Utvoler:
I think the problem has much more to do with the individual conc. blk. themselves. Five, six or seven blocks all in the same course would very likely have come off the same pallet, from the CMU manuf’er. I’ve seen this problem before too, and usually it has to do with a fairly small number of blks. which likely all came from the same manuf’er. lot, were laid at the same time and most likely from the same lot of blks. Blks. immediately around the mushy blks. seem to be o.k., as is the case in your photo, which I assume is upside down as attached. I’m not discounting moisture as a deteriorating aspect, nor freeze-thaw or some reactive soils condition, or some salts of one sort or another, however, the next course up shows much the same moisture patterns (not full height) and would be more susceptible to freezing, but doesn’t show the same blk. deterioration. Some sort of soil reactivity would not likely be so well confined/defined as to start/stop at a horiz. mortar joint. It does seem that the face shells go first, and that cross shells last a little longer, and of course the mortar is a whole diff. animal. I think there was a small portion of a conc. batch which was bad, poorly mixed, reactive aggregate, missing some percentage of some ingredients, etc. or that as the raw batch mix materials were placed in the mold and compacted something went haywire, or that something about the curing process went wrong. It would be interesting to see some sampling and testing of the conc. blk., surrounding soils, ground water, etc. to see if anything stood out.