Output range of an opamp

I have made an opamp and i would calculate its output range in spice. I have non idea how to do it in accurate and precision way without confusing output range with icmr. I've found as method an inverting configuration with feedback gain greater than one: with this method my opamp have a certain Vout range. If i measeure |Vds|-|Vgs|+|Vth|>0 for either output mosfet, i have a smaller range. How can i calculate Vout range? Thanks in advance

Reply to
lionelgreenstreet
Loading thread data ...

Output range is a pretty fuzzy spec. However, the measurement should never depend on knowing the voltages across internal devices. In theory, a customer could do incoming inspection (not done much anymore) on your black box of a device, so no way would they know what is happening inside. A well written datasheet would have the exact test configuration, but things aren't what they used to be.

What I've seen done on ATE is put the op amp in an inverting gain of one configuration, apply the datasheet limit at the input (registor leg), and measure THD at the test frequency, In ATE, you want go.no-go tests. Assuming you have a tester with DSP, the THD test is pretty simple. I never worked anywhere where DC performance was an issue, but they may do something different.

For characterization, you can measure full power bandwidth, that is, sweep the frequency and find the output swing at a specific distortion level. The AP can do this.

For design, why not just make sure the vdsat of the drivers is low enough for your spec under all temperature and process conditions?

Reply to
miso

Thanks for the infos.... What do you think about this method (for simulation)?

1)Open loop opamp with Vin-=0 and Vin+=Vdc 2)dc sweep of Vdc 3)Vout range is the condition where all output transistor are in saturation range (is this right?); so if i consider this condition: | Vds|>|Vgs|-|Vth|==>|Vds|-|Vgs|+|Vth|>0. 4)plot this curve for any output transistor in dc sweep range and find zeros 5)Voutrange is the region where this condition is met for all transistors Thanks in advance
Reply to
lionelgreenstreet

I don't like any test that depends on knowing the state of internal devices. In life, you design something, and then you test it on the bench. I can't see a test that requires knowing the state of internal devices being useful since it can't be compared to the physical device characteristics.

If you don't like my AC tests, assuming you have the common mode range on the input, put the amp in a follower configuration and come up with an acceptable softening of the DC as you reach the rails, subtracting out offset

Generally you design for DC by looking at operating points under worst case conditions, generally the strong file at max and min temperature, and then the weak file at maximum temperature.

Reply to
miso

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.