AS RT PO for failsafe I/Os

Dears,

I need to configure a PCS7 system with the following configuration:

CPU 410H

500 Failsafe digital Inputs (redundant)

500 Failsafe digital outputs (redundant)

One OS Single Station

One ES

I know I have to get the S7 F Systems RT license to use failsafe CFC blocks.

My question is: How can I calculate the required AS PO, OS PO and Expansion Card PO?

Hi AMAZINGAHMED

In general you can use the normal PO calculation guidelines and estimate your logic.  How are the licenses for the AS and OS counted in PCS 7 V7.0 onwards?

Extra considerations related to safety:

Do you plan to use Safety Matrix or just create your own custom yellow code in CFC?  If using Safey Matrix and enabling alarms for causes/effects, then each cause/effect with messaging enabled counts as a PO, in addition to the matrix.  So don't forget to count those in the calculation.

If you are doing just your own custom logic, then you'll have to default back to the manual counting as defined in the link.

AS SEC PO just needs to be unlocked higher than your AS PO count.  So you if you calculate 1150  AS RT PO for example, then you need a AS with AS SEC PO unlocked to 1200 per controller.  (so you can order AS 1000 bundle and add 200 extra per each controller - if redundant this means you need to order 4x 100 adders)  Or you can order AS 1600 bundle and have more spare capacity.

Thank you, RVS.wick. Your answer was very helpful, yet I need more clarification, so I'll try to ask the right questions:

I'll use an example to understand better.

In a PCS7 project, I have a blow off valve (F-DO), and I have 3 pressure transmitters (F-AI). I need to open the valve if 2oo3 pressure transmitters values are high high. I need to display these on an OS.

Case 1: Using CFC safety blocks

• I use one F-DO and three F_AI channel drivers (0 AS PO - 0 OS PO)
• I use AND/OR yellow blocks to connect input to output (0 AS PO - 0 OS PO)
• I can use 1 DIgMon and 3 AnaMon blocks (0 AS PO - 4 OS PO) OR transfer the individual tags manually to OS (0 AS PO - 1 OS PO)

Case 2: Using Safety Matrix

• I use one F-DO and three F_AI channel drivers (0 AS PO - 0 OS PO)
  
• I use safety matrix to define the cause/effect with messaging (1 AS PO - 0 OS PO)

Are the above correct? I feel I have a few mistakes. Could you please advice?

Based on my calculations I think your numbers are slightly off.  I remember having a lot of trouble estimating early on as well.

For my own experience and some historical word of mouth from various other experts I use these rules which can relate to the official method but is less accurate.  (It will not be 100% accurate, but it will be close enough to buy licensing on most projects.)

Using APL or APL like blocks: 1 controllable object = 1 AS PO +1 OS PO (APL object block like MonDiL, not Intlk04, or OR4)
Each SFC instance 1 AS + 1 OS PO +  1 OS PO for each 
+10% factor to OS PO for things like interlock blocks if you are doing BPCS programming. - This accounts for the OS tag basis factor in the published guide.
Baseline to add safety to a non-safety project is ~ 5 AS + 5 OS PO
Safety Matrix object ~ 2 AS PO + 2 OS PO
Each Safety Cause/Effect 1 AS PO + 1 OS PO

So for your examples using my estimation method results in:

Case 1: Using CFC safety blocks

• I use one F-DO and three F_AI channel drivers (0 AS PO - 0 OS PO)
• I use AND/OR yellow blocks to connect input to output (0 AS PO - 0 OS PO)
• I can use 1 DIgMon and 3 AnaMon blocks (4 AS PO - 4 OS PO)
  OR transfer the individual tags manually to OS (1 AS PO - 1 OS PO - this assumes you build a custom block to pass all the from the logic blocks data to OS, if you use the APL object blocks at all you will still have the original PO count on the AS side)
• +4 AS +4 OS PO for F-systems background setup by the compiler

Case 2: Using Safety Matrix

• I use one F-DO and three F_AI channel drivers (0 AS PO - 0 OS PO)
  
• I use safety matrix to define the cause/effect with messaging ( 5/5 baseline + matrix 2/2 + 1 cause 1/1 + 1 effect 1/1 should result in 9 AS + 9 OS PO)

If you actually build Case 2 in a totally blank test project the actual PO usage is 12 AS PO and 11 OS PO, so I think my safety baseline estimate is slightly off or there are other blocks I've just never accounted for which are maybe part of the non-safety compiler setup.  However it will scale up from there if you add more causes/effects to the same matrix according to the estimate methods.  

Maybe another expert can find the hole in my estimation but this is typically good enough for order of magnitude.  In my experience you never want to be right up against your PO limit because the customer/user will always ask for one more thing.