveform in post #4.......will the opamp be able to do this accurately ?


...its just that i worry about the effect of GBW product..................

OPA335 GBW = 2MHz so with a gain of 100, i'll have about 20kHz bandwidth, the amplifier isn't fast enough to even see the waveform.........................is this a problem?

....i thought the capacitors C1 and C2 in the schem in #4 post would filter the current sense resistor waveform, and result in the average vaue of the current sense resistor voltage being accurately amplified?.......................am i wrong in this?