Several studies with all bases having super high quality (100% question mark ASCII)

[jorondo]

I ran into several studies where the samples all seem to have "perfect" quality score of 63 (? over all bases). First time, I thought the samples quality had been tampered with, but then I found 6 more studies sequenced on various illumina platforms that had the same feature. Some examples: SRP410115, SRP403740, SRP423365, SRP424298, SRP425931, SRP434700.

@SRR22424534.1 1/1
GTACTGATACACAAAGTAATGACCTGCAATCCACGATGATTGCTAGGCTTATATTGTTTCTTTTTCATTTTTAAAAATCTTTACTAACTTAAAACCAAAAGGGCACAATGCCCCAAATTGTCTGCAAAGATACGAATAAACGAATAGATA
+
??????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????
@SRR22424534.2 2/1
AGCGTATTGGAAAGGATTATTGTCCTCGCTATAATTCTTACTAGAATACACCTTGTCTGAGTGAATAAGGCTCAACTGATTGCCATCAGCACGTAAGAAATAAGCATACGGAATAATGAACGGAGTGTGATCAGGCTCAATACATTGGG
+
??????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????


I've heard of quality binning but what I've found doesn't explain a score that high, nor does it explain why ALL bases are perfect; in some cases I still find a decent amount of human DNAs in the samples so I don't expect them to have been QC'd. I can't find anything on the illumina website. Anyone has an explanation?