INA146: ESD protection on INA146 inputs

Hi Markus,

The INA146 can survive 1kV out of circuit ESD events without external protection. This may differ from its in circuit performance though. To better understand what you need can you share some more information about your application? Do you need bipolar protection? What is your common-mode voltage and does this stay fixed or does it move? How much of a concern is leakage current?

Something you may consider is using multiple TVS3300 or TVS2200 diodes stacked in series. These are not traditional TVS diodes and have a much sharper knee resulting in a much flatter clamping voltage. The TVS3300 for example has a breakdown voltage of 34V and a clamping voltage of only 40V. They are unidirectional devices though so I'm not sure if this would work for you. There may be other devices on the market that are better for this.

Hi Zak,

Thanks for the information and for the ideas! Unfortunately I need bidirectional protection and I would measure 75V input voltage with maximum of 20V common-mode voltage. I have came to the conclusion that I have to make the differential amplifier from discrete components to raise the common-mode voltage range up to around 200V. This way I can easily protect the circuit from an ESD event with a general TVS diode.

Best regards,
Tamas

Hi Markus,

I fully agree with Zak: It's not wise to fabricate the difference amplifier with discrete components by yourself, because it's not simple at all to achieve the needed good CMRR (common mode rejection ratio). It's often overlooked what sensationally good CMRR these INAs from TI provide! To achieve 90dB CMRR allows resistance tolerances in the difference amplifer of only <0.02%, including temperature drift and long term drift. This is very very difficult to achieve (read impossible) with discretes. Only the monolithic construction on a die can guarantee this.

Kai

Hi Zak,

Unfortunately I can not use INA148 or INA149 as I need some attenuation to measure input voltages up to 75V with the presence of common-mode voltage.

Regards,
Tamas

Hi Tamas,

I apologize, it sounds like I misinterpreted what you are after. 75V is too large of a differential voltage to measure directly. Perhaps we should take a step back. What exactly are you measuring the voltage of and what kind of range are you trying to scale things to? Is your common-mode input voltage constant or does it fluctuate, and if it does, how much?  

There are a couple different approaches you could take here. The first is to divide down the signal to a more suitable differential level and run this into a difference amplifier. You have to be careful doing this though because the internal resistors of our difference amplifiers are very precisely matched in ratio, but not necessarily in absolute value so external input resistance could severely impact the CMRR. As long as you use precision resistors and keep the values small compared to the input impedance of the diff amp you wont degrade the CMRR to heavily. If you do this, you don't necessarily need to use a high common-mode input amplifier, you could get away with something like the INA132 like in the circuit I've attached.

INA132 Divider Circuit.TSC

Alternatively you could just use a true instrumentation amplifier after the divider to have a high input impedance and preserve your CMRR. The viability of this option depends on how much fluctuation you have in your input common-mode voltage.