Leeham News and Analysis
There's more to real news than a news release.
Leeham News and Analysis
- Boeing 3Q deliveries up despite strike October 8, 2024
- Strike slows Boeing’s march toward improving safety culture October 7, 2024
- Aerospace supply chain still ailing; Sustainable Aviation Fuel isn’t the answer to go green October 4, 2024
- Bjorn’s Corner: New engine development. Part 27. Long developments October 4, 2024
- Longshoreman strike adds to Airbus, Boeing woes October 1, 2024
Bjorn’s Corner: New engine development. Part 25. New versus old, CFM56 vs. LEAP
By Bjorn Fehrm
September 20, 2024, ©. Leeham News: We do an article series about engine development and why it has longer timelines than airframe development. It also carries larger risks of product maturity problems when it enters service than the airframe of an airliner.
In our look at examples of recent developments with problems and these put in a historical perspective, we compare the CFM56 to the LEAP, comparing their reliability and durability.
Figure 1. The CFM56 with its mid-span shrouded titanium fan. Source: CFM.
The CFM56 compared with the LEAP
The industry and the airlines complain about the dispatch reliability and durability (time on wing) of the new LEAP engines for the Airbus A320/A321 and Boeing 737 MAX, referring to the CFM56 as the benchmark for both. Memory is indeed short. The CFM56 had more initial problems than the LEAP, including some nasty ones.
CFM56
The CFM56 was the product of CFM International, a 50-50 joint venture between GE Aviation and SNECMA (today SAFRAN engines). The engine first ran in 1974. By April 1979, the joint venture had not received any orders and was weeks away from being dissolved.
SNECMA managed to get the French Air Force to request a CFM56 re-engine of 11 KC-135 tankers, which led to the US Air Force following suit. A re-engine program for the Douglas DC-8 (the -70 variant) led to airlines flying the engine. Boeing then chose the engine for the 737 Classic series (-300, -400, -500), and the rest is history. Today, more than 34,000 CFM56s are delivered, and it is considered a bulletproof engine that stays on the wing until Life Limited Parts (LLPs) need replacing.
This was not always the case:
- The engine had a flaw in its certification requirements. The flutter characteristics of the titanium fan were proven by static ground tests. Subsequently, flutter-induced fatigue separations occurred for fan blades, causing the first fatal accident of the 737 Classic (Kegworth disaster). Subsequent flight envelope testing led to a redesign that fixed the problem.
- The engine could flame out when it passed rain or hail storms. This occurred on both engines for TACA flight 110, which was glide-flown to a safe landing at a river bank (Figure 2). The solution was changes to the fan and spinner to deflect rain and hail from the core and changed booster bleed system to better expel any remaining rain/ice to the bypass channel.
- The CFM56 also had its share of High turbine nozzle and blade problems, with cracking and deterioration reducing time on the wing. Four generations of HPT blades were needed until problems were fully solved with single crystal blades.
Figure 2. TACA flight 110 had dual CFM56 flameout. After unsuccessful restarts, the Captain glide landed the 737 on a river bank. Source: Reddit.
LEAP
The CFM LEAP, a further development of the direct drive CFM56, has had less initial reliability and durability problems than the competing Pratt & Whitney GTF. But it has not been trouble-free:
- Initially, the thermal coating on the CMC HPT shrouds wore off fast. The fix was to change the bonding material between the coating and the CMC.
- A persistent problem has been fuel nozzle coking, caused by residual heat after shutdown, which has coked the remaining fuel in the nozzles. The fix is adding a cooling fan that blows cooling air to the combustor after shut down, using the engine bleed manifold on the last compressor stages (Figure 3). By closing the bleed supply valve and opening the cooling air valve, the air backflows into the compressor and combustor, cooling the nozzles. It’s a retrofit solution.
- The HPT blades’ cooling channels have been clogged in dusty environments. New blades have been developed and are being retrofitted to engines in the field (Figure 3).
Figure 3. CFM LEAP main problems with fixes. Source: CFM.
Conclusion
The above shows how high the bar has become for new developments on the engine side. The market demands the engines to be reliable (which they have been) and durable (which they have not been) from the start. The previous generation engines were neither reliable nor durable, but this is now forgotten.
The problems of fixing the durability issues for the present engines have been amplified by the extreme ramp of engine deliveries for the A320/A321neo and 737 MAX. A problem detected in initial field use for the previous generation would mean less than 200 engines needed refits. Today, 1000 engines have been delivered and are in use after a few years.
The solution for the engine OEMs is to prolong the development part of the engine lifecycle by including a year or two of maturity testing in the program plan. Most of the problems we discussed above were detected within a year or so of in-service use.
To be fair, even with “including a year or two of maturity testing in the program plan”, OEMs and airlines will face initial reliability and durability troubles.
The simple reason is that just you can’t exhaustively reproduce real life conditions in ground testing (especially combination of adverse conditions).
Flight testing get you closer to real life but still quite far from a operational one (flight frequency, maintenance practices, environmental conditions variety…).
OEMs, airlines and the public should just to be aware of it (to the reasonable extent that safety is not compromised of course)
I agree with your analyses. Endurance tests are neccessary but cannot replace real operationnal conditions and all environmental conditions encountered worldwide
Reliability and durability also depend on the readiness of new technologies and of their integration into the propulsion system. Integration is key
old timer CFM
+1
With maturity testing, we mean operational-style flying with the correct flight hour-to-cycle ratio and 7-10 sorties per day in normal and harsh environments. The test aircraft would operate WW to fly in the different environments.
It’s not ground testing and not flying around the home base of the engine or airframe OEM.
Airbus and Rolls-Royce did it for the Trent XWB. We will discuss the result in a future Corner.
Something that was pointed out to me I knew but had not thought of in terms of actually how many hours in a year. 8000 more or less.
Flying 18 hours a day (you got to have fueling and some maint time) call it 6500 hours.
That means 3 years to get relevant time on an engine (or as I was looking at, a new compressor). A lot easier for a Wide Body as the cycles are not the same as a single aisle.
I think a help would be to get what you can hours wise, then keep your test bed flying.
But the test bed aircraft may not show up all the issues you would find in a particular region. You would need at least two, dessert aka ME and Norther Climes. Hot and humid might be a third.
What is not answered is inserting new stuff into existing engines and like the P&W, a lot worse because it was not tested. Nothing new about a seal design, ahhh well it was not a problem before.
And are airlines willing to pay for that? If its a mandate then all have to adhere, but no engine mfg can afford to go it alone.
Airlines never learn, so it is a repeat even though they should know better.
Some context is also healthy. The longest TOWs cited by the CFM are likely A319 or 737-700 installations. Neither installation is sold with meaningful volume today.
As stated, CFM had a much longer gestation period, plus it had the advantage of being a derivative of the F101/F110 core engine. The early CFM56 engine cores are piece part identical to those engines. This significantly reduced the production and program cost to GE and helped save the program. This meant the core had the benefit of testing and experience on the bomber and fighter fleets, something not available to the LEAP. Lessons learned from the common core helped improve the early CFM56s. Also, since the core was originally designed for the more challenging military missions, it is a bit over engineered for the CFM56. This results in a very robust engine when used in the civilian applications. If GE had went with a clean sheet CFM core, the program would have probably ended before 1979.
Sometimes you are good, sometimes you get lucky, the CFM story is a bit of both.
Besides having the military core it is designed to be pretty cost effective to maintain with only one HPT stage. The LPT system was new and the intershaft bearing got beaten from both rotors.
Retired GEAE:
Thank you for the background. While generally familiar with the CFM-56, I had not realized its early history. You and Bjorn present great detail. As a mechanic/technician (building mechanical systems (boiler, pumps, fire pumps, generators, Switchegear etc) those are the things that drove my life and decisions (and recommendations)
Except for safety aspects, I think this is just part of the reality.
As I recall the P&W GTF has had some extremly perilous incidents and with two engines that had the same issue.
I think the safety part was more luck than the system.
P&W not alone, the Trent 1000 had all those issues and they were using models to predict not actually. The loss of a lower hour (not by much) engine out of Rome when its partner engine had moire hours and suddenly its asked to go full power till they landed.
India mandated that you had to have at least one engine on the aircraft that had been fixed or was so new it had no issues.
I felt they were far ahead of EASA/FAA or anyone else. You should never have two questionable engines on an aircraft.
Also believing models when they missed the root cause on the Trent 1000 is grossly stupid.
Maybe mandate a severed test before you allow models to be used, ie confirm it.
@TW
If we are bracketing this as fundamental safety risks associated with design and not quality (e.g., powdered metal), there are a few items that come to mind the #3 Brg lift-off seal and an Aft Hub Seal. The Aft Hub Seal situation effectively grounded the fleet until the retrofit was in place. There have been other issues but those are two biggest from the EIS time period that comes to mind.
Most of the real safety related issues are under control; it is mostly durability improvements now.
I agree P&W has solved the issues on the GTF (or more accurately the rest of the engine). A lot of P&W choices have been made lately and I believe that is due to the problems being solved.
Getting all the engines up to snuff with all the fixes leaves an over full maint system though I gather its improved a bit there. New builds have been solid since sometime in 2023. But the backlog is still there for previous engines.
That said some of those issues were engine failures and dual failures on a twin is disastrous (or usually)
I seem to recall a couple of dual shutdowns after landing maybe.
I don’t think the AHJs really get it. India did and mandated a known good engine on jets operating in India.
You get back to the AHJ deferring to operation of aircraft and airlines to keep things moving when they should be far more cautious.
Ergo MEL. I can see not declaring an emergency but as far as I am concerned, all systems should be 100% per the factory roll out before takeoff or it should not fly.
Engines should be treated as sacrosanct. At the very least not having two engines on an aircraft that the engines have known problems with should not be allowed.
The story is about CFM single aisle engines. How did it jump to widebodies ?
https://simpleflying.com/cfm-airbus-a320neo-leap-engine-defect/
They even got a period of time to fix !
Theres even a summary of the Leaps various issues- which I bet would never cross your screen
https://www.linkedin.com/posts/othman-abdelhamid-6031a51a2_analysis-of-main-issues-in-leap-1a-engine-activity-7123210352934215680-5Nn2/
GE has extremely effective media protection.
The whiff of GE issues immediately releases a volley
of $competitor engine is unsafe,falling apart, premature wear, dangerous in general stories.
Traditionally forwarded by a range of coopted influencers ( samples on site )
Yes. I note the cosy relationship that GE Aviation has with Bloomberg- whose articles get republished by almost everyone.
Boeing gets strong coverage from the Seattle papers and all the US aviation press.
GE Aviation just has Bloomberg …doing puff pieces like this !
‘GE’s CEO Change at Its Cash Cow Is Surprising But Smart”