TLV4333: What the Output behavior is when VCC < 1.8V ?

Hi Martin,

We do not characterize devices outside of their intended operating range, but it is common for op amps to exhibit this behavior on startup because prior to entering the proper bias region, the output can be in a state of uncertainty like you have mentioned. As of now, we do not have a straightforward discrete solution for addressing this problem. Your best option would be to use the TLV3691 as it is a dedicated comparator with power on reset (POR) functionality and will be guaranteed not to exhibit this startup behavior.

Generally speaking, it is not a good idea to use devices like the TLV333 as comparators because they have internal back-to-back, or anti-parallel input diodes that prevent the inputs from being pulled too far apart. If your hysteresis is set such that you never try to pull the inputs more than a diode drop apart then this is not an issue, otherwise you may see excessive currents flowing between your input pins and you could potentially damage the device.

...Or the TLV7011 or TLV7021, which also has an internal POR circuit for a known start-up condition. It is faster than the nanopower TLV3691 if you need more speed.

When Vs < 1.6V:

The TLV7011 (Push-Pull) output will be low

The TLV7021 (Open Drain) output will be high impedance ("Hi-Z", or "floating")

Hello Zak and Paul,

Very appreciate for your advice, however, customer requires low Vos (less than 1mV) because smaller sense resistor is used to minimize power losses for OCP purpose, I tried to filter out the comparator parts but seems no parts can meet, can you help find out other proper solution for my customer's request?

Besides, I'm wondering how to filter out the comparator with POR feature? should I check datasheet one by one?

Thanks.

Martin

Hi Martin,

If you can't increase the shunt resistance, then your best option would be to use the channel of the TLV4333 that you are currently using as a comparator as an amplifier instead and follow it with a dedicated comparator with POR. I believe the comparators Paul and I have mentioned are the only ones that incorporate this functionality, but I do not support comparators so I will leave it to Paul to correct me if I am wrong!

One thing to note is that the comparator and TLV4333 may not have the same minimum operating voltage. If the min supply on the comparator is the lower of the 2 then it will be functioning properly before the TLV4333 and will still pass the glitch and the POR will not protect you. To avoid this you will want to make sure that either the supply of the comparator comes up more slowly (this could be accomplished with a small rc filter on the supply pin) or the inputs to the comparator are filtered so that it doesn't respond fast enough to react to the glitch on the TLV4333 output.

Hi Zak,
Could you help me out with understanding what you idea is by a real schematic?
Thanks a lot.
Martin

Hi Martin,

I'm not sure what thresholds your customer would want or the bandwidth needed, but here is a quick example that gains up a 1mV signal from the shunt to about 520mV at full scale. The comparator is scaled to have a reference of 500mV with about 20mV of hysteresis. While sim doesn't seem to run the transient correctly, by placing an RC on the power line of the TLV7021 you can ensure it comes up a little slower and the POR function will keep its output low until it reaches its operating voltage. The customer will need to determine just how much delay needs to be added.

TLV333 plus comparator.TSC

If you need assistance with adjusting the comparator, then you may find the following presentation from Thomas Kuehl to be useful: Effective Comparator Application_TEK_Apr_20_2016.pdf