Op-Amp Input Current

Can you give us a little schematics? what is w.r.t point? is it ac or dc volt? share with us, so we will discuss about it.

The dig is as shown I want to read vtg across 540 ohm which is abt 440mVolt .Ip is 240Volt.

I can not see clearly the value of resistors, if you use R1 ( v+ op-amp ), R2 ( v- op-amp ), R3 ( v+ op-amp to ground ) and R4 ( feedback ) same value (100 k ohm) the output is hard to measure because it should be 1mv, try to change R3, R4 with 2,2 M ohm, it should be 22mv output, then try to change R1, R2 with lower value or R3, R4 with higher value until you get the output that you need.

Note : R1 should be same value with R2, R3 same value with R4.

Yu got my ckt.

I am using differential amplifier with all resistors as 100k ,I will try with above solution.I can't read it with multimeter or oscilloscope because it is apart of power supply in my design & i want to check the voltage at that point ,for the purpose I will give o/p of OPamp to ADC i/p of microcontroller & after reading it I will turn on the LED by microcontroller to say that if the test is OK or fail.(Or voltage is within the limit across resistor).

Yes the 240V GND & OPAMP GND is common .Will it create problem?

I think isolation is needed.

What is the op amp type & supply voltage? It is not legible on your circuit.

the opamp is LM358 & supply is 5Volts

Rupaliprayag, i know it's impossible to make differential op-amp with 240vac input with 5vdc for vcc and i never have found circuit or experience like that, but i like to try to give you a little support and who knows if you are the first one that can make this impossible circuit and work.

As long as i know, a differential op-amp circuit that you made is for low voltage dc input and the output is unstable for ac input ( low voltage ), to make the output stable for ac input, usually use a capacitor parallel with feedback resistor (try it) about 1 nF - 1 micro depend to input voltage and output voltage that you want.

Try to use capacitor before input resistor ( small value/ ...nF with 400v ) to make a lower ac input.

I know it's almost impossible, but there's a hope to make it and who knows your experiment is working.

One thing that i want to tell you, be careful with high voltage, and also make a best earth ground for safety at your circuit.

... I know Jusup said a little schematics, but can you please make it a bit bigger? What is the amplifier type? What supply voltage?

I'm sorry with my English, i don't mean "a little" schematic as "a small size" schematic but i mean not "all" of schematic but "a part" of schematic.

I'm puzzled mentioned 240V. Is your GND connected to Common pin of amplifier? Theoretically it might be so, but applying the differential mode won't bring you remedy. Make it clear, e.g. put off your digital multimeter and apply analog one either oscilloscope.

I do agree with previous comment that is a low[near or less than 1] gain factor could be a reason.

try looking at eidusa.com/kitsmax1236.

read vtg across 540 ohm which is abt 440mVolt .Ip is 240Volt.

do you know what is differencial voltage? what are you going ot measure?

Your output is very lower as you drawed below, what is more, it might be 0 voltage!

I can tell you, your circuit is only a inverter amplifier. and work point is syn pin volitage which is divided by two its resistors. thats why you got very lower nearly to o voltage.

if you hope to amplify differential voltage, you have to change yhour circuit, or change connect way at input into diffential input.

tell me details, I cn offer you one reference circuit which not difficulty.

Hi Rupaliprayag. You said: "can anyone suggest me how to measure this differential voltage?"...

Well, there is no need to use a differential amplifier in order to do such a differential measurement. There are other (and easier ways) to do this.

If you have a d.c. volrtage to measure (across the 500R resistor) then a multimeter (voltmeter) is suitable for this kind of measurement.

If you have to measure a.c. voltage then this measurement is more difficult. In this case there are two ways to do this using an osciloscope:

1) You make a "differential measurement" using the two channels of the osciloscope. You set "invert" to the channel 2 and you set "add" mode (sum of ch1 & ch2). So you have [ch1]+[-ch2]=ch1-ch2. Then you put the probe of ch1 on the one pin of the resistor and the probe of ch2 on the other pin of the resistor. What do you see on oscilocope's screen is the differential voltage across the resistor.

2) You isolate the osciloscope's main (220V) plug from the "earth-ground". I.e. on a schouko plug you can do this by putting an isolation tape around the plug (perimetric). In this way, when you connect the plug into the outlet, the gnd contact of the plug doesn't attach the corresponding gnd contact of the outlet (because there is the tape between them). Now you are free to connect the gnd clip of the probe to every point of your circuit. Then you connect the gnd clip to the one pin of the resistor (the new "reference" point) and the pin of the probe to the other pin of the resistor. Again, in this way, you measure the differential voltage across the resistor.

George maybe should add that option 2) is inherently dangerous as the chassis of the oscilloscope is now LIVE and will potentially kill the user if he so much as touches the thing..

and so far as option 1) goes, an oscilloscope can only accurately display the voltage if the attenuator is set to display the voltage correctly on both channels.In this case it will need to be set for 50V/div .. the difference will also then be displayed at 50V/div, if you are attempting to measure mV or even low volts then this method is useless and you will get a nice flat line. more likely differences in the calibration between channels will mean that you get several volts ripple displayed in error.

This will also be a problem when attempting to use a differential amp to display the voltage, differences in the resistor dividers will cause errors unless they are matched or calibrated with a variable resistor when no current is flowing. (no offset present)

Stoney, you are right about the 2nd option. There is a risk as the chassis of the oscilloscope is floating (not connected to earth-ground).

About the 1st method: Of course, it's nessecary to set the same V/div for both channels. But it's not nessecary (in this case) to set 50V/div. You can set as low as 5V/div (or even 2V/div) and then you can measure voltage differences of several volts. (Usually you cannot set at lower scales as you'll get completely wrong measurements.) Of course, this is not (usually) an accurate measurement as you'll have an offset (i.e. several mV) between the two channels and probes.And, of course, you cannot measure voltage differences of just a few mV.

[I just did this by using the two outputs of a P.S. one at 35V and the other at 32V. So I expect to measure 2V volt. diff. between them. By using two X1 probes (and by using this method) I measured 1,8V at scale 5V/div and (amazingly) exactly 2V at scale 2V/div. (At lower V/div scales the measurements were completely wrong or sth like 0V). By using x10 probes measurements at scales 0,5V/div and 0,2V/div were less accurate though, as I had an offset of 300mV between the two probes.]

Just try this yourself. But, as you see, you can use this method just for a rough measurement.

Most cro's have enough overhead internally to only cope with signals they can display .. and a few divisions more.

If you feed a 240V signal into a CRO with the attenuator at 2V/div it will clip the signal in the front end, the subtraction is done after the preamp, so the result will be erroneous as the data is already distorted.

A good quality oscilloscope has much higher voltages than 'normal' cros in the front end and can cope with the higher amplitudes that cannot be displayed. These may be capable of quite a bit more, my old tektroniks valve CRO could cope with massive overdriven signals in the front end.

Even so, you could only use 2V/div on 32V at best.. 16 divisions (unipolar), so 240V DC is going to be limited to 240/16 = 15V/div .. I assume he is talking 240V AC though. so to measure the 330V peak he is going to need with your CRO 330/16 = 20V/div as I predicted.

The only really accurate way of using a CRO on subtract is to get the signals to display correctly before hitting the subtract (or invert and add) button. Then you can up the attentuators at the same time until the signal is obviously incorrect and back it off one step.

Hi again. Your problem is that your op amp (LM358) cannot handle the common mode voltage of 240V. This voltage is essentialy applied to both the inputs of your op amp and that's why it's considered as the common mode voltage (concerning the "reference" of the op amp which is the voltage on pin 4, which is gnd).

Look at its datasheet: At its max Vcc=30V the common mode voltage is 30V-2V=28V. In your case you have Vcc=5V so the common mode voltage should be sth just below 5V.

Just try this: Your resistor 500R should be on the "lower side" of your divider e.g. it should be connected to the "gnd" point of your input voltage (instead of the "240V" point of your input voltage). In this way, you'll have a low common mode voltage, probably below 5V and your op amp will operate well.

Thanks ,I think you gave me the solution I got simulation in orcad but now I will do it practically I hope I will get a good result.

The problem arrived because I haven't read the Opamp characteristics properly ,I will take care hence forth.