mikling said:
If this is considered, it will be seen that a catchment tank is more practical. If the ink empties into a catchment container then the air within the container will be 100% saturated eventually. This 100% saturation will prevent further evaporation of the ink within the container. The only exception is when there is air displacement via the vent hole in the catchment tank. This will occur during barometric pressure changes if the other end at the printer is sealed or if the temperature fluctuates casuing a pumping effect due to expansion and contraction of the air inside the container.
Translation:
One of the many existing rigid tanks like the Printer Potty kits. The inclusion of a very small diameter vent hole allows the ink and air to flow into the tank while allowing the air to exit leaving just ink and fairly static air.
Such a design allows for any fluctuations in air pressure in both directions. And just to deal with the issue of evaporation, there's virtually none taking place, nor does the vent hole allow for a current of air to pass through, so there's nothing to draw off the moisture laden air in the tank. The only time I've experience such an was with a tank I had sitting in direct sunlight (such as we have in the UK) for nearly 18 months with around 25ml (1oz-ish) of waste ink in the tank and even then the ink was just thicker, rather than dried out.
Both of these conditions can be mitigated if one introduces a serpentine system towards the vent just like a canon sponge cartridge has in its air vent system. The serpentine will be considered overkill (especially on an Epson because Epson has long purge cycles in general) but a long coiled tube through which the air escapes will fullfill that. The catchment tank only holds liquid and no provision and space is needed for the air that a bladder will require.
Perhaps in a perfect system but such a system exists in cartridges primarily because ink is being drawn off, rather than introduced, so any drying action is a bigger threat in cartridges where overkill is definitely a good idea given the shelf life requirement. Waste ink tanks on the other hand are constantly introducing yet more fluid to the tank every few days, weeks
(if in low use) there's always a moisture laden atmosphere in the tank.
The issue of Canon printers and clogging is unresolved and inconclusive. Ironically the iP4000 I installed a kit on, had such a strong pump that it blew off connectors after something constricted, plugged or kinked the tubes. Whether it was ink drying, ink combination or poor/lengthy storage, or some other issue is an unknown because I didn't chase do a post mortem on the unit and it was some years ago. All advice from that experience is naturally based on the "take no ruddy chances" methodology, hence the various dire warnings to The Hat
The bottom line is that there is currently a big hole in terms of information on what actually happens with Canon printers and while theory is all well and good it really requires more practical experimentation and the occasional "inspired"
(read: nuts) individual to give it a go and see what happens.