ghwellsjr said:
I know that at one time you were expressing the belief, which is also expressed on Neil Slade's website, that getting air into the sponge by printing long after the low ink warning appears (signaling an empty reservoir) is the kiss of death to cartridges because once those air pockets or bubbles get in there, the refilled ink will not be able to flush them out. Do you still believe that? If so, how do you account for the fact that a flushed and dried cartridge readily accepts refilled ink?
You are correct that I was following that line of reasoning when I tried vacuum refilling. However, Craig Ross made a comment that the real problem was that the filter and sponge were "gunked up", and suggested trying a few drops of alcohol on the filters or heating the carts in 140 deg F water to break up the gunk. His suggestions improved (but did not completely fix) the problem. The hot water purge technique is just a very aggressive extension of Craig's suggestion. I was prepared to make the first refill with vacuum to get the air out of the sponges, but I was pleasantly surprised to find that ink injected into the ink chamber quickly wicked into the sponge, almost completely eliminating the air. How do I explain this? If the cleaned and dried sponge material has a surface tension sufficiently above the ink's surface tension, it will readily "wet out" without any assistance.
ghwellsjr said:
I have another possible theory as to why banding might occur under heavy printing. We have to understand how these cartridges work in terms of the ink flow and the air flow. Of course ink that comes out of the hole at the bottom of the reservoir migrates along the lower sponge to the outlet filter. As long as that path is unrestricted, there shouldn't be any problem as far as the ink is concerned.
I agree - I have often printed in excess of 25 8x10 prints in one run. As long as the carts are freely delivering ink, there is never a problem with banding, regardless of the size of the print run. Before I started to purge my carts, I did sometimes encounter the situation where light banding occurred on multiple prints. Letting the printer "rest" between prints apparently allows time for ink to flow through the partially clogged sponge/filter. In these cases, there was no pool of ink on the top of the ink pickup when the offending cart was pulled immediately after printing.
ghwellsjr said:
But there is also a required air flow from the vent on the top of the cartridge, through the upper sponge, down to the groves in the wall separating the two compartments, and finally down to that same hole at the bottom of the reservoir. The slight pressure drop that is causing air to flow along that path is actually created at the outlet port and it's possible that it could suck air directly into the outlet port instead of indirectly sucking it into the reservoir hole. In fact, when you think about it, you might be amazed that the air would actually go into the hole at the bottom of the reservoir instead of out the outlet port which, it would seem is a much closer path. If that did happen, it's possible that the cartridge would not recover without extraordinary intervention such as a flush or maybe vacuum filling.
You are correct that it isn't obvious why the air should go into the ink chamber instead of into the sponge. I suspect that the answer again lies in the relative surface tensions of the ink/sponge/air, and that the sponge's surface "prefers" ink to air
ghwellsjr said:
I was able to create this situation in a third party cartridge with a little experiment I did. I took a completely filled cartridge and attached a tube to the outlet port. I submerged the tube in a sink full of water and raised the cartridge an inch or so above the water level to create a negative pressure on the outlet port. This would drain ink out of the cartridge at a rate that I could control by raising or lowering the cartridge. I was trying to create a situation similar to what a cartridge experiences in the printer. I did this a great many times with many cartridges. One G&G cartridge with a single sponge got into a mode where air was flowing down from the vent through the outlet port, even though the reservoir still had plenty of ink in it. In other words, it acted like an empty cartridge.
That's an interesting experiment. Is it possible that that one cart had the sponge raised slightly off the bottom of the sponge chamber, providing a large leakage path that could override the pull of surface tension in the sponge?
ghwellsjr said:
So what I'm suggesting is that air in the lower sponge causes a problem not because it restricts the flow of ink but because it provides the wrong path for the air to flow. Instead of the air flowing along the upper sponge material to the groves that go half way up the wall between the two compartments and down to the hole so that it can bubble up into the reservoir, it goes directly down through the lower sponge material to the outlet port and gives the nozzles air instead of ink and that is maybe one reason why banding occurs.
That's one possible explanation. However, I have found that when I blow into the vent port to get a cart to drip that various carts can have large differences in how much pressure it takes to cause the cart to drip. When I have seen banding, the offending cart always takes a high pressure to cause this drip. In fact, monitoring this pressure determines when I purge the carts. This is why I suspect that the problem is usually a clogged filter/sponge (as first suggested by Craig).
At first glance, these carts seem to be simple boxes of ink. As we study them, it becomes obvious that they are actually quite complicated. As an example, Canonfodder has done some experiments that explain how the simple passing of a low pressure cell during a storm or an increase in the room temperature can cause partially empty carts to drip and possibly cause cross-contamination.
BTW - this thread seems to have completely deviated from its original topic. Should we ask Rob to break it apart to put the Canon discussion in its own thread?
