I know...wtf, people still use Access? Save the posts about something else, for reasons beyond dumb it is all I have.

I'm setting up a test bench where we want to record results of each test event. The test event will have things like test station, serial number tested, run time, overall result, etc. Each individual test could have pass/fail judgments based on one limit (>, <, >=, etc) or 2 limits, or it could be a string comparison, or a Boolean comparison, or a non-test logging of data only. We do data acquisition and I also want to store the raw samples (on the order of 100,000 of thousands) in the database with each test run. We currently only save .jpg's of the plots, which are compact (5-10kb) but we can't go back and zoom in to the result if we need.

I have no formal database training so what follows is probably stupid, help.

For the database, I'm thinking one table (Test Events) lists the test event with that pertinent information and a key field that points at the Test Results table. In the Test Results table, I'm thinking each row would be a description, and possible some other data (time result was logged, overall result P/F) and then another key to point at a table specific to that test type. If it is a 2 limit test it will point at a 2 limit table that would have a upper and lower limit, if a 1 limit test then the table would be a single limit.

Other than the fact that it is Access, which itself is an abortion of thought, am I on the right track?

10/28/2016 1:46:41 PM

Postgres is awesome and free.

pgadmin is great and you can plug Excel into postgres if you really want to dirty things up.

10/28/2016 2:00:34 PM

simple answer: you're on the right track. you'll end up having more tables like one for test station info, etc., but I think conceptually you're looking at unique key fields that identify the test-case and that's where you want to be.

pro-tip: add an audit date/time field so you know the original insert datetime of the record and an "active" field (int 1/0, char 'y'/'n', bit, whatevs) so you can eliminate bogus data without deleting from the table and you query it later.

[Edited on October 28, 2016 at 2:02 PM. Reason : WHY AREN'T YOU USING SQL/EXPRESS???....jk ]

10/28/2016 2:01:40 PM