RECD Faults After a Fuel Delivery: Building a Timeline

The tanker came on Monday. By Thursday, the DG set’s fuel filter was dirty, the engine was surging a little at load and the retrofit emission control device (RECD) controller was reporting a differential pressure that was creeping up. By the weekend the facility team had decided that the diesel was bad. There was a delivery note, there was a date and there was a sudden change in behaviour. The story told itself.

Sometimes the story is right. Fuel does cause problems, and a delivery is a real event that can disturb a tank, bring in water or change the fuel in the tank. But a convincing story is not the same as a proven one. A dealer service group that studied generator no-run complaints reports that technicians often blame the fuel without checking, and that when samples are tested the fuel frequently turns out to be fine (Bell Performance public service announcement). Blaming fuel when it is not the cause leaves the real fault in place. Missing a genuine fuel problem lets it spread through the system.

This article shows how to settle the question with a timeline. We lay out the events from a delivery onward on four parallel tracks, fuel, engine, exhaust and controller, all on the same clock, and then use delay windows to ask which symptoms could plausibly have been caused by the fuel, and when. We cover the seven suspects, a sample-retention routine you can start at the next delivery, and a worked scenario. The advice is general. Your engine maker’s fuel specification and your RECD supplier’s guidance come first.

Why a delivery can change things

A delivery changes a fuel tank in several ways at once, and each has a different effect on the engine and the RECD.

It disturbs what was already there. Over time, sediment, rust, sludge and microbial growth settle at the bottom of a tank, and water sits below the diesel. A fill stirs the tank. Fuel-management guidance notes that water and sediment come from condensation, rain, vent and cap damage and problems during delivery, and that the contaminated layer sits where it can be picked up when the tank is disturbed (Senergy Petroleum). A marine-fuel guide describes the same thing: sludge that has built up in mats breaks apart when disturbed and can overwhelm filters, which is why an engine can run fine and then start to struggle after a tank is topped up (Tom Haringstad’s fuel system notes).

It brings in new fuel. The delivered diesel may differ from the previous batch in sulphur content, density, cetane, additives, ash-forming contaminants, biodiesel content and water. Fuel quality management advice from an equipment dealer notes that fuel can begin to degrade within weeks of refining and that poor storage conditions can accelerate it (Wheeler Cat). See our guides to ultra-low-sulphur diesel and emission control and biodiesel blends in an RECD-equipped generator.

It mixes old and new. Many DG sets use fuel slowly. A tank that is topped up every few months holds a blend of fuel of different ages. Old, oxidised fuel can produce gums and sediment, and a fresh delivery can dissolve and carry some of it.

It can bring in water. From a tanker, a loading point or a hose, especially in wet weather.

It can arrive at a different temperature, which matters for cold-flow in winter and for expansion and vapour in summer.

So a delivery is not just “new fuel”. It is a bundle of changes, and the symptoms depend on which ones took place.

Four tracks, one clock

The timeline method is simple. Draw four horizontal lines, one above the other, with time running left to right from a few days before the delivery to a few weeks after. Mark the delivery as time zero.

TrackWhat goes on itWhere the data comes from
F: FuelDelivery date and time, quantity, supplier, tanker, tank level before and after, any water dip or sample, filter changes, tank cleaningDelivery note, fuel log, maintenance log
E: EngineStarts, run hours, load, smoke, surging, power loss, fuel filter condition, oil changes, injector or turbo workDG controller, operator notes
X: Exhaust and RECDΔP at a reference load, temperatures, regeneration events, cleaning, alarmsRECD controller log
C: Controls and contextWeather, outages, load changes, setting changes, firmware updates, service visitsBuilding log, weather record, change log

The aim is to see whether a symptom on Track E or X begins after an event on Track F and within a plausible delay. Our guide to what to record in an RECD event log and the article on aligning RECD sensor timestamps with generator runtime logs explain how to get the tracks to share a clock.

Do not skip Track C. Many “fuel problems” coincide with something else: a monsoon week, an outage spell, a service visit, a new load. If two events line up with the symptom, the fuel is only one candidate.

What the lines should show

For each symptom, ask three questions.

  1. Did it start after the delivery, or was it already there? Look back at least two to three months for early signs.
  2. Does the size of the symptom follow the fuel in the tank? If the tank holds a mix of old and new fuel, effects often ease as the delivered fuel is used up or as filters are replaced. A cause that does not change with fuel use is probably not the fuel.
  3. Are other sets on the same fuel affected? If a second set fed from the same tank or delivery is fine, the fuel is a weaker suspect. If both show the same change, it is a stronger one. See why one of two similar RECD installations clogs faster for the comparison method.

Delay windows: when could the fuel have done it?

Different fuel effects show up on different time scales. A symptom that appears outside the window for a given cause makes that cause less likely. The table is a thinking tool, not a law. Rates depend on the engine, the filter and the duty.

Delay after deliveryWhat fuel could plausibly doWhere it shows first
Minutes to hoursDisturbed sediment and water reaching the filter and separator; surging or stallingFuel filter, water separator, engine behaviour
Hours to daysFilters loading faster; injector or pump distress; smoke change from different fuel propertiesEngine track, smoke at the stack
Days to a few weeksChanges in soot character and quantity; rising ΔP if combustion changesRECD ΔP trend, regeneration frequency
Weeks to monthsAsh and non-combustible residues accumulating in the filter from fuel contaminants; microbial growth building upRising baseline ΔP after regeneration, rebuild-rate decline
Immediate but at one instant onlyA step change in ΔP at the moment of the first run after deliverySensor, tap, line, or a real blockage by debris (check everything)

If the symptom began months after the delivery, the delivery is unlikely to be the cause on its own. If it began within hours, look at sediment, water and filters. If it began within days to weeks and involves soot and ΔP, ask whether the fuel’s combustion properties changed.

The seven suspects

Suspect 1: Sediment and sludge stirred up in the tank

Signature on the tracks. Fuel track: a fill into a tank that had not been cleaned for a long time. Engine track: fuel filter becomes dirty quickly, surging or power loss at load within hours to days. Exhaust track: usually no change at first.

Evidence. Dark, gelatinous or gritty material in the filter, a dirty separator bowl and sediment in a sample drawn from the tank bottom.

Test. Draw a sample from the tank’s lowest point, inspect it in a clear jar and examine the removed filter element. A guide to troubleshooting no-run complaints suggests cutting open removed filters to inspect the media as a first diagnostic step (Bell Performance). Our fuel filter replacement guide explains how.

Suspect 2: Water in the fuel

Signature. Water in the separator bowl, rough running, white smoke at start, possible corrosion. Water often rides in with a delivery, or with rain and condensation afterward.

Evidence. A clear water layer in a sample jar, a full or alarming water separator and a filter wet with water.

Test. Drain and inspect the separator, take a bottom sample and send a sample to a laboratory for water content if there is doubt. Microbial growth feeds on the fuel-water interface, so water problems can recur. See wet weather faults for the weather angle.

Suspect 3: Microbial growth released or fed

Signature. Black, brown or slimy deposits, rapidly plugging filters and a sour smell. It can follow a disturbance of the tank, as old growth breaks up.

Evidence. Slime on the filter or in the sample, and a laboratory microbial test.

Test. Laboratory testing is the sure way. A fuel supplier’s guide notes that testing is especially useful when plugging continues after basic maintenance, and that sampling, water and sediment testing, microbial testing and laboratory analysis make up a proper fuel testing process (Senergy Petroleum, above).

Suspect 4: Different fuel properties

Signature. Smoke or soot behaviour changes within days, power or fuel consumption shifts, regeneration behaviour changes.

Evidence. A change in smoke or exhaust temperature at the same load, with the same engine settings, from the first run on the new fuel. A change in the delivery paperwork: supplier, grade, density or biodiesel content.

Test. Compare with the delivery note and the specification the fuel was sold to. Ask the supplier for the batch certificate. A laboratory analysis can check cetane, distillation, density and contaminants. If the fuel passes, you have eliminated it, which is as valuable as finding it.

Suspect 5: Biodiesel or additive content

Signature. Filters plugging after a switch, haziness or colour change, changed lubricity. A diesel-technology guide notes that biodiesel has a limited storage life, that degrading biodiesel raises acidity and reduces lubricity and that sediment-related filter plugging can follow until the sediment is used up (DieselHub).

Evidence. Delivery paperwork showing a blend, colour or haziness change, and a pattern of filter plugging that eases as the old fuel is consumed.

Test. Check with the supplier and your RECD supplier whether the blend is permitted. See biodiesel blends in an RECD-equipped generator.

Suspect 6: Ash-forming contaminants in the fuel

Signature. A rise in baseline ΔP after regeneration that does not recover, building over weeks to months, with no change in engine behaviour. This is the slowest of the suspects and the hardest to prove.

Evidence. Cleaning reports showing heavier or different ash. A step in the rebuild rate that matches a change of supplier.

Background. A SAE experimental study found that a fuel containing ash-forming compounds contributes non-regenerable deposits to a diesel particulate filter, that a higher-sulphur fuel had a marginal effect on ash accumulation but a larger effect on exhaust back-pressure, and that more knowledge is needed about the role of contaminated fuel and fuel filters (SAE 2007-01-1925). It is an engine-bench study on a vehicle filter, so treat it as a pointer. Our article on soot versus ash in RECD regeneration explains why ash does not burn away.

Test. Track the baseline ΔP after each regeneration at a reference load, and compare periods before and after the supplier change. Ask the cleaner to describe and photograph the ash.

Suspect 7: It was not the fuel

This one deserves its own place on the list. Possible alternatives include an air filter that was neglected, a weather effect, an injector problem, a change in duty and a service visit that coincided with the delivery. The timeline exists to keep this suspect honest.

A reading table

What you see on the tracksLeans towardNext move
Fuel filter dirty within hours; surging; no RECD changeSediment or waterTank bottom sample, filter inspection, drain separator
Water layer in sample; white smoke at startWaterDrain tank bottom, test fuel, check vents and caps
Slime, sour smell, fast pluggingMicrobial growthLaboratory test, tank cleaning plan
Smoke and ΔP change within days, filters cleanFuel properties or something elseDelivery paperwork, laboratory fuel analysis
Filters plugging after a blend changeBiodiesel or stabilityConfirm blend permission, fuel age
Baseline ΔP floor rising over months after supplier changeAsh-forming contaminants (possible)Rebuild-rate comparison, ash report
Symptoms began before the deliveryNot the fuelLook back through Track C
A twin set on the same fuel is fineNot the fuel (or weak)Compare differences between the sets

How to test the fuel without starting an argument

Fuel disputes become expensive because they involve a supplier, a carrier and a service company. A neutral, documented test avoids most of the argument.

  1. Draw a sample properly. From the tank’s lowest point and from mid-height, into clean, labelled containers. Take duplicates.
  2. Retain a sample from the delivery itself (see below).
  3. Photograph the sample jar against a plain background right away, and again after an hour of settling.
  4. Send one sample to an independent laboratory and ask for the tests that match your suspicion: water and sediment, contamination, microbial growth, and fuel property checks against the specification the fuel was sold to. Ask the supplier which specification applies. In India, automotive diesel is sold to a Bureau of Indian Standards specification, and the supplier should be able to say what applies to your fuel.
  5. Keep the second sample sealed in case of a dispute.
  6. Record the chain of custody: who drew it, when, from where and who handled it.

A test result that says the fuel is within specification does not close the matter. It narrows the field. A result that shows water, sediment or microbial growth moves attention to tank hygiene, as the contamination may be old and only now disturbed. See fuel and engine condition as a confound in RECD tests for how fuel interacts with measurements.

A sample-retention routine for every delivery

Starting this at the next delivery pays for itself the first time a dispute arises.

Before the delivery

  • Dip or gauge the tank and record the level.
  • Drain any water from the bottom and note what came out.
  • Photograph the tank vent and fill cap condition.
  • Note the weather.

During the delivery

  • Record the delivery note: supplier, tanker number, quantity, grade, density if stated, time.
  • Draw a sample from the delivery hose or tanker outlet if the supplier allows, into a clean, labelled jar. Keep it for at least a few months.
  • Watch for a sign of water or dirt in the stream.

After the delivery

  • Wait for the tank to settle before running sensitive loads, where operations allow. A boating guide, for example, advises waiting some hours after a roadside station’s own tanker delivery so that stirred debris and water can settle (BoatTest). The principle is general.
  • Dip the tank again and compare levels with the delivery quantity.
  • Check the fuel filter and separator after the first loaded run, and again after a week.
  • Add the delivery as an event on Track F, with its time.

A worked illustration: the Monday tanker

The following is an illustrative scenario, not a real customer. A hospital annexe has a 320 kVA DG set with an RECD, and its fuel tank is topped up from a bulk tanker about every three months. A delivery arrives on a Monday. By Thursday, the fuel filter is dirty, and by the following Monday the RECD controller shows ΔP at 50% load rising from 28 to 36 mbar.

Track F. One delivery of 3,000 litres. The tank had been last cleaned two years earlier. A bottom sample was not taken before the fill.

Track E. The fuel filter was replaced on Thursday because it was dirty, and the replacement is also dirty a week later. The engine surges slightly at load steps. There is no change in smoke.

Track X. ΔP at the reference load starts creeping up in the second week. The RECD has not been cleaned for eight months. Regeneration events are normal.

Track C. It was a dry week, with two short outages and a service visit by the electrical contractor on Tuesday, who disconnected the DG set’s control cables temporarily. A rise in load of about 10% came from a new chiller installed the previous month.

Reading the tracks. The filter dirt and the surge point to sediment (Suspect 1) as a fast effect of the delivery, in the hours to days window. The ΔP creep begins more than a week later, is modest and has other candidates: eight months since cleaning and 10% more load. The team does not find a link with the fuel’s properties because smoke is unchanged.

Action. They draw a bottom sample, which contains dark sediment and a trace of water, and send a sample to a laboratory, which reports acceptable fuel properties but sediment above the supplier’s usual limit. The tank is cleaned and the fuel polished. The filter is replaced a final time and stays clean. For the RECD, the team checks the sensing line after the electrical visit, compares the reading with an independent gauge, and schedules the overdue cleaning. ΔP returns to 29 mbar after cleaning. The delivery explained the engine-side symptoms but only part of the RECD trend, and the timeline showed which was which.

Questions people ask

Can bad fuel cause an RECD to clog?

It can contribute, through changed soot, ash-forming contaminants or fuel-system trouble that raises soot output. More often, the engine-side symptoms appear first. Use the timeline and delay windows to judge.

How soon after a delivery should I check the fuel filters?

After the first loaded run and again after about a week. A filter that loads quickly after a delivery is an early warning of sediment or water.

Should I change the supplier?

Not on a hunch. Test the fuel, compare with the specification, keep records over several deliveries and ask the supplier for batch certificates. A pattern across deliveries is more convincing than one bad week.

Is a fuel additive the answer?

Additives are sold for many purposes, but they do not remove sediment, and some can complicate matters. Do not add anything that the engine maker and the RECD supplier have not approved. See can diesel fuel additives replace an RECD.

Does the same approach work for a recently relocated set or a new service contract?

Yes. Any single event can be placed on a timeline. See RECD malfunction after generator relocation.

Takeaway

A fuel delivery is a convincing suspect and an easy one to blame. Draw the four tracks on one clock, apply the delay windows, work through the seven suspects, retain a sample from every delivery and test before you accuse. If the fuel is cleared, you have saved the search from a dead end. If it is implicated, you have the records to act on. If you would like help drawing a timeline for your own installation, or setting up a fuel receiving routine, see our retrofit emission control devices, read the FAQ or contact the Aceget team.

Related Reading

Sources



2 Comments

Leave a Reply