Question for the engine guys
 

What causes that low frequency sound (~0:45 into the video) as the engines start? Seems like it correlates with fuel introduction from the puff of smoke, but then it disappears. Only ever noticed it on the 777

I believe it to be the ‘light off’ or the noise of the flame front in the combustion chamber.
Some airframes more pronounced then on others.

It also seems to be more prevalent on the very high bypass engines, like the Pratt's on the 32X neos. I wonder if it's not related to the higher pressure combustion chambers on the newer more efficient engine designs?

I remember the distinctive sound of the light off on L1011’s....it was really like a whooshing sound.
Also on the 738-300’s I used to fly.
Im sure it’s related to particular features of an engine design.
Maybe atmospheric conditions are a factor also.

Interstate bleed valves and / or variable IGV modulating to control rate of engine acceleration.

Ttfn

It's combustor growl, or instability of the flame in the combustor producing pressure waves in the acoustic range.

‘Light-off’. Where did that term ever come from; is it an American term for light-up?

What Dave said. Flame instability produces a low rumbling sound.
Most high bypass engines have some level of combustor growl during start - although the -115B is quite pronounced and the CFM on the 737NG is rather pronounced.
There have been cases where the acoustic rumble was so strong that it could case damage - forcing a redesign to reduce the instability and resultant rumble.

Back when we were doing initial engine runs on the GE90-94B (~1995), engine start was really interesting to listen to. The GE90 has a 'dual annular combustor' (DAC) and during start they'd switch between the two rings of combustors - the resultant frequency change actually sounded like a car shifting gears...

Thanks guys. Now for my second question. What exactly is an unstable flame?

Have you ever used a propane torch? When you light a propane torch (or something similar) - if you have the gas turned up a little to high, it will light but won't be stable. If it's too high the flame will blow out again, but if it's just a bit high the flame will remain but move around and not attach to the nozzle base, and will make a bit of a popping sound as it moves around. You fix it by turning down the gas a bit to allow the flame to attach to the nozzle, after which you can turn the gas back up to get a nice, strong flame to do whatever you want to do with the torch.

Getting the latest low emission combusters to light is surprisingly difficult. We spent a bunch of time working on improving the light off characteristics of the GEnx - both for 'normal' ground starts and in-flight starting.

Do an internet search on "turbine engine combustor instability" and you will see pages ranging from layman's simple explanations to PhD expert presentations with animations.

I would classify it as one that continues burning outside the combustor

It’s also called a Take-Off isn’t it?
:E

Unstable combustion.

Ah. Makes sense. Familiar with that happening in a Bunsen burner. Don't recall it popping, but it's been a few years. As usual, you've prompted another question- why are the latest combusters difficult to light off (or light up)? Or is that something I'll find out about when I look up "turbine engine combustor instability"? Thanks Dave Therhino!

It's amazing how different the knowledge and skillset is between designing, and operating airplanes.

That's a tougher question. Burner and fuel nozzle design is a bit of a black art to begin with, so I suspect that the compromises that low emission requirements drive in the design are hurting the light-off characteristics. The Dual Annular Combustor design in the GE90 was driven by emissions requirements, and the GEnx uses a very elaborate staged fuel nozzle setup that mimics the low emissions of a DAC without the drawbacks of an actual DAC. But why that makes them so hard to light I don't know. What is clear however is that, compared to the late 1980's designed engines such as the CF6-80C2 and the PW4000/94", the latest generation of high bypass engines are much harder to lite - with corresponding reductions in the in-flight start envelopes. Part of it is due to the increasingly high bypass ratios - but not all.

Well if there’s something about an engine that you aren’t sure of, then I certainly don’t need to know it. Thanks as always for the insight.

Part of it is achieving a level of fuel atomization that can be ignited under cold conditions at a low, initial light up/off fuel flow rate with a nozzle system that has to achieve adequate atomization over a fuel flow range that varies over about a 50 to 1 ratio. For example, I remember one engine that had a light off fuel flow rate in the 700 lbs/hr range and takeoff flow rate in the 30,000 lbs/hour range. Even with a two stage nozzle system that is a difficult combustor and nozzle system design problem.

Another major factor is trying to balance what is called igniter immersion to be able to achieve light off in cold-soak starting conditions yet also to not burn up the igniter tip when operating at high power. Testing is performed to determine the positioning of the igniter tip relative to the spray pattern at lightup to just be able to ignite the edge of the spray pattern. If you just stick the igniter in farther to achieve easier light off, the igniter life may be measured in dozens of flight hours rather than hundreds or a few thousands of hours.

Regarding the term "light off," I have always worked in the Boeing Seattle Area world and did a lot of engine relight related work, and that is the term I always heard since the 1980s. It was probably a Boeing-ism. The Rolls-Royce engineers used different terms, but we tolerated each others' terminology differences (just kidding - I remember a bunch of great engineers from back in the 1980s from RR - unlike the the pilots, they were mostly named Ian rather than Nigel ;-). They said light-up and we said light-off. They said auto-relight (which eventually became standard) and we said autoignition, etc.)

Would DAC mean the dual concentric rings of (for want of a better word) nozzles or flameholders shown in this X-Section? Which seems to create an - interesting - pattern of pressures inside the chamber.

Two sets of small torches instead of single larger ones, which may be where the lower emissions come from?

https://www.researchgate.net/figure/...fig6_269050769

Yes - two concentric rings of fuel nozzles. I believe the GE90 was the first engine to use the DAC. They had a lot of trouble making it work right - one particularly troublesome aspect that was - during an accel they'd transition from one ring to both - and there would be a discrete lag before the flame spread to the second ring resulting in a weird discontinuity in the accel characteristics. I left the program before they got it sorted so I don't know how they fixed it (my sneaking suspicion is that they started using both rings at idle in-flight, but I don't really know).
The DAC was an option on the CFM56-7/737NG - it reduced certain emissions but it resulted in a significant increase in fuel burn so very few were actually delivered.