analog input buffer question

[yougarage]

hello I'm going to make an analog input buffer for a microcontroller.
I came out with this solution but I don't know if it's right.
the op amp (rail to rail)is powered from a 5v source

JP1 and JP2 are two jumpers.
JP1 allows me to choose the input range: JP1 installed, 0-12v, JP1 removed, 0-5v
JP2 allows me to change the op amp gain: JP2 installed, gain = 10, JP2 removed, gain = 1
do you think this design could lead me to problems because the resistors R1 and R3 are unuseful with the jumpers removed?
I thought the tlv2371 rail to rail is a good choice for this purpose, what do you think about it?
thanks for your help!

 

[FvM]

Re: anolog input buffer question

It can work. R3/R4 ratio should be 9 rather then 10 for a gain of 10.

 

[yougarage]

Re: anolog input buffer question

thanks a lot!
and what about adding a capacitor right after R1 to ground to create a lowpass filter?
is it safe to share the resistor between the rc network and the voltage divider?
thanks for your help
it's invaluable for a novice like me!

 

[FvM]

Re: anolog input buffer question

and what about adding a capacitor right after R1 to ground to create a lowpass filter?
is it safe to share the resistor between the rc network and the voltage divider?

Yes, it sounds reasonable. The cut-off frequency will be higher in divided 12 V setting, but no problem, I think.

 

[LvW]

Re: anolog input buffer question

thanks a lot!
and what about adding a capacitor right after R1 to ground to create a lowpass filter?

A lowpass in the negative feedback path gives the closed loop response a highpass characteristic!

 

[FvM]

Re: anolog input buffer question

The suggestion is to implement a lowpass in front of the amplifier, if I understand right.

 

[LvW]

Re: anolog input buffer question

The suggestion is to implement a lowpass in front of the amplifier, if I understand right.

May be, but it was not clear to me - my comment was only to avoid misunderstandings.

 

[alexan_e]

Re: anolog input buffer question

The OP said "and what about adding a capacitor right after R1 to ground..." , based on the schematic of the first post this is a capacitor connected from the opamp positive input to the GND.

Alex

 

[yougarage]

Re: anolog input buffer question

Yes the capacitor is at the op amp input
I attached a picture of my final layout.
I wanted to add a zener to prevent input voltage damaging the op amp, I think the clamping voltage should be 16 v but I'm not sure about it.
The datasheet says: supply voltage Vdd max 16.5 v (see attached picture)
differential input voltage +- Vdd
Am I right with the zener value?

 

[FvM]

Re: anolog input buffer question

The OP can be expected to have an internal substrate diode, clamping overvoltages to the 5V supply. This means, that the Z-diode won't ever conduct current, at least with static overvoltages. It can be nevertheless reasonable to absorb ESD, but then you would need a series resistor in the OP's +ve input to make the Z-diode operate independent of the OP substrate diode. It should be considered, that the input substrate diodes offer a considerable level of protection, they are often specified with a current rating for newer OP types, e.g. 10 or 20 mA. For safe operation of the circuit, also possible lifting of the 5V supply node by input overvoltages should be analysed.

 

[alexan_e]

Re: anolog input buffer question

I wanted to add a zener to prevent input voltage damaging the op amp, I think the clamping voltage should be 16 v but I'm not sure about it.The datasheet says: supply voltage Vdd max 16.5 v (see attached picture)

The datasheet says that the max power supply of the opamp can be 16.5v
The max input voltage is specified as –0.2 V to VDD + 0.2 V so when your power supply is 5v you can only go up to 5+0.2v.
I've seen Schottky diode connected from the input to both power supplies to clamp the voltage in ADC inputs, it should do the job in your case too but they usually have a specification of about 0.3 Vf so it is higher than you need unless you can find a Schottky with Vf 0.2v.

I mean a connection like this www.cirrus.com/en/pubs/appNote/an20.pdf

Alex

 

[FvM]

I've seen Schottky diode connected from the input to both power supplies to clamp the voltage in ADC inputs, it should do the job in your case too but they usually have a specification of about 0.3 Vf so it is higher than you need unless you can find a Schottky with Vf 0.2v.

To determine if additional protection means are necessary, you would refer to the OP datasheet in a first order. The said TLV2371 has an maximum current specification of 10 mA for the clamping diode. If it can't be exceeded, no additional protection is required.

The 0.2V margin should be understood related to the devices temperature range and known Si diodes I/V characteristic. A schottky diode will basically have a lower forward voltage at any temperature (assuming equal temperature of both devices), so it will conduct first. But it doesn't necessarily give full protection for the OP due to it's series resistance. Using low resistance schottky diode wouldn't be a good idea because they have large leakage currents. So practically, only a series resistor between schottky clamp and OP input allows a reasonable circuit dimensioning.

As I already mentioned, I would try to rely on the built-in diodes. External clamps are necessary, if either the rated input current can't be kept, or you are using a dual or quad OP and have to assure, that overvoltages at one amplifier don't affect the other channels.

 

[alexan_e]

Your answer was posted while I was writing mine so I only saw it after I posted.
What you said about the internal diodes meant that my answer was void but I still didn't delete my post.
My answer wasn't meant to say that what you said was wrong in any way.

Alex

 

[FvM]

Thanks for clarifying. Suggesting schottky diodes isn't a bad idea. They haven't been mentioned in this thread yet. But they have also disadavantages, as any other overvoltage protection, too.

 

[yougarage]

thank you all
I will try to rely on the internal diodes, without the zener.
I'm going to use the quad version OP but the input will never exceed 14 v so the input current will be lower than 10 ma