The top sieve is not the goal

Long stems look like fiber and often become refusals.

3 min read · 2026-09-07

In this note

2–8%
19 mm
8 mm
4 mm

The Penn State Particle Separator top sieve catches the longest stems. On a high-producing TMR, that 19 mm fraction is meant to stay small — typically about 2–8%.

Broomstick stems look like fiber on the clipboard and get sorted out of the bunk.

Most of the peNDF mass that supports chewing lives on the 8 mm and 4 mm sieves together — particles long enough to help hold a mat without inviting refusals.

The 4 mm screen is the peNDF cutoff used to estimate physically effective fiber; it is not the main scratch layer by itself.

The 2013 update that added the 4 mm sieve changed how peNDF is estimated on farm; the companion guideline is an eatable-length TMR — enough length to chew, not so much that cows refuse it.

If the top sieve runs high and the refusals are stem-heavy, you built a sorting diet, not a safer rumen.