Given the advanced state of operational plans for late Run II (132 nsec bunch spacing) the C0 IR insert should be designed to operate such that it does not impact nominal Tevatron parameters. This implies an entirely localized insert -- one which is completely transparent to the rest of the machine. This condition has several important design implications, some of which are pointed out below. An IR design similar to that employed at CDF and D0 is unacceptable as a C0 candidate. The addition of such a (single) low-{beta} region to the machine raises the tune by a half-integer in each plane, moving them far from the standard operating point and right onto the 21.0 integer resonance. The nominal (fractional) operating point is most elegantly maintained by adding 2 local low-{beta} in each plane, thereby boosting the tunes by a full integer. The B0 and D0 IR's are not optically-isolated entities. Progression through the low-{beta}squeeze involves adjusting, not only the main IR quadrupoles, but also the tune quad strings distributed around the ring. The result is that the nominal lattice functions at any point in the ring, and the phase advances across any section of the ring, are not fixed, but vary with each stage of the squeeze. A new insert must be sufficiently flexible to track these elusive matching conditions. Without collisions at C0 the unit transfer matrix added by the insertion ensures that the incoming and outgoing helices are automatically matched to their nominal Run II values. To maintain this match with collisions at all 3 IP's, however, requires that additional separators be added in the arcs. Space for these separators can only be generated through replacing standard Tevatron arc dipoles by new magnets with enhanced strengths. In the following sections two design variations for an interaction region are presented. The first of these, which incorporates stronger dipoles, meets all of the ideal design criteria outlined above. The result is a truly independent 3rd Tevatron IR capable of supporting simultaneous collisions at all 3 IP's. The second, stripped-down, version includes neither stronger dipoles nor new arc separators. While this insert is still optically transparent to the machine, collisions can only occur at B0 & D0, or just C0, but not all three. The weaker dipoles also result in a significant reduction in the space available for a detector. However, as is demonstrated near the end of this report, if all the B- and C-Sector separators are freed to assist in C0 orbit control, it might be possible to support collisions at B0 and D0, plus C0, with the second design.
