The Old N1 and EPR again!
 

To all those drivers of engines with BOTH N1 and EPR guages - do you have charts to give minimum N1 on T/O for a given EPR and OAT??
If not - why not??

I am not in the same place as my manuals at present. However on the RR535-E4 thrust is normanlly set on the EPR. If the EPR guage was u/s then there is a chart to give you the required N1 against Pressure Altitude and Temperature.

If you are concerned about faulty EPR then this would be found by mismatching parameters between the two engines

Thanks SAM 2M - I drive a couple of BMW-RR 715's attached to a 717 - no charts for Min N1 or N1 if EPR fails - it seems strange to me that this has been allowed to get thru as I have no way of crosschecking that the power required is being attained.

Any other RR drivers out there that wish to comment - have bought it up with RR & Boeing - no reply yet.

Remeber that 737 that had ice on its engine probes and went into the Plotomac (spelling??).
Those guys set the required EPR but due to faulty indications didn't have anywhere near the required thrust.
For this reason its always a good idea to know what sort of N1 indication you'll get for a given EPR and temperature, so you can tell if there is a problem.
Cross-checking engines is good, but remember if one engine is indicating wrong there's every chance that the other one will be doing the same.
On A/C I fly we use JT9D and RB211, the JT9D has a nice chart (marked guidance only) called "Nominal N1%RPM Take Off". The RB211 only has a chart for Go-Around N1% in alternate EEC mode, but I guess it can still give you an idea.

I fly with the JT9D and we always put both T.O. EPR and N1 on the bug card. In addition we call "Thrust set, N1 checked" when T.O. thrust is set.

It is not difficult to calculate the N1 setting, if you already know the limit N1 and the flat-rating temperature.

For example, if a certain high-bypass fan-jet is flat-rated to 32°C at 98.0% N1, then you could determine the limit N1 at any LOWER temperature. The conversion factor is : square root of the ratio of OAT and flat-rating temperature (all temperatures must be in Kelvin).
eg. at 10°C, 98.0% x sqrt((273+10)/(273+32))
= 98.0 x 0.96326
= 94.4% is the limiting N1 (max T/O power)

For deration, example OAT is 15°C and flex temperature is 45°C: if the limit N1 at 45°C is 93.3% N1, then at 15° the N1 should be 93.3% x sqrt((273+15)/(273+45)) = 93.3% x 0.95166 = 88.8% N1
The flex N1 is 88.8% at 15°C (flex temp is 45°C)

You could make a spreadsheet showing the derated N1 against OAT, and take it with you for flight. Use it to cross check the N1 you get on the engine instruments (if your engine is EPR rated). But remember to make separate charts for different pressure altitudes (sea-level, 1000', 2000', etc)

[This message has been edited by Old Dog (edited 20 November 2000).]

Ask one of your guys with a SITA telex connection to contact SEABO7X (Boeing). They should be able to answer.

------------------
rgds Rat

Sounds like something you'd get the Second Officer to work out Old Dog :)

HPSOV, you'll be surprised to learn how many people could determine the correct N1 without a tabulated chart or the FMC. (Very few!)

If the learned and experienced Captains could not, the young and inexperienced effohs won't improve on the statistics very much.

If you care enough, it is best to prepare your own charts, if your company did not provide any.

I might have missed the point of the post, but if you are looking for something to back up your EPR, why not fuel flow?

Or are you looking for an N1 to use with a failed EPR? Will your MEL allow you to dispatch with that?

After all, power is proportional to fuel flow.

You can actually ballpark some aircraft manouevres via fuel flow for the power part of the power + attitude = performance.

Works for piston as well as gas turbine.

In theory I guess, but all engines are not equal, they get tired, dirty, worn, out of trim etc, also when you select a a/c pack on say, the fuel flow increases and the egt increases, but the power is still the same!

757 RB211 E4's

EPR is the main thrust parameter, however we carry in our QRH performance tables based on OAT and thrust..as previously described.

One of the early DC-10 customers was National A/L, and their engineering was very skeptical of N1 power mgmt. so they insisted on EPR. (All other CF6 operators accepted N1 after some convincing...)

But after Pan Am bought them out, they reconfigured the gages to agree with DC-10's in the rest of the world.

Beware the EPR readings in the 737-200. We once had blocked Pt2 inlet reading on both engines due foreign objects finding their way into the sensor tubes. OAT 30C. Obviously not ice. On take off both EPR digital read-outs were equal at 2.18 EPR but somewhere along the take off run it was apparent that the acceleration was not normal for the indicated EPR.

In the darkened cockpit and with a very short runway and as all needles were identical on both engines, no one took a long hard look at the N1 indications. When it was obvious that we were never going to get airborne before the end of the runway, the captain firewalled both engines, rotated quickly, and got airborne at Vr minus 10 knots. On firewalling, both EPR leapt to around 2.27 (max allowable 2.18). Investigation revealed that in all probability, the actual N1 during the take off run was around 89% when for the conditions that night it should have been 101% N1 for 2.18EPR.

Which means of course the N1 is different as well. Quite common to see two unmatched engines producing the same EPR but having to use more N1, higher EGT and more fuel to do it. With degradtion of fan, turbine and compressor clearances etc. the engine has to go round faster to produce the same thrust.

Barit1,

National were not the only ones with EPR on the DC10. SAS also had EPR on theirs, although N1 was the primary parameter.

Might have been the same on other KSSU DC10's as well.

Fuel flow would be the last thing I'd use to cross check a required EPR. F/F is accurate ONLY as it leaves the transmitter. Crack a line between the F/F TX and the nozzles and it may well be the red light and bell that tells you why the EPR has a shortfall.

All the KSSU DC-10's (vertical scale gages) had a common cockpit including EPR IIRC. But EPR was often neglected because it had no operational use.

ATLAS ships had steam gages, with N1 at the top of the engine stack and EPR at the bottom (if it was installed at all...). National was unique for CF6-powered DC-10's in that EPR was used for power mgmt. and was at the top of the stack.

Can't speak for DC-10-40's, but with JT9's I'm sure they were pwr. managed by EPR...

I seem to remember that on the KSSU DC10's the EPR was measure between the pitot probe and the 5.4 station, there was no Pt2 probe. So it wasn't a 'real' EPR anyway.

I believe it was mostly used for condition monitoring. As the 5.4 pressure still is on some ships.

You are right about the -40. Dunno why anybody would choose the JT9-59A over the -50C. Sure it is more solid, but then it also needs to be. And it needs a lot more 'fixes' for airflow control in the compressor.

Aircraft with both EPR and N1, the EPR is used for takeoff setting. In the event of EPR being inoperative, N1 is used for primary thrust setting. Takeoff N1 can be obtained from Aircraft Operations Manual volume 2. If however, a EPR is considered inoperative there may be a large weight penality incurred.

On that airplane the N1 indicators were at the top of the engine instrument stack with EPR being the second one down. When power was applied for takeoff, the pilots referenced the EPR gauges and read what appeared to be the proper EPR, and we know now that it was not correct (because the PT2 probes were clogged, probably with ice – but NOT from what a lot of people believe), and scanning down the instrument stack from that point, as is the habit with most pilots, all the other indications appeared to be “normal” (unfortunately, at that time, normal was essentially understood to be having the needles of both sets of gauges parallel with each other) and the specific N1 indication, which could have been seen as quite low, was not taken into the scan – or at least that is the speculation. However, each engine was producing about 75% of the power that should have been used – 25% less than expected. The piece that goes unnoticed by many observers was that the airplane was, and still is today, certificated to take off with a 50% loss in power. By that relationship, the airplane had 50% more power than it should have needed to get safely airborne.

AirRabbit:
Here's a bit of enlightenment for you: The airplane was cert. to continue TO w/50% thrust AFTER V1.

However, that's a very different situation from a TO w/ 75% of calculated thrust at brake release, coupled with contaminated wing surfaces and who-knows-what on the runway. He didn't reach V1 until he was on the far-end upside-down numbers. I don't have the report at the ready, but I don't think he ever reached Vr.

This accident could have also happened with a gross TOGW error, or dragging brakes, as well as the EPR error. That crew sensed the slow acceleration in the seat of their pants but didn't abort when their common sense told them to.

MEASURED acceleration (from the INS, e.g.) is a cure for such accidents.

We were told that the BR715 fadec continually compares EPR with corresponding N1 and should there be a significant discrepancy then a warning is generated. Decision can be then made to abort (as it will show up as soon as TKOF thrust is set) or set N1 max rated thrust to continue.

Nonetheless, I personally always did a check of the N1 readings to ensure they hit the high 80's

If the EPR gauge is known to be faulty, would it not be safer to take-off using max thrust?

Or is that not possible with modern bypass engines?

On the old things I used to fly, 'full power' take-offs were mandatory with a U/S engine pressure gauge (we used actual P7 values rather than pressure ratio) - and we also had a minimum rpm value acceptable for reduced thrust take-offs of 93% (in earlier days it had been 96%....).

Eagle,
definitely the case - FADEC EPR engines should, in the case of loss of a valid EPR signal, automatically revert to N1 rated mode (or N1 unrated if it all goes bad) and it's always the case when in doubt, push the throttles up to TOGA / Rated N1 thrust.

:ok:

Pretty easy for a FADEC equiped engine ro revert automatically to a safe mode (last commanded) when it detects inconsistencies between readings. Of course the pilot can cancel it and compute whatever he wants on his own

The RR equipped 757 I was flying recently had no EPR indication on one engine. Now I'm fairly new to all this 'grown up' aeroplane stuff, but my initial thought was that we would probably have to use the other engine's EPR gauge as a guide for the u/s one. As it happens that was a totally wrong assumption. In such a situation there are a lot of weight penalties to apply in all phases of flight. No derating allowed and then reference to N1 parameters in the QRH. I don't have the MEL to hand to quote exact figures, but around 7,000Kgs penalty is applied. Use of the auto throttle is not allowed either and VNAV doesn't work. I'm still doing my home work to try and fully understand what was going on because the FD pitch commands, immediately after rotation, went full scale up.

Further more there is no over boost protection from the EEC (which normally look after N1 parameters) in this situation-apparently.

Should anybody have a full explanation about all aspects and implications, I would be most grateful to learn more.

Thanks in anticipation,

SFD

Sorry to be pedantic, ITCZ, but fuel flow is proportional to thrust in a jet engine and power in a piston engine. As power equals thrust times speed, the power of a jet engine is zero at take-off epr when you're stationary at the threshold and then increases as you accelerate (thrust constant). In a piston however, power remains constant and therefore thrust decreases as you accelerate. (Not really related to thread topic but may be of interest to other nerds like me!)

AFAIR, the CVR revealed the captain saying "Real cold, real cold", referring to the engine temperature being lower than it should when setting T/O thrust. I think they concluded that it was the fear of going over the limits of the engine that kept them from firewalling it.

(Edited, because I found the transcript)

15:59:24 TWR Palm 90 cleared for takeoff.

15:59:28 TWR No delay on departure if you will, traffic's two and a half out for the runway.

15:59:32 CAM-1 Okay, your throttles.

15:59:35 [SOUND OF ENGINE SPOOLUP]

15:59:49 CAM-1 Holler if you need the wipers.

15:59:51 CAM-1 It's spooled. Real cold, real cold.


15:59:58 CAM-2 God, look at that thing. That don't seem right, does it? Uh, that's not right.

16:00:09 CAM-1 Yes it is, there's eighty.

16:00:10 CAM-2 Naw, I don't think that's right. Ah, maybe it is.

16:00:21 CAM-1 Hundred and twenty.

16:00:23 CAM-2 I don't know


See http://www.avweb.com/news/safety/182404-1.html and http://www.ntsb.gov/publictn/1982/AAR8208.htm

Actually, the first link above links further to this very interesting article by John Laming: http://www.avweb.com/news/safety/182403-1.html

Spot on to the original question of this thread, even a 737-200 in the case mentioned.

Might I then add the following question:

When it is possible to calculate quite precisely the V1, Vr, ASD, TOD, N1/EPR and all that, why not include a timing as well? Depending on aircraft type, it could be either a fixed time or a fixed speed, ie "After xx seconds, we must have at least yyy kts". This should take care of all erronous engine thrust settings, as well as dragging brakes, blown tires etc, no matter whether EPR, N1, fuel flow or whatever is used.

I know this has been mentioned elsewhere, but I haven't a clue why it isn't used in many companies? (Who does, I never have?)

Currently, with only V1/Vr and thrust setting, faulty engine readings and brake/gear/tire problems is up to only "sixth sense" detection, or you only know when V1 is there when the runway end is closing fast.

Barit 1:

Uhh …. Thanks for the regulatory update that you believe should have been enlightenment for me, sir, but I assure you, your efforts were not necessary.

The difference between accelerating with both engines to V1, losing an engine, and continuing the takeoff with one engine operating at 100% and the case you and I have been bantering about, and the ONLY differenced between them, would be the distance down the runway when rotation occurred. At, and after, that point, the accident airplane should have had 50% MORE thrust and able to fly quite well, thank you very much.

You seem to be very impressed with the NTSB accident report. Well, read it carefully and you’ll find some very interesting “facts” that are just not considered in the final conclusions. I’ll confess that I have no idea about the meaning of your statement “…until he was on the far-end upside down numbers.” I think that if you refer to the report (which you indicate you do have; it just wasn’t at the ready at your last post) you’ll see several things. Among them are the following:

The airplane immediately following the accident airplane in the takeoff queue, took off after the accident airplane with no apparent problems. Consider, if you will, that this airplane had been “exposed to the elements” just as the accident airplane was exposed, and for the same time period of time after being de-iced. Why, I would ask, did this airplane not have similar problems?

You will note the FDR tracing shows that the accident airplane went from a 3-point attitude (all 3 gear on the ground) beyond the stick shaker and into the stall buffet within just over 2 seconds! If you consider that the stall buffet wouldn’t have been encountered until about 22 – 24 degrees of pitch up at that airspeed, this would have made for a lightning quick rotation rate, on the order of 11 to 12 degrees per second. I remind you of two things; normal rotation rate is approximately 3 degrees per second, AND the F/O, who was at the controls, had indicated he was going to “takeoff the nose gear and then just the airplane fly off by itself.” Does that indicate anything inconsistent to you? A F/O wanting to deliberately limit the rotation to just getting the nose gear into the air, the normal rotation being 3 degrees per second – why, then, would the airplane rotate at 12 to 14 degrees per second?

You will also note the time-lapse between when the accident airplane was cleared onto the runway and when the throttles were advanced for takeoff. You will also note that there was an Eastern B727 on final approach to that runway at the same time – the distance out from landing of that B727 was only 2 and a half miles when the accident airplane crossed the hold-short line and was rapidly decreasing as the accident airplane taxied onto the runway and slowly, and deliberately, turned to line up for takeoff. How close did the two airplanes come? Well, estimates differ, but calculations from the tower tape, FDR and CVR recordings from both airplanes indicate that they BOTH were on the runway at the same time! Had the accident airplane aborted on the runway – there was, at least, a very good chance that both airplanes would have been accident airplanes!

You will also note that the “deicing” that took place in “the chocks” was accomplished with an improperly maintained Trump Deicing truck – that provided an incorrect mixture of heated water (to 160 degrees) and glycol – to the point that at high volume flow from the nozzle provided a whopping 3% glycol solution, NOT the 30% that was expected! Just a quick question, since you know so much about airplane performance, and such, what do you think the aerodynamic changes would be with a very smooth coating of clear ice all along the leading edge of a B737 wing? So that you won’t strain too much, let me help you, just a little. With very little deformation of the leading edge, the outboard portion of the wing will not produce lift until reaching a much higher airspeed than if there was no deformation. The wing roots will produce lift, although slightly less than with no deformation. Since the wing roots are farther forward, ahead of the cg, as soon as the flight crew relaxed forward control column pressure, the inboard portion of the wings, generating lift, rotated the airplane, and rotated it quickly to a pitch attitude of 22 – 24 degrees. The only thing that could have saved them at that point, and there was and is absolutely NO guarantee about that, was by moving the horizontal stabilizer to a more nose-down position. But since the airplane was airborne for just a bit over 15 seconds, it is doubtful that any stab movement would have been successful.

And finally, for this post, and in reference to my comment in my earlier post about why the PT2 probes were clogged with ice, but not from what most people believe … the PT2 probes were probably clogged with ice, but it was from the 160-degree water sprayed on them about an hour and 12 minutes earlier. Just enough time, in 22-degree weather, to allow the water filling those probes to freeze and provide inaccurate engine pressure ratio readings.

No sir, the accident did not happen because of a “gross takeoff weight error.” There were no “dragging brakes.” The airplane did reach V1; it is just that due to the error in the deicing procedures, the airplane became airborne beyond the control of the flight crew. And here’s a bit of interesting, and perhaps enlightening, news for you, sir. On the same day, SAS experienced almost the identical problem with one of their B737s operating out of Oslo, Norway. The difference was that the leading edge ice build-up was Mother Nature’s doing, not man’s, and as such, it was asymmetrical. The resulting pitch-up, equally uncontrollable, was also a rolling pitch (due to the asymmetry). Even thought the crew slammed the throttles to the firewall, input full opposite aileron and full opposite rudder, they were unable to control the pitch/roll of the aircraft. Fortunately, the advanced throttles began to accelerate the airplane as the nose dropped back down toward the horizon because of the radical bank angle. As the airplane accelerated, lift was produced over the outboard portions of the wing, including the ailerons, and that allowed the crew to roll back to level flight – but this happened below 100 feet AGL! Had they not had this bit of inconsistency from Mother Nature, the world would have seen B737s on opposite ends of the world crash on the same day from the same problem.

If I sound a bit “exercised” over this event – that is because I was, and I still am. I was, shall we say, "very close" to the operations of that company and their training/checking program. The accident investigation was, in my opinion, very loosely conducted and, again in my opinion, willing to over-look key pieces of information that didn’t fit with the conclusions ultimately reached. For what its worth, I am grateful that the issue of deicing procedures and hold-over times have been significantly overhauled and are much better understood and respected today – at least that knowledge has been of benefit to this industry.

And, by the way, I would whole-heartedly support a measured acceleration method (from an INS or any other unaffected source) as you have suggested, in the hope that it would deter or reduce such accidents. The reason I don’t say “cure” such accidents, is that as long as we have humans in the loop, I believe we’ll be prone to suffer human failings.

Now, that's pertinent to the subject of this thread. But, I'm sure the probes were frozen solid LONG BEFORE the TO roll started.

Well put. However, the human body is better adapted to being "in the loop", with systems providing external monitoring, rather than the other way around. Don't let triple-redundant FMS fool you.

I’m relieved that you approve of my comment as being “pertinent.” Of course, I’m sure you’ve ensured that no one else here has deviated from that norm. I would, however, offer a small disagreement with your presumption that the “probes were frozen solid…” It was my impression that if the PT2 probes were “frozen solid” there would be no pressure sensed at their location and any differential pressure between the PT2 and the PT7 probes would have been sensed as an “off-scale” differential pressure reading. Had this been the case, any throttle movement would have been seen as a max-scale deflection of the EPR gauge and would have been noticed. That there was an EPR reading just different enough to be somewhat questionable indicates that the mixture in the PT2 probes was only partially frozen, likely due to the 160-degree water mixed with the 3% glycol used to “de-ice” the airplane.

While pilots are expected to be “in-the-loop,” having additional systems provide even more information presentations in the cockpit is not necessarily useful, unless that information is accurate, unambiguous, and immediately useful to the “in-the-loop” pilot who has to make reactive decisions – there just isn’t the time to logically analyze yet more numbers in relation to even more numbers.

Uh hum, 737-200's, Potomac River etc... We are talking about a modern a/c, B717 and modern motors BR715.

Watchdog is on the right track. The FADEC system is constantly comparing back it's inlet pressures to the a/c probe.

If it's out of whack then the pilot is informed and he can select N1 Mode.

Then who needs a chart? The little man in the FADEC is doing that calculation far quicker than I could do with 10 hand a slide rule!

The better FADECs automatically revert to N1 mode on their own and set a flag to inform the pilot

What am I missing here?
 

Since the FARs/JARs already require a tacho on each rotor system, and since N1 can provide the front office with necessary performance information for a high-bypass fan donk, why does anyone still use EPR? It's at best redundant, and (as the Potomac dunking shows) problematic.

LGB -



In your timing proposition for takeoff, it seems it would be a human factors challenge to add another task to the takeoff workload. But it is an interesting concept. I wonder how much the timing would vary for different weights, runway slope, use of flex or derate, winds, etc. It may be variable enough to require another set of charts.

The time from brake release to V1 should be easy to calculate given the right data (TOGW mainly). If V1 is not reached in XX seconds, your acceleration is subpar, and since you haven't travelled the full accel distance, excess stopping distance is available.

But a better solution is to use a direct acceleration measurement (e.g. INS) - for a much earlier decision point.

They had such a system on the VC10 right?

I don’t know what I have to say to get folks to stop thinking that the B737 accident in Washington DC/Potomac River in 1982 had something to do with blocked PT2 probes or a lack of acceleration. Neither was the cause of the accident. It was a faulty de-icing procedure that unfortunately covered the airplane in water that promptly froze. Due to the ice, the wing was “deformed” and caused an uncontrollable pitch-up from which the crew could not recover.