Hi,

-Do you know what does the 0.1% means in the test conditions of settling time ? (see below)

Settling Time : ts : AV = 10, VIN = 10 V p-p, 0.1%        typ:4µs.

- I did not understand figure 30, 61,92 of settling time.

For Figure 92, Vsy=+/-15V, Input switch to -5V to +5V. Output finishes at 0V from which value ? It's not an Av of 10 ?

BR,

Mathieu • Hi,

The 0.1% is the error band.

For this instance on Figure 92, Av= 10, Vin = 10Vpp, 0.1%.

To calculate for the error band, it is equal to 0.1 % multiplied to the Vin, and multiplied to the Av.

The error band for this test condition is 0.001 * 10 * 10 = 0.1V or 100mV.

This means, you have 100mV error band and if your output is within this band, this is considered as settled already.

The measurement of settling time is from where your input signal been triggered.

For the Figure 92, the time 0s starts at the trigger of the input signal which is 10Vpp. And at ~3us, you can see that the output  swings on the

negative side and is over your error band of 0.1V. At 4us (typical), both sides of the output is already within your error band.

Hope this helps.

Thanks.

• Hi,

Thanks to answer so quickly !

I search "0.1% is the error band settling time" and I find Analog devices test method !

https://wiki.analog.com/university/courses/electronics/electronics-lab-1st

So, I guess in figure 92, OUTPUT is not the output of the op-amp : it's the signal at the "false summing node" ?

An other question for figure 92, Vsy=+/-15V and Input goes from -5V to +5V, -> it's impossible to use a voltage gain of 10, because OUTPUT is limited to supply voltage ? If the gain is not 10 : it changes error band ?

BR,

Mathieu

• Hello,

On Figure 92, the Output is just normalized to 0V.

The error band is affected by the gain. As mentioned earlier, it is the product of Vin, Gain, and the tolerance.

Here is a sample computation and a picture of the error band.

Hope this will help. Regards.

• Ok, i did not see it's normalized !

Thank you for everything !

BR,

Mathieu