I'm tempted to use AD8675/AD8676 but I have a couple of open questions:
In the parametric search, the input voltage headroom for AD8675/AD8676 is given as 1.5V (typ) and 2.5V (max).
Under what conditions do I have to consider the max specification? Is it production dependent, temperature or something else?
If I simulate the headroom in LTSPICE I get 3.5V headroom which is even higher, also offset is higher than in the datasheet.
Is this model correct? There seems to have been an issue with it in the past:
Voltage Noise density in the datasheet is only plotted up to 30kHz. Can I assume that it stays at 2.9nv/sqrt(Hz) up to 10MHz?
I have tried ADISimPE to simulate the Noise of AD8676 but it seems that the SPICE model also does not match the datasheet.
Any input would be appreciated
"Under what conditions do I have to consider the max specification? Is it production dependent, temperature or something else?"
Yes its production dependent and temperature also. This is a bipolar input opamp, in which Vbe could move depending on the temperature.
"If I simulate the headroom in LTSPICE I get 3.5V headroom which is even higher, also offset is higher than in the datasheet."
I checked the model and it could have problems in terms of the input voltage headroom but I'm not seeing the 3.5V. So, I also checked the past thread link you mentioned about the spice model issue of this part, but can you attach a clearer image of your LTspice ckt on how you tested the input voltage headroom? I just would like to confirm if you are doing a similar testing on how we test our models for this parameter.
About noise, It's hard to assume if it's not given in the datasheet. The model has also problems in terms of noise model.
We are currently in the process of doing a revamp on our op amp spice macromodels in LTspice.
Thank you for your reply. I uploaded the screenshot with higher resolution, sorry for that.
I took the headroom as the input voltage where the difference between input and output starts to increase. This is at approximately V+ - 3.5V.
I then double checked the simulation with the AD8672 where the result corresponds with the typical headroom specification. But my interpretation of headroom might be naive.
I could not quite understand what you are trying to do here for headroom but we revised the model for AD8675.
Just update your LTspice release by clicking Tools->Sync Release.
The input voltage headroom is properly modeled in this new version,