747-400 engine differences

[alex_767]

Hello,
As an aviation enthusiast and a 747-400 fan, I would like to know what are the differences between the GE, Pratt and the Rolls in terms of fuel burn, operating range, reliability, weight, mainenance costs... And also, as pilots, your preference, if any !
Thank you !
Alex

[MarkerInbound]

Never flew Rolls powered 74s and don’t know much of that geeky stuff but operating GE engines was easier. We could start 2 at a time versus one Pratt at a time. And GE just uses N1 and not EPR so less to look at.

[tdracer]

The GE had the better fuel burn and reliability - especially early on - and the market share reflects that GE getting the lions share (IIRC, the CF6-80C2 powered more 747-400s than Pratt and Rolls combined).
Pratt upgraded the PW4000/94" with the ring case compressor configuration (mandated by AD due to takeoff surge issues with the original segmented case compressor configuration) which matched up very favorably with the CF6-80C2 installation. But by the time the RCC came on-line in the mid-naughts, the 747-400 was past its prime and was only being sold as a freighter.
The Rolls wasn't bad - fuel burn on a par with the pre-RCC Pratt - but it had issues with reliability, especially on the engine control side. Unlike the Pratt and GE, the RB211-524G/H wasn't a true FADEC - it was a "FAFC" - Full Authority Fuel Control. The vanes and bleeds retained the fluidic control from the earlier marks of the engine - and both the FAFC and fluid vane and bleed controls proved to be very problematic.

[alex_767]

Thank you for your answers ! Very interesting and very much appreciated.

[KRviator]

The GE had the better fuel burn and reliability - especially early on - and the market share reflects that GE getting the lions share (IIRC, the CF6-80C2 powered more 747-400s than Pratt and Rolls combined).Were there any meaningful differences in the CF6 that "made it" more reliable than the Pratt or Rolls offerings? Interesting to see the note GE out-sold both PW& Rolls put together!

[tdracer]

Were there any meaningful differences in the CF6 that "made it" more reliable than the Pratt or Rolls offerings? Interesting to see the note GE out-sold both PW& Rolls put together!

GE's big advantage with the CF6 is that it was a major derivative - not an all-new engine. Normally, a clean sheet new engine gives better fuel burn - but generally at the price of reliability early on until the bugs get ironed out. However in this case, GE pretty much matched the fuel burn of the PW4000/94" and had outstanding reliability at EIS. Worse, Pratt had issues performance retention - a brand new PW4000 had marginally better fuel burn than a CF6-80C2, but the Pratt rapidly deteriorated such that after 1,000 hours the CF6 had the fuel burn advantage. Pratt finally addressed that with the Ring Case Compressor upgrade, but that didn't come for another 15 years - too little, too late.
The RB211-524G/H was a good, reliable engine (aside from the previously mentioned engine control issues) and had very good performance retention (big advantage of the 3 spool design - less bending moments that rub seals) but was heavier than the other two and was a bit worse on fuel burn.
In short, GE really nailed the design of the CF6-80C2 - even though it was a derivative it gave nothing away to the 'state of the art' PW4000 and was more reliable to boot. Ironically, Boeing thought the PW4000 would have the dominate market share and the CF6 wasn't going to be a big seller. As a result, Boeing elected not to invest in developing the engine nacelle for the CF6-80C2, instead buying the nacelle from GE (at a much higher price than an in-house design would have cost). An expensive mistake.

[megan]

td, you could write a most interesting book giving an insight into your career.

[Check Airman]

td, you could write a most interesting book giving an insight into your career.

I'd read it!

[KRviator]

GE's big advantage with the CF6 is that it was a major derivative - not an all-new engine.....An awesome & comprehensive reply, muchly appreciated, tdracer. :ok:

[tdracer]

td, you could write a most interesting book giving an insight into your career.
It would never sell - at least not enough to make it worth the effort. But thanks for the thought.

[Telekon]

Some sample numbers from the book to illustrate:

For a 390T aircraft to climb to optimum CRZ alt:
CF6-80C2B1F: 18mins & 7500kg
PW4056-3: 20mins & 8100kg
RB211-524H2: 19mins & 8100kg

Long Range Cruise at optimum alt for 390T (The PW and RR fall behind slightly further for lighter weights):
CF6-80C2B1F: 13144kg/h
PW4056-3: 13184kg/h
RB211-524H2: 13328kg/h

The GE and PW are pretty close overall in the cruise with the Rolls lagging behind by about 1.5%.
Interestingly in climb performance the CF6 really shines for some reason.

Having flown all three engine types, I have to say the CF6 is a joy to start, very quick to spool up while the PW and RR take an age to get going (God help you if its summer and you're in Nairobi or Addis).

PW very reliable though, once had an ENG OIL FILT contamination message which would have meant retarding the thrust and possible shutdown in a CF6 but PW decided having a secondary filter was worth it, no action required.

The Rolls does have one thing going for it however, it by far sounds the best of the three!

Telekon

[Downwind_Left]

The Rolls does have one thing going for it however, it by far sounds the best of the three!

Have to 100% agree with you on that one. The RB211 sounds amazing, no matter what it’s attached to.

[main_dog]

Except, very little residual thrust on the RR… needed to add quite a handful of thrust to get going on taxy if heavy, and also woe betide the pilot used to PW/CF6s who retarded the thrust levers on an RB211 too soon during the flare :}

[fire wall]

Fortunate to have flown all 3
GEnX is my pick, it just delivers.
Eng Ice Crystal Icing events a concern but suspect much the same for all power plants.
Interested to hear your thoughts on this tdtracer.

[tdracer]

Fortunate to have flown all 3
GEnX is my pick, it just delivers.
Eng Ice Crystal Icing events a concern but suspect much the same for all power plants.
Interested to hear your thoughts on this tdtracer.
I assume you mean the CF6-80C2 (unless you've also flown the 747-8)?
Ice Crystal Icing (ICI) is an interesting phenomenon. Our understanding of ICI is way better than it was 30+ years ago when it first raised it's ugly head on the CF6-80C2, but I still wouldn't say that it's well understood. For the uninitiated, 'conventional' icing involves super-cooled liquid water droplets - liquid water that is well below freezing so it's just looking for something to freeze to like a nice cold wing or inlet leading edge. It's readily addressed by adding heat to the susceptible surfaces. ICI is fundamentally different - ICI is literally very small ice crystals - and they simply bounce off cold surfaces. BUT, warm surfaces can present an issue - the IC hit the warm surface and melt. If the ice crystals are present in sufficient quantities, additional crystals hit the liquid water and cool it sufficiently that it re-freezes - resulting in ice accumulation on surfaces that would normally be considered way too warm to ice. In testing, we saw ice accumulation in areas of the compressor where the air temp was as high as 85 deg F/30 deg C!:eek:
There seems to be something about the GE compressor design practices that make it more prone to ICI than Pratt or RR. For a long time, the CF6-80C2 was the only engine that had issues with ICI, although eventually the PW2000 suffered a few ICI events (but nowhere near as bad as the CF6. The GE90 had minor issues with ICI (no shutdowns, but some minor compressor damage events) but ICI hit the GEnx hard. We had a 747-8/GEnx-2B event fairly early on that - although the engines re-lit and ran fine for the balance of the flight, there was severe compressor damage on 3 engines. GE came up with a fairly ingenious method of detecting the presence of ICI and opening the compressor bleeds to send most of the ice overboard and that effectively solved the ICI issue with the GEnx.

[Le Flaneur]

NASA and the FAA did ice crystal icing research with a Falcon 20 and DC-8.

https://www.nasa.gov/aero/nasa-completes-high-ice-water-content-radar-flight-campaign-in-florida

[RB Thruster]

The RB211-524H2 was only ever meant to be a stepping stone to the H3 engine. The H3 was to use the latest technology FADEC from the then new Trent series just coming online. However the H3 was a bridge too far in terms of meeting performance targets, and it ran too hot and fast. The final straw was an almighty bang on a H3 dev engine one nightshift in the Derby test beds, debris all over the test bed floor, and not long afterwards the H3 was canned, (along with similar developments of the 535 as well the E5/F5) So the 747-400 was stuck with the slightly older technology -524H2.
Very interesting times at RR, with the birth of the trents, V2500 and RB211 developments. There was a final development, with fitting the Trent 700 HP module, which actually gave a marked improvement of fuel burn and TGT.

[alex_767]

Thanks a lot guys, very interesting to read! And special thanks to you tdracer for your very informative posts! :)

One question just popped. At the very beginning Cargolux had some 400s with GE and later switched to RR. Besides financial reasons what were the other considerations, if any, that could have justified this choice ?

[HOVIS]

Just my tuppence worth.
From a line maintenance point of view the RR has always been easier to service. The gearbox is on the fan case with all the accessories. Starter motor, IDG, Fuel and oil pump packs etc. Access is simple, just open the fan cowl and away you go. The GE and PW philosophy was to cram as much as possible around the hot core section which means you need a pump to open the C Ducts/ Thrust Reverser halves. More time consuming and in my opinion more difficult access.
The GE CF6 is a brilliant bit of kit though. Incredibly reliable.

[alex_767]

Just my tuppence worth.
From a line maintenance point of view the RR has always been easier to service. The gearbox is on the fan case with all the accessories. Starter motor, IDG, Fuel and oil pump packs etc. Access is simple, just open the fan cowl and away you go. The GE and PW philosophy was to cram as much as possible around the hot core section which means you need a pump to open the C Ducts/ Thrust Reverser halves. More time consuming and in my opinion more difficult access.
The GE CF6 is a brilliant bit of kit though. Incredibly reliable.
Thanks HOVIS for your response. Very informative.

[punkalouver]

It would never sell - at least not enough to make it worth the effort. But thanks for the thought.
It is not the money that matters. It is the fact that you were able to get information out that will be forever gone. it is like the test pilot books I am reading. Fantastic information about companies and aircraft that we would otherwise never get to know.

[punkalouver]

Fortunate to have flown all 3
Eng Ice Crystal Icing events a concern but suspect much the same for all power plants.

GE90-115. It is nice to not have to worry about such minor things as an all engine flameout.

[punkalouver]

Except, very little residual thrust on the RR… needed to add quite a handful of thrust to get going on taxy if heavy,

Beats riding the brakes on a light aircraft.

[punkalouver]

I would be interested to know why two PW4000's cannot be started at the same time while CF-6 has dual starts.

[jettison valve]

Just my tuppence worth.
From a line maintenance point of view the RR has always been easier to service. The gearbox is on the fan case with all the accessories. Starter motor, IDG, Fuel and oil pump packs etc. Access is simple, just open the fan cowl and away you go. The GE and PW philosophy was to cram as much as possible around the hot core section which means you need a pump to open the C Ducts/ Thrust Reverser halves. More time consuming and in my opinion more difficult access.
The GE CF6 is a brilliant bit of kit though. Incredibly reliable.

I agree, and would like to add another view:
Although being rather a desk driver and havn´t touched hardware for a long (too long?) time, I always found the wiring on the T700 (and other RR engines) superior to CF6 / PW for my part of the job - proper routings, markings, colour codes. It always seemed to me as if in GE/PW, the wiring guy came last in the design phase and had to find his ways along all the obstacles along his routes. And I somehow got along with the RR manuals much better.

[Zar_1]

Never flew Rolls powered 74s and don’t know much of that geeky stuff but operating GE engines was easier. We could start 2 at a time versus one Pratt at a time. And GE just uses N1 and not EPR so less to look at.

Ahh the annoyance towards EPR spans across manufacturers... it seems :)

In the IAE A320 or the 340-600, I find the EPR to be rather unintuitive to use... especially since I started studying the 737NG first.

[EEngr]

It would never sell

It would if you could get it translated into Chinese.:E

[Check Airman]

Ahh the annoyance towards EPR spans across manufacturers... it seems :)

In the IAE A320 or the 340-600, I find the EPR to be rather unintuitive to use... especially since I started studying the 737NG first.

In the IAE 320, I’m only vaguely aware that it’s there. I use N1 for just about everything.

[tdracer]

Wow, an old thread comes back to life :ok:
OK, starting (must have missed this first time around). In general, the PW4000 was harder to start than the CF6-80C2. It was standard practice to spin the PW4000 up to close to max motoring before putting the fuel on - the CF6 you could turn on fuel at 15% N2 and get a reliably good start. So the Pratt took more air to start. When Boeing built the LCF (aka Dreamlifter) for hauling around 787 parts, they removed the APU (putting a fuel line through where they spit the fuselage to load/unload was too big a challenge), so ground carts were needed for starting - turns out it needed to be pretty good ground cart to get enough air to reliably start.
EPR vs. N1 is a long standing controversy. N1 has a huge advantage of being an easy measurement relative to EPR which requires reliable measurement of two pressures - one in an area that can be prone to icing (inlet EPR probes require ~ 500 watts of power to heat - which also makes reliable temperature measurements tricky). OTOH, EPR is better related to actual thrust - and the thrust/EPR relationship is pretty much constant (TO EPR is pretty much constant below cornerpoint temperature). N1 has that whole messy square root of theta temperature meaning thrust set N1 changes with temperature.
Wiring (and to a lesser extent plumbing) - Rolls certainly looked like it was designed. Pratt and Rolls both just sort of added it when everything else was done - it looks like a mess.

[Zar_1]

Yeah N1 is definitely easier to calculate, since it's kinda a Tachometer by design... and yes, EPR is the more accurate of the 2.

An innovation that I've seen (ig in the A350) is the presence of 'calculated thrust' gauges which are meant to be more intuitive to use.

So, the PW4000, if given inadequate bleed... would give you a nasty Hot Start or the FADECs would auto-cut the fuel and turn the start into a Dry Crank Motoring phase?

[condor17]

TD , a belated thanks to you [ and all engineers ] for safely designing , making , servicing all our flying machines for the last 52 yrs .
I'd buy your book ..
It's a delight to read your explantions and stories . You have the knack of making them understandable and enjoyable to an old tractor driver with wings .
Seem to remember we could auto start 2 Rolls on 744s until you got Hot 'n High .. Then 1 . Then Manual start .
Agree on the Rolls sound , whether inside or out the -22bs on our by then old Tristars forever embedded . More like a big diesel winding up before fuel goes in . And 'wot a privilege to have handled 4 Rb 211-524Hs on t/o .

rgds condor .

[Zar_1]

Is the manual start done so that the N1 is higher than normal, when fuel is injected?

[IBMJunkman]

https://blog.kotobee.com/self-publishing-on-amazon-pros-and-cons/

It would never sell - at least not enough to make it worth the effort. But thanks for the thought.

[tdracer]

Yeah N1 is definitely easier to calculate, since it's kinda a Tachometer by design... and yes, EPR is the more accurate of the 2.

An innovation that I've seen (ig in the A350) is the presence of 'calculated thrust' gauges which are meant to be more intuitive to use.

So, the PW4000, if given inadequate bleed... would give you a nasty Hot Start or the FADECs would auto-cut the fuel and turn the start into a Dry Crank Motoring phase?

"Calculated Thrust" is simply measured EPR compared to the max rated EPR for those conditions. Perhaps slightly more intuitive, but still has all the potential issues as EPR. I had some friends that spent a lot of time working on a "0-100" thrust measurement. They finally decided that - while there were various ways to make it dimensionless, in the end it really wasn't any better than simply using N1 and/or EPR.

Autostart wasn't basic on the PW4000/747-400 - autostart was an add-on box called an SCU (Supplemental Control Unit). It didn't work very well though...

Junkman - I'd still have to make all the effort to write the darned thing...