Why EPR?

[Jwscud]

I have been thinking about EPR for thrust setting since reading the Swift Air accident report. It is not the only accident I can instantly think of where inlet probe icing led to incorrect thrust limiting and setting (Air Florida comes instantly to mind) and I believe that Boeing removed EPR from the 737 in favour of N1 for precisely that reason.

EPR also seems a far less user friendly method of thrust setting - I have only flown in EPR equipped aircraft in the sim (732 and 744) but did not find setting 1.17 EPR on a tape anywhere near as simple as setting 60% N1. I have also seen the EPR equipped 320s with a 3 decimal place EPR as primary thrust setting indicator.

I am aware that with everything working, EPR actually tells you how much thrust the engine is producing whereas N1 only tells you how fast the fan is spinning.

Is the advantage of knowing precise thrust output enough to outweigh the limitations I've set out? I don't have enough experience with EPR to answer the questions (or know if I'm asking the right ones) so would welcome some input from those who do.

[Meikleour]

Jwscud: EPR does have one quirky advantage.

Say with a 4-eng aircraft the final approach power was 1.25 EPR This equates to a 4 x .25 increment ie. +1 therefore a 3-eng approach would require +1/3 ie. 1.33 EPR for final.
2-eng aircraft. 1.25 EPR would require 1.5 EPR for approach.
Simples!! It relates directly to thrust whereas N1 is not linear wrt thrust.

[gearpins]

EPR is a direct measure of thrust, whereas N1 / n2 is indicative of corresponding thrust.
when instruments are suspected to be unreliable all aircrafts revert to N1 as the base vale.
however under normal operations EPR provides a more precise measure which helps FADEC.
guess each has its good and bad.

[FullWings]

I’ve had a fair amount of experience with EPR and N1 as the primary reference and on the whole, I prefer N1 now.

The older engines seemed to have a “truer” EPR display, i.e. 1.0 was zero net thrust and 1.6 was twice as much as 1.3. The RR Trent is happy pushing 250T of aircraft along at 250kts with an EPR<1. How does that work? Also, due to the huge variation in required TO EPR with altitude and temperature, there are no “ballpark” settings - at least none that I can remember. Doing circuits N1 is a lot better for a initial stab at a power setting.

With thrust synced between engines on EPR, you get that annoying phasing vibration that you don’t get with synced N1s. N1 is also a more robust measurement in that probe icing, etc. doesn’t affect the reading. I hear RR might be going to N1 as primary in the near future, anyway...

[Tu.114]

There also was an accident that showed the value of EPR over N1. In 2004, a F70 of Austrian suffered a forced landing on a field not far from EDDM (https://aviation-safety.net/database/record.php?id=20040105-0) (Report) (https://aviation-safety.net/get.php?http://www.bfu-web.de/DE/Publikationen/Untersuchungsberichte/2004/Bericht_04_AX001-0.pdf?__blob=publicationFile) due to a rather peculiar double engine failure: in severe icing, the constant barrage of ice shed by the fan and thrown at the protective strips just behind the fan caused them to come loose. These strips were of a size and material that did not allow them to depart the engine down the normal airpath; they accumulated in front of the fan stator vanes and blocked the air path.

So the engines were both spinning as they were expected to. N1, N2 and EGT was indicated and within normal ranges, but what was lacking was EPR and consequently the expected thrust.

Both approaches have their benefits and downsides. It is possible for both indications to show normal values with not nearly enough thrust being developed, and to show this behaviour on all engines installed on an aircraft at once.

EPR is simply one more measurement showing the state of an engine. Also if it is not used to set power ratings, I think it is a valuable information to have in order to get a rounded picture and to see that everything is in order.

[lomapaseo]

and then there is the engine that controls to EPR with a mixed flow between the fan duct and the turbine exhaust and when the fan goes south ends up over-boosting the core compressors ala the SK MD80 ice ingestion.

you can't have your cake and eat it to

but we've had this discussion over and over in this section over the years and the pilots seem to be able to handle the differences :)

[fantom]

EPR? Goodness no.

Gimme N1 anytime.

[tdracer]

As others have noted, EPR is a 'more' direct measurement of actual thrust than N1 (although EPR has become a less direct measurement as fan bypass ratios have increased). As an engine guy, the biggest advantage to EPR is that it's temperature independent - at a given pressure altitude EPR will give you pretty much the same thrust regardless of TAT (something you can see in the power setting charts - TO EPR will be pretty much constant below the corner point temperature).
N1 correlates to thrust as a function of air density - and air density varies with temperature and humidity. Temperature can be accounted for with the square root Theta term (aka "root Theta") but humidity is generally an unknown (and can be significant on a hot day), so power setting charts tend to assume worst case for humidity (if it's a hot, dry day you'll get some extra thrust). N1 is also very bad in the event of fan damage (e.g. bird strike) - N1 will go up while thrust is going down.

All that being said, at least from the design engineer standpoint N1 is easier to deal with and has fewer failure modes.

At least to date, pretty much all GE/CFM engines use N1, Pratt and Rolls use EPR. That may change.

[Speedwinner]

Sorry I don't get it. The aircraft is operated by a pilot. The pilot needs to know the values for power and set it. And then we get airplanes with a .00334422 indication which is so small and not easy to handle. And in case of a fault you fly n1. So he don't you fly always n1. Is the epr system better for auto thrust?

[Tu.114]

After all, a figure is a figure. If Your engine requires one indication to be 93 of something or another figure to show 1.73 of something else for it to deliver the required takeoff thrust, what is wrong with just setting it and coping with what the engineers have thought most appropriate for their engine?

True, every method has its downsides and there are situations in which the value shown has little relation to the available thrust that either method might lead into. But if one knows the pitfalls and does not only look at the primary measurement, but also at all the other secondary indications for a plausibility check, one is not that easily caught by something nasty. An EPR or N1 indication in the takeoff range simply does not come with a low EGT or hardly noticeable N2.

[FE Hoppy]

EPR is for Flight engineers. You pilots should have N1 and pretend it means the same thing.
Just set the number and don't think. Or better still let the autothrottle do it.

[Turbine D]

tdracer says it best:
All that being said, at least from the design engineer standpoint N1 is easier to deal with and has fewer failure modes.

[Thridle Op Des]

Like FullWings, I flew an engine where we had an EPR of 0.982 or something equally meaningless (RR Trent 500), since then, I have always preferred N1, not as a direct measure of performance, but rather as a reliable reference I can come back to. I now operate the GP7000 engine which has a EPR variant called 'Thrust'. It's very clever (perhaps too clever), 100% 'Thrust' is exactly that, nice idea but relies on multiple sensors. For an unreliable speed reference, I am still noting the cruise N1 and pitch attitude as my reference in case the clever system becomes confused.

[barit1]

More trivia: FAR's (& JAR's I surmise) require each rotor system to have a tachometer indication for the crew.

So EPR, if installed, is an additional requirement, not a replacement for N1. :=

(Not much of an issue on modern displays, but I recall when steam gages ruled the skies)

[lomapaseo]

So EPR, if installed, is an additional requirement, not a replacement for N1.

Yup, EPR is for the fuel control to know what it's supposed to do in the Brayton cycle and it gives the pilot something to watch and control on :E

[Romasik]

Like FullWings, I flew an engine where we had an EPR of 0.982 or something equally meaningless (RR Trent 500), since then, I have always preferred N1, not as a direct measure of performance, but rather as a reliable reference I can come back to. I now operate the GP7000 engine which has a EPR variant called 'Thrust'. It's very clever (perhaps too clever), 100% 'Thrust' is exactly that, nice idea but relies on multiple sensors. For an unreliable speed reference, I am still noting the cruise N1 and pitch attitude as my reference in case the clever system becomes confused.

EPR values less than 1 are not meaningless. They mean that the thing is producing drag, not thrust. Something that you never get from N1 indication. Could be usefull in understanding what's the difference between idle thrust and dead engine. Of course you need to know EPR at the same speed on dead engine. We may check it in the SIM.

[mnttech]

EPR is for Flight engineers. You pilots should have N1 and pretend it means the same thing.
Just set the number and don't think. Or better still let the autothrottle do it.
FE Hoppy, you just made my day.
Back when I was the cruise control for the 707, that was about true. They set it for takeoff, and after that it was mine until approach time. Then, for the most part, the two drivers used fuel flow... 3K a engine worked just fine until those stupid stage 3 cowlings came along. Flaps 25 on a -300 was guessing game. The 3 holer (727) used 3K a lot, big fat numbers to try and find.

[Thridle Op Des]

I'm sorry, for this poor pilot an EPR of .982 in the cruise, level flight at FL410 at 220 tonnes IS meaningless. Maybe I should have been more specific about the conditions I was describing.

[oceancrosser]

I have been thinking about EPR for thrust setting since reading the Swift Air accident report. It is not the only accident I can instantly think of where inlet probe icing led to incorrect thrust limiting and setting (Air Florida comes instantly to mind) and I believe that Boeing removed EPR from the 737 in favour of N1 for precisely that reason.


I don´t think Boeing "removed EPR" from the 737 Classic. Rather they chose an engine, CFM56 that used N1 as primary power parameter in line with GE philosophy. The CFM56 predates the 737 Classic by almost 10 years.

[Jwscud]

Fair enough. That was the reason I was given in relation to Air Florida. Always good to learn the more common sense explanation!

Setting thrust by fuel flow these days is rather hard as it's a tiny box on the secondary indications, whereas N1 is a nice big user-friendly dial for idiots like me who can just about set thrust to the nearest 1%.

[barit1]

oceancrosser:I don´t think Boeing "removed EPR" from the 737 Classic. Rather they chose an engine, CFM56 that used N1 as primary power parameter in line with GE philosophy. The CFM56 predates the 737 Classic by almost 10 years.

And the CF6 used N1 when it first flew on a B-52 testbed - late 60s. http://www.aviastar.org/pictures/usa/boeing_b-52.jpg

[Uplinker]

I think if EPR gauge scales were shown as a percentage, instead of 1.387 or whatever, us pilots would find them much easier to use.

We want ballparks which we then fine tune. Setting 55% N1 is easy. Setting (and remembering) 1.432 or whatever is not.

Lose the decimal point and the digits after it. Have the scale read -10% to 110% say.

[Check Airman]

I think if EPR gauge scales were shown as a percentage, instead of 1.387 or whatever, us pilots would find them much easier to use.

We want ballparks which we then fine tune. Setting 55% N1 is easy. Setting (and remembering) 1.432 or whatever is not.

Lose the decimal point and the digits after it. Have the scale read -10% to 110% say.

I believe that's what Airbus did with the "Thrust" indication.

I don't see the point of the EPR indication either.

[Uplinker]

Ah, right. It is not on the 320/321/330 I fly, but it sounds sensible.

[tdracer]

We (Boeing) have done a lot of work over the years with regard to a generic "thrust" indication - something that would go from 0 to 100%. Short story is when it works, it's great, but the failure modes are nasty.
BTW, the reaction to Air Florida was not to get rid of EPR, it was to make the inlet probe heat automatic (e.g. JT9D on the 767, PW4000 on the 767, 747-400, and 777), RB211-524G/H (767 and 747-400) and Trent (777 and 787). Loss of inlet probe heat will automatically result in alternate (N1) mode operation.
Oh, and EPR less than 1.0 is far from meaningless (and doesn't mean negative thrust). At cruise airspeed, EPR can easily get as low as 0.7 with an engine making positive thrust due to the ram component (I don't think I've ever seen a scenario with a running engine that the fan pressure rise or ratio (FPR) wasn't positive - and that means positive thrust). Further, even if the primary exhaust pressure is the same as ambient (static) pressure, the much higher temperature of the exhaust means positive thrust.

[FullWings]

For me, the issue is non-intuitive EPR values. For example, on the GE90-115, full thrust is 100-105% over a large range of altitudes and temperatures. If you’re on a derate, <85% probably means something is wrong. In the circuit, F5, 55% and 5degs NU is a good starting point.

With EPR on the RR Trent, who knows? 1.372, 1.594, 1.193...? It’s not very memorable and changes much more with environmental conditions. Is 1.228 a good thrust setting for a derate? Gross error checking isn’t as easy.

At the end of the day, the AT doesn’t care if the units are nano furlongs per ounce second squared or walnuts per light year: if it’s slow it adds a bit and if it’s fast, it takes some away. :)

[FGD135]

EPR represents the thrust you actually have.

N1 represents what you should have.

[Cough]

We (Boeing) have done a lot of work over the years with regard to a generic "thrust" indication - something that would go from 0 to 100%. Short story is when it works, it's great, but the failure modes are nasty.

I presume the issues were overcome to produce TPR on the 787?

[A Squared]

EPR values less than 1 are not meaningless. They mean that the thing is producing drag, not thrust. Something that you never get from N1 indication. Could be usefull in understanding what's the difference between idle thrust and dead engine. Of course you need to know EPR at the same speed on dead engine. We may check it in the SIM.

Well, yeah, You'd think. But, If I'm reading this correctly, there's at least 2 posters speaking of having all EPRs <1 in steady state cruise. Which indicates thrust being produced at an EPR less than one. Dunno, I don't fly jets, so I'm just trying to follow the conversation.

[FullWings]

Here we are from the QRH unreliable airspeed tables for RR Trent:

15,000’, 270kts, 160T, EPR = 0.978, N1 = 64.2%

I can’t remember what the EPR is on a failed engine, I’ll have a look next time I’m in the sim.

[Meikleour]

Regarding EPR versus N1 for power achieved: I once had an experience of fan blade damage to two engine on a CFM56 powered A340. The resulting N1 settings required to make the approach were very much higher than usual since the two affected engines were not producing the normal thrust expected from the N1 readings.
When talking about the Trent engine EPR values, iirc these are Integrated EPR since the engine has three spools. So ,one frequently saw values less than 1 although I am quite sure the engine was not "sucking" at the time!!!!

[FlyingStone]

It's interesting that Airbus doesn't even mention EPR setting for unreliable airspeed or severe turbulence (IAE powered A320s) and uses N1 instead, even though this is said to be primary thrust setting indicator.

My personal opinion - EPR information is very nice to have, but N1 as primary thrust indicator/setting wins hands down. Very few failure modes and very reliable. In the very rare event that you have some engine damage (bird strike etc.), you could always refer to EPR that could give you some more clue.