SFB4 in FC Step 7

Martin Nielsen · Sep 28, 2013

[FONT=&quot]Hi All
I managed to use SFB4 in a FC using the IN_OUT interface, but I would like to improve if possible.

Can I determine where the "timer area" starts?

The code seems to work perfect, but if the UDT is changed I would like the offset to change as well with a minimum of code change.

I want to get rid of those lines in my code:

L P#14.0
L P#22.0

I prefer "L First Timer" or similar

My UDT

[/FONT]

Martin Nielsen · Sep 28, 2013

I had trouble uploading pic.

Here is my comment
Pic 1: My UDT
PIC 2: [FONT=&quot]My FC Interface with the UDT
Pic 3: [/FONT][FONT=&quot]My FC with first STL Code and then the ladder logic
Pic 4: [/FONT][FONT=&quot]My FC calls

And thanks to L D[AR2,P#0.0] since it was an old answer from him, to a tread regarding SFB4, that bring me in this direction

[/FONT]

L D[AR2P#0.0] · Sep 28, 2013

If you add the udt to the temp area, you can call another FC to determine the offsets.
I've only transferred the offsets to AR1 for monitoring purposes.

L D[AR2P#0.0] · Sep 28, 2013

Source for FC8

Code:

FUNCTION FC 8 : DWORD
TITLE =
VERSION : 0.1


VAR_INPUT
  pBaseAddress : POINTER ;    
  pElementAddress : POINTER ;    
END_VAR
VAR_TEMP
  dwBase : DWORD ;    
  dwElement : DWORD ;    
END_VAR
BEGIN
NETWORK
TITLE =calc offset (in pointer format) from base to element

      L     P##pElementAddress; 
      LAR1  ; 
      L     D [AR1,P#2.0]; 
      L     P##pBaseAddress; 
      LAR1  ; 
      TAK   ; 
      L     D [AR1,P#2.0]; 
      -D    ; 
      T     #RET_VAL; 
      SET   ; 
      SAVE  ; 


END_FUNCTION

Martin Nielsen · Sep 29, 2013

Thanks a lot L D[AR2,P#0.0], it works great.
It took me some time to figure out how you do it, but I think I almost got it now. You get the offset from the V-Stack?

L D[AR1,P#2.0] //
Are you reading point value in the V-Stack from the calling FC?

L D[AR2P#0.0] · Sep 29, 2013

Yes.

When you call a function with a pointer, the editor generates the absolute address of the variable to pass to the function.

I notice from your code that you are relying on the contants of the accumulators when you call your FCs. You should not do this but instead read the DB and area pointer from the V area.

Martin Nielsen · Sep 29, 2013

Yes I get this thought as well.
My code need to be in top to work.
I am also a little concerned if an interrupt appear, will I then could trust the content of the ACCU's ?

L D[AR2P#0.0] · Sep 29, 2013

Interrupts will not be a problem.

Martin Nielsen · Sep 29, 2013

How powerful this is.

I improved my code so I don't depend on the contents of my ACCU's as you suggest

Code:

      L     P##T_InOut
      LAR1  
      L     D [AR1,P#2.0]               // Read Start Address
      T     #Start_Address              // Transfer start address to pointer
      L     W [AR1,P#0.0]               // Read Instance DB No.
      T     #DB_No                      // Transfer actual DB to DB pointer
      OPN   DI [#DB_No]                 // Open actual instance Data Block

L D[AR2P#0.0] · Sep 29, 2013

...

SFB4 in FC Step 7

Martin Nielsen

Member

Martin Nielsen

Member

L D[AR2P#0.0]

Lifetime Supporting Member

L D[AR2P#0.0]

Lifetime Supporting Member

Martin Nielsen

Member

L D[AR2P#0.0]

Lifetime Supporting Member

Martin Nielsen

Member

L D[AR2P#0.0]

Lifetime Supporting Member

Martin Nielsen

Member

L D[AR2P#0.0]

Lifetime Supporting Member

Similar Topics