If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

# OPA4314: Offset data in Figure 5 in the datasheet

Part Number: OPA4314
Other Parts Discussed in Thread: OPA2990
Hello,

My customer has some questions for Figure 5. Offset Voltage Production Distribution in the datasheet(SBOS563G).

Q1
This graph shows data in a range of +-1.4mV while max offset is defined as 2.5mV.
They want to understand the meaning of this graph.
I think you just evaluated a certain amount of devices and all devices were within this +-1.4mV.
How many devices did you evaluate to make this graph?

Q2
What percentage of the total is showed in the range of this graph in our actual production?

Regards,
Oba
Q1
This graph shows data in a range of +-1.4mV while max offset is defined as 2.5mV.
How many devices did you evaluate to make this graph?

Q2
What percentage of the total is showed in the range of this graph?

• Hello Oba-san,

Here are the replies to your customer's questions.

Q1:  The purpose of this graph is to give a meaningful representation of the distribution of offset for this part.  The number of devices evaluated will be large enough to be statistically significant.  In the cases where the number of tested devices is included in the figure, I have seen this number be quite large.  For example, the data sheet for the OPA2990 has the number of tested devices in figure 1 of the data sheet and more than 15,500 device results were included in the figure.

Since the maximum value of offset for this device is 2.5mV (at tested conditions), all units will have an offset of less than 2.5mV.  However, this does not mean that an offset of 2.4mV is likely.  In fact, it is quite unlikely.  The typical offset for this device is +/-0.5mV.  This corresponds to a one-sigma value.  So about 67% of all devices will have an offset that is 0.5mV or smaller.  If we multiply by 3, we get a three sigma value of +/-1.5mV.  So, the probability that a device will have an offset that is less than +/-1.5mV is 99.7%.  So, it is not unreasonable for the plot to be cut off at 1.4mV, since the probability of a device beyond that region is quite small.  To be clear, it is not impossible but small.

Q2:  I am not sure that I understand this question.  Is the customer asking about the total devices that have ever been made?  The data will be taken at device characterization prior to RTM.  Then, the data will be put into the plot.  Again, it is possible that there are devices with an offset greater than 1.4mV.  This is why the maximum specification is 2.5mV.  However, the possibility is very small.  Thus, the plot is cut off to give greater granularity to the regions of higher probability.

Regards,
Daniel