For the archives:

First of all: Thanks to all the persons hinting me in the right
direction.

It was wrong to blame Finnigan for the more flexible acid needle design.
The problem were located elsewhere (I suspect (4)).
I am not sure what did the trick at the end, so I list all things we
did:

(1) Cleaning the capillary from the acid reservoir to the pump head with
compressed air.
(2) Heating up the acid needle and capillary in hot water in an
ultrasonic bath. We connected one end to compressed air and "pushed" out
the remaining aicd thriugh the needle into a water bath.
(3) We dismanteled the pump head and cleaned everything.
(4) We assembled everything. It turned out that one have to make sure
that the capillaries are pushed through BOTH viton O-rings of the
fitting on each side. However, they must not enter the pump head, of
course. The only way to get this properly done is to fix the capillaries
in the fitting with open pump head. Here you can look inside the pumpe
head and can push the capillaries ends in the correct position.
Afterwards the pump head can be mounted to the pump motor.
(5) We started pumping with water, diluted acid and the real stuff, as
recommended.
(6) Finally we need about 10 to 12 turns per drop at 250 skale divisions
(about half of the maximum pumping volume).

It's still not perfect (I just get the measurement neddle clogged by
some piece of a septa which had acid underneath...), but the acid neddle
seems to be back at work.

Regards,
Robert
 
-- 
Dr. Robert van Geldern
Leibniz Institute for Applied Geosciences (GGA)
Section 3: Geochronology and Isotope Hydrology
Stilleweg 2, 30655 Hannover, Germany
phone: ++49-511-643-2313
mailto:[log in to unmask]