SLC Queue Shuffle

Hi there,

this is my first ever post on this site. Currently, I am using a SLC5/05. I would like your recomedations on the following:

I have 12 Clarifiers into a desludge queue. The Desludge queue is prioritised in the following order:
1) Reason 1
2) Reason 2
3) Reason 3
When the Clarifier enters the queue, the queue is reshuffled in order of priority.

My question is: which is in your opinion the best way of creating queue shuffles on AB SLC's (Indirect addressing,FIFO, etc).

The way i have done so far it works but it is too long and untidy.
Your opinion and your help is appreciated.

Care to show us what you are doing now so we have someplace to start?

Queue Shuffles

Hi there,

All i am doing at the moment is just compares, file copies and moves. I have 236 Rungs at the moment on that file. I bet there is faster way of doing this. For some reason I am currently struggling to send you a copy of what i've done so far.

tol-tol
Lets see if I have your system right
One desludge system
12 clarifers (what ever they are)
When a clarifer is used it needs to get cleaned, only one clarifer can be cleaned at a time, depending on the clarifer usage (reason 1,2,3) determines the order of precedence of cleaning

Me, I would use 3 FIFO queues of length 12 each (12 clarifiers)
One FIFO queue for each reason
If Reason 1 is empty then select from the queue for reason 2

For short queues like this: I use a indirect addressing to load and a COP to unload (you could use FLL FLU)
For Long Queues I use indexed read and write pointers in a circular buffer.

Sorting the buffer - I keep away from
MG

Hi,

When I do clarifier, filter or membrane unit queueing I set out the logic as follows:

1. If new sort is required, Copy up to date clarifier data to a working area for sorting.

2. Make a single bubble sort pass through the working area, using indirect addressing.

3. If no swaps were made in last pass data is now in order, Copy working area to sorted clarifier data area, set flag for new sort required.

4. Wait for the next scan...

Why like this?

1. It's simple and the code is short.

2. Doing a full sort each scan on an SLC, particularly with the overhead of indirect addressing, can have an adverse affect on scan times.