Does Index parameter control execution order of loops?

[comcon]

I'm configuring a complex combustion control strategy using an SRx
IFC.

There are up to 5 cascading control loops.

I'm pretty sure that the "Index" parameter of loop points control the
execution order of these points, but nowhere can I find this for
sure. Some DCS platforms (Honeywell TDC for example) go in reverse
order, so that higher number "index" points execute prior to lower
ones. Most other's go in direct order.

The customer has just always let this index be auto-assigned.

If however they were in the wrong order, at a one second scan and a 5
level cascade loop, it would introduce a 5 second lag in the overall
control response.

Does anyone here now the answer to this?

Thanks,

Chris Henderson

[Tim Alosi]

To the best of my knowledge (using Provox for 20 years) Index does NOT
guaranty execution order. It is a memory location within the
controller.

I will have to dust off my Provox manuals, because I forget exactly
where to configure it, but you can specify related loops (such as
cascade loops) in Envox and that will link the execution order.

Similarly, using input and output LCPs will ensure that the Input LCP
executes before the Loop with the Output LCP executing after.

If you have a complex strategy, using input and output LCPs help
ensure execution order.

I'll post the cascade info when I dig it up.

Tim

[omcamp]

Hi, Tim,
Chris,

In my 20 plus years of PRoVOX I have found that the initial
distribution of points to the 0.1 second ticks is in the index
increasing order (at least mostly) on an initializing total download.
But later partial downloads can change the order of point execution.

If you must have 5 levels of cascade you can split them into two pairs
of primary/secondary and find the correct tab (Pseudo FST?) to link
them. Then they will get executed in three, probably separate, 0.1
ticks. I would expect between .6 and 2 seconds of deadtime from loop
execution. (See the manual for IO timing.)
In my complex combustion control, each fuel and each air stream (and
furnace pressure where needed) had its own loop with input (and output
-where needed) LCP for PV calculation and output splitting and
characterization. We could not calculate setpoints in the loop LCP
without overload issues with the SR90 IFCs, so they were in an air
setpoint LCP and a fuel setpoint LCP.