Using Substructure for contact modelling in Abaqus

I have a
problem to use substructure for contact anaylsis of two gears. I have
already read the example in abaqus Example problem manual for help.
Well, to explain my problem in detail, I have two nastron files of two
gears and later using FORTRAN I am able to write the programm to
generate the *.inp files. What I tried to do is
1. Substrcture generation of two gears in seperate files. Now I have two
sets of substructures available for individual gears. While generating
the substruture, I have retained the DOF for the nodes which are on the
flanks because I wish to use these nodes to write the conatact between
two gears.

2. In another file, I write down the nodes and elements only related to
the flanks of the gears OR the retained nodes. Now comes the tricky
part, I have used till now different type of combinations to write the
elements but I could not do it successfully till now. Every time, I
recieve a different error. The combinations I have tried to use for
writing the surfaces for contact are:
a. Both the elements are written using surface elemtns like SFM3D8, so
an element based surface to write the contact between two gears.
b. Both the elements are written using membrane elements like M3D8, so
an element based surface to write the contact between two gears.
c. One surface is used with SFM3D8 elements and another is used with node based surface to define contact between two gears.
d. Only one gear is used for substructre generation and other gear is
kept as it is with original C3D20 elements. So for one gear the surface
is written as element based surface and for another gear it is written
as node based surface.
If anyone has worked with a similar problem, it would be really
grateful if someone can suggest something about it. Also if someone has
any clue about the follwing errors I am getting during the simulation,
they can please reply me with the root cause of the error. The errors
are:
1. The volume proportional damping factor could not be calculated.
2. Too many attempts made for this increment.
3. The static stabilization factor could not be calculated.

If someone is interested in going through the files, I can upload them and thank you in advance for help.

Regards,
Marko