Input impedance of a (generic) IC? LM 339?

[conducteur]

Hello,
This isn't mentioned in the datasheet i found of the LM339 quad comparator, but what is a typical input impedance of an input pin on such a device? To what extent will this input impedance affect for example a simple RC first order low pass filter with it's output connected to the + or - input? If the resistor of the RC lowpass filter is to high, I assume it will become more or less a voltage divider, rather than a lowpass filter?

[Alti]

The inputs are current sources.
The input bias current is specified to be around 20nA typical, offset current +-3nA. So the input cannot be replaced by a voltage divider, in general.

[T3sl4co1l]

Right, the impedance is high.  It's not infinite, there is a slope to the current.  Like maybe, Idunno, 1GΩ or so?  It goes down substantially for differential voltages near zero (where the current shifts from one input to the other; ~MΩ?), and near the supply rails where either the current drops towards zero (or still further out, reverses), or the substrate diode turns on.

Tim

[magic]

(Dynamic) input resistance is sometimes specified in opamp datasheets. LM358 (similar to LM393) is missing it, but typical numbers for bipolar inputs are hundreds of kΩ to single digit MΩ.

edit
T3sl4co1l is right, it's not going to be that simple for a comparator. But we have the schematic: inputs are PNP emitter follower with active loads, driving a hidden differential pair inside. Any variation in input current will be due to Early effect in the follower and its load. Off the top of my head, I'm not sure how much to expect. It probably depends on the (unknown) standing current of this stage too? Good news is that it should be largely independent of what the other input pin is doing.

[duak]

Page 8 of the LM358 data sheet from TI shows Zin: http://www.ti.com/lit/ds/symlink/lm358.pdf

Both  the '339 & '358 have Darlington PNP input circuits with similar bias currents so I wouldn't be surprised if they had similar Zin, at least in the linear range.

I had remembered values of Zin for emitter followers in the range of 100K to 1M from h-parameter analysis in college (before SPICE).  The values in the data sheet make sense give the Darlington inputs.

Back to conducteur's question:  The comparator's Zin will be in parallel with the RC network's capacitor.  This will cause the signal at that point to be attenuated to approximately  Zin/(R +Zin) in addition to the frequency selective attenuation of the RC network.  The corner frequency (fc) of the RC network will be increased in frequency because the Thevenin equivalent resistance at that point will be equal to R in parallel with Zin.

for example lets say R = 100K and Zin = 1M0.  The signal applied to the comparator input will be approximately 90% of what it is without the comparator loading.  The corner frequency of the filter will be increased by about 10%.

[Zero999]

Page 8 of the LM358 data sheet from TI shows Zin: http://www.ti.com/lit/ds/symlink/lm358.pdf

Both  the '339 & '358 have Darlington PNP input circuits

No they don't have Darlington input stages. They have a differential pair, with an emitter follower on each input, which both reduces the bias current, as well as enabling them to work down to the negative rail.

[magic]

Page 8 of the LM358 data sheet from TI shows Zin: http://www.ti.com/lit/ds/symlink/lm358.pdf

Page 8 is for LM358B, "the next generation version of the industry standard LM358". I don't know what it is inside but it might be meaningfully different.

[duak]

Points taken.  I jumped to a conclusion and agree the circuit does not use Darlingtons.

About the actual Zin, I said I wouldn't be surprised if they had similar Zin.  However, I used the value of 1M0 in my example.  It may be more or less but I think it's on the order of that.  Perhaps some buckaroo could determine what it actually is?

[Zero999]

Points taken.  I jumped to a conclusion and agree the circuit does not use Darlingtons.

About the actual Zin, I said I wouldn't be surprised if they had similar Zin.  However, I used the value of 1M0 in my example.  It may be more or less but I think it's on the order of that.  Perhaps some buckaroo could determine what it actually is?

It seems to be an easy mistake to make: you're not the first person I've corrected. I suppose when one sees the emitter connected to the base of the other transistor it's all to easy to think Darlington and overlook the collectors not being connected. If it was a Darlington, it probably won't quite work down to the negative rail, since Q2's collector could go no lower than Q5's V_BE.

As far as the bias current/input impedance is concerned. It will be non-linear, as the bias current will change depending on which input is higher. The input at the higher voltage will turn the corresponding transistor in the differential pair off, thus reducing the emitter current along with the base current of the transistor on the input.

You're right that the LM358 has a similar input stage.

[duak]

In spite of more important things to do, I spent some time on measuring Zin.  My high school physics teacher, Mr. Peterson would say "how do we know that g is 9.8 m/sec^2? We measure it!"

To do so, I set up a Fairchild uA339PC on a breadboard.  I biased pin 6 (-input) to half of VCC (5 V) with a voltage divider of two 10K resistors.  Relative to that, I applied a 1 kHz sine wave of various amplitudes to pin 7 (+input) through a 1M0 resistor.  I used a Tek TX3 4.5 digit DMM to monitor the voltage after after the resistor both unloaded and loaded.  I then determined the value of Zin with simple algebra.  Note, the TX3 has an input impedance of 10 Mohm in parallel with approximately 100 pF.  I neglected the capacitance and did not treat the meter as a complex impedance in the calculations.  I've attached a graph of the results.

Based on one sample, it appears that the LM339 has a differential input impedance in the 10's of megohms for inputs between 25 mV and 2 V RMS.  Above 2 V, I would expect to see a reduced impedance as the LM339 inputs are driven beyond their specified range.  The reduction of impedance at levels of 25 and 50 mV is probably due to noise.  Even though the breadboard and meter are on a metal panel connected to the circuit common, there is noise induced, most likely from the function generator and power supply that are in a stack and not as close as I would like.  If there's enough interest, I could revisit the setup and properly isolate the drive and also isolate the meter loading from the circuit.

I also tested the bias current as a function of common mode and differential voltage.  The TX3 has a resolution of 0.01 uA or 10 nA on its most sensitive range. This makes reading 50 to 100 nA somewhat problematic.  However, in spite of a 20 - 30 nA bobble in the reading,  I did not see a material change in the +input bias current as I varied both inputs from about 0 to 3 V.  It dropped off above that.  I also didn't see a material change when setting the -input to 2.00 V and varying the +input from 0 to 3.5 V.

This makes sense given the design of the input stage.  Refering to the simplified schematic posted above, Q1 & Q4 are PNP emitter followers with their emitters biased to V+ with separate current sources.  Neglecting the diff amp Q2 & Q3 for the time being, the base currents of Q1 & Q4 must be 3.5 uA dvided by their respective HFEs.  This applies for base voltages greater than when their base-collector junctions are forward biased and less than when there is insufficient voltage across the current source for it to operate.

Bringing the diff amp Q2 & Q3 back in, the common emitters are biased to V+ with a 100uA current source.  Depending on the relationship between the input voltages, the base currents of Q2 & Q3 can vary from 0 to 100 uA / HFE.  If the HFE is 100, the bias current to the emitter followers can vary by 1 uA from 3.5 to 4.5 uA.  This will manifest as a variation in the input bias currents but I can't see it with my setup.   It would be nice to have a more sensitive meter.  Anyone got a VNA that will work at low frequencies?

Anyway, the bottom line is that as long as the inputs are operated within their limits, the bias current is fairly constant and the input impedance is quite high for a bipolar input device.  The emitter followers are doing a good job of buffering the input pins from the differential amplifier.  Also, looking at the graph, Zin actually seems to increase as the differential input voltage increase.  I suspect this is due to the differential amplifier transistors  alternately being biased off and not contributing base current to the emitter follower stages.

[magic]

Bringing the diff amp Q2 & Q3 back in, the common emitters are biased to V+ with a 100uA current source.  Depending on the relationship between the input voltages, the base currents of Q2 & Q3 can vary from 0 to 100 uA / HFE.  If the HFE is 100, the bias current to the emitter followers can vary by 1 uA from 3.5 to 4.5 uA.  This will manifest as a variation in the input bias currents but I can't see it with my setup.

You see it as the dip in impedance near zero amplitude, when both inputs are at 2.5V and the diffpair transitions from one branch to another and base currents of Q2 and Q3 vary significantly with input voltage. At this point the comparator behaves essentially like an opamp and its input characteristics should be similar.

And yes, I was wrong before. I didn't realize that emitter current is so much higher in Q2,Q3 than in Q1,Q4.

Good news is that it should be largely independent of what the other input pin is doing.