Essay: Inside the Design of China’s Yuan-class Submarine

August 31, 2015 2:13 PM - Updated: August 31, 2015 5:39 PM
People's Liberation Army's Navy (PLAN) Yuan-class submarine.
People’s Liberation Army’s Navy (PLAN) Yuan-class submarine.

China’s evolving submarine force is a topic worthy of sober examination and debate. However, for the discussion to be useful in informing both national policy-makers and citizenry alike, the content must be based on accurate data and sound analysis. Unfortunately, both are often found wanting when it comes to English-language reporting on the People’s Liberation Army Navy (PLAN). A recent USNI News essay, continues this trend. In the essay, Henry Holst, argues that the Type 039A/B Yuan-class submarine was “designed primarily as an anti-ship cruise missile (ASCM) platform capable of hiding submerged for long periods of time in difficult to access shallow littorals.”

He bases his conclusions on the Yuan’s “small” size, air-independent power (AIP) system, and long-range ASCM capability, which make it ideal to operate in shallow, coastal waters. While I wholeheartedly agree that the Type 039A/B is a fine anti-surface platform, the main points of the essay’s argument is based on a misunderstanding of largely inaccurate data. This rebuttal will examine the key points of the essay’s argument and show that the Type 039 A/B Yuan-class is, in fact, an open-ocean submarine designed to meet the needs of the PLAN’s near-seas active defense aspect of their maritime strategy, and not primarily a boat to operate in Taiwan’s coastal waters.

Yuan-Class Submarine Size

The assessment the Type 039A/B Yuan-class is a small submarine, designed to operate in the shallow littorals, is arguably the linchpin of the essay’s conclusion. A key point used in support of this position is a comparison of the Yuan-class with Japan’s Soryu-class, another AIP equipped submarine, where it is argued the Yuan is comparatively “far smaller.”

In particular, Holst focuses in on the Soryu’s “draft” of 10.3 meters. This figure is highly suspect. When a submarine’s draft exceeds its beam, you might want to check your data. The 10.3-meter figure is not the draft of the Soryu class. Rather, it is the submarine’s “depth”— the vertical distance from its keel, to the top of the freeboard deck, measured at mid-length of the vessel. A submarine’s draft is included in its depth. A number of open sources list the Soryu’s draft as 8.5 meters, and a review of online photos of the fore and aft draft markings shows the draft as 8.3 meters, very close to the published value.

The description of the Type 039A/B Yuan-class submarine’s dimensions is also inaccurate. However, that reflects more the inadequacies of traditional Western sources that do not address PLAN platform and systems very well. In his essay, Holst specifically states, “PLAN naval architects deliberately maintained the Song-class’s size even with the installation of an AIP system.” While the author recognizes that trade-offs in the Type 039A/B Yuan design would have to be made, there is no attempt to suggest what they might have been. The author merely expresses his faith that the PLAN naval architects would have found a way to carve out the necessary space. This unqualified assertion is naïve at best.

Submarines are, by design, compact and severely limited in internal volume. One cannot just wave a magic wand and rearrange large propulsion plant components and squeeze out the considerable volume an AIP system requires. In particular, the cryogenic oxygen tank takes up a great deal of space—and the longer the submerged endurance, the bigger the tank. If the Song-class had that much spare volume to begin with, it would have been built smaller. There is no evidence the engineering spaces on a Song-class submarine have any extra unused volume. Video excerpts of the Type 039G show these submarines are just as cramped and constrained as one would expect. And since both the Song and Yuan use double-hull construction, a designer can’t even go after the fuel load, as the vast majority of the fuel is carried in tanks external to the pressure hull. Since the Yuan has an AIP system, it has to be bigger.

A review of Google Earth and hand-held photography indicates this is indeed the case. Google Earth imagery of both submarine classes berthed near each other shows the Yuan has a larger beam than the Song-class. This strongly argues that the sources that hold to the narrower beam of 7.5 meters for the Type 039G Song are probably closer to the truth. Furthermore, analysis of hand-held imagery indicates that the Yuan is not only longer, but also has a deeper draft than the Song-class.

Soryu-class submarine, Hakuryu during a visit to Guam in 2013. Note the bow draft markings show the submarine’s draft is about 8.3 meters. US Navy Photo
Soryu-class submarine, Hakuryu during a visit to Guam in 2013. Note the bow draft markings show the submarine’s draft is about 8.3 meters. US Navy Photo

Chinese submarines use the Russian system for draft markings. This means the markings do not show the draft by direct measurement, but rather it shows the deviation from an established draft. On Chinese submarines, the longer white line in the draft markings shows the submarine’s normal surface waterline, with deviations in the draft noted in 0.2-meter increments. From hand-held, broad aspect photographs of the Type 039A/B submarine, both in the water and out, one can accurately measure the waterline length and the length overall. The often-quoted length of 72 meters for the Yuan class is actually its waterline length. This value is also consistent with Google Earth measurements. The submarine’s overall length from the analysis is just over 77.2 meters, which is consistent with numerous Chinese-language websites that list the length as 77.6 meters. The same can be said of the normal surface draft that comes in at about 6.7 meters, greater than the 5.5-meter value held by many open Western source references.

Putting all of this together shows the Yuan is a large conventional submarine, only marginally smaller than a Soryu-class boat—on the order of 15 percent smaller. Perhaps a better comparison would be with the PLAN’s other large conventional submarine, the Russian-built Project 636 Kilo. That comparison shows the Yuan comes out as being slightly bigger than a late model Kilo. The table below lists the basic physical characteristics of the four submarines discussed in this article. Project 636 Kilo and Soryu-class data come from official sources, while Yuan and Song data are largely derived from the analysis mentioned above.

Thus, in stark contrast to the essay’s conclusion, the Type 039 A/B Yuan is not a small submarine at all. It is one of the largest conventional combat submarines in the PLAN inventory, and is no more maneuverable in shallow water than other large conventional submarine designs, such as the Kilo or Soryu-classes. If a navy truly wishes to invest in a “coastal submarine,” or SSC, then it would look at submarines like the German Type 205 and 206, and the North Korean Sango, all of which come in at less than 500 tons submerged displacement.

Type 039B Yuan-class submarine during rollout at the Jiangnan Shipyard on Changxing Island. Note the long white line in the draft markings, which designates the submarine’s normal surface waterline. Also note the low-frequency passive flank array just above the keel blocks.
Type 039B Yuan-class submarine during rollout at the Jiangnan Shipyard on Changxing Island. Note the long white line in the draft markings, which designates the submarine’s normal surface waterline. Also note the low-frequency passive flank array just above the keel blocks.
Project 636 Kilo Type 039A/B Yuan Type 039GSong Soryu
Length Overall 73.8 m 77.6 m1 74.9 m 84.0 m
Beam 9.9 m 8.4 m 7.5 m 9.1 m
Draft 6.6 m 6.7 m2 5.7 m4 8.3 m5
Surface Displacement 2,350 tons 2,725 tons3 1,727 tons 2,947 tons
Submerged Displacement 3,125 tons 3,600 tons 2,286 tons 4,100 tons
1) The often-quoted Type 039A/B length of 72 meters is waterline length, not length overall.
2) Type 039A/B draft is larger than the reported 5.5 meters that is nearly identical to the smaller Type 039G Song-class.
3) Yuan-class surfaced and submerged displacements come from Chinese language websites, of which there is some confusion on surface displacement. The value given on most Chinese websites (2,300 tons) would result in a reserve buoyancy of 50+ percent, which is not realistic. The estimated surface displacement in the table reflects a reserve buoyancy of 32 percent that is consistent with earlier Type 035 and 039G designs.
4) The reported 5.3-meter value for the Type 039G’s draft is suspect. Hand-held photos of this submarine in drydock suggest the draft is probably closer to 5.7 meters.
5) The draft of the Soryu-class is often listed as 8.5 meters, however, numerous photos of fore and aft draft markings show it is closer to 8.3 meters.

Shallow Water Environment

Designing a submarine to operate in very shallow water has other problems beyond just maneuvering. Holst correctly points out that the acoustic environment in coastal areas is chaotic and difficult, making it challenging for an antisubmarine platform to find a Yuan ensconced in such waters. But what is good for the gander is also good for the goose.

Radiated noise from shipping is far louder, and even with multiple bounces off the bottom and sea surface, a lot of the acoustic energy will still reach a submerged submarine’s sonar. With shipping, biologic, and wave noise coming in from both near and far, a Yuan would be hard pressed to detect, track, and identify a target of interest; particularly as surface combatants tend to be quieter than civilian merchants. In other words, it will be very difficult to find and obtain an accurate fire control solution on a desired target without using a periscope to sort out the tactical picture. Doing so, however, would increase the submarine’s chances of being detected by radar or electro-optical sensors. Therefore, a submarine hugging the bottom in shallow coastal waters will be vexed by the same problem that an ASW ship has to deal with in looking for the submarine. But what is even more curious—if Chinese designers had intended from the very beginning for the Yuan-class to be a shallow water boat—why was a passive low frequency flank array put on these submarines?

The H/SQG-207 is a line of individual hydrophones mounted to the hull, and is designed to provide long-range detection against noisy ships—low frequency noise suffers lower absorption losses and travels further in water. The problem is this kind of array is most effective in deeper water where interference with the bottom is limited. Such an array would be severely degraded in very shallow water, offering little, if any benefit, beyond the capabilities of the medium frequency bow array. The fact that the H/SQG-207 array is on the Yuan-class argues strongly that its design operating areas are in deeper waters where this passive sonar can serve as the primary sensor.

PLAN Submarine Weaponry

Holst’s treatment of PLAN submarine weaponry shows it perpetuates a longstanding misunderstanding when it asserts that the Type 039A/B Yuan carries the C-802 ASCM. The fundamental problem with this is the C-802 is not a submarine-launched missile. In fact, the “C-802” designation is for an export surface, air, and land-based ASCM with a range of 120 kilometers, rather than the 180 kilometers stated in the essay. Of interest, there is no evidence the C802 was ever accepted by the PLAN. Lastly, it isn’t the same missile that the PLAN has fielded on the Type 039G, 039A/B, 091, and 093 submarines—the YJ-82. The YJ-82 is a solid-rocket propelled missile based on the YJ-8/8A ship-launched ASCM, but without the booster. The YJ-82 is launched in a buoyant capsule that is virtually identical to the U.S. submarine-launched Harpoon. Normally, the range of the YJ-8/8A is only 42 kilometers. But without the booster, the YJ-82’s range will be even less, possibly as short as 30 to 34 kilometers (16–18 nautical miles). This very short range means the launch will almost certainly be seen by its target, or an escort, as the missile will be within the radar horizon of most warships by the time it reaches ten meters in altitude. With such a short engagement range, the firing submarine’s location will be fixed quite quickly, with a counter-attack likely following shortly thereafter if a warship is nearby—flaming datums have a bad habit of attracting that sort of response. That is one of the key reasons why the PLAN submarine force is eagerly awaiting the fielding of the YJ-18, which reportedly has a maximum range of up to 220 kilometers.

With respect to the authors the Department of Defense’s, 2015 Annual Report to Congress, the cited range of 290nm (550km) for the YJ-18 in (p. 10) is undoubtedly a typographic error. It is virtually impossible for a missile that is very likely smaller than the SS-N-27B Sizzler to have a range that is almost two and a half times greater. The ranges given in the 2015 report for the YJ-8A and YJ-62 are also incorrect and reflect a reliance on inaccurate open source information.

The YJ-82 is the submarine launched version of the solid-rocket propelled YJ-8/8A. CPMIEC
The YJ-82 is the submarine launched version of the solid-rocket propelled YJ-8/8A. China Precision Machinery Import-Export Corporation Photo

Then there is essay’s explicit claim that the Yuan was designed “primarily as an anti-ship cruise missile (ASCM) platform.” While IHS Jane’s applies the “G” to the designation of any vessel that can physically fire an ASCM, this doesn’t make the submarine “primarily” a cruise missile platform. The U.S. intelligence community, NATO, and Russian designation systems require a submarine to be fitted with dedicated launchers, not just the torpedo tubes, to be called a guided-missile submarine. That is why there is such a furious debate as to whether or not the new Type 093B or 095 will have a vertical launch system—perhaps as many as 16 tubes. This will clearly indicate when the cruise missile (both anti-ship and land attack) has eclipsed the torpedo as the submarine’s main weapon.

Another point of interest is that the weapons carried by the Type 039A/B Yuan are identical to those on the Type 039G Song, to include both the current YJ-82 and the future YJ-18 ASCMs. And yet, Holst makes no mention of the Song-class having the exact same weapons capability, both in terms of the number of torpedo tubes and weapons carried. The Type 039A/B Yuan-class will undoubtedly be the more effective ASCM platform; due to its enhanced sonar suite and the tactical flexibility provided by the AIP system, but in both instances the ASCM is a secondary weapon because of the small load out of missiles in the torpedo room, and the low salvo-size driven by torpedo tube limitations. If the Chinese continue to follow Russian tactical concepts, two of the six torpedo tubes will be loaded with YU-6 torpedoes for self-defense against an unexpected appearance by a submarine or surface ship. And while a salvo of four YJ-18 ASCMs is nothing to sneeze at, it is probably insufficient to overwhelm a modern warship’s hard and soft kill air defenses. The Mach 3 speed of the YJ-18’s sprint vehicle is impressive, and will seriously reduce a ship’s reaction time, but numbers are still needed to saturate todays modern air defense systems.


In sum, I believe Holst has drawn incorrect conclusions on the Type 039A/B Yuan submarine design basis because of inaccurate technical data and inadequate analysis. The Type 039A/B Yuan is a large submarine, particularly for a non-nuclear boat, and is comparable in size to Russia’s Kilo and Japan’s Soryu-classes. The sonar suite of the Yuan is tailored more for deep water where it can use the low frequency flank array to make long-range detections against noisy ships. The lack of a vertical launch system means the Yuan, and Song-class, are limited by the number of torpedo tubes that can be allocated to ASCMs; making it very hard to saturate a ship’s air defenses with only four, or at most five, missile salvos. And given the current short-ranged YJ-82 ASCM, a Type 039A/B Yuan-class submarine is better off attempting to close inside 15 kilometers and engaging the target with YU-6 torpedoes. But even after the introduction of the YJ-18, the restrictive factors of the torpedo room’s capacity and the small number of torpedo tubes remains. The PLAN appears to appreciate this constraint, as the discussion of future nuclear submarine designs having as many as 16 vertical launch tubes suggests.

The design aspects of the Type 039A/B Yuan-class submarine point toward deep-water operating areas in the near seas, to include the approaches to Taiwan, where their improved sonar and AIP capabilities will aid the submarine in detecting, tracking, and engaging targets of interest. And while a Type 039A/B Yuan could soon be loaded with a more effective, long-range ASCM, the submarine’s design limitations will continue to rely heavily on the torpedo.

Capt. Christopher P. Carlson, USN (Retired)

Capt. Christopher P. Carlson, USN (Retired)

Christopher P. Carlson is a co-designer of the Harpoon tactical naval wargame and a bestselling author.

Get USNI News updates delivered to your inbox