Abstract #688

# 688
Using parlor data to map liner performance.
John F. Penry*1, Stefania Leonardi2, John Upton3,1, Paul D. Thompson1, Douglas J. Reinemann1, 1University of Wisconsin-Madison, Madison, WI, 2Universita delgi Studi di Milano, Milan, Lombardia, Italy, 3Animal & Grassland Research & Innovation Centre, Teagasc, Moorepark, Co. Cork, Ireland.

Liner performance can be described in terms of milking gentleness, speed and completeness of milk-out, with gentleness being the most important. It is widely accepted that peak milking speed will be increased as vacuum and the milking phase of pulsation are increased, but it is also known that raising the vacuum and b-phase duration increases teat end congestion. Increasing liner compression (LC) also results in higher milk flow rates while also elevating the risk of teat end hyperkeratosis. The aim of this experiment was to characterize the average milk flow rate of 8 liners, representing differing LC estimates, across a range of pulsation and vacuum settings. The 36-d trial involved an 80-cow herd milking 2× at the UW-Madison Dairy Cattle Centre. The parlor was fitted with 8 commercial liners (round, triangular, vented and non-vented models), which were rotated through all stalls during the trial. Treatments were a combination of selected system vacuum and pulsation settings with a fixed 295ms d-phase. Nine treatments were used representing commercially applied settings for vacuum and pulsation applied over 3 equal periods in a central composite experimental design. Treatment settings for system vacuum level ranged from 36 to 49 kPa and pulsator ratios from 50:50 to 70:30. During the course of each 9 treatment cycle, the central point (42.3 kPa and 60:40 pulsator ratio) was applied every third day allowing for an estimate of within treatment variability. Parlor average milk flow (AMF) data were analyzed using a MIXED model in SAS 9.3. This model assessed the effect of liner, treatment, milking stall, milking time and milker. The SAS RSREG procedure was used to produce individual liner response surfaces. Liners with lower LC did not produce as high an AMF under high vacuum and long b-phase compared with a high LC liner due to the effects of teat end congestion during each individual pulsation cycle. Knowledge of individual liner compression estimates and, where available a response surface, is highly useful for determining the optimum vacuum and pulsation settings without compromising gentleness of milking.

Key Words: liner, performance, compression