principle & function testing

[Archie]

Hello, what is the difference between 'principle & functional' testing?

[Peter]

Hello, what is the difference between 'principle & functional' testing?

PJW would be able to give you a better definition than me, but essentially, functional testing is checking that something performs at it is expected to in response to a set of control inputs, principles testing is making sure that it does what it does when it is supposed to do it. Another way of looking at it is that the functional testing checks that it has been built right, principles testing checks that the right thing has been built.

For example, functional testing of a set of points will make sure that the points move normal when the relevant relays are up, and when it is there, the right detection relays are in place. Principles testing would test that the points only move when the conditions are right, that other things cannot happen when the points are not in the position they should be.

[PJW]

Hello, what is the difference between 'principle & functional' testing?

Like many terms in railway signalling one has to be careful as they don

[Archie]

So what is the process of testing, with regards to the two different types of engineer, or can the same engineer do both or are both practices to be carried out independently by two different engineers?

[Peter]

So what is the process of testing, with regards to the two different types of engineer, or can the same engineer do both or are both practices to be carried out independently by two different engineers?

There is no reason that the two parts cannot be carried out by the same person assuming they are competent in both fields.

[PJW]

So what is the process of testing, with regards to the two different types of engineer, or can the same engineer do both or are both practices to be carried out independently by two different engineers?

Different IRSE licences but both can be held by same person and no restriction against doing both on the same project / site, though generally would be different people purely because of logistics.

It is a good question though; having "strap and functioned" a route relay interlocking, have you lost your independence to do the Principles? Actually I think not and indeed I am far more comfortable that way, but actually you do tend to do a lot of your "Principles thinking" when circuit testing.

Nowadays most testing of New Work is data driven and off-site and there is no "functional testing" of the interlocking itself. My practice is generally to do the first pass of the data testing largely without looking at the data, but then when waiting for the Test Logs raised from that pass to be actioned, start studying the data to make sure had been written as one had originally assumed to ensure had actually done sufficient testing. Effectively have to look at the data when get the 2nd pass of the data, in order to determine quite how much needs to be tested on that pass and what has been left unchanged from the first pass; have to generate a "difference list" to highlight the area of change. Unless there is some form of automated regresssion testing, do need to restrict the extent of re-test.

Yes independence is good, but only to a degree. A certain level of knowledge of how something has been achieved is needed to test it sufficiently in a reasonable timescale; if the interlocking is a totally black "black box" then how can anyone know what to assume about its working and therefore what is worth investigating and what can be taken for granted?

Given the number of inputs and therefre permutations which are possible, to say nothing of sequential logic and timers, one has to accept that it is completely impossible to "totally test" an interlocking. Need to focus on what it is worth doing and unless you understand something about its inner workings, then can't possibly do this. It is possible, particularly if you are an awkward pedantic sod like me, to maintain a healthy level of scepticism and hence independence of view even if not insulated from knowledge of the design.