That is why I prefer not to have any sand. also from your above statement I think it is important to preserve the gene pool that you are talking about even if they are crossbred.This is an interesting one actually, one which is too quickly dismissed on this forum.
For chelonoidis-in general, some localities of carbonaria are exceptions-, erosa/homeana, impressa, emys, indotestudo etc. , they wouldn't come across what we consider sand in the wild. However, chilensis, elegans, pardalis, testudo genus, etc. would all come across 'sand' in the wild. This would probably be primarily silica, usually in quartz form. This is the kind of continental sand that you'd most likely find in the Russian Steppe, Sub-Saharan Africa, Patagonia, etc. This isn't necessarily baking hot and dry, it can be a mild climate and there can be plenty of grasses, but we would usually classify the small particles of rocks and minerals as sand, rather than rocky sediment. This 'sand' however is quite different to what we get for tortoises in captivity, which I suspect is mainly of carbonate origin, eroded from soft rocks. However, while this 'sand' is very different to the 'sand' found in habitats where most of these tortoises are from, I doubt that there would be much difference in the problems that people have experienced with impaction. Also, at least in theory, the Sonoran Desert sand would probably be very similar to some of the stuff found in the tortoises' habitat even for non-American species, but I don't know how often this is sold as your sand in the States???
The main problem here is a lack of clarity over what we define sand as, and the rocks which this sand came from.