quality control charts

#1
Hello,
I try to understand the relation between average run length and sample time of an quality control chart.
I think about the following problem:
I should choose an exponentially weighted control chart with a predefined weighting factor. In the fault free case there should be an average run length of e.g. 371. Now there is a shift in mean which can be detected after e.g. 200 samples. This means, after 200 minutes (with sampling time of 1 minute) the fault will be detected.
So far so good.
My Signal recorded originally with a sample time of 1 second. If I use a sample time of 1 second, the fault will be detected after 200 seconds?
Is it allowed to reduce the sample time to 1 second?
Is there a limit for such a reduction?
Thanks for Help, Ben
 

Miner

TS Contributor
#2
You will have better success getting an answer at the Elsmar Cove Quality Forum. There are a lot of SPC (Statistical Process Control) experts there. While I have used many control charts over the years, I do not have experience with the exponentially weighted moving average control chart. It is rather specialized, and is not often used.
 
Last edited: