S7 315-2DP lags when PG online

[Nepiter]

Hi!

I have never cross this kind of problem..

I have kind of working program on 315-2dp cpu.
Program has been working nearly 20 years like this, but now customer needs some changes.
But.. I can't go online, because plc program "hangs".
When setting PG to online, status bits on vat table are flashing randomly, and program status remain same state on every bit, even they should change status.. untill I switch PG to offline, then everything is ok again.

There is profibus network configured correctly, couple of VLT's, and couplers, all ok.
And MPI- bus configured to HMI. That works too.
No errors on diagnostic buffer, no error leds..

Resources are fine to my eye, 1 PG, 1 OP, 12 basics, and online: used 1 PG, 1OP, 0 basics.
And I have tested with 3 different PG's, and via MPI and Profibus. Same thing.

Only thing what I noticed, all programs are located in FB's, even they do not need to be FB's. And they are called in ob1 with UC..
Could that cause this problem?

Or what..

 

[JesperMP]

When setting PG to online, status bits on vat table are flashing randomly, and program status remain same state on every bit, even they should change status.. untill I switch PG to offline, then everything is ok again.

How can you know that the bits are OK if you are offline ?
When you are online, what is the CPU main cycle time. How much memory reserve ?

 

[Nepiter]

How can you know that the bits are OK if you are offline ?
When you are online, what is the CPU main cycle time. How much memory reserve ?

Thanks for Your reply!

I just assume bits are OK, because system has worked 20 yrs, and after switching PG offline, system runs like it has ran 20 yrs.

When I try to monitor program, everything kind of freezes on that state. Except bits on PG, they are flashing.
By freezing, I mean, motors run, even they should stop, lamps remain same state, stop button and op- panel does not react, etc. until PG is Offline..

Cycle time is around 14-19ms and memory used 10%..

I apologize for my poor English expression..

 

[JesperMP]

When I try to monitor program, everything kind of freezes on that state.
[..]
By freezing, I mean, motors run, even they should stop, lamps remain same state, stop button and op- panel does not react, etc. until PG is Offline..

OK, the PLC outputs seems to freeze.
I remember vaguely having heard about that before, but cannot find a link to it.

I suggest update the firmware to latest.

 

[Nepiter]

OK, the PLC outputs seems to freeze.
I remember vaguely having heard about that before, but cannot find a link to it.

I suggest update the firmware to latest.

Yes, I was thinking about something like that too, thank You for confirming!

It is funny, if motor is not running, I can go online, and i can see bit status correctly.
But when motor starts, and I am online in that FB, which controls that motor, it freezes.. I can be online in another FB, but not in that, which controls running motor..

 

[JesperMP]

Are you using 'monitor with look-up path' ?

 

[Nepiter]

Are you using 'monitor with look-up path' ?

..No, just normal monitoring. That's something new to Me..

 

[Nepiter]

Ok, I just googled 'monitor with look up path'..

I think there's no need to use 'monitor with look-up path' function..

These FB's are not used like they should be.
FB's are called just once at ob1, by UC call.

 

[JesperMP]

If you dont use monitor with look-up path, then you will not see an individual instance of the FB, but random values from all instances.
That may explain some of the weirdness you are seeing.
It doesnt explain the freezing of the outputs though.

If you did use 'monitor with look-up path', you would have to activate 'test operation' first, and that could possible affect the CPU load, and that could possibly have to do with the weirdness. But .. since you do not then that is a dead end.

 

[Nepiter]

If you dont use monitor with look-up path, then you will not see an individual instance of the FB, but random values from all instances.

Just to clarify things, that I understand what this mean:

If I create FB, which needs db to work individually, I need to monitor that with monitor with look-up path?

But in this case, FB does not use db at all. FB is called in OB 1 like this: UC FB7 (no instance db call ).

At FB7, there is no parameters at FB interface, it has only networks programmed, with global variables and IO's.

But .. since you do not then that is a dead end.

I checked from CPU parameters, and there is test mode selected. I have not chosen that option. Do You think that could cause the problem? If i remember right, most of CPU's which I have seen, is this test mode selected.

 

[adfox]

This is copied from S7 Help, take a look. Test mode has a very significant impact on scan cycle.

Irrelevant for S7-400 and CPU 318-2, all S7-300 CPUs with a firmware version V3.0 or higher and for the safety mode of an F-CPU with a safety program.

In process operation the test functions such as program status or monitor/modify variables are restricted so that the set permitted scan cycle time increase is not exceeded. Testing with breakpoints and sequential program execution (single-step mode) are not possible.

In test operation all test functions via programming device/PC can be used without restriction that may cause considerable increases to the length of the scan cycle time.

Important: If the CPU is in test operation, you must ensure that the CPU or the process can "cope" with large increases in scan cycle time.

By selecting the "Process Mode" option, you can prevent the test mode and the call environment from being selected in the program editor.

 

[Nepiter]

Thanks for the advices!!

Sorry for late reply, but I did not had a chance to try suggested tricks until now.

And problem kind of solved. CPU was faulty.

Upgrading firmware was not an option, because it was already 2.6.1, latest downloadable version.
Now there is new cpu, with version 2.6.6, and it's possible to monitor online, without lagging whole system.

The first "broken" S7-300 CPU, which I have seen in 25 years.
And some of 300- CPU's which customer has, are around 25 year old, and still running without problems..

..Actually, there's some S5- CPU's too, which are still running. Back to the 80's..