LOG114: Curent below 100nA cannot be measured

Hi Pravin,

Basically, you are using the single-supply LOG114 EVM circuit which should be capable of responding correctly to a current level below 100 nA. You can seen for the EVM circuit in the EVM documentation, Figure 13, LOG114EVM Measured Results.

Please explain what you mean by "I tried with LOG114 but current below 100nA is not possible to measure." How is the VLOGOUT voltage behaving over the 5 pA to 5 mA input current range?

Regards, Thomas

Precision Amplifiers Applications Engineering

Hi Thomas,

For input current > 100nA approx the output is as expected as mentioned in datasheet.
When input current is less than 100nA, the measured current stuck on 18pA. Out put is noisy.

PCB layout is same as LOG114EVM.

Hi Thomas,

LOG114EVM userguide shows results from 10nA to 10mA.
Is this circuit capable of masuring low currents in pA?
Do I need to make any change in the circuit or layout ?

Hello Pravin,

So it appears that the LOG114 circuit is working as expected for an input current > 100 nA. That is a good indicator that the circuit and LOG114 are likely okay, and that something unexpected is preventing optimal operation at the lower current levels. That is not an unusual thing to occur.

Often, the issue behind obtaining incorrect results at low currents is due to PC board and/or component contamination. Excessive leakage currents from solder flux contamination can completely disturb the log amplifier's operation at lower current levels. We recommend that the circuit be assembled using solder having water soluble flux. Once assembled, we run the PC board containing the LOG114 circuit though two separate cycles of ultrasonic cleaning using fresh DI water each time. Then, the boards are dried in an oven or by a hot air gun. Flux can get trapped under the package and that is one area that brush cleaning doesn't clean well.

Also note that flux contamination at the input sensor circuit can degrade performance at low current levels as well. Therefore, all of that circuitry must be ultra clean as well.

Understand that as the input current is reduced the transimpedance gain of the LOG114 increases to very high levels. Because of that very high gain increase at low input current levels external noise coupled into the circuit becomes a big issue. Keeping external noise out of the LOG114 input circuitry is difficult, but must be done if you plan to use it with very low input current levels. The LOG114 PC board should be enclosed in a closed metal enclosure, and any wires in and out of the enclosure should be shielded. We sometimes power the LOG114 from a battery inside that same metal enclosure to avoid bringing noisy power supply lines into the box.

I do see that you are using an OPA335 for the reference buffer in your circuit. The original EVM used an OPA365 which uses a completely different design and has a higher output current capability than the OPA335 suggesting a lower open-loop output impedance (Zo). The OPA335 may work fine in the application, but just wanted to mention that as an observed difference between you circuit and the EVM.

Regards, Thomas

Precision Amplifiers Applications Engineering

Hello Thomas,

1) Is there any alternative for ultrasoinc cleaning, as I do not have this setup. ?

2) Using battery powered cicuit is not feasible in commercial product. Still I will check with Li-ion battery.

3) Can LOG114EVM measure below 10nA?

Thanks for your suppor.

Regards
Pravin Khisti

Hi Pravin,

haven't we discussed this issue some time ago?

Kai

Hi Pravin,

1) Is there any alternative for ultrasonic cleaning, as I do not have this setup. ?

I do not have any other suggestions, other than to do a web search regarding recommended cleaning procedures for the particular solder you are using. Even then, the recommended procedure may not provide the level of cleanliness needed for a very low leakage current environment.

2) Using battery powered cicuit is not feasible in commercial product. Still I will check with Li-ion battery.

Okay, just be aware that the power supply wires coming from the outside world may bring noise into the LOG114 environment.

3) Can LOG114EVM measure below 10nA?

Ideally, the LOG114 VLOGOUT will be +1.25 V for I2 = 10 nA, and I1 = 1 uA. This drops to +0.875 V for I2 = 1 nA. The VLOGOUT lower limit is (V-) + 600 mV so it isn't advised to try and push too close to that limit. The EVM document in Figure 13 shows some deviation in the LOG conformity below 100 nA, which probably worsens below 10 nA. It doesn't appear that the EVM was enclosed for this test, but it almost certainly was carefully cleaned as recommended. The EVM is usable down to 10 nA and possibly a little less, but the EVM document doesn't indicate it was tested below 10 nA.

Regards, Thomas

Precision Amplifiers Applications Engineering

Hi Kai,

Thanks a lot.
Yes we desicussed but thread was closed before we can reach some conclusion.
So I thought to reinitiate it as I have not made any progress even after that discussion.
Regards
Pravin Khisti