The field dilemma and risk
A refinery changes demulsifier because residual water, salt or rag is unsatisfactory. The laboratory result looked better, but the operating unit shows little benefit. Increasing the dose without closing the evidence chain can raise cost, mask a delivery fault, change brine quality, or distract from temperature, mixing, interface, wash-water or electrical limitations.
What is easily misdiagnosed
- “The new product passed the bottle test, so the unit should improve.” A bottle test is useful for screening, but it does not reproduce the full electric field, distribution, residence and withdrawal system.
- “No improvement means the chemistry is weak.” Wrong concentration, blocked injection, poor calibration or an unsuitable injection location can create the same result.
- “A higher dose must be safer.” Demulsifier performance is selective and can plateau or deteriorate beyond its useful range.
- “Fast water drop proves success.” Desalted-oil quality, interface/rag and brine oil/solids must also be acceptable.
Decision mechanism
Chemistry-fit branch
The supplied chemical reaches the process as intended, but the current crude blend, emulsion film, solids or ageing differs from the evaluation sample. Re-screen with representative material and evaluate oil, water and interface together.
Delivery-and-process branch
The formulation may be suitable, but dose preparation, injection, mixing, temperature, residence, water load, interface or electrical availability prevents it from working in the field.
Step-by-step troubleshooting path
1. Freeze the comparison
Record the before/after dates, crude blend, throughput, temperature, wash water, mixer differential pressure, interface/rag, voltage/current, product analysis and brine condition. Do not compare unsynchronized averages.
2. Prove chemical identity and delivery
Verify product identity, batch, storage condition, prepared concentration, tank mixing, metering-pump calibration, stroke response, line condition, check valves, injection quill and actual injection point. Reconcile consumption with measured throughput.
3. Confirm that the evaluation sample is representative
Compare the screened sample with the current crude blend and its storage/transport history. Changes in crude, solids and emulsion ageing can invalidate a previous ranking.
4. Evaluate the whole separation outcome
Trend desalted-crude salt and water together with separation rate, interface/rag thickness, brine oil/solids and electrical stability. A chemical that drops water quickly but creates dirty brine or persistent rag is not automatically the best choice.
5. Check the operating window
Review temperature, viscosity evidence, wash-water amount and quality, injection sequence, mixing severity, residence, distribution and brine withdrawal. A chemistry trial cannot compensate for an unverified hydraulic or electrical constraint.
6. Run a controlled comparison
Use representative laboratory screening to narrow candidates, then define a controlled field trial with a baseline, stabilization period, one-factor-at-a-time changes, sampling plan and stop criteria. Do not convert historical book values into a site target.
Cause-to-direction map
- Consumption is lower than the pump calculation: inspect preparation, calibration, suction, valves, blockage and leakage.
- Laboratory response is good but the field is unchanged: verify sample representativeness, injection location, contact time, mixing and distribution.
- Water drops faster but rag remains thick: compare interface evolution, solids and current crude condition; do not judge on water-drop speed alone.
- Oil quality improves while brine oil worsens: optimize against both outlet streams, not one laboratory number.
- Performance changes with crude switching: re-screen or conditionally blend chemistry; a universal formulation should not be assumed.
Trial checklist
- Product identity, batch, storage and prepared concentration
- Metering-pump calibration, real consumption and injection point
- Representative crude blend and sample handling history
- Temperature, throughput, wash water and mixing differential pressure
- Interface/rag trend and independent level evidence
- Desalted-oil salt/water and brine oil/solids using consistent methods
- Voltage/current behaviour, stabilization time and stop criteria
Applicability boundary
This framework does not specify a product, dose, dilution, injection point, temperature, mixing setting, settling time or outlet guarantee. Those values require representative crude testing, project hydraulics, chemical-provider data, plant instrumentation and an approved trial plan. Bottle tests support screening; they do not by themselves prove full-scale performance.
Source note
Primary sources: Crude Oil Electrostatic Desalting and Dehydration Technology, professionally reconstructed and two-round reviewed edition V0.5, PDF pages 35 and 100–101 (printed pages 23 and 88–89), plus PDF page 162 (printed page 150) for evaluation logic; and Atmospheric and Vacuum Distillation Technical Q&A, Chapter 4, refined electronic edition, PDF pages 8–9 (printed pages 52–53). Historical dose examples and test recipes in the source were deliberately excluded from current recommendations.
Need to review a chemical trial?
Provide synchronized operating data, actual chemical delivery records and both oil- and brine-side results.