In response to our control loop special, reader Richard Compére believes that loop optimisation is often a case of too little, too late. He has observed increased divergence between the disciplines involved. Richard writes:
"The tendency for these disciplines to work as individual entities rather than to work as a whole has become very much the norm, and in my opinion has led to many 'problems' when it comes to commissioning and eventual successful running of the plant.
The lack of professionally 'experienced' people within these disciplines is also of huge concern. I read with interest Richard Mackay's problem of trying to tune a density loop. Let us do a little imaginative role-playing here, and we will move backwards from the commissioning stage where Richard was having a problem. The first 'problem' might be that when the guy doing the commissioning in the field goes to the output of the I/O card, even before it gets to the I/P on the valve, he finds that the signal there is 'wrong'. After checking, it is found that no internal software scaling has been done on the output of the PID loop within the control system. Once that is corrected the second 'problem' then emerges. Now for 0 to 100% output on the controller there is a corresponding 4 to 20 mA output on the output card, which drives the valve from fully closed to fully open within a few seconds, flooding everything in sight, and causing even more consternation. Now the third problem sticks its ugly head out: the software guy has to try and set up the PID loop to control the valve, which cannot be done because the valve is hopelessly oversized. When the instrument engineer is then asked why such a 'big' control valve was used or specified his reply is that that is where the process engineer wanted it on the P&ID and that was the line size, yet two feet from the valves outlet the line reduces anyway (P&IDs are not HAZOPED anymore). Now we are back at the previous point and we sit with a loop that will 'work' but never in its lifetime control efficiently."
Richard then mentions the 'jobs for pals' - 'it is not what you know, but who you know...' factor as a significant contributor to the level of incompetence that is evidenced in such instances. He continues:
"...the indirect consequence of this is that the commissioning process is put under undue pressure to meet deadlines and that instead of loops being commissioned correctly, they are set up to 'work' and are never ever really looked at again.
So when somebody writes an article saying that over 80% of loops are inefficiently controlled, I can well believe that, but I do not believe that going back, after the fact and trying to rectify these problems, is the correct approach. I believe, we as engineering or project houses, should take a long hard look at how the project is engineered from the outset and ensure that when the loop is designed, engineered and commissioned, it is designed, engineered and commissioned correctly, the first time."
Quite right Richard - in an ideal world we should not need to go and fix things up after they have been done properly. But, besides pressure from competing plants (if any), is there any other mechanism in place to encourage companies to both take on - and appropriately reward - the 'right person for the job'?
Some are complaining that the affirmative action legislation has been making it difficult to employ experienced, qualified people. We hear tales of experienced, qualified people having to emigrate to find employment. Whether this is as severe as the migration of our nurses, I do not know - but I believe that this will improve as more non-whites emerge from the tertiary institutions and gain experience in industry. This is an incredible adjustment we are making in SA.
I believe that there will always be a measure of incompetence in any industry, so there will always be a need for the 'panel-beaters' of the control system world to occasionally come in and elevate things to a commercially viable state. There other factors too - like plain bad management and top-heavy organisations. But all is not bad. SA industry has some fine examples. We still have great potential that is not being tapped.
Thank you Richard for your letter. Perhaps somebody out there could come up with some good ideas that would help speed up the 'sharpening of SA's industrial pencil'?
German trip
Last month I had the privilege of touring some parts of Germany (with Siemens) and visiting a number of Siemens manufacturing plants - as well as an impressive automated warehouse. If there is one thing about the Germans that struck me the most, it had to be their fanatical drive to 'do it right'.
The Hannover industrial trade show was immense. The Siemens stand was a live cool drink manufacturing and bottling plant. Capable of producing hideous quantities of product, the plant was deliberately run very slowly so that visitors were able to drink the product as the bottles came off the end. Russia had some impressive stands. One stand had a working scramjet on display. Other halls included MAGLEV technology and superconductors. See 'Siemens' Hannover Messe and German tour 2005' to see some pictures from the Siemens tour.
John Gibbs, editor
Send your letters to the Editor, [email protected]
Do all your I&C shopping at www.ibg.co.za
Do your fluid power shopping at www.mcbg.co.za
| Tel: | +27 11 543 5800 |
| Email: | [email protected] |
| www: | www.technews.co.za |
| Articles: | More information and articles about Technews Publishing (SA Instrumentation & Control) |
© Technews Publishing (Pty) Ltd | All Rights Reserved
printer friendly version