G:G98: consistency failure oligomeric conjugated systems

 Dear Ning,
 In our research on analogous polyDIacetylenes and oligomeric derivatives of
 this, convergence could -as in many other cases that one wants to study
 with large basis sets- be best achieved in a stepwise manner: start with a
 HF/3-21G single point, follow-up with guess=read with a B3LYP/6-31G* &
 loose scf criteria, and finish off with the B3LYP/6-311G** computation.
 (Do use the polarization functions, as the description will be
 significantly worse without them.)  Once you have a B3LYP/6-311G**
 wavefunction on the starting geometry, our experience is that this will
 then rarely yield problems further on in the optimization in G94 (Gaussian
 98 will likely be similar), even with much larger systems (upto C80).
 Feel free to contact me if you continue to have problems!
 Best regards,
 Han Zuilhof
 -------
 Original question of Ning Xie:
 >Hi, all,
 >I'm trying to use BLYP/6-311g** to calculate the thermodynamic properties of
 >polyacetylenes (C2H2 to C20H2 or even higher with respect to the number of
 >carbons). The job is terminated shortly after it is started and gives a
 >"consistency failure #2 in CalDSu" message in the log file. I
 changed the
 >basis set to 6-311g and use SCF=QC and it still does not work.
 >Any suggestion is appreciated.
 >Thank you very much.
 >Regards,
 >ning
 >
 >The last part of the log file is
 >
 > Requested convergence on RMS density matrix=1.00D-08 within  64 cycles.
 > Requested convergence on MAX density matrix=1.00D-06.
 > Integral accuracy reduced to 1.0D-05 until final iterations.
 > Problem detected with inexpensive integrals.
 > Switching to full accuracy and repeating last cycle.
 > Warning!  Spurious integrated density:
 > NE=   74 NElCor=    0 Integral=    0.00000 Tolerance=1.00D-03
 > Consistency failure #2 in CalDSu.
 > Error termination via Lnk1e in /usr2/g98/l502.exe.
 > Job cpu time:  0 days  0 hours  0 minutes 38.5 seconds.
 > File lengths (MBytes):  RWF=   19 Int=    0 D2E=    0 Chk=    1 Scr=    1
 >
 >The input file is:
 >
 > $ RunGauss
 >%mem=8000000
 >%Chk=C12H2.chk
 ># BLYP/6-311G FOpt SCF Geom=(NoDistance,NoAngle)
 >
 >C12H2
 >
 >0 1
 >geometry specification
 ______________________________________________________________________
 Han Zuilhof                                   |  E-mail:  ZUILHOF -AatT-
 SG1.OC.WAU.NL
 Laboratory of Organic Chemistry  |  FAX   :  31-317-484914
 Wageningen University                 |  phone :  31-317-482367
 Dreijenplein 8                                |
 6703 HB  Wageningen                  |
 The Netherlands                            |
 http://www.spb.wau.nl/oc/perspages/han/index.htm