Chinese Engine Development

Skywatcher

Captain
Must have test fired the TBCC, wonder how long until the first test flight?

Incidentally, didn't Henri K mention that another Chinese research institute had successfully flown a TBCC last month?
 

jobjed

Captain
Here are some background on the WS-10 and monocrystal turbine blade fabrication in China, as well as WS-15 updates and predictions.


About the WS-10, sources:
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
(
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
):
  1. Its OPR is overly ambitious for a country starting off weak in material science, at around 32 whereas the AL-31's is around 24. A higher OPR means the turbine-inlet temperature and the stresses on the compressor stages are higher, putting greater demand on material quality. The OPR was that high because WS-10's core came from the CFM56, the same core as the F101 developed by GE which had a high OPR. Now GE didn't mind a higher OPR because they had the material science to back up their ambitions but the same could not be said for 606. The F110 uses monocrystal blades whereas the WS-10 uses directionally-solidified blades which are hugely inferior to moncrystals and don't last nearly as long meaning WS-10s today have a lifespan of ~1500 hours while F110s go upwards of 5000 hours. This resulted from China not having monocrystal turbine blade fabrication capability when WS-10 was under development. Additionally, 606 had to strengthen the WS-10's compressor stages to counter the stresses of a high OPR, making the WS-10 significantly heavier than the AL-31, some 250kg heavier according to gongke.
    • Another effect of the high turbine-inlet temp is the incorporation of large cooling channels in the WS-10's turbine blades. These channels bring up cool air to cushion the blades from the hot air of the combustion chamber but every litre of air used for cooling is a litre of air not optimally used to produce thrust. Therefore, most engine manufacturers try to minimise the amount of air passed through these channels to maximise efficiency. 606, on the other hand, had to enlarge these cooling channels to compensate for their directionally-solidified blades' inferior heat resistance. This means a relatively high portion of the WS-10's airflow is used for cooling and isn't optimally combusted.
  2. In the 1980s, 606 couldn't design components of an advanced engine independently so they copied the core of an American engine but adopted Russian design standards for the low-pressure compressor stage and afterburn section. This seems fine on the surface until you realise the Americans spin counterclockwise and the Russians spin clockwise. Why does this matter? In 4th-gen engines, the low-pressure compressor shaft is contained concentrically within the high-pressure compressor shaft meaning you'd want to minimise the relative velocity of the two shaft surfaces to reduce friction. In an engine where both the LP and HP shafts are spinning in one direction, the relative RPM between the two shafts would be something like 5000. For the WS-10, where the LP and HP shafts spin in opposite directions, the relative RPM is more like 20,000. For comparison, the F110's LP section has max RPM of 8500 and its HP section, ~15000, giving a relative RPM of 6500. If this was replicated on the WS-10, the relative RPM would be a ridiculous 23,500. This means the WS-10's compressors don't have as much margin to increase its RPM. As engine RPM increases, the bypass ratio decreases, see Figure 1. Unfortunately for the WS-10, it can't spin its engines above the RPM corresponding to a bypass ratio of 0.84 because its LP and HP shafts spin in opposite directions and a higher RPM would destroy the shafts, destroying the engine/aircraft. This means a minimum bypass ratio of 0.84, the highest of its contemporaries, is what the WS-10 has to work with. This isn't ideal because high-altitude cruising is most efficient with a bypass ratio as close to 0 as possible.
  3. 606 attempted to design the WS-10 to a modular architecture to facilitate rapid disassembly which is beneficial for maintenance, especially in the field and not the depot. This is also how GE designed the F101 and its derivatives. However, 606 didn't get everything right and the WS-10 ended up being only half-modular. This means while depot maintenance was simplified, field maintenance remains a hassle.
gvVBpZ7.png


Figure 1: WS-10 Bypass Ratio vs RPM, x-axis 1.00 = 100% RPM








Monocrystal turbine blade situation in China, sources:
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
.
  1. As stated above, active service WS-10 do not use monocrystals. A few examples were built using monocrystals manufactured by 170 Factory but test results were disappointing and nothing came of it. The WS-10IPE is an attempt to use better materials including new monocrystals to squeeze more performance out of the WS-10 but the improvements are not significant enough to convince the PLA to invest more into the project.
  2. Chinese turboshaft technology is the least behind of all aeroengine areas. In fact, turboshafts are most prominent users of monocrystal blades in China at the moment. The WZ-9 uses monocrystal turbine blades manufactured by the 331 Factory, or AECC South Industry Company Ltd. Their blades are pretty good quality and their yield factor is decent. However, the small physical size of their products limit their use to turboshafts.
  3. Guizhou 170 Factory, now acquired by 621 Institute, achieves a monocrystal-growing batch yield factor of around 70% and is 606th's monocrystal supplier. For comparison, Western engine manufacturers typically achieve 95% yields.
  4. The Chinese Academy of Sciences' Institute of Metal Research also manufactures monocrystal turbine blades with a yield factor of around 70%.
  5. A private company, the Hangyu Superalloy Technology Company Ltd (航宇超合金技术有限公司) achieves the highest yield factors in China (~90%), comparable to Western manufacturers. However, their catalogue range is smaller than 170 and Institute of Metal Research's so their products aren't as widely used.
  6. 170 Factory's catalogue includes the latest 3rd-generation monocrystal blades such as DD32. The WS-15 does not use the DD32 because its design predates the arrival of the 3rd-gen blades. However, it definitely uses monocrystals of some kind, probably 2nd-gens.







Now, updates about the WS-15:
  1. Half a year ago, Dr Liu Daxiang predicted the WS-15 will achieve design certification within 3 - 5 years. Gongke also said an engine can achieve design certification 3 - 5 years after first flight if testing goes well. This means the WS-15, if it wants to achieve design certification within 5 years as predicted, should be flying within 2.
  2. An experienced Chinese aeroengine enthusiast
    Please, Log in or Register to view URLs content!
    (see Figure 2) that got posted last month. The statement about Batch 3 of prototype WS-15s is the main attraction. Looking back at the WS-10 program for hints, its Batch 1 and 2 prototypes were exclusively for ground tests, beginning delivery in 1995 while Batch 3, consisting of seven prototypes, was delivered from 2000 - 2001. In June 2001, a J-11 with its starboard AL-31F replaced by a WS-10 took off, marking the first flight of the WS-10. Flight verification of the WS-10 was subsequently conducted using Batch 3 and Batch 3S engines. Beginning in March 2004, the WS-10 began its design certification process, eventually being granted design certification in October 2005. In other words, a few months after WS-10 Batch 3 was delivered, the WS-10 had its first flight and after little more than four years, it got design certification. And now we know the WS-15 Batch 3 has or will soon be completed and delivered. However, we must remember that the testing of the WS-15 is far more rigorous and comprehensive than the WS-10's so the visible progress may be slower as 606 takes their time doing things right this time.
  3. In March 2018, someone
    Please, Log in or Register to view URLs content!
    if the WS-15 was at a stage where 606 could just cram it in an airframe and make it fly for ten minutes. Gongke replied it definitely could. From this we know the WS-15 was already in flyable state almost a year ago, and probably considerably sooner than that taking into account the travel delay of news from 606 to Liyang.
The conclusion I get from this is we should expect a first flight of the WS-15 aboard the Il-76LL testbed within two years, quite possibly occurring this year.

OmP99nj.jpg


Figure 2: Billboard of achievements by young aeroengine designers
 

ZeEa5KPul

Colonel
Registered Member
A fantastic post, jobjed. Thank you for taking the time and effort to make it.
Guizhou 170 Factory, now acquired by 621 Institute, achieves a monocrystal-growing batch yield factor of around 70% and is 606th's monocrystal supplier. For comparison, Western engine manufacturers typically achieve 95% yields.
Could you explain the term "monocrystal-growing batch yield factor"? Is it how much monocrystal you get for a certain quantity of input materials? Does it just mean that Chinese manufacturers waste a lot more input material than Western manufacturers to create a product of equal quality or are Chinese monocrystals also compromised by China's manufacturing process?

Do you know of any efforts to improve the yield?
170 Factory's catalogue includes the latest 3rd-generation monocrystal blades such as DD32. The WS-15 does not use the DD32 because its design predates the arrival of the 3rd-gen blades. However, it definitely uses monocrystals of some kind, probably 2nd-gens.
Do you foresee future variants of WS-15 using DD32 (like WS-10IPE used a monocrystal)?
 

SinoSoldier

Colonel
Here are some background on the WS-10 and monocrystal turbine blade fabrication in China, as well as WS-15 updates and predictions.


About the WS-10, sources:
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
(
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
):
  1. Its OPR is overly ambitious for a country starting off weak in material science, at around 32 whereas the AL-31's is around 24. A higher OPR means the turbine-inlet temperature and the stresses on the compressor stages are higher, putting greater demand on material quality. The OPR was that high because WS-10's core came from the CFM56, the same core as the F101 developed by GE which had a high OPR. Now GE didn't mind a higher OPR because they had the material science to back up their ambitions but the same could not be said for 606. The F110 uses monocrystal blades whereas the WS-10 uses directionally-solidified blades which are hugely inferior to moncrystals and don't last nearly as long meaning WS-10s today have a lifespan of ~1500 hours while F110s go upwards of 5000 hours. This resulted from China not having monocrystal turbine blade fabrication capability when WS-10 was under development. Additionally, 606 had to strengthen the WS-10's compressor stages to counter the stresses of a high OPR, making the WS-10 significantly heavier than the AL-31, some 250kg heavier according to gongke.
    • Another effect of the high turbine-inlet temp is the incorporation of large cooling channels in the WS-10's turbine blades. These channels bring up cool air to cushion the blades from the hot air of the combustion chamber but every litre of air used for cooling is a litre of air not optimally used to produce thrust. Therefore, most engine manufacturers try to minimise the amount of air passed through these channels to maximise efficiency. 606, on the other hand, had to enlarge these cooling channels to compensate for their directionally-solidified blades' inferior heat resistance. This means a relatively high portion of the WS-10's airflow is used for cooling and isn't optimally combusted.
  2. In the 1980s, 606 couldn't design components of an advanced engine independently so they copied the core of an American engine but adopted Russian design standards for the low-pressure compressor stage and afterburn section. This seems fine on the surface until you realise the Americans spin counterclockwise and the Russians spin clockwise. Why does this matter? In 4th-gen engines, the low-pressure compressor shaft is contained concentrically within the high-pressure compressor shaft meaning you'd want to minimise the relative velocity of the two shaft surfaces to reduce friction. In an engine where both the LP and HP shafts are spinning in one direction, the relative RPM between the two shafts would be something like 5000. For the WS-10, where the LP and HP shafts spin in opposite directions, the relative RPM is more like 20,000. For comparison, the F110's LP section has max RPM of 8500 and its HP section, ~15000, giving a relative RPM of 6500. If this was replicated on the WS-10, the relative RPM would be a ridiculous 23,500. This means the WS-10's compressors don't have as much margin to increase its RPM. As engine RPM increases, the bypass ratio decreases, see Figure 1. Unfortunately for the WS-10, it can't spin its engines above the RPM corresponding to a bypass ratio of 0.84 because its LP and HP shafts spin in opposite directions and a higher RPM would destroy the shafts, destroying the engine/aircraft. This means a minimum bypass ratio of 0.84, the highest of its contemporaries, is what the WS-10 has to work with. This isn't ideal because high-altitude cruising is most efficient with a bypass ratio as close to 0 as possible.
  3. 606 attempted to design the WS-10 to a modular architecture to facilitate rapid disassembly which is beneficial for maintenance, especially in the field and not the depot. This is also how GE designed the F101 and its derivatives. However, 606 didn't get everything right and the WS-10 ended up being only half-modular. This means while depot maintenance was simplified, field maintenance remains a hassle.
gvVBpZ7.png


Figure 1: WS-10 Bypass Ratio vs RPM, x-axis 1.00 = 100% RPM








Monocrystal turbine blade situation in China, sources:
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
,
Please, Log in or Register to view URLs content!
.
  1. As stated above, active service WS-10 do not use monocrystals. A few examples were built using monocrystals manufactured by 170 Factory but test results were disappointing and nothing came of it. The WS-10IPE is an attempt to use better materials including new monocrystals to squeeze more performance out of the WS-10 but the improvements are not significant enough to convince the PLA to invest more into the project.
  2. Chinese turboshaft technology is the least behind of all aeroengine areas. In fact, turboshafts are most prominent users of monocrystal blades in China at the moment. The WZ-9 uses monocrystal turbine blades manufactured by the 331 Factory, or AECC South Industry Company Ltd. Their blades are pretty good quality and their yield factor is decent. However, the small physical size of their products limit their use to turboshafts.
  3. Guizhou 170 Factory, now acquired by 621 Institute, achieves a monocrystal-growing batch yield factor of around 70% and is 606th's monocrystal supplier. For comparison, Western engine manufacturers typically achieve 95% yields.
  4. The Chinese Academy of Sciences' Institute of Metal Research also manufactures monocrystal turbine blades with a yield factor of around 70%.
  5. A private company, the Hangyu Superalloy Technology Company Ltd (航宇超合金技术有限公司) achieves the highest yield factors in China (~90%), comparable to Western manufacturers. However, their catalogue range is smaller than 170 and Institute of Metal Research's so their products aren't as widely used.
  6. 170 Factory's catalogue includes the latest 3rd-generation monocrystal blades such as DD32. The WS-15 does not use the DD32 because its design predates the arrival of the 3rd-gen blades. However, it definitely uses monocrystals of some kind, probably 2nd-gens.







Now, updates about the WS-15:
  1. Half a year ago, Dr Liu Daxiang predicted the WS-15 will achieve design certification within 3 - 5 years. Gongke also said an engine can achieve design certification 3 - 5 years after first flight if testing goes well. This means the WS-15, if it wants to achieve design certification within 5 years as predicted, should be flying within 2.
  2. An experienced Chinese aeroengine enthusiast
    Please, Log in or Register to view URLs content!
    (see Figure 2) that got posted last month. The statement about Batch 3 of prototype WS-15s is the main attraction. Looking back at the WS-10 program for hints, its Batch 1 and 2 prototypes were exclusively for ground tests, beginning delivery in 1995 while Batch 3, consisting of seven prototypes, was delivered from 2000 - 2001. In June 2001, a J-11 with its starboard AL-31F replaced by a WS-10 took off, marking the first flight of the WS-10. Flight verification of the WS-10 was subsequently conducted using Batch 3 and Batch 3S engines. Beginning in March 2004, the WS-10 began its design certification process, eventually being granted design certification in October 2005. In other words, a few months after WS-10 Batch 3 was delivered, the WS-10 had its first flight and after little more than four years, it got design certification. And now we know the WS-15 Batch 3 has or will soon be completed and delivered. However, we must remember that the testing of the WS-15 is far more rigorous and comprehensive than the WS-10's so the visible progress may be slower as 606 takes their time doing things right this time.
  3. In March 2018, someone
    Please, Log in or Register to view URLs content!
    if the WS-15 was at a stage where 606 could just cram it in an airframe and make it fly for ten minutes. Gongke replied it definitely could. From this we know the WS-15 was already in flyable state almost a year ago, and probably considerably sooner than that taking into account the travel delay of news from 606 to Liyang.
The conclusion I get from this is we should expect a first flight of the WS-15 aboard the Il-76LL testbed within two years, quite possibly occurring this year.

OmP99nj.jpg


Figure 2: Billboard of achievements by young aeroengine designers

If this were Reddit you'd deserve a gold star. Great info.
 

gelgoog

Brigadier
Registered Member
...
Could you explain the term "monocrystal-growing batch yield factor"? Is it how much monocrystal you get for a certain quantity of input materials? Does it just mean that Chinese manufacturers waste a lot more input material than Western manufacturers to create a product of equal quality or are Chinese monocrystals also compromised by China's manufacturing process?
...

If it is anything like silicon crystal growing it should be a highly energy intensive and also possibly time consuming process.
The worst the yield factor the more you discard. Cost gets higher and production rate is lower.

It also might indicate that the crystals might have more defects on average than other producers.

I also do not understand why they won't simply redesign the WS-10.
Since they have so many aircraft using that size of engine.
Perhaps too many design resources are focused on the WS-15.
 

jobjed

Captain
A fantastic post, jobjed. Thank you for taking the time and effort to make it.

Could you explain the term "monocrystal-growing batch yield factor"? Is it how much monocrystal you get for a certain quantity of input materials? Does it just mean that Chinese manufacturers waste a lot more input material than Western manufacturers to create a product of equal quality or are Chinese monocrystals also compromised by China's manufacturing process?

Do you know of any efforts to improve the yield?

You're welcome, the term I used probably isn't industry standard. It just means what percentage of crystal-growing attempts bear fruit. For 170, three times out of ten, it's wasted effort whereas for Western manufacturers, only once out of twenty is wasted.

The yield rate can be improved if the manufacturer gathers more experience and is dedicated to rolling out improvements to their production process. The issue with 170 and the Metal Institute is that 170 and the Metal Institute's parent organisations, 621 and Chinese Academy of Sciences, are both research oriented and spend their effort discovering newer, better materials. 331 and Hangyu, on the other hand, are production focused and spend all their effort finding ways to scale up production.

This handicap will continue until AECC grows to a sufficient scale where its subsidiaries have to sacrifice neither innovation nor production. Funding for the aeroengine sector improved not too long ago and those seeds haven't yet sprouted.

Do you foresee future variants of WS-15 using DD32 (like WS-10IPE used a monocrystal)?

Hell yeah. After vanilla WS-15 is done, I'd be pissed if they didn't immediately start full-scale development of an improved variant with 3rd or even 4th-gen blades.
 

latenlazy

Brigadier
Hell yeah. After vanilla WS-15 is done, I'd be pissed if they didn't immediately start full-scale development of an improved variant with 3rd or even 4th-gen blades.
If the billboard and some other rumors are anything to go by, a revision of the WS-15 is already in the works.
 

Totoro

Major
VIP Professional
While the rest of the text sounds plausible enough, the bit about two parts of the engine designed to turn into different directions sounds extremely silly to me. No one forced the Chinese engineers to copy the US core literally. I see no reason not to be able to copy something as a mirror image, and change the spin direction, without losing efficiency.
 

Inst

Captain
The assumption is that the Chinese aren't going to rush the WS-15 out and certify it early. Given the strategic situation, such a choice might be understandable.
 

Hyperwarp

Captain

588.gif


***
As stated above, active service WS-10 do not use monocrystals. A few examples were built using monocrystals manufactured by 170 Factory but test results were disappointing and nothing came of it. The WS-10IPE is an attempt to use better materials including new monocrystals to squeeze more performance out of the WS-10 but the improvements are not significant enough to convince the PLA to invest more into the project.
***

So the PLAAF won't be adopting the WS-10IPE?
 
Top