hI Marvin--

Did you clean the source a couple of months ago?  If not, does it need 
to be cleaned? ;)

It can be a failing electrical component--but it might be a loose 
connection.  One of the wires may have popped off the source, or the 
source may not have been tensioned properly when re-assembled (check 
pins holding rods in place).  You can check connectivity (hV OFF!!!) to 
make sure there are no shorts before taking the source out too.  IF one 
of the heater bulbs fell out, those wires can move a bit and cause a 
short too.

Not sure if  the electronics changed much going to 253+.  IF you can 
peak at the lens control boards, check the LED's and components and look 
for obvious problems, e.g. a chip blew it's top.  The instruments are 
generally pretty good at handling arc's, but it's possible that one 
fried a component.

good luck!

gerry


On 10/26/2016 9:45 AM, Marvin Bobrowski wrote:
> Hello,
> We are working with a Gasbench II connected to a 253Plus and since a 
> couple of month ago the signal has become very unstable.
>
> I got suspicious because my linearity for d18O is much higher than 
> recommended by thermo (is:0.12 %o/V [should:0.05%o/V]) and every once 
> in a while the signal changes over one order of magnitude for a short 
> period of time.
> Sometimes I even lose almost all of the signal over night and will 
> have to refocus to get it back to where it was before.
> Also, measurements through the dual inlet show an irregular behavior 
> leading me to think that there might be a malfunction in the electr. 
> control unit or a dirty source and not something in the Gasbench itself.
>
> We checked water (m/z18 below 1V)and the system seems to be leak-free. 
> The filament is rather new, box and trap appear to be stable. We run 
> it with a current of 1.5mA
>
> Did anyone of you have had a similar experience with his/her 253 or 
> has heard of a malfunctioning source control unit?
>
> Thanks,
> Marvin
>