Resistance to Instrumentation

Resistance to the Effective Use of Instrumentation III: Over-Design

Here is the latest in the Resistance  to the Effective Use of Instrumentation series by Water Industry Process Automation and Control’s (WIPAC) Oliver Grievson:

In this week’s continuation of the blog series on the resistance to instrumentation I am going to talk about a relatively sensitive subject that most people wouldn’t touch on and that is the over-design.

In all my time in the Water Industry I have found numerous examples of overdesign as I am sure that everyone in the Water Industry has. I will restrict this blog to a few examples that have either resulted in the instrument being switched off or unused or a whole control system that was put in at great expense just simply not being used.

The first example that I will use is that of an instrument which in this case was multi-parameter and thus a good investment. The interface with the operator on site who of course needed to undertake calibration routines was basically a personal computer in a box, the instrument itself was incredibly sophisticated and was very useful. The calibration routines were multi-layered and if continually repeated would result in the instrument being calibrated to the point that its operating range was close to being correct and the response very poor. This is surely a case of incorrect operation but the instrument was so hard to calibrate that the poor person who had to calibrate it at its routine frequencies in the end had to give up. That particular instrument was removed from service and replaced by a more operable instrument.

The second example actually featured in the first Question of the Week although in a veiled manner, and it was the case of a multi instrument control system in an activated sludge lane multiplied across the four or five lanes that the treatment works had. The system was based upon the advanced control of the aeration system by mapping the ammonia concentration across the system. This resulted in the treatment works having approximately 20-25 ammonia monitors all in place. The theory being that with a correct map of the ammonia across the lane that the aeration system could be ramped up and down with differences in ammonia concentration and thus saving aeration costs. The actual practice of course is that that amount of ammonia probes is very difficult to maintain and the control system is eventually doomed to failure when the operational staff can’t maintain that number.

This second example is often repeated with dissolved oxygen probes and is especially a problem in theUKwhere staffing levels aren’t as plentiful as in other countries. On a treatment works where there is either multiple activated sludge plants, or multiple lanes or multiple instruments per lane, resulting in numerous probes (off the top of my head I can think of at least) several plants with over 20 DO probes on sites ranging from 100-600,000 population equivalents. On none of those sites were the DO probes maintained properly because of their sheer number. The actual probes were working which is a testament to the design of the instrument manufacturer.

The third example is one of data blindness and refers more to the control system and how it can get out of the control and refers to the integration of the instrument into telemetry or SCADA. This example comes from my dim and murky past when I was accessing a site remotely through SCADA to respond to an emergency. I was confronted with the trends for the site that numbered well over 50 in number. It was a big and complicated site but in an emergency situation for the life of me couldn’t locate the trends that I required for a good few minutes. An operation that should have taken 10 minutes took close to ½ hour. The error in this example was that there was too much data & not enough information. An operator responding to that example would have been more practised but when it is three in the morning and he is tired all of the lines start to blur

From this there are some factors to pick up for the engineers among us (and I include myself in this)

  • What data, information or operation do I hope (as an engineer) to get from a particular instrument. Does this help the operator, maintenance engineer or company to do its job of running the plant.
  • Is the instrument that is being installed operable, can a calibration routine be conducted within a matter of two to three minutes with the right equipment (i.e. a bucket of known mixed liquors or a handy supply of oxygen).
  • Is the system maintainable for the facility that you are designing for? Don’t install multiple instruments at a facility that is not going to be manned or only manned by a few people. In this case go for instruments that needs minimum maintenance but may cost a bit a little bit more. The trade off of high CAPEX –v- Low OPEX will pay dividends in the end.
  • Make any system of process control & automation easy to operate, and easy to understand. The works will end up in manual at some point in its life and the operator will need to be able to run the works.
  • When integrating instrumentation into the works operating system (for large works) ensure that there is a balance between what is on site and what is needed for operation and what can be transferred to a centralised control, i.e. tailor the data & information for the different type of people within the water company. The operator, manager, engineer and chief executive of the company need different types of data/information, which is different to what is needed on site, in the office or in the boardroom.

In the next blog I shall be discussing further about the poor use of data and its management and what is needed data, information or both


About noahmorgenstern

Entrepreneurial Warlock, mCouponing evangelist, NFC Rabbi, Innovation and Business Intelligence Imam, Secular World Shaker, and General All Around Good Guy


One thought on “Resistance to the Effective Use of Instrumentation III: Over-Design

  1. Oliver,

    Terrific examples! I’m sure you could have given dozens more. One cannot over-emphasize the importance of obtaining the right data over the right time interval, along with effective synthesis of the data to yield valuable information for decision-making. This is no easy task, and many engineers simply aren’t cognizant of operational needs, so we see the common practice of over-designing unit processes to compensate. There is a drawback of increased capital costs, but everyone desires to have a bullet-proof design. Of course, engineers are very good at designing unit processes, but some are not so good at process control. Here is where a combination of collegial work among the operational staff and instrumentation vendors is required at a minimum.

    Posted by John B Cook | February 1, 2012, 4:12 pm

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: