Two experiments were conducted to test the hypothesis that different combinations of conditioning and expansion of soybean expellers increases nutritional value. Non-heat-treated soybean expellers (L-1) and soybean expellers conditioned for 60 s at 90ºC followed by expansion at 110ºC (L-2) were used. Two additional sources of soybean expellers (L-3 and L-4) were processed as L-2 with the exception that the initial conditioning was followed by long-term conditioning for 12 or 48 min at 100ºC before expansion. Analyzed trypsin inhibitor activity in L-1, L-2, L-3, and L-4 was 34.0, 23.1, 4.2, and 2.4 mg/g, respectively. In experiment 1, 10 cannulated barrows (54.22 ± 4.54 kg) were allotted to a replicated 5 × 4 Youden square with 5 diets and 4 periods and 8 replicates per diet. Each source of soybean expellers was included in one diet, and a N-free diet was also used. Data were analyzed by ANOVA using the Mixed Procedure of SAS. The standardized ileal digestibility (SID) of all amino acids (AA) in L-1 was less (P < 0.01) compared with L-2, L-3, and L-4 (Table 1), and SID of all AA in L-2 was less (P < 0.01) than in L-3 or L-4. In experiment 2, 40 barrows (17.52 ± 1.63 kg) were housed in metabolism crates and fed a corn diet or 4 diets based on corn and each source of soybean expellers. Feces and urine were collected using the marker-to-marker approach with 5-d adaptation and 4-d collection periods. Data were analyzed as in Exp. 1. The metabolizable energy (ME) in L-1 was less (P < 0.01) than in L-2, L-3, and L-4 (Table 1).