We have discovered a third quasar absorption system which is consistent with a low deuterium to hydrogen abundance ratio, D/H = 3.4 times 10^-5. The z ~ 2.8 partial Lyman limit system towards QSO 0130-4021 provides the strongest evidence to date against large D/H ratios because the H I absorption, which consists of a single high column density component with unsaturated high order Lyman series lines, is readily modeled -- a task which is more complex in other D/H systems. We have obtained twenty-two hours of spectra from the HIRES spectrograph on the W.M. Keck telescope, which allow a detailed description of the Hydrogen. We see excess absorption on the blue wing of the H I Lyman alpha line, near the expected position of Deuterium. However, we find that Deuterium cannot explain all of the excess absorption, and hence there must be contamination by additional absorption, probably H I. This extra H I can account for most or all of the absorption at the D position, and hence D/H = 0 is allowed. We find an upper limit of D/H < 6.7 times 10^-5 in this system, consistent with the value of D/H ~ 3.4 times 10^-5 deduced towards QSO 1009+2956 and QSO 1937-1009 by Burles and Tytler (1998a, 1998b). This absorption system shows only weak metal line absorption, and we estimate [Si/H] < -2.6 -- indicating that the D/H ratio of the system is likely primordial. All four of the known high redshift absorption line systems simple enough to provide useful limits on D are consistent with D/H = 3.4 +/- 0.25 times 10^-5. Conversely, this QSO provides the third case which is inconsistent with much larger values.