Hi, I am doing a small experiment about historian deadband/compression and I am trying to see if there is a real Ignition use case worth looking into.
The question is basically: are there cases where two different original signal histories end up generating the same stored historian points, but would give different answers to a question somebody later wants to ask about the original signal?
For example if they asked how long a value was above a certain threshold, or if it was continuously above a threshold for at least a certain amount of time.
I have tested this already in a synthetic example but I don’t want to just make another one, I want to see if this could matter in a real historian workflow.
I’m looking for an anonymised example from an analog tag where there is some way of comparing the timestamped values before historian deadband/compression is applied with the raw points actually stored in the Ignition historian. I’d also need the relevant history settings, such as sampling mode, deadband mode/value and any minimum or maximum sample timing. An important part for me is also having a real question that somebody actually needs to answer from that data.
I’m not trying to find a case where the historian has definitely lost information; even an outcome showing that the stored data is sufficient would be helpful.
I don’t need an entire plant dataset; even a short period from one anonymised tag could be enough. I have no need for company information, plant names, or anything commercially sensitive.
Has anyone here done any comparison of values seen before historian deadband with the stored history that comes out, or done a short parallel capture for testing? I would also be interested to see any examples of retrospective questions you routinely need to answer from historical process data.