TLV758P: Vout accurancy, TLV75801PDRVR

Hi Kento,

I understand you would like a setpoint design for your customer.
I assumed a nominal voltage of 1.525, which sits right in the middle of the lower (1.4V) and upper (1.65V).

The design requires 1.525 +/- 8.2%.  This is very achievable with the TLV758.
Below I will describe an option of R1 and R2 resistors for the setpoint.
The other parameters which fit in the +/- 8.2% tolerance window are in the datasheet.
These include steady state parameters (line regulation, load regulation) as well as transient conditions (line transient and load transient).

Let's set the upper resistor R1 to 24k and R2 to 13.7k. 
Let's run a 100,000 monte carlo analysis on the output and plot the +/- 3 sigma values.
Here are the results:

Vout Average: 1.514704V
Vout Standard Deviation: 8.838mV
Vout 3 sigma: 8.838m*3 = 26.514mV

Vout Nominal Desired: 1.525V
Set the limit lines to be +/- 3% of the nominal desired value.
Lower limit: 1.525 * 97% = 1.47925V
Upper limit: 1.525*1.03 = 1.57075V

Here is the histogram plotted as a line chart of the results.

Thanks,

- Stephen

Hi Kento,

Please let me know if I misunderstood the customers question in my reply below.

The simulation was completed across temperature. 

So at exactly 95C it will only get better as the Vref tolerance will be closer to +1% than +1.5% on the high end.
The datasheet lists Vref tolerance at 25C, -40C to +85C, and -40C to +125C.  There is no specific data right at 95C.

As you can see below, the Vref test data was collected with the following conditions. Transients are not included (line transient, load transient), load regulation is not included (Iout is only 1mA), and line regulation is not included.  So you should add those in to your calculations.

Thanks,

- Stephen

Hi Masaharu,

・could you try the simulation in case of 95℃?

There is insufficient data in the datasheet to review the tolerance exactly at 95C.
The analysis will get better if the temperature range is limited to 95C than what I provided.

・is below correct ?

The output voltage regulation variation (±1.5%) was collected with the following conditions. Transients are not included (line transient, load transient), load regulation is not included (Iout is only 1mA), and line regulation is not included.  So you should add those in to your calculations.

Thanks,

- Stephen

Hi Masa,

Unfortunately limiting the temperature range to 95C will not reduce the amount of drift the LDO's output will experience. This is because the temperature drift of an LDO is dominated by the temp drift of the reference and in general bandgap references tend to reach their highest voltage somewhere in the range of 65C-85C. So the worst case drift will almost always be from -40C to ~75C, and temperatures higher than ~75C will actually show better drift performance. 

So, since the customer's temperature range still includes 75C the results of the analysis that Stephen did still remains the same. 

hi, Kyle-san.

thanks your support.

i believe that i can close the question from customer by showing our conversation here.

thanks so much