High Impedance circuit

[Yves]

Hello

I have a build circuit using the MCP607 opamp which has I believe an impedance higher that 1T Ohms when used as buffer of gain 1. When I check the input with my multimeter I read only 27k Ohms in the input . I don't need very high amplification and the frequency of the external floating signal is very low and can be consider as DC. The output will feed a A/D MCP3421 which will have its Vin- connected to the ground. Why I am not achieving the high impedance? I'm pulling my hair and don't were I'm going wrong. I have attached a Proteus 8.13 schematic.

[david]

My guess is that you're likely turning on the internal ESD diodes with the multimeter voltage.  You could try reversing the leads to see if the reading is different but the datasheet reads like even the reverse drop of the diodes is quite low and may act like a low voltage zener.   If reversing the leads gives a different reading then you've definitely got a junction in the way.

Cheers,
David

[Stephen Moss]

When I check the input with my multimeter I read only 27k Ohms in the input

I may be wrong but I don't think you can accurately measure the resistance/impeadence of a semiconductor with a multimeter, at least I have never managed to do so.

It is a constant source of both amusement and despair at work when undergrads woefully miscalculate the power dissipation in device such as a voltage regulator because they measure the resistance across it and perform an I2R calculation rather than a VI calculation and end up with something ridiculous like 270mW when it is actually 7W and then wonder why it is getting really hot. 

I am sure a semiconductor materials expert could explain why using a multi-meter is not accurate and how to measure it correctly/infer it.

[Yves]

Well let leave the multimeter it may have issues.  I have setup a small voltage source giving 1000mV through 0 Ohm when I place is series resistor of 100 kilo ohm I get only 267 mV. Is that not relatively low impedance?

Yves

[Yves]

It is not the protective diodes as the test I did above giving me the same results but in negative range.

[david]

Unless your meter is faulty the reading of 267mV is way too low.  Definitely a problem there somewhere.

[david]

So is this something like your test circuit with R2 being the suspected input impedance?  Sorry I couldn't resolve your files  - do you have a simple picture (jpg, gif  etc)

[John Lawton]

The MCP607 is a dual op-amp, how is the other half connected?
What is your supply voltage?

[Yves]

That is the circuit

[david]

Still not sure what you're trying to achieve there.  It looks like a voltage follower biased to 1.25V and a battery supply that can add an additional voltage via RV1 and R2.  Even if RV1 is set to 0V additional signal such that U6A only has the 1.25V bias I think you will find that the output at R8 is already siting at 3.75V.  Adding a small +ve signal via RV1 of say 100mV, on top of the 1.25v bias (1.35V total) will set the output at R8 to 3.55V due to the inverting gain of U6B being 2x .  Is this what you require?

[Yves]

Thank you all for your help. I found the problem (at long last). The circuit I published above worked as indented giving a dc impedance in the upper giga Ohm. The problem was when I soldered the small length of the coaxial cable leading to the BNC socket, I burned some of the insulation so it was not making  a short but a sort of resistor(about 27K) between the wires. The cell in schematic is just a for the simulation of a ORP electrode (Oxido Reduction Potential), they are use to measure potential changes in chemistry. Those electrode requires  very high impedance amplifiers. I'm more verse in chemistry than electronic and programing pics, lol. I still believe you can have an idea to measure a DC impedance using using the Ohm function of a multimeter but maybe it is debatable...Now with the circuit above the multimeter reads over range in the input.
Thank you very much for your help. It is a great forum with good and helpful people.

Yves

[Yves]

The MCP607 is a dual op-amp, how is the other half connected?
What is your supply voltage?

not it is not the MCP607 is a double 5V IC and the middle point is created by a virtual ground. If you look at the schematic the floating voltage source or electrode  "seats" on the virtual ground, so it can swing below the virtual ground. The opamp still see it as a positive signal.

Yves

Yves