LMH32401: Is it OK to use a positive APD bias but ac couple the input?

Hello Ian,

Here is a thread with discussion on this topic. Are you looking into using the ambient light cancellation functionality of this device?

Thank you,

Sima

Hi Sima, 

I had already checked this thread but I do not think it is quite the same. My APD will be reversed biassed to about 60V but to get the current flow out of the IN terminal I propose to use ac coupling. Just wondering if anyone ha done this and if they had found any issues. 

Regards

Hi Ian,

I would run a TINA-TI simulation.

Kai

Hello Ian,

  I talked to our test team and we haven't tested for this specific situation in practice. For the first part, we have tested a similar type of setup:

When the photodiode sources the photocurrent (cathode is biased to a positive voltage and anode is tied to the amplifier input) a soft clamp activates when the amplifier input is overloaded. When the soft clamp activates, the amplifier takes longer to recover. The recovery time depends on the level of input overload. (Section 8.2.2 for more information)

  As you noted, in this configuration, the photodiode will not sink current. However, if ALC is not needed for your application, then this method will work. Or, I would recommend the OPA857 which will also work for this configuration. 

  In simulation, for your specific configuration (remove the series resistor R28), I could get the correct output with ALC on or with the AC coupling. Both might be redundant (watch out for input current saturation). External AC coupling with TIAs comes with a few disadvantages such as degradation in saturation recovery, increase in recovery/charging time, and a voltage divider will form which will depend on value of the capacitor vs input impedance. 

Thank you,

Sima

Hi Sima,

I had included R28 as the datasheet for the LMH32401 shows this on page 29 to add damping for stray capacitance and bond wire inductance. There seems to be nothing in the DS about how to calculate this resistor value. 

With reference to the ds, Fig 8-7 is with the device sinking current, (ie with the APD positively biassed). For my application, the extra pulse widening would not be an issue at these current levels. Does the input current clamp still work up to 100mA with the amplifier sinking current? What sort of pulse widening would be expected at 100mA input current with a positively biassed APD? 

I can probably manage without the ambient light cancellation. 

Thanks

Ian    

Hello Ian,

  Thank you for pointing this information out in the DS. This isolation resistor is optional and is helpful if the trace is long enough between the photodiode source and the input of the amplifier. The reason is as you noted due to long traces introducing parasitic inductance which can also isolate the input capacitor and cause instability when coupled with the input/feedback of the amplifier. This is addressed in another device's datasheet, the OPA855 DS page 25 which suggests about 10-20 Ohms. 

   Device would be sinking current if APD is positively biased and anode is connected at the amplifier input. The clamp which aids in fast recovery does not work with device sinking current, which is why the pulse stretching/extending occurs as the overload current increases. 

Thank you,
Sima

Hi Sima,

So if I positively bias the APD, does the maximum allowable input current of 100mA still apply or not?

Regards

Ian

Hello Ian,

 In either configuration (sinking or sourcing), the linear current range of the device is around 65 µA in the high-gain configuration and 650 µA in the low-gain configuration. This range will be slightly lower for the device sinking current. The 100mA integrated clamp protects the device from input currents exceeding the linear current range of the device. Higher input currents saturates the amplifier which increases amplifier recovery time and pulse stretching. The clamp is there to mitigate these issues to less than a few ns. More information can be found under section 7.3.2 on page 21 in the datasheet. Therefore, technically, the maximum allowable range is the linear spec. Do you need to allow currents up to 100mA in a linear transimpedance conversion? 

Best Regards,
Sima 

Hi Sima,

I do not mind the pulse extension for the amplifier sinking currents (positive APD bias). Do you have any data for the pulse width increase when you have input currents closer to the 100mA clamp level in the soft clamp mode? Also, I am still not clear what the 100mA limit is referring to? Is this the clamping capability of the clamp circuitry whether hard or soft clamping mode? What happens above this level of current, for example if the lidar beam strikes a retroreflective target and a very high currnet is generated in the APD and fed to the amplifier? My normal application could see current pulses of up to 50mA going into the amplifier from the APD. Retro-reflectors produce higher pulses. 

Thanks again, I am finding this discussion very useful. 

Ian   

Hi Ian,

a usual TIA is designed to have the photodiode directly connected to the -input. But in your design there's a AC coupling cap with a small series resistance between the photodiode and the TIA. This actually changes the topology and you no longer have a pure TIA but an inverting amplifier with AC coupling driven by the complex source impedance of the photodiode.

Kai

Kai,

I can remove the capacitor and use the positively biassed APD directly coupled to the amplifier input but I need some more information as per my previous post please. 

Thanks Ian

Hello Ian,

  Sorry for the delay in response, I will talk to our design and test engineers and get back to you. That is correct, clamping condition in general. The 100mA limit refers to the amount of current the device can handle, but in the sourcing condition. I believe without the clamp the absolute maximum the device can take is 25mA. I will also double check this with the design engineer.

Thank you,
Sima 

Hello Ian,

Here is the feedback from the designer: 

In the sourcing configuration, the clamp will protect the device to 100mA. The 100mA is a current pulse of a few tens of nA, probably up to 100ns for low duty cycle. The 25mA is a continuous rating; that is, it could take this for the life of the part.

In the sinking configuration, the device can also withstand a 100mA pulse but will it will not recover quickly. It can also absorb 25mA continuous.

Thank you,

Sima