LOG114: Precision current bias circuit design

Hello,

It is somewhat difficult to follow exactly what you are looking to do here; do you want to measure current from 0 - 99.5 uA?  Separate from the LOG114, what is the goal of the circuit?

Regards,
Mike

Thank you for your reply, I want to photomultiplier output current for a certain range of truncation in the logarithmic amplifier, the need to truncate the main reason is that we are a wafer defect detection, the wafer without defects when there is a background current, the need to subtract the background current in the logarithmic amplifier for photoelectric conversion, because the logarithmic amplifier for the amplification of smaller currents of the voltage difference is more pronounced, to help us detect the defects of small particles, such as 40nm 28nm.

Case For example, the optical circuit of the laser hit on the wafer without defects, photomultiplier tube current size of 500nA or other values, and the logarithmic amplifier intercept is 1nA, we carry out defect detection, the logarithmic amplifier of the range of 0-500nA does not take advantage of, for our small particles detection there are great difficulties, so we need to wafer without defects to cut off the current (subtracted) and then give the logarithmic amplifier to the processing, can be effective in obtaining small particles of higher signal-to-noise ratio; for this reason, we need to design a weak current bias circuit or subtraction circuit.

The following is the voltage difference between the current when the wafer is defect-free by modifying the intercept and slope of the logarithmic amplifier and subtracting the current when the wafer is defect-free, and it is found that the small-grain signal-to-noise ratio that can be obtained is more accentuated when subtracting the background current when the wafer is defect-free

Regards!

cao

Hi Cao,

Ok, as I understand it; you would like to have 1 nA of current measurement resolution, however you nominally have up to 500 nA of static/DC current (biasing the photomultiplier tube) that you would like to ignore.

This sounds like an AC-coupled solution may be the right approach; is the photomultiplier tube bias current constant?  And the small, 1 nA signal a short period of time/higher frequency?

Can you tell me if the drawing below is correct:

Regards,
Mike

Thank you for your reply and help, your understanding is correct, due to our current system in the research stage, so the bias current of the photomultiplier tube is not yet fully set a specific value, for this reason, I want to design a current bias adjustable circuit for implementation, if this is more difficult to implement, you can first try to bias current is fixed at 500nA or 5uA.

The following diagram is a schematic of the process we need to achieve, for your convenience and reference, thank you for your dedication and help!

Hi cpin,

I think that this is what you may look for, see the attached simulation. 

You would need a 500nA precision constant current reference for the application. Or you will need to have dual beam optical measurement method. You would ratio out the difference and make your delta measurement. Then you can gain it up to your system settings. 

LOG114_PM Subtraction 04132024.TSC

OPA192 precision op amps are good for your sensing range. If you want something better, we have even more precise and low noise op amps. For instance, 1.00V reference will probably need tp be an ultra precision and low drift voltage reference, and we can help you to identify these op amps once you know what the design is going to be.   

Please let us know if this will work for you. 

Best,

Raymond

Hi Cao,

Raymond gave a good example of a current source as well as how to use the LOG amp. with a reference current.  But, I wanted to add some more detail to this in case you were looking for higher sensitivity.  In order to get high sensitivity on the log amp. for currents as low as 1 nA, then the reference current also needs to be in that range.  Consider for example the block below:

If you used the circuit Raymond gave above, and tried to measure 1 nA on top of 500 nA with a 500 nA reference current, then the "measured current" would be 501 nA, the "subtracted current" would be zero, and the "reference current" would be 500 nA.  The voltage would change by approximately 300 uV:

However, if you were able to set the "Subtracted Current" to 500 nA, and change the Reference Current to 10 nA, then the voltage change would be much higher. The math below shows a change from 1 nA to 2 nA with in "Iin1" with the reference current at 10 nA

In the math above, the output change is 112 mV, so the resolution is much improved.  So, you could design a circuit that creates the "Subtracted Current", and achieve very high resolution, BUT it would have to be a highly accurate subtraction current. Keep in mind the current into the log amp. has to be > 100 pA to avoid saturation.  So, if the "Measured Current" - "Subtracted Current" is < 100 pA, the log amp. will go into saturation - this would become very difficult.  To me this is all much easier if it can be AC-coupled.

An AC-coupled solution would look similar to this:

We actually have other AC-coupled solutions using TIA amplifiers that are designed to remove ambient light.  You can find an example reference to that here:

www.ti.com/.../sboa324.pdf

Can you confirm that an AC-coupled solution can work for you? 

Regards,
Mike

嗨，雷蒙德 ，

Thank you for your reply, your suggestion is very helpful to me, but I would like to achieve more is the input current has a bias current in the logarithmic amplifier for use, the size of the bias current can be adjusted according to the system project requirements, currently can be set to 8.4uA,that is, the input current subtracts the bias current 8.4uA after the input to the logarithmic amplifier for logarithmic conversion, the reference current is 1nA, this time, assuming that the input is 15uA, then the logarithmic amplifier's voltage output is  Vlogout =0.375*log((15-8.4)/0.001) ,do not know if you have a recommendation on this side of the circuit design or case study reference.

Here are two pictures that should help you understand:  

I look forward to your reply!

嗨 曹品,

cpin said:

the reference current is 1nA, this time, assuming that the input is 15uA, then the logarithmic amplifier's voltage output is  Vlogout =0.375*log((15-8.4)/0.001)

The most accurate optical measurement is shown below, where this is a simplified block diagram. You probably will need to a chopper to remove 1/f noise of photomultiplier (consider using the solid state PM vs. the tube type these days). 

The image above is configured in transmittance mode, and your application is in reflection mode. It is measured by the Beer's law. 

I provided an ultra precise current source for the application. If the application is using dual beam optical configuration, such the current source is not needed. For a low cost solution, then you would need the constant current source, which is programmable to what the application is called out. 

cpin said:

do not know if you have a recommendation on this side of the circuit design or case study reference.

The dual beam optical configuration is a classic optical measurement and it is used in high performance optical instrumentation. We do not believe that have the reference design for the sensing circuit, but we have similar application note that are similar, see the thread below. And you should spend some time to search our previous E2E, where it may find additional related suggestions and technical know-how.   

https://www.ti.com/lit/an/sbaa482/sbaa482.pdf?ts=1713170751916&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FLMH6629%253Futm_source%253Dgoogle%2526utm_medium%253Dcpc%2526utm_campaign%253Dasc-null-null-GPN_EN-cpc-pf-google-eu%2526utm_content%253DLMH6629%2526ds_k%253DLMH6629%2526DCM%253Dyes%2526gad_source%253D1%2526gclid%253DCjwKCAjwoPOwBhAeEiwAJuXRhxdbnIuufN8Q8D2HSHi2Sd1hMwyD1ooeQ5S64c-jVc5UB9263xR7VBoCr2QQAvD_BwE%2526gclsrc%253Daw.ds

https://e2e.ti.com/support/amplifiers-group/amplifiers/f/amplifiers-forum/901010/opa858-amplifier-for-fast-and-standard-output-of-silicon-photomultipliers

https://www.ti.com/lit/an/snoa606b/snoa606b.pdf?ts=1713193797398&ref_url=https%253A%252F%252Fwww.google.com%252F

https://www.hamamatsu.com/content/dam/hamamatsu-photonics/sites/documents/99_SALES_LIBRARY/etd/PMT_handbook_v4E.pdf

My colleague, Mike is correct, which the signal detection design will need AC signal coupling into the Log114 or Log200 op amps. If the design is using a chopper in the front (to chop the incoming laser beam), then you have to use AC signal coupling approach, since the DC reflection signals consist more noises. The application is only interested the ratiometric differences vs. the reference sample. In addition, the ratiometric measurements will remove common mode noises, temperature, stray light, unwanted reflection and other undesired optical disturbances. 

Please let me know if you have additional questions. 

Best,

Raymond