For example, if a 1k ohm resistor has a quoted temperature coefficient of 5 ppm / C, does that mean it will not vary more than 5 milli-ohms per degree C?
No. It could mean that it is likely to vary by somewhere around 5 milli-ohms per degree C, unless the tempco is said to be a minimum. The datasheet should indicate whether it is specifying expected behaviour or likely worst behaviour. For some applications a designer might want to know that it won't vary by less than, say, 3ppm/C.johngalt47 said:
Sometimes the spec is actually +/- ppm/deg C.
See this, for example. http://www.vishay.com/docs/31018/cmfind.pdf
What I was trying to understand is what the ppm stands for. I understand the acronym means "parts per million" but what does it really mean.
The 'ppm' bit you got right in your first post. For a 1k resistor 5ppm means 5m-ohm.
The issue then, which is what I thought you were asking about, is whether that is a prediction or a limit. So 5ppm/C could mean:
1. the value will not change by more than +5mohm or -5mohm for 1C - zero change is possible.
2. the value will definitely change by around +5mohm for 1C- zero change is excluded.
The issue then, which is what I thought you were asking about, is whether that is a prediction or a limit. So 5ppm/C could mean:
1. the value will not change by more than +5mohm or -5mohm for 1C - zero change is possible.
2. the value will definitely change by around +5mohm for 1C- zero change is excluded.
- Status
- Not open for further replies.