16

2026

-

11

Combine Harvester Grain Loss: Where It Goes and How to Fix It

Author:


A line of grain left behind the machine at the end of a pass is one of the clearest signals in harvesting, and one of the most often misread. Most operators reach for the sieve first. In practice the sieve is rarely the culprit on its own: the settings have drifted out of step with the crop, or they were made for last week’s field and never touched again.

Three different failures all look the same from the cab. Grain can fall off the header before the knife reaches it, it can be carried out with the straw after threshing, or it can be blown off the cleaning shoe. Each has its own causes. Adjusting the wrong one costs half a day and gains nothing, so the first job is not adjustment. It is finding out where the grain leaves the machine.

Find the loss before you touch a setting

Run the machine normally for twenty or thirty metres, stop, and reverse back over the stubble. Lay a sheet, a tray or a frame in four places: under the full width of the header, under the straw discharge at the rear, across the shoe outlet behind the rear sieve, and beside the tailings return. Count the kernels inside a half-square-metre frame, or simply compare the four piles. The heaviest one tells you which part of the machine to work on.

Two numbers make the result useful. Acceptable total loss is usually quoted at around 1.5 percent for wheat, with rice and maize slightly higher; use the standard that applies in your market rather than a figure from a forum. And the loss monitor in the cab is an indicator, not an instrument. If it has not been calibrated for this crop and this moisture, a climbing bar may be reporting straw.

Cause one: drum speed and concave clearance out of balance

Threshing is a compromise between two losses, and the operator sets it. A slow drum with a wide concave sends unthreshed heads out with the straw, visible in the swath, while the tailings auger keeps working all day. A fast drum with a tight concave gives the opposite: cracked kernels, straw rubbed into short fibres, and a cleaning shoe suddenly asked to handle far more material than it was set for. The grain then leaves over the back with the chaff.

Start from the chart for the crop and the moisture in front of you, then move in small steps: ten to fifteen percent on drum speed, two or three millimetres on the concave, followed by a test run. Before changing anything, take a handful of straw from the discharge. Whole heads in it mean the concave is too open or the drum too slow. Short, rubbed straw and cracked grain in the tank mean you have gone past the middle.

Cause two: the machine is being fed faster than it can thresh

Loss that climbs through the day, with the engine note sagging in the heavy patches, is usually feed rate. A full header in a thin crop reads the same on the gauge as a full header in a heavy one, but the walkers and the shoe only see tonnage. When the straw mat gets too thick, separation stops and grain rides out the back.

The reflex is to slow the engine. That is the wrong lever: it drops drum speed and fan speed together, and every setting made earlier is now wrong. Use the gearbox instead. One gear down, throttle at rated rpm, header still full.

Cause three: the fan and the sieve working against each other

Air volume is the most common error in the cleaning system. Too much air and light kernels are carried over the back with the chaff; the trail on the stubble looks like a neat strip of clean grain, because that is exactly what it is. Too little air and the mat sits thick on the sieve, grain rides over the rear edge, and the tank sample turns dirty, so the operator opens the sieve and makes the loss worse.

Set the fan by watching the discharge rather than the dial. Chaff should lift and float just above the sieve while grain stays down and works along it. Then set the chaffer and the sieve to the grain size in front of you: open enough for grain to drop through in the first half of the shoe, closed enough for a clean sample without turning the returns into a grain conveyor. A sieve packed with damp straw, sagging in the middle, patched at one edge or rubbed through will do exactly the work of a badly set one, so look at it before adjusting it.

Cause four: crop moisture and field condition

Two fields on the same afternoon can want different settings. Grain above about eighteen percent moisture threshes badly and wraps the drum, and the tailings load climbs sharply. Grain below about twelve percent is brittle: kernels crack, the lighter ones are easy for the fan to lift, and the header shatters heads that would otherwise have gone into the tank.

Dew, green weed patches, lodged crop and uneven stands change the feed rate by more than any settings change can absorb. In practice that means respecting the dew. Starting an hour later in the morning is often cheaper than resetting a machine twice. When the sample shows more chaff and the monitor climbs at four in the afternoon, the answer is usually the clock, not the wrench.

Cause five: wear, leaks and damage nobody looked for

A good share of what gets called excessive spillage never had a chance to be threshed. Rasp bars worn round at the edges, concave bars worn flat, a grain pan with a hole rubbed through where the return drops, a tailings auger flight worn away, a tank seam that opens when it is full, an elevator boot leaking onto the ground. After a few passes, park the machine and look underneath. Grain on the ground under a stationary machine is a leak, and no setting will fix it.

The walk-around belongs to the back of the machine as well. A straw chopper deflector set to throw wide will scatter grain from the straw across the full cutting width, where it disappears into the stubble and later gets blamed on the shoe.

Cause six: slope, wind and the attitude of the machine

On side slopes grain slides to one side of the shoe, and the loss appears only there, usually diagnosed as a sieve fault. Levelling the machine, working across the slope instead of straight up and down, and slowing where the header has to float will each recover more than a sieve tweak. A strong crosswind does something similar to the fan’s work, pushing the light fraction off centre before it reaches the chaffer. If the pattern of loss changes with wind direction, the wind is part of the answer.

A sequence that works in the field

Change one thing at a time. Two changes in one run tell you nothing, and this is where most of a day disappears. The order that costs least time:

  1. Walk around the machine with the engine off. Look for leaks and worn parts, and fix those first.
  2. Set the drum speed and concave clearance for the crop and the moisture in front of you.
  3. Set the fan so the chaff just floats. Leave the sieve alone for now.
  4. Set the chaffer and the sieve to the grain size, then check the tank sample.
  5. Set ground speed last, with the engine held at rated rpm.
  6. Run the tray test again at the four positions and write the settings down.
  7. Keep that note with the field. Next season, at the same moisture, you start from a known point instead of from scratch.

The last step is the one nobody does. A note in the cab, with crop, moisture, drum, concave, fan, sieve, ground speed and the result, turns next season’s first pass into a five-minute job instead of a day of guessing.

When the problem is the machine, not the settings

There is a limit to what adjustment can do. Two machines set identically will still lose different amounts if their cleaning shoes are different sizes, if one has a single fan and the other a double fan, if grain is not spread evenly across the full width of the sieve, or if the returns are dropped back into the drum instead of being handled separately. At high throughput, the shoe area and the distribution of material across it decide the result well before the operator’s settings do.

The view from the factory

Golddafeng has been building harvesters and tractors for 32 years. Around 85 percent of the core components in our machines are produced in our own workshops rather than bought in, and drum, concave, sieve and fan are matched to each other at the design stage. A substitute part with slightly different geometry can quietly undo that match. Every machine is tested before it leaves the factory.

Outside China, the service side counts as much as the hardware. Our dealer and service network in Africa, South Asia, Southeast Asia, Central Asia and South America exists for the first harvest as much as the tenth. A technician who can run a tray test in the customer’s field and reset the shoe is worth more than a page of specifications. Where the crop is different, lowland and upland rice, irrigated wheat, maize taken at high moisture, we build to the crop rather than asking the farmer to adapt.

Questions we are asked

Is some grain loss unavoidable?

Yes. Threshing, separating and cleaning all leave a fraction behind, and around 1 to 1.5 percent total is normal for wheat on a well-set machine. The aim is not zero. It is to know what your machine should be doing, and to notice when it stops doing it.

Should I close the sieve to stop grain going out the back?

Usually the opposite. Closing it pushes more material toward the rear edge and can dump grain in a narrower band. Check the air first, then the feed rate, then the sieve.

Does driving more slowly always reduce loss?

Only when feed rate is the limit. If the drum, concave or fan are wrong, slowing down hides the problem and costs you hectares. Fix the settings, then find the speed the machine will actually take.

Why does the loss get worse in the afternoon?

The crop dries, kernels become brittle, and the light fraction gets easier for the fan to lift. Check the moisture, and adjust the fan and drum to it instead of fighting the same settings all day.

Can I trust the loss monitor?

As a trend, yes. As a number, only after calibration. Calibrate it for each crop and moisture range, and confirm the reading with a tray test at least once a season.

If your machine is throwing grain and the tray test points at the hardware rather than the settings, that is a conversation worth having before next season, with your dealer or with the factory that built it.