DP Compressed Air Pt.1 - Configuration

DP Compressed Air Pt.1 - Configuration

Introduction:  Dynamic positioning (DP) vessel compressed air systems store and distribute energy.  Even though the pipes may be passive, the transferred pressure is active.  This is a bit like bus bars being passive copper, but the generators, loads, and power transfers across them being active.  Like its electrical equivalent, compressed air distribution systems need to be split or properly protected, if they have DP vital loads.  Everyone knew this in the past, when vital pneumatic loads were common, but with almost all controls electric, people have forgotten this and equipment with vital pneumatic controls are sometimes fed by common compressed air systems.  DP vital pneumatic loads are assumed to no longer exist, but some still do.  This article will look at vital loads and configuration, while the next will look at operation, protections, and testing.

 

I could go into more detail about maintainability and operation but that’s next week - just redundant configuration right now. Assume there are unshown pressure relief valves in each system.

Configuration:  Compressed air systems normally consist of compressors, possibly dryers, air reservoirs, and the distribution.  There is normally a high pressure start air system and a lower pressure work or instrument air system.  Start air systems are often about 30 bar and their power is mostly sized to provide sufficient engine starts.  Work/instrument air systems are often about 8 bar and typically supply support systems.  The work system(s) might be supplied from the start air systems through pressure reducers or from dedicated work air compressors.  The instrument air system is normally differentiated from the work air system by dryer(s) and separate reservoir(s) and tends to supply control equipment requiring better air quality.  The start, work, or instrument air systems can be common if there are no DP vital loads or they are properly protected, but they should otherwise be split according to the redundancy group – 2 groups for 2 split, 3 for 3 split, and 4 for 4 split.  I have never seen a real redundancy design greater than 4 split, 2 splits are most common, and I have seen fake 3 splits (switched 2 split with common failure modes). 

 

Something you don’t want to happen to the work air system

Vents:  Speaking of splitting dissimilar things – please use separate high and low pressure relief lines.  I see too many designs where the start air and work air over pressure relief valves go to the same line.  This is already dangerous if the systems are independent, but some systems have one system feeding the other, so a start air overpressure causes a work air overpressure through a frozen or leaking pressure regulator, and a poor vent line then connects start air to work air and creates a greater work air overpressure.

 

The most vital pneumatic load is sometimes neglected.

Loads:  The DP vital loads are what determines if the system needs split or extra protection.  In the days of pneumatic governors on each engine, it was obvious that the control air supply needed to be split or each engine needed a dedicated big enough air reservoir fed by the check valve so a loss of pressure or pressure pulse on the main distribution wouldn’t immediately effect the running engines.  You might think that such pneumatic engine controls had gone the way of the dodo, but that is sometimes wrong.  The electrical controls are better and commonly used, but occasionally someone is nostalgic for pneumatic controls, and engine pollution limits means games need played with combustion air and fuel injection.  It was a design by a nostalgic engineer that first brought this to my attention.  He decided to go for the pneumatic governors that he remembered so well, but forgot that those old ships had split air systems, and forgot to split the system or add individual air reservoirs.  The split or bottles were an obvious feature when you walked the old ships, but the newer standard common design isn’t built to support DP vital loads from different redundancy groups.

 

“100% load step? No problem. Challenge me.” Modern engines need a pneumatic exoskeleton to provide a third of the same response.

Engines:  People, who remember engines that only needed compressed air to start, sometimes forget all the shenanigans weaker modern engines need to go through to avoid putting out black smoke or choking under load changes.  Without the ability to throw fuel at the power changes, this requires a lot of playing with air flow and fuel injection.  Old engines would just take a 50%-100% load step and smile, while putting out a cloud of black smoke.  Now, everything needs to be adjusted to avoid that smoke, and compressed air is used heavily in some engines’ adjustments.  It is sometimes used to alter valve timing in response to load, sometime used to boost fuel injection, sometimes used to boost combustion air, and sometimes used to adjust combustion air flow with waste and recirculation gates.  In addition, some engines use compressed air to turn off the fuel for emergency stop.  In normal, healthy conditions, loss of air to these running engines doesn’t seem to have much effect, but the emergency stop and the ability to handle major load variations has been lost.  So, is that just loss of redundancy?  Sometimes, but the engines are active pneumatic loads and a catastrophic engine fault can affect both electrical load sharing and pneumatic supply.  Sometimes, engines can also be vulnerable to pressure pulses from the common bus and shutdown.  The classic examples for this are crankcase pressure detectors and oil mist detectors, but even the engine controls can be sensitive.  So far, we have only looked at engines as vital loads that may need separated, but this is a big ticket problem that modern designers seem to forget about the most.  Common systems can be tested and appear to pass, but split or protection is preferred, as hidden or developing problems will not always be revealed by the testing.  Some engine designs are very air dependent and others are independent.  DPEs should know how their engines work, and the people designing the air system and performing the FMEAs should do the same.  One project, which I’m looking at, knew about the problem and specified engine air reservoirs, but somewhere someone forgot about this and the shipyard talked them out of the “extra” expense.  Now, they have to put them back in or prove the impossible.  Do it right during design, when it’s cheap.  None of this description has differentiated engine start or instrument air, as design varies and some engines make their own control air from the start air.

 

That’s a big tire. And well worn.

Other Loads:  While the engines are a big and increasingly forgotten area of concern, other DP vital loads can exist, such as quick closing valves (QCV), thruster seal oil tanks, thruster brakes, purifiers, dampers, control valves, selective catalytic convertors, pumps/motors, and other potentially vital automatic control loads.  All the vital control systems and equipment need dug through, so you can find out someone used pneumatic controls in the internal VSD cooling systems and that loss of air endangers all drives.  That is anomalous, but people are creative, and you need to keep an eye out for that.  Remember that compressed air is there when the lights go out, and some designers think of it like a UPS and run unexpected services off of it.  Such hidden surprises need looked for, but the usual suspects are:

  • QCV – The quick closing valve system should be air to close, but I see a lot of systems that are air to open.  Obviously, loss of pressure in such a system closes all the valves and blacks out the vessel.  It’s usually safer for the vessel to have power when performing high risk operations, so people who prioritize shutdown over safe operation have made a dangerous decision.  Air to open is only OK if there is a separate QCV system for every redundancy group and they all are fed by non-return valves, and have internal reservoirs, pressure relief valves, and pressure alarms.
  • Seal Oil – Thruster oil usually isn’t a problem, so long as each pressurized tank is fed by its own pressure regulator/check valve, and has a pressure relief valve.  Usually loss of pressure is only a long term problem and overpressure is protected against.  Some people try to be cheap and have a single regulator and relief valve.  It’s a bad idea.
  • Brake – Thruster brakes are normally air to engage and thrusters are normally running, so this isn’t usually a problem.  The opposite arrangement requires split systems – either at the source or via dedicated reservoir fed by check valve.  High integrity systems may have slow leakage but won’t stay that way over the years.  Ships are made for the long haul.
  • Purifiers – Some purifiers don’t need air and others shutdown with loss of air.  Purifier operation isn’t usually a short term critical item.
  • Dampers – Critical dampers, such as the Engine Room supply and exhaust dampers, can’t be air to open unless they aren’t gas proof fire dampers.  Loss of common air cannot be allowed to close all Engine Room dampers and kill the engines and crew by pulling a vacuum.  Gas tight fire dampers are great for fire fighting, but no common fault should close them.  Air to open requires a split air system.
  • Control Valves – Pneumatic control valves used to be common and can still be found.  It is vital that all vital valves reliably fail safe on loss of pressure.  If not, it requires a split air system.  Some systems with pneumatic valves are not DP-vital or not immediately vital (chillers).
  • SCR – Selective catalytic reduction systems that safely shutdown on loss of air pressure, do not malfunction on pressure pulses, and are protected from over pressure are OK.  It’s sometimes hard to get that information, because the SCR vendors aren’t used to thinking about malfunctions.  Some SCR system have potential pneumatic malfunctions and need a split air system.
  • Motor – DP critical pneumatic motors are vanishingly rare, but some designs have important pneumatic backup pumps.  These are normally for blackout restart.  Officially, we try not to blackout, but we know it will happen and we need a robust recovery or ride through.  That was one of the major problems with the Dali hitting the bridge earlier this year.  The DGs died on fuel starvation.  If the blackout recovery or ride through depends on pneumatic pumps, then they should be split according to the redundancy group.  Yes, it is fault recovery, but we want at least one group to come back, and if we need pneumatic pumps, then we need split air systems.
  • Other – You don’t see too many pneumatic clutches anymore.  Large crane systems or gangways may have accumulators for emergency lift off.  The old active heave compensators could sometimes cause problems and this might be true of some work loads.  Look at everything that attaches to the air system and see if it can be DP vital or produce pressure pulses.


Sorry for the previous wall of text. I ran out of time for pictures.

 

Conclusion:  If any of these concerns are true, then the air system should be split or additional protections added.  Typical additional protections include appropriately sized (30min endurance) check valve fed reservoirs or accumulators to absorb disruptive pressure pulses.  We are increasingly used to seeing common compressed air systems, but that can only be accepted for redundant DP if no single failure of the system (high pressure, low pressure, shockwave, malfunction, rupture) can cause loss of DP or more than one redundancy group.  That’s not a given and is assumed far too much.  People need to understand the equipment attached to produce appropriate redundant systems.


Silly Summary:

"Two, Three, or one," you see,

That is the question.

Whether it is nobler to split,

Or throw all upon a sea of commonality.

Per Stensønes Kjelby

Sr. Marine Advisor | IMCA DP Trials & Assurance Practioner

2w

One more item which is overlooked in FMEA’s and testing is the Jet Assist which can be found on MAN engines, and license built MAN engines out of Korea.

Paul Kerr

Engineering Management Professional | Experienced, Practical, Registered Professional Engineer | Dynamic Positioning Subject Matter Expert (DP SME)

2w

Added a silly summary at the end of the article.

Like
Reply
Damir Radan

PhD, Power System Technical Manager/Consultant - Power System

2w

Excellent article Paul. Thanks for these FMEA leasons learn findings. We can learn and prevent this. We also need to be organized in a proper way.

To view or add a comment, sign in

Insights from the community

Others also viewed

Explore topics