There was a period where I considered DIY for such machines impractical. I now think that to diy all the way through is impractical, the design process is too laborious even for those suitably skilled. However once the design is worked out it is actually pretty practical, with a 3d printer. You definitely have to own the printer, they charge way too much to have things printed at cloud services for some reason.

I have pubished my STL files for an earlier version of the TW4 energy recovery ventilator. It's totally practical to make your own with a printer, but it would take you a few days of labor time for sure to get everything working nicely.

Reliability is very much on offer even with diy. In reality I have looked at many commercial units and they have serious reliability problems for long term like 5 years or more, and we should be planning for 50 years wherever building tech is concerned, imo.