I could probably do an entire thread on that subject
. The CF6-80C2 became the engine of choice for the 747-400 - for good reason. It had the best fuel burn (by a small margin), the best reliability, and the best FADEC. The biggest problem the CF6 had was Ice Crystal Icing issues which caused a number of shutdowns (all re-startable). Eventually the ICI issue was pretty much solved with software, but it took 20 years to get it right. GE didn't want to do a FADEC version of the CF6-80C2 - they didn't think it was cost effective (and the CF6-80C2 PMC had the best, most sophisticated hydromechanical/supervisory electric control I ever came across). But Boeing said they were not putting throttle cables on the 747-400 - if you want to be on the -400 you need FADEC. So GE developed the FADEC version - I'm not sure if it was because they developed it after the PW4000 and were able to learn from Pratt's mistakes along with a later generation of electronics - but they ended up with a
much better FADEC control (including having autostart resident in the FADEC - unlike the PW4000 which required another box. There were a few issues with the CF6-80C2 FADEC that may trace back to FADEC being a bit of an afterthought - the most significant was their Permanent Magnet Alternator (PMA) - which is the primary electrical power source for the FADEC - was troublesome and sometimes ate itself (the gearbox setup could allow the PMA rotor to go into a whorl mode which would rub with the stator). However since the CF6 had aircraft backup power basic (PW4000 didn't) it never caused a shutdown.
During the 747-400 development, the popular wisdom was that the PW4000 - being a new engine - would be the dominate engine while the derivative CF6-80C2 would be a more of a niche engine. As a result, Boeing built the PW4000 nacelle in-house, while they purchased the CF6-80C2 nacelle from GE (Middle River), believing they wouldn't sell enough CF6-80C2 to justify the cost of developing the nacelle - bad move since the CF6 easily outsold the PW4000. Not surprisingly, being a completely new engine, the PW4000 had early problems. Pratt didn't think autostart would be popular and didn't design for it - then had to scramble to come up with the Supplemental Control Unit (SCU) to provide the autostart, engine monitoring, and backup power functions that GE had basic. The SCU was a kluge - it cost extra, weighed a lot, and didn't work very well. But what really ended up hurting the PW4000 was the compressor stability issues - particularly the takeoff surge. The thermal characteristics of the compressor caused it to lose surge margin something like 90 seconds after takeoff power set - at which time it could surge - right in the middle of the initial climb (often not recoverable)
. Pratt put together this big song and dance for the FAA, basically claiming the probability of two engines surging on the same takeoff was less than 10-9/takeoff - so of course it then happened on an Airbus twin. Fortunately one engine recovered and they were able to limp back and land. Pratt tried a number of fixes to no avail - finally they ended up redesigning the entire compressor (based on the PW4000/112" compressor) but they didn't certify the fix until 2004 or 2005 - then it took several years to retrofit the entire worldwide fleet of PW4000/94" engines with new compressors - and it cost Pratt a fortune. The resultant engine was a major improvement - and was arguably a better engine than the CF6-80C2 - but by that time it was too little too late. Between the 767 and the 747-400, the CF6-80C2 outsold the PW4000/94" by almost two to one.
The RB211-524G/H wasn't a bad engine, although it was heavier with marginally worse fuel burn than either GE or Pratt and it only the traditional Rolls customers bought it. But it was hamstrung by some typical "what were they thinking" Rolls decisions. Probably the biggest of those was the engine control - it wasn't a FADEC, it was an "FAFC" - Full Authority Fuel Control - the inlet guide vanes and bleed valves were controlled using the same archaic pneumatic system that they'd used on the RB211 on the L1011. We were having major problems with the vane controls during the flight test program (with at least one N2 overspeed) and it continued to be a headache up to the point where I retired. There were also early issues with the "Pump and Governor" unit - the engine had a mechanical N3 overspeed governor incorporated into the Main Engine Fuel Pump - and it had been designed with ridiculously tight clearances - as a result the governor would sometimes seize which would fail the fuel pump and shutdown the engine. Ironically, about 15 years later Rolls was proposing a new fuel metering unit for the Trent 800 on the 777 - and the 777 guys pulled me into a design review due to my experience with the RB211 PAG. They were proposing using the
same ridiculously tight tolerances for the new fuel metering valve that had proved so troublesome on the RB211 - and I reminded them of that. The response was along the line of 'we're smarter now, we can do this'. Sure enough, first time the did a cold fuel shock test, they new metering valve seized up and they had to open the clearances to make it work...