GE engines not displaying EPR
Thread Starter
GE engines not displaying EPR
I’ve operated P&W, RR and GE power plants, all except GE display EPR and it is the primary thrust reference
Always wondered why GE chooses to not display this parameter, using N1 as primary instead ?
Always wondered why GE chooses to not display this parameter, using N1 as primary instead ?
If all the fan blades are there and undamaged, N1 will give a good indication of thrust. The early A300s had EPR gauges but they were removed at some point. They were on the FE panel so were not used for setting thrust. I guess having EPR was useful for cross checking other engine parameters. They did away with engine vibration gauges as well.
In the days of pure jet and low bypass engines you needed EPR of P7 or similar.
As to why GE went down this route I haven't a clue.
In the days of pure jet and low bypass engines you needed EPR of P7 or similar.
As to why GE went down this route I haven't a clue.
N1 is a simple parameter that's far less prone to measurement errors than EPR. For EPR, you need highly reliable inlet and exit pressure measurement - including probe heat for the inlet probe (and the electrical power required for probe heat is significant - around 500-600 watts per engine). Further, heated pressure probes have still been known to ice up and give erroneous indications. With N1, you need to measure the rotor speed anyway so no additional instrumentation is required, and it's a simple, highly reliable measurement.
Newer EPR engines not only cross compare inlet pressure with aircraft total pressure, they calculate a 'synthetic' exit pressure based on other engine parameters, and if anything is out of a fairly narrow tolerance, the FADEC invalidates EPR and you set thrust based on - that's right - N1.
Now, there are some advantages to EPR - assuming it's accurately measured it correlates a little better with thrust then N1 does, and it's not as affected by high humidity as N1 (N1 power setting charts basically have to assume the humidity is very high - which can make a meaningful difference in thrust at high ambient temperatures and humidity. As a result, N1 rated engines need more margin and tend to 'give away' a little thrust relative to EPR engines (i.e. an average N1 will give you a little more thrust than an EPR engine at a specific thrust set). EPR for a specific thrust setting is usually independent of temperature - so for example takeoff EPR is a constant below the corner point temperature - where as N1 varies with temperature due to that messy "root Theta" term to convert Corrected N1 to Physical N1. With fan damage (e.g. birdstrike), N1 goes up while thrust goes down - EPR engines are much less affected by that issue.
It's worth noting, aircraft have crashed due to EPR measurement errors. I can't think of any that have crashed due to N1 measurement errors.
Newer EPR engines not only cross compare inlet pressure with aircraft total pressure, they calculate a 'synthetic' exit pressure based on other engine parameters, and if anything is out of a fairly narrow tolerance, the FADEC invalidates EPR and you set thrust based on - that's right - N1.
Now, there are some advantages to EPR - assuming it's accurately measured it correlates a little better with thrust then N1 does, and it's not as affected by high humidity as N1 (N1 power setting charts basically have to assume the humidity is very high - which can make a meaningful difference in thrust at high ambient temperatures and humidity. As a result, N1 rated engines need more margin and tend to 'give away' a little thrust relative to EPR engines (i.e. an average N1 will give you a little more thrust than an EPR engine at a specific thrust set). EPR for a specific thrust setting is usually independent of temperature - so for example takeoff EPR is a constant below the corner point temperature - where as N1 varies with temperature due to that messy "root Theta" term to convert Corrected N1 to Physical N1. With fan damage (e.g. birdstrike), N1 goes up while thrust goes down - EPR engines are much less affected by that issue.
It's worth noting, aircraft have crashed due to EPR measurement errors. I can't think of any that have crashed due to N1 measurement errors.
It's worth noting, aircraft have crashed due to EPR measurement errors. I can't think of any that have crashed due to N1 measurement errors
The Air Florida Boeing 737-200 takeoff crash into the Potomac River at Washington was a tragic example of EPR errors that were a contributory cause. See: https://en.wikipedia.org/wiki/Air_Florida_Flight_90
Plus, this writer had personal experience as jump seat observer during a night take off from a Pacific island in a 737-200. Both Pt 2 sensors were blocked by foreign objects ( insects and coral dust). Indicated EPR on take-off was 2.18 (JT8D-17 P&W engines). Actual EPR was probably around 2.10.
The difference in N1 between 100% (planned) and actual N1 95%, is so small on the gauge - around 5mm - that the crew never picked it with low instrument lighting on the dark night. Both engines EPR displayed 2.18 EPR, in other words significant over-reading from the true EPR. At the last few seconds of the takeoff run on the short 5600 ft runway the captain realised the aircraft would not get airborne by the end of the runway. An abort would have been fatal into the rocks and sea at the end of the runway.
The captain simultaneously firewalled the thrust levers and pulled back on the control column and was able to get airborne just below V1. Confirming - just below V1. The increase in acceleration at the moment of firewalling the thrust levers was very noticeable. A classic example of a correct instant decision to GO and firewall the thrust levers that resulted in the aircraft just scraping airborne. This incident showed the vital importance of cross-checking EPR readings with planned N1 during takeoff. If any doubt use N1 as the final arbiter. .
Last edited by Centaurus; 20th Jul 2020 at 02:47.
EPR was invented by and is proprietary to RR. When PW made a licence agreement with RR to produce engines in USA, the EPR technology was included.
hence why GE does not use EPR.
When the FADEC on a RR decides there is a problem, N1 becomes the default control instrument.
The airplane drivers do not need EPR to get airborne, but you do need N1. See Centaurus's illuminating post. After Palm 90, at least one operator decreed that N1would be calculated and bugged on one engine, The other was left at EPR.
hence why GE does not use EPR.
When the FADEC on a RR decides there is a problem, N1 becomes the default control instrument.
The airplane drivers do not need EPR to get airborne, but you do need N1. See Centaurus's illuminating post. After Palm 90, at least one operator decreed that N1would be calculated and bugged on one engine, The other was left at EPR.
Last edited by Three Wire; 20th Jul 2020 at 05:01. Reason: Additional info
Join Date: Mar 2006
Location: USA
Posts: 2,527
Likes: 0
Received 0 Likes
on
0 Posts
N1 is a simple parameter that's far less prone to measurement errors than EPR. For EPR, you need highly reliable inlet and exit pressure measurement - including probe heat for the inlet probe (and the electrical power required for probe heat is significant - around 500-600 watts per engine). Further, heated pressure probes have still been known to ice up and give erroneous indications. With N1, you need to measure the rotor speed anyway so no additional instrumentation is required, and it's a simple, highly reliable measurement.
Newer EPR engines not only cross compare inlet pressure with aircraft total pressure, they calculate a 'synthetic' exit pressure based on other engine parameters, and if anything is out of a fairly narrow tolerance, the FADEC invalidates EPR and you set thrust based on - that's right - N1.
Now, there are some advantages to EPR - assuming it's accurately measured it correlates a little better with thrust then N1 does, and it's not as affected by high humidity as N1 (N1 power setting charts basically have to assume the humidity is very high - which can make a meaningful difference in thrust at high ambient temperatures and humidity. As a result, N1 rated engines need more margin and tend to 'give away' a little thrust relative to EPR engines (i.e. an average N1 will give you a little more thrust than an EPR engine at a specific thrust set). EPR for a specific thrust setting is usually independent of temperature - so for example takeoff EPR is a constant below the corner point temperature - where as N1 varies with temperature due to that messy "root Theta" term to convert Corrected N1 to Physical N1. With fan damage (e.g. birdstrike), N1 goes up while thrust goes down - EPR engines are much less affected by that issue.
It's worth noting, aircraft have crashed due to EPR measurement errors. I can't think of any that have crashed due to N1 measurement errors.
Newer EPR engines not only cross compare inlet pressure with aircraft total pressure, they calculate a 'synthetic' exit pressure based on other engine parameters, and if anything is out of a fairly narrow tolerance, the FADEC invalidates EPR and you set thrust based on - that's right - N1.
Now, there are some advantages to EPR - assuming it's accurately measured it correlates a little better with thrust then N1 does, and it's not as affected by high humidity as N1 (N1 power setting charts basically have to assume the humidity is very high - which can make a meaningful difference in thrust at high ambient temperatures and humidity. As a result, N1 rated engines need more margin and tend to 'give away' a little thrust relative to EPR engines (i.e. an average N1 will give you a little more thrust than an EPR engine at a specific thrust set). EPR for a specific thrust setting is usually independent of temperature - so for example takeoff EPR is a constant below the corner point temperature - where as N1 varies with temperature due to that messy "root Theta" term to convert Corrected N1 to Physical N1. With fan damage (e.g. birdstrike), N1 goes up while thrust goes down - EPR engines are much less affected by that issue.
It's worth noting, aircraft have crashed due to EPR measurement errors. I can't think of any that have crashed due to N1 measurement errors.
Join Date: Apr 2019
Location: EDSP
Posts: 334
Likes: 0
Received 0 Likes
on
0 Posts
This may be part of the historical reason in the first place - however as a patent is only good for 20 years this would not have been a cause for the non usage of EPR by GE for the last 40+ years.
This is an extremely good thread, I learned a lot... especially thanks to TdRacer
People are probably wondering why I care as I no longer fly any jets...the answer is because I just like learning about airplanes (and to some extent) Helicopters
People are probably wondering why I care as I no longer fly any jets...the answer is because I just like learning about airplanes (and to some extent) Helicopters
Join Date: Sep 2011
Location: FL390
Posts: 241
Likes: 0
Received 0 Likes
on
0 Posts
TPR is slightly more useful because it works on a scale of roughly 0-100. The only benefit of either EPR or TPR over straight N1 is that they correlate directly to thrust. Two engines producing the same EPR/TPR are matched in power even though their N1 speeds might differ due to wear or damage.
Ultimately I think it's pointless and would be far happier if all manufacturers adopted the Airbus metric of percentage thrust. 100% = everything the engine can give.
Ultimately I think it's pointless and would be far happier if all manufacturers adopted the Airbus metric of percentage thrust. 100% = everything the engine can give.
TPR is slightly more useful because it works on a scale of roughly 0-100. The only benefit of either EPR or TPR over straight N1 is that they correlate directly to thrust. Two engines producing the same EPR/TPR are matched in power even though their N1 speeds might differ due to wear or damage.
Ultimately I think it's pointless and would be far happier if all manufacturers adopted the Airbus metric of percentage thrust. 100% = everything the engine can give.
Ultimately I think it's pointless and would be far happier if all manufacturers adopted the Airbus metric of percentage thrust. 100% = everything the engine can give.
Join Date: Mar 2006
Location: USA
Posts: 2,527
Likes: 0
Received 0 Likes
on
0 Posts
The A350 (and 787 I think) have TPR as primary. It displays engine output as % thrust, so full power with bleed and anti ice off will be 100. I wanted Tdracer’s take on any engineering advantages or disadvantages to this system. Obviously, from a crew perspective, it’s a lot more useful.
Join Date: Sep 2011
Location: FL390
Posts: 241
Likes: 0
Received 0 Likes
on
0 Posts
From RR:
The Trent 1000 engine uses TPR as the normal (NORM) parameter for thrust control. TPR is a measure of the power available at the LP turbine, used to drive the fan, which provides most of the thrust of the engine.
TPR=(P30 × √EGT)/(P20 × √T20)
The 100 isn’t a hard limit, climb TPR for a cruise climb is about 106.
The Trent 1000 engine uses TPR as the normal (NORM) parameter for thrust control. TPR is a measure of the power available at the LP turbine, used to drive the fan, which provides most of the thrust of the engine.
TPR=(P30 × √EGT)/(P20 × √T20)
- P30 is the HPC delivery pressure
- EGT is the Exhaust Gas Temperature
- P20 is the engine air intake pressure
- T20 is the engine air intake temperature
The 100 isn’t a hard limit, climb TPR for a cruise climb is about 106.
I have no first hand experience with TPR - but it would have similar issues as EPR - namely it requires high integrity measurement of the inlet and compressor exit pressures - plus inlet and exhaust temps. Now, compressor exit pressure is a major engine control parameter anyway (the fuel flow schedule is directly tied to the compressor exit pressure) and it's hot enough that probe icing is not a concern - but it can still fail (faulty P30 is a fairly common cause of shutdowns and loss of thrust control events - some FADECs can synthesize P30 in a 'get home' mode if it's detected as failed). And inlet pressure (P20) is always going to be problematic due to probe icing and other types of probe contamination. So, you're still going to need a backup way of setting power when one of the measurements is corrupted (along with rather sophisticated error detection logic to validate the inputs) - and that backup is going to be N1.
I had some friends several years ago who were working on a 'generic' thrust setting parameter - something that went from zero (idle) to 100 (max rated thrust). The big problem they came up with was the failure modes - you'd still need some sort of backup power setting (probably N1) - so you'll always end up with two sets of power setting charts. One for normal operation, and another for backup. So long as you still need the backup, it takes away much of the advantage.
I had some friends several years ago who were working on a 'generic' thrust setting parameter - something that went from zero (idle) to 100 (max rated thrust). The big problem they came up with was the failure modes - you'd still need some sort of backup power setting (probably N1) - so you'll always end up with two sets of power setting charts. One for normal operation, and another for backup. So long as you still need the backup, it takes away much of the advantage.
Join Date: Jul 2013
Location: Within AM radio broadcast range of downtown Chicago
Age: 71
Posts: 851
Received 0 Likes
on
0 Posts
SLF here, and probably worse for a Tech Log thread - attorney, expanding practice and career into aviation (public and private international air law, specifically).
I recall the Air Florida Washington D.C. accident very well; rare is the accident without drama, yet this one was especially poignant. How ironic, or something of a similar nature, that an event nearly 40 years ago could hold relevance for understanding present-day information relating to engine operation. This might be the most important thread I've read on the site, all included. (Yes I looked up EPR and N1, not trusting intuitive guesses.)
Why? Because one of my projects, so to speak, is to try to take the curriculum in programs like the LL.M. in Air and Space Law at a certain Canadian university known for its proximity to ICAO, IATA, CANSO, IFALPA and others in Montreal, and get it back to having a strong, solid connection to aviation fundamentals. Maybe not aviating (navigating, communicating) as such, but so much of the legal profession which does stuff with heavy, even profound influence and/or impact on how aviation actually operates - so much of the profession has no idea what EPR and N1 are about. No idea how well Air Florida 90 on that snowy day in Washington illustrates how fundamental this stuff really is.
I recall the Air Florida Washington D.C. accident very well; rare is the accident without drama, yet this one was especially poignant. How ironic, or something of a similar nature, that an event nearly 40 years ago could hold relevance for understanding present-day information relating to engine operation. This might be the most important thread I've read on the site, all included. (Yes I looked up EPR and N1, not trusting intuitive guesses.)
Why? Because one of my projects, so to speak, is to try to take the curriculum in programs like the LL.M. in Air and Space Law at a certain Canadian university known for its proximity to ICAO, IATA, CANSO, IFALPA and others in Montreal, and get it back to having a strong, solid connection to aviation fundamentals. Maybe not aviating (navigating, communicating) as such, but so much of the legal profession which does stuff with heavy, even profound influence and/or impact on how aviation actually operates - so much of the profession has no idea what EPR and N1 are about. No idea how well Air Florida 90 on that snowy day in Washington illustrates how fundamental this stuff really is.
Oh, one other thing I failed to mention as a plus for EPR. The EPR relationship with thrust is fairly linear - so during takeoff 1.3 EPR would pretty close to half the thrust of 1.6 EPR. At altitude, idle EPR can be significantly less than 1.0 (I've seen as low as 0.7) so the EPR relationship with thrust is less intuitive, but it's still fairly linear.
OTOH, N1 is highly non-linear with thrust below around 85% N1 - and while at high N1 the relationship is more linear, it's not 1 to 1. The number I remember was about 2.5% thrust per 1% N1 between ~90% and 100% N1.
OTOH, N1 is highly non-linear with thrust below around 85% N1 - and while at high N1 the relationship is more linear, it's not 1 to 1. The number I remember was about 2.5% thrust per 1% N1 between ~90% and 100% N1.
IIRC the RR Spey -511 and -512 on the BAC 1-11 used EPR, but the gauges were calibrated in % and an index number was set to correct for ambient temp and pressure so that T.O. thrust indication was always 100%.
I think the early Spey -506 had a P7 gauge, but I can't remember the indications.
I think the early Spey -506 had a P7 gauge, but I can't remember the indications.
Join Date: Mar 2006
Location: USA
Posts: 2,527
Likes: 0
Received 0 Likes
on
0 Posts
I have no first hand experience with TPR - but it would have similar issues as EPR - namely it requires high integrity measurement of the inlet and compressor exit pressures - plus inlet and exhaust temps. Now, compressor exit pressure is a major engine control parameter anyway (the fuel flow schedule is directly tied to the compressor exit pressure) and it's hot enough that probe icing is not a concern - but it can still fail (faulty P30 is a fairly common cause of shutdowns and loss of thrust control events - some FADECs can synthesize P30 in a 'get home' mode if it's detected as failed). And inlet pressure (P20) is always going to be problematic due to probe icing and other types of probe contamination. So, you're still going to need a backup way of setting power when one of the measurements is corrupted (along with rather sophisticated error detection logic to validate the inputs) - and that backup is going to be N1.
I had some friends several years ago who were working on a 'generic' thrust setting parameter - something that went from zero (idle) to 100 (max rated thrust). The big problem they came up with was the failure modes - you'd still need some sort of backup power setting (probably N1) - so you'll always end up with two sets of power setting charts. One for normal operation, and another for backup. So long as you still need the backup, it takes away much of the advantage.
I had some friends several years ago who were working on a 'generic' thrust setting parameter - something that went from zero (idle) to 100 (max rated thrust). The big problem they came up with was the failure modes - you'd still need some sort of backup power setting (probably N1) - so you'll always end up with two sets of power setting charts. One for normal operation, and another for backup. So long as you still need the backup, it takes away much of the advantage.
Join Date: Mar 2006
Location: USA
Posts: 2,527
Likes: 0
Received 0 Likes
on
0 Posts
Oh, one other thing I failed to mention as a plus for EPR. The EPR relationship with thrust is fairly linear - so during takeoff 1.3 EPR would pretty close to half the thrust of 1.6 EPR. At altitude, idle EPR can be significantly less than 1.0 (I've seen as low as 0.7) so the EPR relationship with thrust is less intuitive, but it's still fairly linear.
OTOH, N1 is highly non-linear with thrust below around 85% N1 - and while at high N1 the relationship is more linear, it's not 1 to 1. The number I remember was about 2.5% thrust per 1% N1 between ~90% and 100% N1.
OTOH, N1 is highly non-linear with thrust below around 85% N1 - and while at high N1 the relationship is more linear, it's not 1 to 1. The number I remember was about 2.5% thrust per 1% N1 between ~90% and 100% N1.
At best, I can give you that at idle, the EPR is ~1, but the only way I can get 1.3 being half the thrust is if full power is 2.0, which is quite often not the case. Sometimes max is 1.7, maybe 1.8. Who knows?
If you say that’s linear, I believe you, but that’s engineer-linear. Not pilot-linear
Thanks as always for your insight.
Unless my math is bad, how’s 1.3 half of 1.6?
So do that and the thinking is that .3 is half .6
Having probably made an idiot of myself I'll get back in my box now.