Siemens: copy specific word address in DB

Good evening

Hope I can get a little help with this one

When I scroll true my DB and find a certain value, I would like to copy the actual address in the DB to a tag, is there a function I can use or is more simple then that ?

Can you give a specific example - I'm not clear what you are asking.

Sorry LD for the late reply

I was trying to spool down true a DB to find a specific value and then transfer the DB address of this value to a tag.

Well I will probably have to step-up and use indexed addressing anyway to do the Search, so when I get that right I will have the pointer as my value.

First I will have to fill the DB,
so I am working on a routine that will insert data into the next available address in the DB until its full.
This must have been done a million times before, Do You know of any sample project that I can have a look at ?

Thanks
/Holmux

Are you filling the DB with a fixed value or a different value for each location?

Hi LD

I am feeding the DB with different values, I have a DB with DINT where a fix't part is filled with different values.

Out of these values I need a Minimum, Maximum and Average,
I have found a way to sort the values so I have minimum and the first address and Maximum at the last address, (my first day with STL, so I have to cheat where ever i can )

And i found a code for generating an Average of the data area, so in theory a am able to put my project together, the last part I am working on is combining the two codes to safe a loop, maybe I am going to deep in my surge for scan time

Here are the codes I am trying to combine:

//Average
L P##values_to_average
LAR1
L W [AR1,P#2.0]
T #elem_count
L W [AR1,P#4.0]
T #db_num
OPN DB [#db_num]
L D [AR1,P#6.0]
LAR1

L 0
T #sum
L #elem_count
_loo: T #loop_temp

L W [AR1,P#0.0]
L #sum
+I
T #sum

L 16
+AR1

L #loop_temp
LOOP _loo

L #sum
L #elem_count
/I
T #RET_VAL

//Sorting
OPN #Data_DB //Open Data Block
M002: LAR1 P#DBX 0.0
SET //LET #Sort_done = TRUE
S #Sort_done
L #DB_Length
L 1
-I
Loop: T #Count //FOR INDEX = Count TO DB_length
L W [AR1,P#0.0]
L #Average // load sum of all previous elements
+D // add these two together !!!
T #Average // store it in the sum
L W [AR1,P#0.0] //IF M(INDEX) > M(INDEX+1) THEN
L W [AR1,P#2.0]
<=I
JC M003
// change M(INDEX+1) for M(INDEX)
L W [AR1,P#0.0] //LET TEMP = M(INDEX)
T #Temp
L W [AR1,P#2.0] //LET M(INDEX) = M(INDEX+1)
T W [AR1,P#0.0]
L #Temp //LET M(INDEX+1) = TEMP
T W [AR1,P#2.0]
SET
R #Sort_done //LET #Sort_done = FALSE
M003: TAR1
L 16 //NEXT INDEX
+D
LAR1
L #Count
LOOP Loop //End of sorting pass
AN #Sort_done //If sorting is not done
JC M002 //Jump Conditionally to M002
L #Average
L #DB_Length
/D
T #Average
BE

Your code is summating swapped data so the average will be incorrect. No need to sort the data - set up a min/max and compare as you summate.Your code is also operating on integers, not DINTs as your post specifies.
Hint - prefix your variable names with their data type to minimise coding errors.

Here's my implementation. I've calculated the average in a separate network to make it easier to monitor.

I've posted source code that can be cut/pasted into a STL source code container and compiled to generate the block container.