Constant-gm constant-slew-rate high-bandwidth low-voltage rail-to-rail cmos input stage for vlsi cell libraries

gm-Extraction for rail-to-rail input stage linearization

“…We notice that the very low degree of distortion achieved within the ampliÿer dynamic range dramatically overcame g m correction obtainable with other techniques recently presented [9].…”

Bulk-driven flipped voltage follower

“…The traditional design technique for rail-to-rail operational amplifiers is the deployment of complementary differential pairs with the tail current being controlled with current switches to keep the gm constant [8]. However, due to the fact that the mobility ratio of the complementary pairs (µn/µp) is process-and temperature dependent, causing the gm 978-1-4244-3828-0/09/$25.00 ©2009 IEEEvariation to deviate by approximately 12% [9], there exist circuit topologies which use only a single type of the differential pair, where one of them is the bulk-driven one. Figure 2(a) illustrates a bulk-driven differential pair that uses p-type devices only.…”

Achieving rail-to-rail input operation using level-shift multiplexing technique for all CMOS op-amps

“…When using complementary input pairs, on the other hand, designers must deal with the mobility ratio of the pairs (µ n /µ p ), which is process-and temperature-dependent and may deviate by approximately 12% [7]. To avoid investigating this in the first place, there are some available techniques that only use a single type of differential pair, such as level-shifting [7], floating-gate [8], and bulk-driven [9].…”

“…To avoid investigating this in the first place, there are some available techniques that only use a single type of differential pair, such as level-shifting [7], floating-gate [8], and bulk-driven [9]. Each of these techniques, however, possesses several drawbacks.…”

“…Each of these techniques, however, possesses several drawbacks. The level-shifting technique proposed in [7] utilizes two identical differential pairs, each biased with the same amount of the tail current, and the input of one of the differential pairs is level-shifted using source followers. The FeedForward (FF) cancelling section is used to ensure only one differential pair contributes to the effective g m over the rail-to-rail CM input.…”