TLV9152: Pull-down value on SHDN pin

Clément Letonnelier

Part Number: TLV9152
Other Parts Discussed in Thread: , UCD9090

Hello,

We read in the datasheet that a pull-down was present on the shutdown pin.

We intend to do the following schematic to ensure we have no voltage on our CPU until the 3V3 is present (3V3_aux is present way ahead).

The output of the UCD not being piloted until the POR is ended, we want to put an external pull-up to P3V3_aux on the signal that drives the shutdown pin.

However, the datasheet does not provide information on the pull-down value, only a typical current over the whole voltage range so it's not helping much.

Do you have more information about it ?

Is a 10k on R1 enough to ensure the TLV9152 is shutdown at board startup until the UCD drives a 0 logic level ?

Best regards,

Clément

over 2 years ago

0 Clemens Ladisch over 2 years ago

Guru 271320 points

With the typical current, the 10 kΩ resistor would generate a voltage drop of 5 mV. So this is safe even if we assume that the maximum SHDN current is ten times that.

0 Clément Letonnelier over 2 years ago in reply to Clemens Ladisch

Expert 2436 points

If we consider the current yes for sure but as it is stated that the device embeds a pull-down, I was more thinking about a potential voltage divider induced by both our R1 and the pull-down inside the device.

That's why I was highlighting the datasheet was unclear as specifying a current over the whole device voltage range does not make totally sense.

0 Caro Walter over 2 years ago in reply to Clément Letonnelier

TI__Mastermind 18907 points

Hey Clément,

I just tested this in lab and the presence of R1 does not change the shutdown functionality of the device. The concern below does not apply.

Clément Letonnelier said:
I was more thinking about a potential voltage divider induced by both our R1 and the pull-down inside the device.

Device 3.3V ON with SD 3.3V ON in series with 10k, output is disabled.
Device 3.3V ON with SD 3.3V OFF in series with 10k, output is enabled.

All the best,
Carolina

0 Clément Letonnelier over 2 years ago in reply to Caro Walter

Expert 2436 points

Hello Caro,

Thank you for your answer.

A test is however not sufficient for us as we need to justify design choices.

Are you able to provide us with tangible data ?

Best regards,

Clément

0 Caro Walter over 2 years ago in reply to Clément Letonnelier

TI__Mastermind 18907 points

Hello Clément,

I don't understand, which design choice are you truing to justify? The op amp or the R1?

All the best,
Carolina

0 Clément Letonnelier over 2 years ago in reply to Caro Walter

Expert 2436 points

Carolina,

That the value of R1 is compatible with internal pull-down inside the op amp which is not specified per se.

Clément

0 Caro Walter over 2 years ago in reply to Clément Letonnelier

TI__Mastermind 18907 points

Hey Clément,

Understood, I can go into lab again on Monday and confirm what the voltage at the intersection of resistor (R1) and the SD pin is.
Please allow until June 6th by EOD.

All the best,
Carolina

0 Clément Letonnelier over 2 years ago in reply to Caro Walter

Expert 2436 points

Carolina,

Still this would mean it's test data on a single setup, are you able to provide engineering data through the design team ?

Clément

0 Caro Walter over 2 years ago in reply to Clément Letonnelier

TI__Mastermind 18907 points

Hey Clément,

Understood, I have inquired internally if this is something I am allowed to share.
In the meantime, could you specify if you are using the automotive version (TLV9152-Q1) or the commercial version (TLV9152) of the device?

All the best,
Carolina

0 Ron Michallick over 2 years ago in reply to Clément Letonnelier

TI__Guru*** 145766 points

Hello Clement,

You want SDEN pin to be at least 1.1V ; that is 3.3V - 1.1V / 10k = 220uA flow alliable. The typical current demand is 0.5uA all the up to 16V!

10k to 3.3V is absolutely fine as a pull up.

0 Clément Letonnelier over 2 years ago in reply to Ron Michallick

Expert 2436 points

Hello,

Caro Walter: We are using the commercial version.

Ron Michallick: if I take into account the typical current it sure works, I am not questioning that, and yes the available flow is way above it.

However, for worst case analysis purpose (required in the aeronautical field for critical equipments), I need to justify based on tangible data, I can't justify by saying "current is way above typical, it will work in all conditions", even if I know it's the case from engineering judgment.

That's why I am insisting as the max. for that bias current is not provided.

Sorry to bother you

Clément

0 Ron Michallick over 2 years ago in reply to Clément Letonnelier

TI__Guru*** 145766 points

Clement,

I understand your point, but is this the only problem? About half of the TLV9152 specifications do not have limits. A few specs don't include temperature. Do you need me to suggest a device with more defined limits?

0 Clément Letonnelier over 2 years ago in reply to Ron Michallick

Expert 2436 points

Ron,

I indeed must admit I only focused on that particular item.

I discussed this with a colleague and he told me he usually applies a x3 coefficient when there is only typical information available, do you see this as an acceptable thumb rule for all non limited items ?

If you have in mind a device with more defined limits in a similar board space, I could indeed be interesting. Considering that it needs to have a shutdown/enable pin that can be set externally with a pull-up/pull-down at board power-up until we drive the pin with the UCD9090.

Thank you,

Clément

0 Caro Walter over 2 years ago in reply to Clément Letonnelier

TI__Mastermind 18907 points

Hey Clément,

There is a statistics section in the datasheet that helps build a more robust design.

It is similar to what your colleague had described to you. I will let Ron provide his recommendation.

All the best,
Carolina

+1 Ron Michallick over 2 years ago in reply to Caro Walter

TI__Guru*** 145766 points

Clément

Here is sample data for your application of 3.3V. The initial portion is linear until the FET acting as a resistor has its channel pinched. After that the current becomes constant. ISD versus voltage across temperature range.

ISD for different supply voltages [0V, 3.3V, 16V] is below

ISD vs Vs is below. VSD = 16V pinched channel (current limited)

The resistance of the pull down FET is based on a controlled process.

Your colleges 3X rule of thumb is a reasonable in this case. Knowing how the resistor performs should give confidence in using the SD pin.

I didn't find a just as small, shutdown, similar device with more parameters with limits.

0 Clément Letonnelier over 2 years ago in reply to Ron Michallick

Expert 2436 points

Hi Ron,

Thank you for the additional data.

Best regards,

Clément

Amplifiers

Amplifiers forum

TLV9152: Pull-down value on SHDN pin