This thread starting me wondering - so I did a test mash today in a 1 quart drink cooler.
All mineral and acid measurements were weighed on a gram-weight scale, and the grain to the nearest tenth of ounce. I mashed in first with RO water - intentionally added the tiny mineral adjustments to the mash, along with a small acid addition. I was attempting to strike at 156F with a pH of 5.4, and a 1.5 qt/lb water/grist ratio. After a few minutes of stirring, visually well mixed with no dough balls, I sealed the vessel and gave it a good shaking. Letting that sits for a few minutes, I took some pipettes and drew 5 samples into shot glasses from various levels in the mash, as well as measured temperatures in various levels and areas of the mash with a NIST traceable digital thermometer. I was frankly very shocked at the distribution of values. The pH varied from 4.8 to 5.7, and the temperature varied from 166F to 150F.
I decided to stir again vigorously, this time with a whisk. I whisked for at least 3 more minutes, which caused a temperature drop. I added hot water to bring up the temperature to about 1.75 qt/lb ratio, which stabilized the mean temperatures. The pH measurements continued to range from 5.0 to 5.6. I closed up the vessel and let it sit for another 30 minutes. At that point, I checked temperatures, which only varied about 1-2 degrees between locations. I only took 3 pH readings from various levels in the mash, which ranged from 5.4 to 5.55. However, conversion tests (iodine on paper towels) showed conversion was not complete at that point (about 40-45 minutes into the mash). At 60 minutes, I checked the mash again, drained the mash into a pot and the first runnings were right at 5.43, and right onto my expected gravity of 1.058. I should add all samples were cooled to around 70F.
I now wish I took better notes about where in the mash each sample was taken, and a more scientific approach. I did not write down all of the measurements - nor could I take a temperature reading at the sampling location accurately. So no precise repeatability, and of course, a single set of data. Take the above with a grain of salt. Also - the iodine test is not terribly reliable, and I did not test the conversion samples on a refractometer.
My point? It seems that static mashes are highly susceptible to stratification of temperatures (which I knew) but also pH, and I assume concentration of enzymes or their activity. I am wondering if some of the inconsistency we generally see in mash readings are from readings from a mash site that is not representative of the entire mash, either concentrated or diluted. I now wonder where the ideal sampling location might be - if there is one at all, perhaps a vorlauf-like procedure in the middle of the mash with a larger wort sample? I got a consistent reading on the runnings of this mash as it was collected and homogenized. I brew on a RIMS system, and never really worried about liquor mineral distribution as I generally pump pretty quickly with a course crush. Martin's comment about acid staying on the top of his RIMS mash is what triggered me to check. While not scientific - my observation seems to uphold Martin's assertion of treating the strike liquor before mash-in to maximize the distribution of minerals and acid to the grist, but also support that mash-in requires fairly rigorous stirring to ensure complete hydration and temperature distribution beyond just eliminating dough balls.
Perhaps someone else is doing a test mash soon that could verify my observations?