Across the dozen detergent powder plants I have walked through in north India over the last four years, the same three failures show up on the daily QC log. Bulk density out of band. Active matter under-spec. Caking after eight weeks in the warehouse. Three failures, one root cause each, each fixable, almost none fixed.

This post is for the plant manager or the QA head who is tired of the same three findings showing up on the morning report.

Failure 1: bulk density drift

Detergent powder is sold by weight; consumers measure it by volume (a scoop, a cup, a cap). The whole pricing model assumes a stable bulk density — typically 0.45 to 0.65 g/cc for spray-dried; 0.7 to 0.95 g/cc for agglomerated. When bulk density drifts, the consumer either over-doses (and complains about residue) or under-doses (and complains about wash performance). Either way the brand pays.

The root cause is almost never the formulation. It is the spray-drying tower's atomisation pressure — and specifically, drift in atomisation pressure across a single shift. When the night shift runs the tower at 38 bar and the morning shift runs it at 42 bar because the operator was trained to "compensate" for variable slurry temperature, the bulk density swings by 0.05 g/cc inside one production day. That is a 10% volumetric error at the consumer end.

The fix is not better operators. It is a closed-loop slurry-temperature control on the feed tank, with the atomisation pressure derived from temperature rather than set manually. It is a one-time instrumentation change — about ₹4–6 lakh installed — and it eliminates the bulk density finding from the daily report within two weeks.

Failure 2: active matter under-spec

Active matter (the surfactant content, typically expressed as %) is the spec line every QA lab tests first, and the one most likely to come back 1–3% under spec on month two of a new formulation. Brands react by raising the formulation by 1–3%. The underlying problem rarely gets diagnosed.

The actual diagnosis, almost every time, is hydrolysis loss in the slurry tank. LAS (linear alkylbenzene sulphonate), the workhorse anionic surfactant in most Indian detergent powders, hydrolyses slowly in alkaline conditions above 70°C. If the slurry sits in the feed tank at 75–80°C waiting for tower availability, it loses about 0.4% active matter per hour. Over a typical 4-hour wait, that is 1.6% — exactly the gap most brands are seeing on the QC certificate.

The fix has two parts: reduce the slurry tank residence time below 90 minutes (which is a scheduling problem, not a chemistry problem), and add a slurry-side cooler to hold the tank at 60°C between tower batches. The cooler costs about ₹2.5 lakh. The scheduling change costs nothing except a conversation with the tower operator.

Failure 3: caking after warehouse storage

The third failure is the most expensive because it shows up in the field, not in the QC lab. A pack that scoops cleanly on day one of dispatch is, by week eight in a Chennai warehouse with no humidity control, a brick that the consumer hits with the back of a knife.

Caking has two causes that get conflated: deliquescence (the powder absorbing atmospheric moisture and dissolving into a paste) and recrystallisation (sodium sulfate or sodium carbonate crystals re-forming bridges between particles). Deliquescence is fixed at the packaging end (a barrier-laminate inner liner solves it). Recrystallisation is fixed at the formulation end (the ratio of sodium sulfate to filler matters; >35% sodium sulfate practically guarantees a cake by week six in Chennai humidity).

Most plants try to fix caking with packaging alone. If the formulation is over-sulfated, packaging only delays the failure by a couple of weeks. The reverse is also true. Both have to be addressed.

A short cost-of-quality table

For a plant doing 30 MT/day, here is what the three failures cost annually if untreated, versus what the fixes cost once.

These are figures from a 30 MT/day plant we audited in early 2026. Your numbers will scale roughly linearly with capacity.

Why these failures persist

Most plants do not fix the three failures because the QA team and the production team report up different chains. QA flags the failure. Production owns the fix. The capex sits with neither. As long as the brand is willing to absorb the cost of over-formulation and the warehouse is willing to absorb the cost of returns, nobody up the chain has to make a decision.

The brands that fix this are the ones that have moved their daily QC review meeting to include both heads, with the plant CFO present. The conversation changes within a quarter.

If you are running a plant

Get in touch. We have walked through enough plants in similar straits that the diagnosis is usually faster than a fresh consultancy engagement, and we have the equipment vendors on speed dial for the closed-loop and slurry-cooler installs.