S7 - Is this another V5.4 funny?

When I installed V5.4 a few weeks ago, I didn't bother to look at the settings in the STL/FB/Ladder editor, so I wasn't aware that it had defaulted to "create source file on saving".

That, I discovered today, when I got the following fault screen when I tried to save. The funny thing is that I can switch between FB (which is what I'm programming in for this customer) and STL without any problems, which I assume means that the FC is compiling OK and that the problem has purely to do with the creation of a source file.



I'm not quite at the point where I can load and run the program to see if it actually works or not, but I should get there later this afternoon. In the meantime, has anybody seen this problem before?

If anybody can make anything out of the error messages, that is NW17 which is onscreen (the 1st error message), I had to squeeze things a bit to stay under the 750 pixel mark.

Is there any additional information when you click the "Hilfetext" button ?

Did you install SP1 for v5.4 ?

Any chance of knowing what those messages are in English ?

Is there any additional information when you click the "Hilfetext" button ?

Click to expand...

I didn't look at the time, although I guess it probably won't be much more than is shown in the window in the bottom part of the message window (with the exclamation mark on the yellow triangle).

Unfortunately, I've evidently modified something significant in the meantime, becase I can no longer reproduce the fault!

Did you install SP1 for v5.4 ?

Click to expand...

No, I meant to after seeing your comment a few days ago, but I couldn't find it on the Siemens web-site on a quick visit and haven't got back round to it since then.

Any chance of knowing what those messages are in English ?

Click to expand...

All variations on the theme:
"Reduction (optimisation?) of NWxx (don't know what the bit in brackets is, presumably the line number in STL or something similar) in Block FC13 failed."

In addition 4 of the entries, including the first and last have a comment in brackets saying:
"Abs.(olute?) l.(ocal) var.(iable)!"

By the way I'm now testing the program and it seems to run fine.

Link for V5.4 SP1 as follows:

http://support.automation.siemens.com/WW/view/en/23849953

Thanks Simon.

Is there any additional information when you click the "Hilfetext" button ?

Click to expand...

I didn't look at the time, although I guess it probably won't be much more than is shown in the window in the bottom part of the message window (with the exclamation mark on the yellow triangle).

Click to expand...

Sometimes that button (sometimes called "Details" in the english version) doesnt give any more useful information.
But quite often you actually get a lot more in-depth information to what may have caused the error. There may be a whole list of possible causes.

I've now found a condition when I can force the fault to come. It's not quite the same as before because under these circumstances it only happens once and then on a second attempt it saves OK.

This happens when I modify and save a DB, perhaps inserting a new line and therefore shoving other existing entries down a few bytes. If I'm then editing an FC and enter a parameter using the newly defined address, in the FC is still a copy of the original DB so the new address doesn't convert immediately to the new symbol (this was always the case in older versions of S7). Only when I attempt to store the FC does S7 update to the new structure of the DB and that's when I now get the fault. However, after clicking away the error message, the Symbol then updates correctly to the new entry and a second attempt to save then works OK.

As I said, with the original fault when the error came it was permanent, but I assume it was in some way related.