Summary

This project benchmarked the efficacy of commercially available harvest weed seed control (HWSC) mills – Seed Terminator, iHSD, and HC Seed Terminator – under controlled and comparable conditions. Results showed that annual ryegrass with smaller seed size, such as those from background field populations, was harder to control than larger, commercial seed sourced for testing. Seed Terminator achieved the highest control across all weed types. Mill efficacy was influenced by chaff feed rate, mill RPM, and mill wear. Power consumption was highest in Seed Terminator and increased with increasing chaff flow for all mills. However power consumption was strongly correlated with weed seed kill, with higher control requiring significantly more energy. Findings will help growers better understand performance limits of HWSC technology in real-world farming conditions.

Background

Harvest weed seed control (HWSC) is an important tool in integrated weed management strategies, aiming to reduce the weed seed bank by capturing and destroying weed seeds during harvest. Impact mills, such as the Seed Terminator, Harrington Seed Destructor (iHSD), and HC Seed Terminator, are increasingly used to target problematic species like annual ryegrass (Lolium rigidum), wild radish (Raphanus raphanistrum), and small-flowered mallow (Malva parviflora). Previous studies have reported high levels of control, often exceeding 90 per cent for annual ryegrass. However, factors such as seed size, mill wear, chaff feed rate, and power input can influence mill efficacy. Understanding these variables is essential for optimising HWSC in modern cropping systems.

Research Aims

The core objectives of the project were to:

• Independently compare efficacy of all commercial HWSC mills.
• Assess impact of feed rate, mill speed, and wear on weed seed control.
• Evaluate kill rates for various weed seed types.
• Understand the treatment effects on energy use and its relationship with efficacy.

In The Field

Testing was conducted using the iHSD, Seed Terminator, and HC Seed Terminator mills under consistent protocols. All mills were mounted on a test stand that allowed measurement of power consumption and control of variables such as chaff feed rate and mill speed. Chaff was processed for a fixed duration, and seed control was measured by comparing the germination of seed after passing through the mill with untreated seed.

Each mill was trialled with different chaff feed rates (1-2 kg/s), mill speeds (100 per cent and 90 per cent), and wear levels (new and worn components). Ryegrass seed, including a mixture of 63 per cent natural background population and 37 per cent commercial seed, was used to mimic field conditions. Seeds of small-flowered mallow and wild radish were also tested. Power use was logged to correlate energy input with seed kill rates.

Background ryegrass seed were found to be smaller than commercial seed, averaging 2.1 milligrams (mg) in weight, 38 per cent smaller than the 2.9mg commercial seed.

Results

Annual ryegrass control averaged 77 per cent across all mills and test treatments, small-flowered mallow 76 per cent, and wild radish 99 per cent. Maximum control of annual ryegrass was 97, 89 and 85 per cent for the Seed Terminator, iHSD and HC Seed Terminator, respectively, achieved with new mills operating at recommended RPM and 1kg/s chaff feed rate.

This control level on ryegrass was lower than previous studies using only commercial ryegrass seed. The size of the seed is hypothesised to be a significant factor in this difference, with testing finding that the efficacy for smaller background seeds was 5 per cent lower compared to the mixture of background and commercial seed. The importance of ryegrass seed source and size is being investigated in a follow up project TCO6424. Wild radish was consistently easy to control, supporting previous studies.

Energy consumption closely correlated with kill rates within a given treatment: more power meant better control but this can be at a cost to harvester efficiency. For example, a 10 per cent increase in control from 80 to 90 per cent consumed 20 per cent more power.

Worn mills trended lower, but wear measured by component weight was not a good predictor. The shape and blunting of impact surfaces may better explain performance losses although this is difficult to compare between manufacturers.

Feed rates of 2kg/s and lower mill speed drastically reduced control, especially for ryegrass and mallow.

Seed Terminator produced the highest kill rates in most scenarios.