EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: Simon on November 16, 2017, 12:22:11 pm
-
So, I am looking at this nice little opamp. I'm checking the input offset just to be sure. So it's max 7mV at 25C, "typically" is varies by 7.0uV/C - don't forget the .0 just to clarify how significant that 0.1uV is - and then you see that actually at 85C it can be max 13mV, so you work out the actual max deviation with temperature and low and behold discover that next to that dead accurate 7.0uV/C (don't forget we are being accurate to 0.1uV) the max should say 100mV !!!!!! nice try ON Semiconductor:
http://docs-europe.electrocomponents.com/webdocs/14b2/0900766b814b2dcf.pdf (http://docs-europe.electrocomponents.com/webdocs/14b2/0900766b814b2dcf.pdf)
-
I wouldn't say they are bending the facts, rather that you are reading something into the specs that are not there.
The average temperature coefficient of the input offset voltage is 7 uV/C.
This means that the manufacturer measures the offset voltage at the highest and the lowest temperatures and divides the difference with the temperature span. It says nothing about offset voltage excursions in between, which can be larger due to non-linearity.
It doesn't really matter which manufacturer, the datasheets all read the same.
-
Yes i know all of the datasheets are pretty much the same. They omitted to fill in the max box and instead created a separate spec to say the same thing putting it in a different light which avoids saying that the temperature drift can be anywhere from 7 to 100uV, when you put those two number side by side you see the reality, when you just see 7mV at 25C and 13mV at Thigh you can assume it's the 125C or 150C the part is rated for not 85C
-
Datasheet states the offset is 5,7 or 9mV (depending on which particular chips you have) at Tmax. Period. That is the guaranteed value you have to take into account.
If you want a part with specified tempco of offset, you have to search somewhere else.