By Greg McMillan and Stan Weiner, PE
Stan: The title of the June puzzler, Why Did the Electrode Change When it was Inserted?, posed a different problem than the puzzler, which asked, why did an electrode in a recirculation line show a change in vessel pH before an electrode inserted into the vessel? Since they are both good questions that generated a lot of interest, we are going to take it easy for a while and let some of the readers do the talking. To leave room for a Top Ten list, we will include just the first four replies, which were from Eric Marcelo, Bruce Seaba, Phil Spray, and Gary Steinbaugh.
Eric: I don't know what the exact configuration of the system is, but it might be similar to my experience. A neutralization tank had a slow turning mixing propeller. Acidic influent was injected from the side of the tank in such a way that it countered the natural swirl created by the mixer. This had the effect of negating the swirl, such that the tank's contents were practically being mixed super slowly. The caustic feed was fed from the top and, since the mixing was super slow, it took a long time for the neutralized mix to reach the bottom of the tank where the pH probe was. They wondered why the pH suddenly went from acidic to caustic in seconds. They decided to install a sampling line and moved the pH measurement there, along with the caustic feed for neutralization. Unfortunately, they put the sampling inlet close to the sampling return. Now, they wondered why the manual pH reading at the tank's outlet was different from the in-line probe. After a couple of piping changes and installation of a faster mixer (and baffles) they finally got things working.
Bruce: The electrode in the recirculation line is probably part of the pH control loop downstream of the acid or base addition (maybe downstream of the recirculation pump for better mixing). This means that the pH of the vessel is being adjusted to that of the recirculation loop. It could take a significant amount of time to bring the entire tank volume pH to the pH of the flow being recycled (the entire tank volume would need to be turned over). Another possibility is coating of the element within the vessel due to no flow or lack of agitation could also impair the response within the tank. I experienced this while working at a large chemical manufacturer in the cornfields of Iowa.
Phil: The vessel was most probably not well agitated. It may have been that the recirculation line was the agitator. If the location of the vessel electrode was in a dead spot of the agitation the recirculation line would show a change before the vessel would. It would be even more pronounced if the reagent that caused the pH change were fed into the recirculation line, as it most probably would be if the recirculation line were the agitator.
Gary: Most venerable personages: There are three, and only three, possible answers to this question:
- pH is intrinsically wicked.
- All the electrode's KCl leaked out, but "pH at recirculation electrode" was misread as "phat recirculation electrode," so a bad sensor was never considered.
- The turbulence in the recirculation line caused better mixing of the reagent with the process fluid, thus a faster response.
My money is on 1.
Greg: Even if the vessel was classified as well mixed, the bulk velocity is less than 1 fps second unless the electrode is perilously close to the impeller tip, which would be OK if your setpoint was 7 pH and you wanted to sit phat dumb and happy with a chopped electrode. The velocity in a pipeline is usually 5 to 7 fps which means the electrode response is several times faster. The higher velocity helps keep the electrodes clean and avoids the disastrously slow responses from even thin coatings on the glass bulb. If you throw into the mix the fact that reagent added to the suction of the recirculation pump creates a fast cheap inline system that gets around the problems of reagent short circuiting, back filled reagent dip tubes, and dead zones in supposedly well designed vessels, you are way ahead of the game with a pipeline electrode.
Stan: This assumes the insertion electrode is located relatively close to the pump discharge and the process fluid is nonabrasive and conductive.
Greg: The reading will have some fast noise but this is easier to filter out than the slow junk associated with vessel installations. The velocity should be kept fairly constant and below 8 fps.
Stan: Working on pH applications could be an addition to the following Top Ten Reasons not to Retire from Glenn Mertz.
Top Ten Reasons not to Retire
10. You would really miss those two-hour staff meetings every Friday morning.
9. You would have trouble finding how to use those hours you spend on your commute.
8. You can't bear the thought of not starting out your day with a delicious office cup of instant coffee under flickering fluorescent lights with the crackle of office sounds.
7. You would have too much trouble finding reasons to use PowerPoint.
6. You could not live without an annual performance review.
5. Your fitness would be compromised by missing the strength training of carrying your laptop all over the world.
4. It is too difficult to figure out what you would do with your office wardrobe.
3. You are afraid of withdrawal symptoms if you can't spend eight hours a day on the phone and doing Email.
2. You don't like the idea of having 52 weeks of vacation every year.
1. The company canteen vending machines have the best cold hot dogs in town.
This Month's Puzzler:
The Case of the Deadly Design?