Executive Summary

Flare system design is intricate, governed by API standards, operator philosophies, and sophisticated radiation modeling. Engineers produce detailed reports showing acceptable heat flux levels at critical locations <  2 kW/m² at normally manned areas.  The calculations look perfect on paper. But here’s the reality: at < 2 kW/m², people are already running. Vessels are too hot to approach. That ‘acceptable’ number on your radiation plot? Nobody volunteers to stand there when the flare is actually roaring. Process upsets don’t last seconds – they can rage for 30+ minutes or even hours while operators scramble. And when a helicopter is landing and your largest blocked discharge case occurs with unfavorable wind, those carefully calculated contour lines become meaningless.

Even more troubling: the disconnect often starts earlier in the project lifecycle. HAZID workshops generate actions like ‘ensure helideck availability 100% of the time – size flare boom accordingly.’ Months later, detailed engineering reveals the platform can only guarantee 90% availability due to wind variability and blocked discharge scenarios. That missing 10%? It represents the exact moment when a helicopter is landing and an emergency occurs. Yet this critical gap frequently goes unnoticed because action tracking relies on email chains and Excel spreadsheets instead of integrated engineering systems.

This article examines the uncomfortable gap between flare design theory and operational reality, particularly for high-producing platforms squeezing everything onto a single structure. We’ll explore why 800+ MMSCFD platforms can have onboard flare towers, what happens when design assumptions meet real process upsets, how critical safety requirements get lost in translation between project phases, and the cost-risk trade-offs engineers rarely discuss openly.

Too Hot to Handle? The Action Tracking Problem

Before we dive into the technical complexities of flare design, we need to address a systemic problem that plagues large engineering projects: the gap between safety study requirements and detailed design reality. Consider a typical HAZID workshop for an offshore platform. The team identifies radiation hazards from the flare system. An action is raised:

Action #247: Ensure helideck remains available for emergency landing operations 100% of the time, regardless of flaring conditions. Size flare boom and adjust platform layout accordingly. Owner: Flare Lead Engineer. Due: FEED completion.

The action gets logged in an Excel tracker. Emails circulate. Months pass. Detailed engineering progresses. The flare engineer runs API 521 radiation calculations for all credible scenarios – planned blowdown, blocked discharge, multiple PSV lifts, various wind directions.

The final radiation study concludes: ‘Helideck radiation levels are within acceptable limits per company philosophy for 90% of flaring scenarios. During unfavorable wind conditions with blocked discharge events, helideck availability may be temporarily restricted per emergency response procedures.’

That 10% gap? It’s not trivial. It represents exactly this scenario:

• Helicopter inbound, landing clearance given
• Compressor anti-surge valve fails, 800 MMSCFD goes to flare
• Wind was in the unfavorable quadrant
• Helideck radiation spikes to about 2kw
• Pilot has 90 seconds to decide: land in extreme heat or divert 40 minutes to nearest alternate

Is it too hot for the chopper to land?

The real question is: did HAZID Action #247 ever get properly closed? Did anyone verify that ‘100% availability’ requirement against the detailed radiation study showing ‘90% availability’? Who reconciled this discrepancy?

In our experience across dozens of major projects, the answer is often: somebody has checked but its not been approved by client 

HAZID actions get logged in Excel. Email trails grow long and confusing. Different engineering disciplines use different tracking systems. The flare engineer closes the action based on ‘meeting radiation criteria per company philosophy.’ The HAZID chair assumes 100% availability was achieved. The operations team inherits a platform where the emergency response plan and the design basis don’t quite align.

We’ve seen this pattern repeatedly throughout large engineering projects: critical safety requirements established in early-phase studies get diluted, reinterpreted, or lost entirely as they cascade through FEED, detailed design, and construction. Not through malice or incompetence, but through the limitations of email and Excel-based action tracking in complex, multi-discipline projects.

Action tracking is critical. Yet we still see emails and Excel spreadsheets as the primary tools, even on billion-dollar projects where a single missed action can compromise safety or operability.

This brings us to the deeper technical question: beyond the action tracking problem, are our flare design calculations themselves aligned with operational reality? Let’s examine the gaps.

Challenging the Conventional Wisdom

Conventional wisdom in the offshore industry suggests that high-producing gas platforms – particularly those around 800 MMSCFD – should route their relief loads to remote flare bridges or dedicated flare platforms. The logic seems sound: massive gas volumes create intense radiation that endangers personnel and equipment on the main platform.

But the real answer is more nuanced. You can operate an 800+ MMSCFD platform with an onboard flare tower. The question isn’t whether it’s possible – it’s whether you understand and accept the operational constraints that come with it. And more importantly, whether your engineering calculations acknowledge what actually happens when that flare ignites.

The Two Governing Design Cases

Flare system design is governed by two fundamentally different scenarios, each with distinct characteristics and safety implications. Understanding which case controls your design is critical – and often, it’s not the one you expect.

Case 1:  Blowdown

This is the ESD activation, planned depressuring, or turnaround operations. Characteristics include:

• Lower instantaneous flowrates compared to blocked discharge
• Manageable radiation levels not governing though
• Rarely governs minimum safe distance or tower height

Blowdown cases are important for overall system sizing and knockout drum design. We will have another article to state why blowdown rates should be increased past API 521

Case 2: Blocked Discharge (Process Upset)

This is normally your governing

A pressure safety valve lifts due to blocked outlet – compressor discharge isolation valve closed by mistake, export line blockage, downstream vessel full, or instrumentation failure. What happens:

• Much higher instantaneous flowrates – its the full flow of the platform
• Can be sustained for extended period
• Multiple PSVs may lift simultaneously – cascading effects
• Heat flux must not exceed that 4 ish kW/m2
• Potential for liquid carry-over if knockout drum is overwhelmed

Blocked discharge scenarios, not production capacity, should drive your flare location decision. This is often the governing case for flare tower height, separation distance, and whether an onboard flare is even feasible. Yet many platforms are designed primarily around blowdown cases because they’re easier to model and more predictable.

High Producing Platforms with Onboard Flares: The Reality

Can you operate an 800+ MMSCFD gas platform with an onboard flare tower? Its been done, with constraints

Why It Can Work

Several factors can make onboard flares viable even for high production:

• Adequate flare tower height reduces ground-level radiation – 120-150m length towers aren’t uncommon
• Well-managed blocked discharge cases through proper overpressure protection philosophy
• Strategic platform layout – critical equipment and personnel areas positioned away from radiation zones
• API 521 radiation modeling showing ‘acceptable’ exposure levels

The key phrase above is ‘acceptable’ – and that’s where theory diverges from practice.

The 100% Availability Myth

Here’s what nobody wants to write in the safety case: If your flare is on the same platform, you sometimes cannot have 100% availability all the time, maybe at the helideck and other sensitive operating areas

 

Option 1: Dedicated Flare Platform

The Gold Standard

Complete separation means no radiation impact on living quarters, helideck, or process areas. 100% availability on the main platform during flaring events. Clearer safety case, easier barrier management, simpler maintenance access.

The brutal reality:

• Capital cost: $150-300M+ for structure, jacket design, piling, installation
• Marine spread costs for installation
• Additional pipeline/riser tie-ins and subsea infrastructure
• Separate maintenance campaign requirements

Only economically viable for large field developments, major processing hubs, or when regulatory requirements absolutely demand it.

 

Option 2: Platform Edge

Flare boom extends 80-150m from the platform edge, reducing but not eliminating radiation impact. Still affects some deck areas, helideck availability may be limited during extreme flaring. Structural complexity from cantilever loading and wind moments.

Typical cost: $30-60M additional over a simple tower. Provides meaningful separation without the full expense of a dedicated platform.

What Actually Happens at < 2 kW/m²

At < 2 kW/m², people are already running. Vessels are too hot to approach. That ‘acceptable’ number on your radiation plot? Try standing there when it’s actually flaring.

 

Engineering assumption: Tolerable for brief periods, personnel can maintain station for essential operations.

 

Operational reality:

• Exposed skin becomes uncomfortable in 20-30 seconds even at < 2Kw
• Instrument reading impossible – you can’t stay to monitor
• Radio communication difficult – people backing away instinctively
• Any maintenance activity? Forget it
• Even with full PPE, exposure feels dangerous and unsustainable

Who volunteers to stand at the < 2 kW/m² contour line to verify our calculations during a real flare event?

What the Calculations Say

Helideck radiation exposure: 1.6 kW/m² – within acceptable limits per operator philosophy. Safety case approved.

What Actually Happens

• Intense radiant heat washing over the helideck
• Helicopter fuel tanks heating up
• Pilot and crew feeling the blast, making split-second decisions
• Passengers in survival suits sweating profusely
• Do you take off immediately? (risky during active process upset)
• Do you shut down and evacuate to TEMPSC? (where exactly – everywhere is hot)
• Communications with platform difficult (crew evacuating radiation zones)

The unwritten procedure: If flare ignites during helicopter operations, pilot decides whether to evacuate immediately or shelter. Platform cannot guarantee helideck availability – even though the design says it’s acceptable.

Paper vs Reality: The Institutionalized Gap

The offshore industry has institutionalized a disconnect between what engineering reports say is acceptable and what operations actually do.

On Paper (Engineering Deliverables)

• Radiation contour plots showing precise 2 kW/m² boundary lines
• ‘Helideck meets radiation criteria’ – checked and approved
• ‘Escape routes available’ – verified on drawings
• ‘Muster areas protected’ – within acceptable limits
• Safety case approved by authorities

 

In Reality (3 AM Process Upset)

• Alarms blaring, flare roaring with 80 MMSCFD discharge
• Helideck unusable despite ‘meeting criteria’ – operators evacuated
• Primary muster point too hot, people relocating to secondary location
• Nobody verifying vessel conditions in the radiation zone
• Medevac helicopter 40 minutes away – can’t land safely until flaring subsides
• Operations manual clearly states: evacuate these areas during flaring

Our calculations show acceptable radiation levels. Our operations manual says evacuate those areas during flaring. Which one is the real design basis?

Why the Gap Exists

Several factors create and perpetuate this disconnect:

• Duration assumptions – Calculations assume brief transients, reality shows sustained events
• PPE limitations – Nomex coveralls aren’t heat shields for extended exposure
• Psychological factors – Large visible flame triggers instinctive retreat regardless of numbers
• Concurrent hazards – Noise, potential liquid carryover, flame stability concerns compound the threat
• Escape route anxiety – ‘If this gets worse, where do I go?’ drives conservative behavior

 

Conclusion: Making Peace with Reality

The 800 MMSCFD question isn’t really about flow rate. It’s about what you’re willing to accept:

1. Accept the cost – Build a dedicated flare platform, sleep soundly knowing you have true separation and 100% availability.
2. Accept the constraints – Onboard flare tower with full acknowledgment of operational limitations. Manage them rigorously, train personnel extensively, plan for restricted access during flaring.

The dangerous middle ground is when you squeeze the design but don’t fully acknowledge and plan for the operational constraints. That’s when your safety case has gaps between what the calculations say and what actually happens during that 3 AM blocked discharge event with personnel scrambling to evacuate the helideck.

And perhaps most importantly: when HAZID actions demanding 100% helideck availability get marked ‘closed’ while detailed engineering shows only 90% is achievable, and nobody connects those dots because action tracking lives in Excel spreadsheets and email chains.

I’ve reviewed dozens of flare designs over my career. The best ones made a clear choice and owned the consequences. The worst ones tried to have it both ways – claiming 100% availability while the radiation calculations and operations procedures told different stories. The most dangerous ones never reconciled the requirements from early-phase safety studies with the realities of detailed design.

The numbers don’t lie – but sometimes they don’t tell the whole truth either. When your radiation plot says 2 kW/m² is acceptable but your operators won’t stand there, when your HAZID demanded 100% but your design delivers 90%, you need to ask yourself: which one represents reality?