Thanks Shabubu, I clearly see the problem now.
Most probably the piston seal - which also functions as shut-off valve & check valve- tends to stick to its shut-off surface on the section.
The axially low flexible piston seal is fitted axially floating on a conical seat over a restricted stroke of 1 -1.5mm , being only tight when pushing the plunger down. This feature is not shown on the Visconti simulation, it is however clearly shown on the patent drawing ( patent US 6,250,832 B1- sheet 3)
When pulling the plunger out, the piston seal will normally get stuck radially as soon as the seal leaves the enlarged filling chamber and enters the normal - narrower- piston chamber. At this moment only the plunger moves , till - after a further 1-1.5mm stroke -the front plunger collar contacts the seal & both move further backwards again. At this moment the ink in the reserve chamber -behind the seal - can flow back to the filling chamber (hence out of the pen) through the conical gap created between the seal bore and its plunger seat. In fact the axially floating piston seal acts like a check-valve.
Potential draw back- proven by your experiences- is the fact that the seal can stick to its shut-off back-up surface, since one needs a 1 -1.5 mm stroke to pull the seal effectively from its back-up surface on the section.
A little confusing though, since one should expect the ink can still flow through the conical gap between the seal bore & its seat when screwing the plunger 1mm out, even with a sticking shut-off.
Given your experience it looks the seal not only sticks on its back-up surface on the section, but also sticks on its conical plunger seat.
Aiming for a positive shut-off, while coping with barrel & plunger rod length tolerance variations, Visconti most probably applied a nominal 0.2/0.3mm axial pretension of the piston seal on its barrel back-up surface. This logically implies that the plunger has to be screwed open at least 0.3 mm before the seal can break from its seat. And even then, given the potential "double" sticking situation, the rubber seal will only axially deform a little when opening the plunger 0.3mm -remaining sticking on both surfaces- and not allowing any ink flow.
Only screwing the plunger further out will break the sticking , at least on one of the "sticking" surfaces.
Both sticking surfaces will only break positively loose after the axial plunger collar effectively pulls the seal backwards.
I hope this clarifies the problems you encounter, problems which seem effectively inherent to the "triple function combination" of piston seal, check valve & shut-off valve.
Splitting these functions- as Onoto did & I would do using modern sealing techniques- would surely overcome these problems.
Thanks again for your very informative posting, it really opened my eyes!
Regards, Francis
Edited for completion
QUOTE(Shabubu @ Mar 22 2008, 10:17 PM) [snapback]554100[/snapback]
The feed is standard, but the nib is reground from a broad to a itallifine. It was similar prior to the customisation, but is more pronounced now. The plunger on the filler doesn't seem to break the seal after being fully unscrewed and so flow is only really good enough when the blind cap is pulled an extra 2-3mm. Basically the flow gets really stingy after two paragraphs, and so I just tend to write with the blind cap fully unscrewed. I find the same with emptying the front resevoir, it requires an extra few millimeters of pull on the blind cap after it's fully unscrewed in order to allow the ink back out. I can see what they were hoping to achieve with the system, but you have to empty the resevoir prior to travel (or flying), and as such I cannot see what is gained above the standard vacuum fill and ink shutoff of the onoto (and subsequent Pilot 823) system.