Two hundred grams per cow per day. It sounds like nothing. A handful of pellets, a rough scoop, a scale that drifts when the battery gets low. But run that number across 300 cows for a full 305-day lactation and the maths stops looking trivial very quickly.
This piece is for dairy nutritionists and farm managers who want to understand what imprecise concentrate measurement actually costs, and what specification to demand from the equipment responsible for it. The numbers below use current South African input prices, so the rand figures are grounded in what producers are paying right now, not in some textbook scenario.
The Rand Cost of a 200 g Daily Error
Start with the basics. A 200 g/cow/day over-measurement of concentrate across a 300-cow herd is 60 kg of concentrate dispensed above the recipe every single day. Over a 305-day lactation, that is 18,300 kg of concentrate fed in excess. At a blended commercial concentrate price in the region of R7,500 per tonne, the total unnecessary spend comes to roughly R137,250 per lactation. That is not a rounding error. That is a significant input cost that most farms would treat as a crisis if it appeared on a single invoice.
The under-measurement version is a different kind of damage. Feed 200 g less concentrate per cow per day across the same herd and the energy deficit compounds through mid-lactation. Peak yield drops, body condition suffers, and cows that should be cycling by 60 days post-calving are lagging. The financial hit is harder to line-item because it shows up in reduced milk income, extended calving intervals, and higher culling rates, but it is no smaller than the over-feed scenario.
With drought conditions grinding through the Free State and KwaZulu-Natal right now, concentrate is doing more work than usual. Pasture and forage quality are down, purchased feed costs are up, and the margin for measurement slippage has narrowed accordingly. A 200 g error in a ration that used to be well-buffered by good pasture becomes a much larger proportion of the cow’s energy supply when the pasture is gone.
What Indicator Accuracy Actually Means
Mixer indicator accuracy is quoted as a percentage of full scale. That specification is what tells you how well the machine holds a target weight at any given load. It is not the same as resolution, and it is not the same as repeatability. Full-scale accuracy determines the worst-case error at any point across the weighing range.
A unit rated at ±0.5% of full scale sounds precise enough until you do the arithmetic. If the machine’s load cell capacity is 1,000 kg, ±0.5% full scale means the reading can be wrong by up to 5 kg in either direction at any load. At 500 kg of concentrate, that 5 kg error is a 1% deviation from recipe. At 200 kg of concentrate, it is 2.5%. Those errors are systematic. They do not average out over time. Every batch is wrong in the same direction.
The Topcon Digi-Star scale systems fitted as standard on LEO Agriculture‘s concentrate mixers carry an accuracy specification of less than ±0.01% of full scale. At a 1,000 kg full-scale capacity, that translates to an error of 100 g or less across the entire weighing range. The practical difference between a ±0.01% indicator and a ±0.5% indicator is not theoretical. It is the difference between a 200 g daily error and a 50 g one, across every batch, every day, for the life of the machine.
The 2104 and 2106 models in the 21 Series concentrate mixer range are the machines where this specification matters most in a commercial context. These are horizontal-auger units designed specifically to blend concentrates, grains, molasses, oil cakes, and mineral packs into a uniform supplement. They are not full TMR machines. Their job is to produce a consistent, accurately weighed concentrate batch that can be added to a TMR or fed separately, and indicator accuracy is the single specification that determines whether they do that job reliably.
Batch-to-Batch Repeatability Is a Different Problem
Single-batch accuracy and batch-to-batch repeatability are related but not the same thing. A machine can hit its target weight on Monday and drift 300 g off it by Thursday without the indicator ever failing a calibration check. Repeatability is what cows actually experience. They respond to consistency in their ration before they respond to the absolute quantity. A cow that gets 4.0 kg of concentrate three days running and then 3.7 kg on day four will show it in her milk before the weekly weigh shows it on paper.
Two things erode repeatability over time. The first is load-cell overload history. Load cells have a rated capacity. Push them beyond that, even briefly, and the zero point shifts. The cell still reads, but it reads wrong relative to where it started. This happens more often than most operators realise, usually when an ingredient is loaded aggressively or a delivery auger dumps a slug load rather than metering smoothly. LEO Agriculture fits 10,000 kg (22,046 lb) capacity load cells as standard on their vertical TMR machines, specifically to reduce the overload risk. On concentrate mixers working with lower batch weights, overload is less likely to be a physical issue, but the principle still applies: a cell that has been stressed reads less consistently than one that has not.
The second factor is recipe lock integrity. A 20-recipe programmable scale, like the Digi-Star system referenced above, stores target weights for each ingredient in each ration. If those stored values are accidentally edited, or if the operator manually overrides a target weight without updating the recipe, the machine continues producing what it thinks is the correct batch while the actual ration has drifted. Checking recipe integrity against the nutritionist’s current formulation should be a monthly task, not an annual one. Ingredient prices change, rations get reformulated, and the scale recipe list can quietly fall out of sync with the feeding programme without triggering any alarm.
On the inspection side, there are four things to check before spring feeding ramps up. Zero the scale with an empty machine and confirm the reading holds stable for two minutes under normal engine vibration. Check the load-cell mounting bolts for movement, since a loose cell introduces variability that no calibration will fix. Load a known test weight and compare the reading against a certified reference. Then run three consecutive batches at the same recipe setting and compare the recorded batch weights: if they vary by more than 0.5% from each other, the cell or the indicator needs attention before it costs you a full lactation’s worth of drift.
If you want to understand how ISOBUS-integrated weighing extends this kind of control into the full TMR mix cycle, the article on how ISOBUS changes the way you manage TMR ration accuracy covers the data-logging and recipe-control side in detail. And if you are assessing concentrate mixer capacity alongside a larger vertical TMR machine, the guide on sizing a vertical TMR mixer when your herd is still growing addresses how to match machine volume to your current and near-term herd.
What to Demand from a Concentrate Mixer Specification Sheet
When you are evaluating a concentrate mixer, four specifications carry real weight. First, indicator accuracy expressed as ±% of full scale. Anything above ±0.1% full scale is a liability at commercial concentrate volumes. Second, load-cell rated capacity relative to your maximum batch weight: the ratio matters more than the absolute number. Third, programmable recipe count: 20 recipes covers most commercial operations without forcing operators to improvise on the floor. Fourth, protection class for the indicator: IP68 dustproof and waterproof rating means the display and electronics survive the kind of environment concentrate handling actually creates.
Resolution matters too. An indicator that displays in 0.5 kg increments cannot confirm a 200 g target weight. An indicator displaying in 100 g increments can. That difference in display resolution is what separates a machine that enforces your nutritionist’s ration from one that approximates it.
The maths at the start of this piece used round numbers to keep the arithmetic clean. The actual cost on your operation depends on your concentrate price, your exact herd size, and how far your current system drifts from recipe. But the structure of the calculation does not change. Small daily errors, multiplied by herd size, multiplied by lactation length, produce large annual costs. The specification that prevents it is already on the data sheet. You just have to ask for it.
