310 injection problems

[jia]

Hello,

I would really appreciate advice from anyone who has ever encourntered the
following problems on 310s. 

I have been using a 310 for a year and the results are pretty consistant.
However, we had injection problems in the last several weeks. Our major
problem is the inconsistance of the results. When we run a group of 
samples,
some of them worked just fine and some didn't work at all. The samples 
that
didn't work just give us flat lines. If I reload the same sample, we will
get just nice sequences. This has been going on for a while and getting
worse and worse. A quarter of our samples need to be reload to get 
results.
For the sample that actually worked, the signal strength has been 
decreased
dramatically compare to what we used to get.  

ABI technical people told me to check the samples and the septa. We know
some septa are defected and do not use them at all and people do sequeeze
the septa before use them. The samples are usually stored at a freezer and
votexed and spun down before loading on the analyzer. The TSR volume is 
25ul. 

Our service engineer did come here last week. He did a through check and
claimed the instrucment is in very good shape and nothing is wrong. He
believed the problem is caused by bad samples and the way we run it ( 
e.g. I
only change the buffer once week, the capillary should be a little bit 
lower
). He set up the instrument in his way and set up one primer standard and
did two injection from that same sample. He left when he see the first
injection worked. However, the second injection gave us flat lines again.
The service engineer then told me to change the capillary and set the
capillary a little bit higher than he did (?). We set up 9 injections from
one new primer standard and it worked fine. We then set up three control
cycle sequencing and did 5 injections on each, which make the total
injeciton number 15. The control samples didn't work really well. The 
signal
strength in several samples are very low ( 40-80 ) and looks bad. To make
things more complicated, the new capillary went bad during the last 
injection. 

The service engineer is confident that the instrument is good. But we know
somethng is wrong. My personal guess would be a injection problem. But I
can't convice him.



Thanks

Jia