Dear Ben,

I have a trouble to run a tgcm15 job. It always stops at onle 6 mins model
run. The script files been used are located in directory tiso/cmh/tgcm15.
It is highly appreciated, if you would check it for me.

With best regards,

Chien-Ming


From foster@@hao.ucar.edu Mon Nov 11 09:40 MST 2002
Received: from vishnu (vishnu.hao.ucar.edu [128.117.16.89])
	by jabba.hao.ucar.edu (8.9.0/8.9.0) with SMTP id JAA24766;
	Mon, 11 Nov 2002 09:40:37 -0700 (MST)
Message-Id: <200211111640.JAA24766@@jabba.hao.ucar.edu>
Date: Mon, 11 Nov 2002 09:36:30 -0700 (MST)
From: Ben Foster <foster@@hao.ucar.edu>
Reply-To: Ben Foster <foster@@hao.ucar.edu>
Subject: Re: Your Message Sent on Mon, 11 Nov 2002 00:20:55 -0700 (MST)
To: cmh@@hao.ucar.edu
Cc: roble@@jabba.hao.ucar.edu, foster@@jabba.hao.ucar.edu
MIME-Version: 1.0
Content-MD5: aNtalJjDMgErq686WHgy7Q==
X-Mailer: dtmail 1.3.0 @@(#)CDE Version 1.4.2 SunOS 5.8 sun4u sparc 
Content-Type: TEXT/plain; charset=us-ascii
Content-Length: 1695
Status: RO
X-Status: $$$$
X-UID: 


Chien-Ming:

There are some upper boundary problems in O+, NE, and TI in
/CMH/tgcm15/80_temp.nc, but I'm not sure that is causing your
crash. I would suggest turning off a loss rate calculation used
in settei.F, by commenting out the statement starting at line 388, 
and substituting zero, as follows:

!
! Gpi runs may stop here with overflow in exp() when TE gets too hot.
! Sometimes S10 may already be -INF from above.
!
!       S10(I,1) = 1.57E-12*exp((2.4E4+0.3*(F(I,NTEK)-1500.)-1.947E-5
!    1  *(F(I,NTEK)-1500.)*(F(I,NTEK)-4000.))*(F(I,NTEK)-3000.)/(3000.
!    2  *F(I,NTEK)))*S10(I,1)
        S10(I,1) = 0.
        
If it still crashes, try starting from a known stable equinox history, 
e.g., from Ray's series in /ROBLE/tgcm15r, with this change. 

If you would like to get better diagnostic from the IBM, change 
"-O3 -qstrict" to "-g" in the Makefile and recompile the code. Maybe 
also run it with 4 processors instead of 12. Then you should get a line 
number in the source code where it is crashing (check all 4 output files
after execution). Let me know if you need help with this.

--Ben 

>
>Dear Ben,
>
>I have a trouble to run a tgcm15 job. It always stops at onle 6 mins model
>run. The script files been used are located in directory tiso/cmh/tgcm15.
>It is highly appreciated, if you would check it for me.
>
>With best regards,
>
>Chien-Ming
>

-----------------------------------------------------------------------
Ben Foster		      	High Altitude Observatory (HAO)
foster@@ucar.edu			phone: 303-497-1595  fax: 303-497-1589  
Nat. Center for Atmos. Res.     P.O. Box 3000 Boulder CO 80307 USA
-----------------------------------------------------------------------


@