Engine efficiency has been limited by the need to avoid potentially damaging peak pressures. But supposing you could alter those pressures on the hoof, adjusting for load while the engine was operating?
It’s the idea behind Variable Compression Ratio (VCR) systems, which aim to change the reach of the piston rod. “If you have low loads, you want to burn a smaller amount of fuel. So moving the piston’s top position upward in the cylinder raises the compression ratio and makes combustion more efficient,” explains Dominik Schneiter of WinGD. “A higher load requires a larger amount of fuel, so then you lower the piston’s top position, helping avoid thermal overload.”
Admittedly, “the idea is an old one”, says Schneiter, but there’s a renewed surge of interest, driven by the increasing focus on efficiency and pollution control. Firstly, it could yield better engine performance across a broader operating range. Marine main engines are typically run at intermediate (50% to 75%) loads while hardly ever getting above 90% of their maximum power rating.
Secondly, it would be an advantage to be able to raise or lower compression ratio (CR) to allow, for example, adaptation to humid ambient conditions which can adversely affect oxygen levels, he explains. Further, new alternative more greenhouse gas friendly fuel composition would potentially result in altered combustion characteristics causing unexpected peaks in cylinder pressure or incomplete burn, increasing both wear and emissions: here again, a VCR system could mitigate the issues.
However, dual-fuel engines of all stripes are an obvious application. As Professor Stefan Pischinger of the Institute for Combustion Engines at VKA RWTH Aachen University, explains, compared to diesel fuel combustion, gas demands far lower compression ratios. This creates “significant efficiency drawbacks” for liquid fuel operation.
Schneiter elucidates WinGD’s issue: “In gas mode our X-DF engines are on a pre mixing, lean-burn cycle; here we cannot have a high compression ratio without risking knock or combustion instability”.
Therefore, as Pischinger underlines, VCR addresses both sides by “maintaining good fuel consumption in diesel mode and also a favourable efficiency in part-load gas operation” through raised CR, while a reduced compression ratio setting allows “safe, high load gas operation”.
So, how is it achieved?
The two-stroke crosshead engine concept now being taken up by WinGD was initially developed by IHI Corporation in conjunction with IHI Power System (which recently merged with Diesel United).
It focuses on installing a pump-supplied hydraulic cylinder inside the ‘eye’ of the crosshead pin where it connects with the piston rod, pushing it further out and upward.
Research on one cylinder of an X72DF test unit at IHI Power System’s facilities have been promising: the compression ratio was increased from 12.0 to 18.0 by lifting the piston 100mm.
However, a marketable system needed something more in the way of reliability and responsiveness. So, together with IHI Power System and IHI, WinGD continued the technology’s development towards a VCR ready for implementation on regular production engines.
Therefore, the commercial design has seen the plunger-type pump swapped for a proportional valve which controls the flow of a common-rail servo oil feed to the hydraulic chamber which raises the piston. Supply doesn’t have to struggle against particularly high forces as the oil is only admitted at the bottom, lower-pressure part of the stroke; consequently, it only needs 5MPa pressure behind it: a one-way valve stops the oil being pushed out again. A separate, timed relief valve drains it away from the chamber to allow the compression ratio to fall.
Another tweak introduced a second chamber above the first as a safety feature: this acts as “a hydraulic ‘collar’ to keep everything in place”, explains Schneiter.
Further, an inherent weakness to the original arrangement lay in the wear on the chamber’s piston rod sealing ring. The only other approach was to rely on a closer fit between rod and housing. However, this involved very tight clearances, alongside keeping the sliding components in parallel even under potentially deforming pressures. A morphology was found that would deal with these competing demands, and a simulation model put the slight-but-inevitable oil leakage at 0.7mm3 per cycle, approximately 0.13% of the overall volume, well within the acceptable range.
There were also challenges in integrating standard WinGD technology: for example, the cooling oil has to drain away from the piston rod without penetrating the lower hydraulic chamber; the supply line also had to be positioned so it wasn’t warmed by the returning flow, lowering viscosity.
Most importantly, the original mechanical cam-plate trigger has been ditched in favour of a sensor-based system “that continually monitors and adjusts the piston’s position relative to the cylinderliner”, says Schneiter.
This allows for the design of control strategies that can respond to real conditions. Therefore, in addition to keeping engine operation inside the necessary safety margins, algorithms that will embrace actual load, firing pressure, gas quality and so on are under development.
But most importantly for the technology, if there’s trouble, it falls back to a fail-safe position at a lower CR; while it may not have the efficiency of the fully operating unit, it continues to be mechanically viable.
So, how much of an advantage could this technology yield? “The benefits do depend on the case,” admits Schneiter. “On a gas operation optimised for 30% to 50% part loads, the savings are only between 2% and 3%, but for diesel mode, you immediately gain 5% to 6%.”
Further, he points out that take-up could be high. “Fundamentally, we want to make this a standard component that can be retrofitted fairly simply.” Its proving ground on WinGD’s X72DF is no coincidence: this engine “now has the biggest market share amongst LNG carriers”, he underlines. Moreover, “since orders are still coming through for this engine, the VCR system could be installed from the start”.
Nevertheless, he believes that if successful, uptake won’t stop with the X72DF and predicts that it may well be rolled out across a swathe of WinGD’s portfolio, although he remarks that “you always need to see how the potential advantages work out in practice for a given engine design”.
So, when will the new VCR system be available? The technology is currently undergoing full-scale tests, focusing on verification of the basic design, functionality and performance of the industrialised VCR system. However, “it’s very close to commercialisation,” says Schneiter, so it should be entering real-world trials fairly soon: “We are already looking for test vessels, between 30,000 and 60,000 dwt, that will allow us to check running behaviour in the field.”
If this starts as expected in 2020, it will enable the first commercial systems to be released as early as 2021.
However, Professor Pischinger is interested in applying the technology to four-strokes, and his department has been working with FEV to see what could be accomplished in this, somewhat different arena.
Here, the VCR mechanism is likewise set in the conrod for a modular design with the potential for easy integration to existing and new DF-engines. But instead of a single hydraulic chamber, an eccentric bushing with four hydraulic chambers are used. All four chambers are charged with oil and discharged via an internal hydraulic circuit inside the conrod.
However, the activation is promisingly elegant: “During the development process of the two-stage VCR conrod, the main target was to create a system that can switch between two different compression ratios without the need for additional external energy for the switching process,” outlines Pischinger.
So, rather neatly, it utilises the gas and mass forces that come from the engine itself: the first from the combustion pressure, the second from the crank train’s inertia.
Between them, these create a kind of pumping action which charges or discharges opposing pairs of cells. A switching valve (on the conrod’s lower, larger eye) triggers the action which initiates the start of rotation of the eccentric bush, its offset position elevating or retracting the piston pin: valves and throttles in the oil circuit prevent it moving backwards.
Keeping it modular to allow for retrofitting was another challenge: therefore most of the components such as the switching valve, check valves, channels and even the continuous oil supply groove necessary for both eccentric bushing and piston cooling system are accommodated inside somewhat larger bearing caps.
It’s also worth noting that when applied to four-strokes, this system could also be utilised in combination with variable valve timing (VVT). Therefore, during both diesel and gas operation, the VCR may be combined with the VVT through a two-stage Miller system (which closes the intake valves before the piston reaches bottom dead centre). Under full-load diesel operation, strong Miller timing in combination with the high compression ratio will reduce cylinder temperature and NOx emissions. On the other hand, for part-loads during gas operation, strong Miller timing in combination with the VCR conrod at a high compression ratio setting assures good fuel efficiency and avoids knocking or other irregular combustion: for full-load gas operation, a low compression ratio with mild Miller timing will also avoid knocking, while maintaining high engine efficiency.
Pischinger too, believes that there could be a broad uptake for VCR technology, piggy-backing on the uptake of duel-fuel engines. After all, IMO Tier III emission limits in North American and UC Caribbean Emission Control Areas (ECAs) are about to spread. “As of January 2021, [these limits] will also apply to the Baltic and North Sea,” he points out. Therefore he says “the use of LNG is becoming increasingly interesting” as it’s combustion yields very low raw exhaust emissions of NOx, SOx and particulates, dodging the need for costly exhaust after-treatment.
By Stevie Knight