Why Colour-Sorting Matters in Loba, and What “Uniformity” Really Costs

When a retail packer or exporter asks for “uniform” loba (alsande / cowpea / black-eyed pea), they are really specifying a discard rate. Colour sorting removes off-colour, stained, mottled, and defective seeds from the accepted lot. What leaves the yard as sorted loba is what remains after that removal. The cost of uniformity is, therefore, largely a function of how much of the incoming lot must be discarded to meet spec, and in our experience at KVM & Co., Mysore, buyers routinely underestimate this number.

What buyers actually mean by “uniform”

“Uniform” is not a single standard. In practice, when processors and packers use the word, they are usually bundling together four separate requirements:

Colour consistency
Predominantly one colour class, white / cream for white loba (karamani), or red-eye for red-eyed varieties, with off-colour seeds typically below 2–3% by count.
Broken percentage
Seeds split or missing more than a quarter of their body. Below 2% for premium lots. BIS IS 1154 (cowpea and related pulses) uses this as a principal quality parameter.
Foreign matter
Stones, soil, weed seeds, husk. Below 1% for retail or export. AGMARK Grade 1 sets the same ceiling; Agmarknet publishes the full schedule.
Weeviled seeds
Seeds with pulse beetle exit holes. Below 1% for domestic retail; export markets frequently require zero tolerance.

A lot can pass on colour and still carry 4% weeviled seeds. Uniformity is the intersection of all four, and each one costs you something.

The uniformity trade-off: what each grade discards

The table below reflects typical outcomes we observe at the yard for mixed-colour loba arrivals. Actual discard varies with crop year and source quality, but these bands are representative.

Grade Sorting method Typical discard Colour pass rate Broken & FM Relative price vs. unsorted
Standard (minimal sort) Destoning + winnowing only ~5% Not guaranteed <4% broken; <2% FM Base
Premium colour-sorted Hand-grading + optical sensor pass 10–15% >97% on-colour <2% broken; <1% FM +₹300–500 / quintal
Export-grade ultra-sorted Optical sort + size screen + manual top-up 20–25% >99% on-colour <1% broken; <0.5% FM; zero weevil +₹700–1,200 / quintal

The discard re-enters trade as feed-grade loba at a lower price. But the lot you dispatched cannot recover that weight, that is the true cost of the spec you agreed to.

“The part you don’t see leaving the yard is often the most expensive part of a uniform lot.”

How colour sorting works at KVM

At KVM & Co., Mysore, we offer colour-based sorting as a value-added service on loba lots at APMC Yard, Bandipalya. Sorting runs in two stages:

  1. Optical sensor sort. Seeds pass through a channel sorter; sensors flag off-colour seeds and ejectors remove them in real time. One pass typically reaches the premium colour-sorted band.
  2. Manual top-up. For tight-spec export or retail-pack orders, hand-graders check the sorted output and remove any seeds the optical pass missed, including broken and weeviled seeds sensors are not calibrated to catch. We apply this step only where the spec genuinely requires it.

Hand-grading alone works for smaller lots with low natural colour variation. For mixed-origin arrivals or late-season loba, the optical pass is faster and more consistent.

When paying for colour sorting is worth it

  • Retail-packed pulses. The consumer sees through the bag; off-colour seeds drive returns. For FMCG packers, the sorting premium is absorbed easily by the retail margin.
  • Export shipments. Gulf, East Africa, and South-East Asian markets carry strict appearance norms. Import rejection on colour grounds costs far more than sorting does here.
  • Food-service catering. Hotels and large caterers buying loba for whole-seed dishes treat colour and size consistency as an operational need, not an aesthetic one.

When it is not worth paying for

  • Industrial grinding. Milled into besan, flour, or extrudate, whole-seed colour is irrelevant. You are paying a sorting premium and discarding it the moment the lot enters the grinder.
  • Large institutional buyers cooking at scale. Pressure-cooked or mashed loba in canteens has no appearance requirement at the seed level. Broken percentage affects cook-time uniformity; colour does not.
  • When the incoming lot is already clean. Single-origin, same-season loba often arrives at 94–96% natural colour consistency. One light optical pass reaches spec at 6–8% discard. Full sorting charges on a visually clean lot are worth questioning.

Writing a uniformity clause that does not cost you yield. The most common mistake is specifying a colour percentage without a measurement method. “95% uniform colour” means nothing unless both parties agree on sample weight and whether it is counted by seed or by weight. A workable spec for sorted loba should state: (1) colour class and acceptable range by seed count on a stated sample weight, (2) broken percentage by weight, (3) foreign matter by weight, (4) weevil tolerance, and (5) what grade the rejected fraction is priced at, because that last point determines who carries the 15–25% discard risk.

Related reading

Technical

Reading a horse gram lot: visual quality markers procurement teams should know

Day in the Life

The hand-grading station: why we still trust eyes over machines

Technical

Destoning standards: how we measure clean