Engine start on ground with Engines Windmilling A320

[Bkdoss]

Is it recommended by Airbus to do a Manual Start in case the engines are already windmilling on ground ? The airlines I fly for insists on doing a MAN start, but I fail to find Engines windmilling among the listed conditions necessitating a MAN start in Supplementary procedures

[NSEU]

I assume the manuals are referring to engines windmilling in the wrong direction?

Some engine types have EECs which monitor rotor* speed and won't allow an autostart if rotor speed is not great enough. This, I believe, is to stop hot starts due to insufficient airflow through the engine.

If the EECs on your aircraft don't monitor rotor speed, then it's up to you to ensure engine airflow is sufficient.

*(EDIT) I mean rotor speed other than the starter-driven rotor. The rotors can be turning in different directions, slowing down the gas flow through the engine.

[lomapaseo]

The engine cares about the rotor that pressurizes the combustor, that pretty much excludes the fan rotor windmilling backwards.

The windmill is insignificant on the ground to any work from the compressors. However any flow through the rear compressors has to pass through into the turbine and the turbine driving the rear compressor pretty much works as designed in turning that rotor in the right direction as long as there is flow through the engine and not just out the fan pipe.

Of course what instruments do and operating manuals say always takes precedence in discussions like this.

[DevX]

QUOTE: "The rotors can be turning in different directions, slowing down the gas flow through the engine".

Not so. All shafts will rotate in the same direction if an engine is windmilling irrespective of whether it has 2 or 3 shafts. The only exception of course are engines that have counter rotating shafts, Trent 1000 etc.

Lomapaseo, you're talking through your "fan pipe"! :yuk:

[Meikleour]

Bkdoss: My experience of CFM56 engines over the last twenty years shows -

The N1 fan can rotate quite fast in a tailwind. If the FADEC is powered one can see readings of up to 2% N1 rpm - however this is in the wrong direction!
The N2 fan does not usually windmill on the ground because of the high drag/load from the ancillary gearbox.

When the engine starter is engaged, the N2 shaft is driven in the correct sense and one can watch the N1 slowing down and then building in the correct sense.
Another good clue is the EGTs increasing and then decreasing as the correct direction of airflow is established through the engine prior to fuel in and lightoff.
The start then progresses normally.

Some large fan engines like the Trent700 have maximum tailwind limitations and with the A330 we used to push to crosswind for a start in these situations.

[Bkdoss]

Thanks everyone for your insights.I'm a beginner on the Bus, so excuse my naivety

Meikleour, glad to hear out the views of veterans like you on this forum.

Reading your replies, I had a couple of other questions sprout up.

a) As you've pointed out, windmilling in the wrong direction can be quite a stress on the engine while starting. In fact, of the two engine variants my airlines has, CFM-5B6 and IAE V-2500, Airbus recommends manual start on IAE V-2500 when starting with tail winds more than 10 kts, but such a condition is non-existent in CFM engines. So would you suggest throwing caution to the winds and invariably using a manual start whenever Engines are windmilling in the right or wrong direction


b) I don't yet understand the fundamental difference between a manual start and an automatic one from an engine point of view. Agreed, the difference basically lies in either FADEC monitoring the start sequence or the pilot taking over control in case of man start choosing when to start the fuel pumps and so on. But in principle, while starting we follow the same sequence as the FADEC does, on when to time the opening of the fuel valves and so forth. In such a scenario, can you explain how doing a MAN start scores over an automatic one in case of tailwinds, or starting at high altitude airfields ?

[Meikleour]

Bkdoss: a) the N1 & N2 shafts rotating in different directions does not "stress" the engine - it just means that the driven(by the starter motor) N2 spool has to run for longer to establish the correct airflow direction through the engine. So, the short answer to your question is NO. Follow the manufacturer's recommendations. ie. for the CFM56 no problem. For the IAE engine - again follow the recommendations. Perhaps your airline has more aircraft with the IAE engine and they have decided to "standardise" procedures for simplicity?

b) With a manual start two things are different: Dual Igniters are used and the fuel is added by pilot input. For instance, with an autostart on the CFM fuel goes in at 22% N2 but for a manual start you may want to wait for a higher value of N2 before introducing the fuel. In addition control of over EGT is entirely the responsibility of the crew cf. the auto start.

Hope this is of help?

[Turbine D]

Meikleour,

Your response to Bkdoss is absolutely correct. As most people know, the fan is connected to the LPT. One can start the fan and LPT rotating using a finger, that is how good and balanced the bearing system is. With a tailwind entering the exit end of the LPT, it turns the LPT in the opposite direction of intended design, thereby turning the fan in the opposite direction as well. Besides the points you mention regarding the HPC and HPT, I don't think the energy of a tailwind ever gets that deep into the engine, the energy is dissipated in the LPT. As you point out there is no mechanical stress on the engine.

[vilas]

In CFM engine manual start when EGT exceeds starting limit before reaching 50% N2, FADEC will abort start. This is the only exception.

[barit1]

If one rolls back the calendar 45 years or so, PanAm's very first 747-100 revenue flight (JFK-LHR) was parked with wind up the tailpipe; it suffered hot start(s) on JT9D-3's, forcing the callup of a backup ship. Ever since then some engines have been sensitive to tailwinds, and others not so much.

I think the feedback in posts above is representative of most of the industry.

[BOAC]

There's always the old trick - crack the reversers.

[Bkdoss]

Meikleour.

Thanks for the elaborate reply . The dual igniter part slipped my mind.

for a manual start you may want to wait for a higher value of N2 before introducing the fuel

I invariably put the Master switch on at 22% or at the max motoring speed i.e when I feel that the acceleration on N2 rpm has flattened out above 20%. Are there any other cues to help me decide when to switch the master ON ?

[Agaricus bisporus]

The airlines I fly for insists on doing a MAN start, but I fail to find Engines windmilling among the listed conditions necessitating a MAN start in Supplementary procedures

Never seen reference to that in either my airline's procedures or in Airbuses. Or on the 737. Last time I came across this was on the Quadropuff.

Someone's making up unnecessary procedures to justify his paygrade, I think.

"Well, we did it on the DC8, so we'll do it here!" :ugh:

If you want to throw away the FADEC's safeguards and unnecessarily revert to an unfamiliar procedure (I've never done a manual start out of the sim) that could be more hazardous I'd say it's a bloody silly policy, but hey ho!

[flyingchanges]

With a tailwind entering the exit end of the LPT, it turns the LPT in the opposite direction of intended design, thereby turning the fan in the opposite direction as well


I would bet the fan is turning the lpt...

[Bkdoss]

Agaricus bisporus

Someone's making up unnecessary procedures to justify his paygrade, I think.

Precisely what I thought.

(I've never done a manual start out of the sim)

On the Bus, really ? Have you ever ended up with start faults when you have have disregarded the man start conditions ?

[Turbine D]

FC,
I would bet the fan is turning the lpt...
Good bet if the wind is blowing into the fan, bad bet if the wind is blowing up the LPT from the rear…

In normal engine operation, do you think the fan drives the LPT, or does the LPT drive the fan?

[flyingchanges]

So, you really think the lpt is turning the fan vs the air going through the bypass turning the fan? Considering none of the core is rotating, how much air do you think is going through the lpt?

Which one has more surface area and access to free flowing wind?

[barit1]

Turbine D:Good bet if the wind is blowing into the fan, bad bet if the wind is blowing up the LPT from the rear…


I respectfully disagree. The fan, even in reverse flow, can generate a lot more torque at low P/P than can the smaller diameter LPT, which is designed for much higher P/P.

And BOAC's remark about deploying the fan reverser makes sense in this context.


In normal engine operation, do you think the fan drives the LPT, or does the LPT drive the fan?

This has nothing to do with the very low P/P case.

[DevX]

QUOTE: "..........about deploying the fan reverser makes sense....."

Had to chuckle, it sounds like the reverser rotates the fan opposite to the normal direction of rotation. It simply diverts the direction of LP airflow forwards out through the cascades (or buckets), but then you knew that, right?. :hmm:

[Agaricus bisporus]

Anyway, arguments on fan rotation vs. p/p (wtf?) etc isn't what the OP asked.

Bkdoss, if Airbus doesn't specify it as a procedure it just isn't necessary. It may or may not be potentially harmful, but anything that reduces the automatics' protection levels must increase potential risk of damage. It does, imo, indicate that the person/persons/department that wrote and approved that requirement is less than ideally qualified for the job. It also suggests a lack of understanding of the Airbus philosophy and probably an excessive belief in their own technical superiority over Airbus themselves, the first of which is a highly undesirable trait in tech management, and the second is pretty damned unlikely.

Beware people like that, they can lead you into grief in several ways.

[NSEU]

BdKoss, on your A320s, do you have one or more the packs running during start?

Getting a good airflow through the engines (all rotors turning in the right directions) may help reduce fuel fumes in the cabin.

As engineers, we mostly use manual starts unless there is something specifically wrong with the Autostart System. We have more control over the start: The process is slower, giving us time to monitor specific parameters at specific times. We get to choose when the fuel/ignition goes on. If fuel has pooled in the combustion chamber, a relatively long dry spin ensures that all the excess fuel is blasted out the back before ignition is turned on. EECs are not that smart.

[barit1]

My P/P remarks are directed to colleague TurbineD - but here's the elucidation: P/P is simply the total pressure ratio, in this case the tailwind upstream pressure divided by downstream pressure. In the tailwind case the number is very low, close to unity, and when applied to the rear end of the LPT won't do much.

But applied to the much larger fan blades, a substantial amount of torque can be had even at very low P/P.

Compris?

And deploying the reverser places the blocker doors in the way of this P/P, closing the valve, spoiling the driving force of the fan rotor.

[Bkdoss]

NSEU,on the ones that I fly, once the Engine mode selector is switched to START, the FADEC powers up and the PACK valves are closed. The Pack valves are opened 30 secs after the second engine start . I don't know if this logic is engine specific or common across the 320 fleet. We have more control over the start

This is the part which baffles me. When you do a MAN start , apart from the dual igniters firing as rightly pointed out by Meikleour in one of the replies, I find no difference in the start procedure as listed in the procedure . The procedure says , place the master switch to ON position, when the N2 reaches 22% which is what the FADEC actually does in an automatic start . So in effect you are just mimicking the actions a FADEC would do. If not, can you please throw light on the other cues that help you in timing the start sequence better than what a FADEC would do ?

[Meikleour]

BKDOSS: Re-read your SUP PRO's again!

For a manual start the fuel MAY BE put in at 22%N2 or MAX MOTORING.

ie. if it is possible to achieve a higher N2 speed on the starter then do that before the fuel is added. This may help with an engine that is prone to over temp. By the same token with an air source that is a bit weak you may not even achieve 22%N2

Do you understand now why MAN STARTs may be used?

[Bkdoss]

Thanks a lot for your explanation Meikleour
All the while I have been switching on masters at N2 of 22% even though the engine was accelerating,as the SUP procedures does not clearly enlist the factors that would help me decide whether to let the N2 flatten out above 22% or start the Masters at 22%. So your suggestion is if the engine has a history of high EGTs during start, then delay opening the fuel flow. The weak bleed source is easier to identify as N2 wouldnt build up sufficiently, so you might want to reduce the load on the APU by isolating the consumers such as fuel pumps etc.

[vilas]

Bkdoss
I think you did not read it carefully. SUP PRO for CFM clearly mentions:
The maximum motoring speed is defined as the speed at which N2 acceleration is less than 1 % in approximately 5 s.
and for IAE
Maximum motoring speed is reached when a significant decrease in N2 acceleration is observed. Do not attempt start unless N2 is at least 15 %

[fruitloop]

As "Meikleour" has pointed out Min motoring "Before" introduction of fuel (22%)and as a lot of engineers look for a positive turning of N1..Fadec during a auto start doesn't care if its turning backwards or forwards and will introduce fuel. The starter limits for operation are fairly tolerant (An example is when doing "Low Pressure starts" with a dodgy start unit..or with a worn out APU..

[barit1]

fruitloop:a lot of engineers look for a positive turning of N1
This bit may not apply to either CFM or V2500, but is a bit of history from earlier high-bypass engines: A new engine, or one with a fresh LPT module, may have too-tight clearances, and the N1 rotor may not turn freely - it may not even turn at ground idle.

The solution applied was to shut down after perhaps 30 seconds, wait a bit for the LPT case to expand from the heat, then re-start. By then the blade-to-shroud clearance should be opened up a bit (the rotor discs are massive and don't heat up as fast).

Your mileage may vary! :uhoh:

[DevX]

Quote: " ....... and the N1 rotor may not turn freely - it may not even turn at ground idle."

Trent engines have logic within the EEC that prevents a start if all 3 shafts aren't rotating so the above scenario isn't even an option.

[Natstrackalpha]

If doing a manual start (CFM`s again) and you decide to abort start (due to whatever reason you decided to) - and if you wanted to then Crank - (for whatever reason you decided to Crank) - then is there a lower speed (N2) that you would wait for like on the L1011 - you had to wait for the turbine speed to get below a certain . . .speed or the starter would crash start the engine at a higher turbine speed = not good. I do not see this written anywhere.

[BOAC]

20% - Normal Procedures, Engine Start.

[DevX]

There's no problem crash re-engaging the starter on the old RB 211-22b (L-1011) as long as the N3 speed is at, or below 10% :ok:

[vilas]

Nats
From FCOM for CFM


Dry crank the engine for 30 s. The start valve automatically reopens when N2 is below 20 %. After the starter cools, and for any subsequent attempt to start the engine, the flight crew must perform a manual engine, or must report the “no start condition” to maintenance for appropriate action.

[FullWings]

I’d have thought it best to do what it says in your manuals, rather than inventing a procedure or copying one from another type.

If you carried out a non-SOP engine start and it spat some of the hot section out onto the tarmac, how would the subsequent discussion of the incident go, especially when asked why you didn’t follow SOPs? The engine was probably going to do that whatever but now you’ve included yourself in the blame loop...

[barit1]

If doing a manual start (CFM`s again) and you decide to abort start (due to whatever reason you decided to) - and if you wanted to then Crank - (for whatever reason you decided to Crank) - then is there a lower speed (N2) that you would wait for like on the L1011 - you had to wait for the turbine speed to get below a certain . . .speed or the starter would crash start the engine at a higher turbine speed = not good. I do not see this written anywhere.

I had this happen to an engine (not CFM) in the test cell a few decades back. The engine had preservative oil in the fuel system, and the test crew did not flush it properly. In fact there was some residual oil pooled in the LPT section. First start attempt he had a puff of flame in the tailpipe; this panicked him and he chopped fuel and hit the starter at the same time.

Not too smart.

As the starter was winding up, N2 was unwinding, and when the starter clutch caught, it overstressed the "shear point" on the tower shaft. We had to cannibalize another engine to resume the test. :yuk:

[DevX]

Barit1, a 30 second dry crank followed by a normal start would have sufficed. There's no need to flush out inhibiting fluid as the engine will start and run quite happily on it. :ok:

[barit1]

Agreed,DevX. This crew panicked when they saw the flame. This engine design had a trap - a small pooling area in the LPT case that could hold a few ounces of fuel or preservative oil that could feed a tailpipe fire.

[TURIN]

Back to the OP.

Do any of you, er, more mature, ladies & gents remember the chap on the headset delivering a "positive N1" remark during start?

I still do it occassionally when starting engines in to a tail wind. The big fan engines will light off and N2/N3 rotors will happily accelerate while the N1 is still going the wrong way. CF6 seems to be the most common. Autostart doesn't seem to worry about negative N1, however, a smaller fan engine such as the V2500, may be more susceptible hence the FCOM requirement.

Oh, and by the way, the fan drives the rotor/LPT when windmilling in either direction.

As BOAC states, the old trick of deploying the reverser is to block the airflow through the cold stream duct therefore removing the driving force on the fan.

My company's policy for engineers is to max motor the engines during start due to an embarrasing incident on a 747 some years ago when 3 donks got trashed during engines tests.