PPRuNe Forums

PPRuNe Forums (https://www.pprune.org/)
-   Rumours & News (https://www.pprune.org/rumours-news-13/)
-   -   A321 AAIB Report: Mayday/emergency landing due to fuel additive error (https://www.pprune.org/rumours-news/631797-a321-aaib-report-mayday-emergency-landing-due-fuel-additive-error.html)

TelsBoy 24th Apr 2020 09:54


Originally Posted by WHBM (Post 10761394)
I do agree that stipulating a heavy engineering operational instruction in ppm is almost deliberately making things difficult. But we can extend our discussion of "cheap" to the maintenance manual production, where once upon a time a professional in house technical documentation author would never have simplistically written it like that. Instead outsource the production, again to the low bidder, and they will just whack in the product manufacturer's spec sheet without consideration.

I worked in a previous life for an avionics manufacturer that used to have professional spec writers, employed solely to write test procedures, manuals etc. In other words, they'd take what the boffins and bods would give them and translate it into something legible for those of us of a more... average... intelligence. That's long gone and when I was there the production engineers themselves had to write the specs, procedures and manuals. I think its the same across all industries.

Uplinker 24th Apr 2020 10:53

Without wanting to blame the hapless engineer, Googling something and then using a random calculator found somewhere on the internet, but without fully understanding the terminology...........Why did he not have access to a proper procedure and a proper, validated calculator?

Engine critical procedures, such as opening the cowl flap locks, the engine oil caps etc., have mandated dual inspections. Why not for fuel additives? Fuel is pretty critical !

Engineers attending aircraft I have flown, have print-outs of the procedures, and often phone another authority if they are not familiar with a particular procedure, to check. And that is fine and sensible.

Why this guy did not phone his boss or another engineer to check his figures is beyond me.


Sallyann1234 24th Apr 2020 11:04

Didn't want to admit/display his ignorance ?

esscee 24th Apr 2020 11:24

But who was the supervisor that oversigned/stamped the task? If he did the task and able to countersign/stamp it then that is even worse, in that someone with that authority did not know/understand the task details and what he was meant to do! For heaven's sake who was the MRO management/QA people allowing such a person to be working on aircraft? Absolutely frightening for the paying customers and equally for the crew flying the aircraft!

TelsBoy 24th Apr 2020 11:35


Originally Posted by esscee (Post 10761505)
But who was the supervisor that oversigned/stamped the task? If he did the task and able to countersign/stamp it then that is even worse, in that someone with that authority did not know/understand the task details and what he was meant to do! For heaven's sake who was the MRO management/QA people allowing such a person to be working on aircraft? Absolutely frightening for the paying customers and equally for the crew flying the aircraft!

One of my key points. Who's checking the work, ongoing staff competency/recency, training, procedures etc. If the engineer didn't feel they could ask a question/confirm something, why? Wasn't there a supervisor or colleague available, if not why not, if so then why couldn't they ask them? Was the engineer pressured into finishing a job at all costs rather than step back and leave it if unsure? Is there an unsafe culture where staff cannot ask questions freely without reproach or judgment? Are unsafe or incorrect practices challenged? Sounds like some simply want to lay the blame on the engineer rather than deal with the much wider and more pressing issues.

lomapaseo 24th Apr 2020 12:17

In order to keep a balance in the discussion we should also look beyond the blame of a person and also look at safety nets beyond the initial mistake. Perfection is not possible, we also depend on mitigation or shielding (redundancy, detection etc.)

pilotmike 24th Apr 2020 12:53


Originally Posted by Fursty Ferret (Post 10761355)
Even the very best people make simple mistakes when dealing with parts per million (because it adds an extra step to a calculation)

Really? How does 'per million' (divide by 1e6) "add an extra step to a calculation", where 'per cent' (divide by 1e2) does not?

Indeed, in such a high dilution ratio as this (1: 0.0001, or 0.001% or 100PPM), it is arguable that PPM is the clearest way to express it, as it uses only whole numbers. Also, being a multiple of 1e3 (1,000), it works very well with standard x1,000 multiples {eg. (k)g, (m)l etc.}, where % is x100, which prevents such easy multiples and the relatively simple use of k(ilo) m(illi) prefixes.

Both PPM and % are valid ways for engineers / scientists to express dilution ratios. Neither should be more or less likely to be prone to error, and certainly not by "add[ing] an extra step to a calculation", as you claim. Competent engineers should be expected to be able to work accurately with either.

cashash 24th Apr 2020 13:14


Originally Posted by esscee (Post 10761505)
But who was the supervisor that oversigned/stamped the task? If he did the task and able to countersign/stamp it then that is even worse, in that someone with that authority did not know/understand the task details and what he was meant to do! For heaven's sake who was the MRO management/QA people allowing such a person to be working on aircraft? Absolutely frightening for the paying customers and equally for the crew flying the aircraft!

I would assume that the certified the task himself if he was a Licensed Engineer. With regards to the screw up over the calculation, as the MRO seems to be based in Cyprus perhaps English was not his first language and the instructions in the AMM were not totally understood - it is a situation that I have seen numerous times overseas especially with Airbus manuals that seems to be written by a committee when compared to those from Boeing.




Webby737 24th Apr 2020 14:11


Originally Posted by cashash (Post 10761640)
I would assume that the certified the task himself if he was a Licensed Engineer. With regards to the screw up over the calculation, as the MRO seems to be based in Cyprus perhaps English was not his first language and the instructions in the AMM were not totally understood - it is a situation that I have seen numerous times overseas especially with Airbus manuals that seems to be written by a committee when compared to those from Boeing.

I agree with this, even for someone like me whose first language is English the Airbus manuals can sometimes be hard work.

One small example of this, if you look in one of the uncustomised manuals (The SRM for example) Airbus will list each and every MSN, one list might be the A319, another the A320 etc, etc. Why couldn't they just state A319, A320 or Effectivity ALL ?
The effectivity can be filtered in Airnav or AirnavX but it doesn't always work.

Boeing make your life easier with "Effectivity ALL" or B737-700 etc.

Chu Chu 24th Apr 2020 22:39

Since the concentration was specified in PPM by volume, but both the fuel and the product are normally measured in KG (and have different densities), there were actually 4 calculations required:

1. Convert the 6200 KG of fuel to 7678 L.
2. Convert 100 PPM to .0001.
3. Multiply 7678 L of fuel by 0001 to get .7278 L of product.
4. Convert .7678 L of product to .799 KG.

And that's after looking up the specific gravities of both the fuel and the product.

If the procedure had said that 1 KG of product treated 7760 KG of fuel, the calculation would have looked like this:

1. 6200 KG divided by 7760 equals 0.799 KG.

Ex Cargo Clown 25th Apr 2020 08:20


Originally Posted by pilotmike (Post 10761614)
Really? How does 'per million' (divide by 1e6) "add an extra step to a calculation", where 'per cent' (divide by 1e2) does not?

Indeed, in such a high dilution ratio as this (1: 0.0001, or 0.001% or 100PPM), it is arguable that PPM is the clearest way to express it, as it uses only whole numbers. Also, being a multiple of 1e3 (1,000), it works very well with standard x1,000 multiples {eg. (k)g, (m)l etc.}, where % is x100, which prevents such easy multiples and the relatively simple use of k(ilo) m(illi) prefixes.

Both PPM and % are valid ways for engineers / scientists to express dilution ratios. Neither should be more or less likely to be prone to error, and certainly not by "add[ing] an extra step to a calculation", as you claim. Competent engineers should be expected to be able to work accurately with either.

Scientists don't really use PPM because it's far too ambiguous. We generally use molar concentrations for liquids. Makes it far easier.

pilotmike 25th Apr 2020 09:48


Originally Posted by Ex Cargo Clown (Post 10762459)
Scientists don't really use PPM because it's far too ambiguous. We generally use molar concentrations for liquids. Makes it far easier.

Do you really believe this engineer, who royally screwed up the mixing, using 40x the correct amount, would have coped with the molecular concentration calculations?

Having demonstrated themseIf to be utterly defeated by 100ppm, it is unreasonable to believe they'd have aced molecular concentrations, never mind finding it "far easier" as you claim.

As the manufacturer knows the density of their own product, and using as standard fuel density, Chu Chu's suggestion of the manufacturer specifying 1kg of product to 7,760kg of fuel is easily the best suggestion, as that is the very simplest calculation by a long way.

Inside every vexing problem is a very simple solution (ha!) struggling to be seen!

IRRenewal 25th Apr 2020 10:20


Originally Posted by Chu Chu (Post 10762182)
3. Multiply 7678 L of fuel by 0001 to get .7278 L of product.

If ever we needed proof how easily these things can go wrong.

Pugilistic Animus 25th Apr 2020 10:23


Originally Posted by ivor toolbox (Post 10759529)
Er, the AAIB is UK equivalent of US NTSB,
the answer is yes to both your points, you might want to go back to the spotters balcony now and leave further discussions to grown ups

That's the funniest statement I've heard here in a long time...to save yourself further embarrassment try reading a few older posts of someone before deriding them. I learn a lot from him reading his post

I'm just saying

old,not bold 25th Apr 2020 10:23

I'm struggling to understand why the manufacturer suspended the use of Kathlon, especially when doing that leaves EU operators with no approved biocide treatment.

It wasn't the product that caused the problem, it was gross misuse of it.

Failing a technical reason, perhaps the manufacturer's insurer is behind it?

On another topic, I hope that the "on-going AAIB investigation" includes a systematic MEDA (or equivalent) analysis of the Human Factors involved, for the industry's benefit as well as the AMO concerned.

EDLB 25th Apr 2020 10:30

Easiest would be if beside the 1kg treats 7.760kg of Jet fuel, a table for typical even masses of fuel is given. Like 129g for 1000kg 1.288kg for 10.000kg etc. At least that allows to check, that your calculated number is in the ballpark. On that product I doubt, that +/- 20% error would matter much, but you should not get it wrong by orders of magnitude.

As pilot you often use rule of thumb calculations for checking if the number given make any sense.

pilotmike 25th Apr 2020 11:52


Originally Posted by EDLB (Post 10762568)
Easiest would be if beside the 1kg treats 7.760kg of Jet fuel, a table for typical even masses of fuel is given.... you should not get it wrong by orders of magnitude.
1kg doesn't treat 7.760kg of fuel.

As pilot you often use rule of thumb calculations for checking if the number given make any sense.

"You should NOT get it wrong by orders of magnitude." - in your very own words!

You should be "checking if the number given make any sense" - in your very own words!

slfie 25th Apr 2020 13:22


Originally Posted by pilotmike (Post 10762629)
1kg does NOT treat 7.760kg of Jet fuel!!!
...

That's possibly because of the continental way of using a decimal point in place of a comma, i.e. UK (and US) would write 7,760Kg but some european countries would write it as 7.760Kg (so there's a potential error of 3 orders of magnitude there). Certainly in the post it's confusing to my (UK) eyes, saying 1000Kg then 10.000Kg for 10 times as much.

Irrespective of the original dosing, the more surprising (and dangerous) thing is it didn't get picked up before it caused loss of thrust at take-off. Mistakes happen, but it does seem this one should have been caught long before it got as far as it did (the potential being there for a considerably worse outcome).

Momoe 25th Apr 2020 14:39

Over-reacting
 
Before anyone else criticises ground maintenance and throws in the "Pay peanuts, get monkeys adage", how many folk at the pointy end have royally screwed weight calculations and had some interesting flex take offs, or failed to configure correctly for take-off.

One common denominator here and that is we're all human.

Salusa 25th Apr 2020 14:44


Originally Posted by Webby737 (Post 10761695)
I agree with this, even for someone like me whose first language is English the Airbus manuals can sometimes be hard work.

One small example of this, if you look in one of the uncustomised manuals (The SRM for example) Airbus will list each and every MSN, one list might be the A319, another the A320 etc, etc. Why couldn't they just state A319, A320 or Effectivity ALL ?
The effectivity can be filtered in Airnav or AirnavX but it doesn't always work.

Boeing make your life easier with "Effectivity ALL" or B737-700 etc.

Agreed 100%

I'm B1. 3 (Turbine Helicopters) and the difference between Airbus/Leonardo vs Bell/Sikorsky is a world apart.

Euro manuals are dumbed down to the point of the lowest common denominator, but somehow manage to complicate it even further!

​​No schematics or system explanation in Euro. Just an idiots guide which you are expected to follow verbatim.

Literally two pages of instructions to remove a Filter with illogical cross references and tasks, rather than "remove the filter".

Now whilst not relevant to this particular incident its a pointer of what is wrong these days.

The qualifications required to obtain a LAME has been dumbed down and now we are seeing the results. EASA standards are not universal across EASA.



​​​​​






​​

Big Pistons Forever 25th Apr 2020 16:49

Nobody has talked about the role of the regulator here. The engineer held a license issued by an aviation authority with skills and knowledge requirements defined by them, arguably significantly dumbed down from what used to be required. The manual he was using and which you could make an argument was not fit for purpose, was approved by the authority.

industry on both sides of the Atlantic have actively pushed to marginalize and neuter regulators. This works great and saves manufacturers and air operators tons of money until the airliner dives itself into the ground or the whole rotor comes off the transport category helicopter at 2000 ft AGL.......

Kit Sanbumps KG 25th Apr 2020 19:29


Originally Posted by Big Pistons Forever (Post 10762893)
Nobody has talked about the role of the regulator here

It was a couple of days ago...


Perhaps ‘execs’, like the travelling public, believe that all the massively costly and inconvenient ‘regulation’ they are subjected to, should mean that the ‘cheap’ ones are also ‘good’ enough.

winglit 26th Apr 2020 06:35

I agree. The amount of fuel in the aircraft is measured in weight, but the amount of additive to be added is in volume.

The AMM writers were a bit sloppy with that procedure. I haven't checked the Airnav glossary recently, but I wonder if PPM is actually there. And in any case, how hard would it have been to clarify and write Parts Per Million?

Airbus manuals are usually full of charts and tables. Why didn't they add one detailing the amount of additive according to fuel tanks contents.

WHBM 26th Apr 2020 10:20


Originally Posted by Webby737 (Post 10761695)
I agree with this, even for someone like me whose first language is English the Airbus manuals can sometimes be hard work.

Bear in mind that at Airbus the Technical Author is unlikely to have English as their first language, as well.

esscee 26th Apr 2020 10:39

So who proves reads the manuals after they are initially written, anyone close to hand, office cleaner? For heaven's sake, to say that a Tech author is unlikely to have English as their first language is ridiculous, employ one who does have English as their first language. Simples!

Webby737 26th Apr 2020 10:46


Originally Posted by WHBM (Post 10763690)
Bear in mind that at Airbus the Technical Author is unlikely to have English as their first language, as well.

Quite true, French is my second language and often it helps when trying to decipher what is required by the AIB manuals.
I'll also add that many of the employees, including I suspect the ones that write the manuals have never worked in aircraft maintenance. A couple of years ago during an Airbus audit I was asked a question, my reply was "we use the SRM"
The next question was "what's an SRM?"

Webby737 26th Apr 2020 10:49


Originally Posted by esscee (Post 10763717)
So who proves reads the manuals after they are initially written, anyone close to hand, office cleaner? For heaven's sake, to say that a Tech author is unlikely to have English as their first language is ridiculous, employ one who does have English as their first language. Simples!

I think they are proof read by someone who might have a good grasp of English but not necessarily technical English.
Still, at least they are better than the Sukhoi manuals, I think they used Google Translate on those !

Webby737 26th Apr 2020 10:51


Originally Posted by winglit (Post 10763475)
The AMM writers were a bit sloppy with that procedure. I haven't checked the Airnav glossary recently, but I wonder if PPM is actually there. And in any case, how hard would it have been to clarify and write Parts Per Million?

Airbus manuals are usually full of charts and tables. Why didn't they add one detailing the amount of additive according to fuel tanks contents.

It was noted somewhere in the report that PPM was not in the Airnav glossary.

pilotmike 26th Apr 2020 13:01


Originally Posted by esscee (Post 10763717)
So who proves reads the manuals after they are initially written, anyone close to hand, office cleaner?!

Who proofread that?

cashash 26th Apr 2020 14:30


Originally Posted by Webby737 (Post 10763733)
It was noted somewhere in the report that PPM was not in the Airnav glossary.

It is in my pdf copy of the MM. I dont now have access to Airnav so cant check that.

lomapaseo 26th Apr 2020 14:35

Airbus is kind of a large company with a very broad employee base in their engineering departments. I wouldn't even consider a shortfall in writing skills towards the benefit of their customer base. What to communicate, how and where.may be argued here.

woptb 26th Apr 2020 14:39


Originally Posted by Momoe (Post 10762791)
Before anyone else criticises ground maintenance and throws in the "Pay peanuts, get monkeys adage", how many folk at the pointy end have royally screwed weight calculations and had some interesting flex take offs, or failed to configure correctly for take-off.

One common denominator here and that is we're all human.

A lot of perfect humans on here!
Passing comment is fine,but passing judgement, without being in possession of all the facts is rather dumb!

Everyone has screwed up royally in the work place,if you never have,it’s just a matter of time! To hear some of the posters pontificating on this is rather irksome.When you screwed up or when you screw up,it doesn’t necessarily make you an idiot,just human! Next you’ll all be telling me you’ve always follow procedures............

Webby737 26th Apr 2020 15:44


Originally Posted by cashash (Post 10763921)
It is in my pdf copy of the MM. I dont now have access to Airnav so cant check that.

I can't check either, I don't have access to Airnav from home but it looks like it may have been revised, here's an extract from page 10 of the report."In addition, the AMM task instructions used the term ‘ppm’ for which there was no definition within the AMM glossary, and no additional guidance was provided of how to perform the biocide fuel dosing calculation."

Webby737 26th Apr 2020 15:46


Originally Posted by woptb (Post 10763926)
A lot of perfect humans on here!
Passing comment is fine,but passing judgement, without being in possession of all the facts is rather dumb!

Everyone has screwed up royally in the work place,if you never have,it’s just a matter of time! To hear some of the posters pontificating on this is rather irksome.When you screwed up or when you screw up,it doesn’t necessarily make you an idiot,just human! Next you’ll all be telling me you’ve always follow procedures............

Quite true,
The only way to never f*ck up is to do nothing !

Winemaker 26th Apr 2020 20:07


Originally Posted by DaveReidUK (Post 10760729)
Or, millilitres per litre - which comes to the same thing, but may be a bit more intuitive. Those are the units quoted in the FAA bulletin.

Yes, the way I calculate additives in the winery is 1 ppm = 1 mg/L. This makes it very easy to do rough calculations on the fly and is a good check before dumping stuff in.

FlightDetent 27th Apr 2020 14:57

While I can attest that operating a contemporary jet is not an exact science, certainly there is hope that the engineering would be. PPM stands for 1 x 10^(-6), no units given - it's a simple multiplier.

1 ppm = 1 mg/L
That does not taste right. Perhaps mg/kg or ml/m^3, anyone?

A case of a point for the voices saying ppm is hard to grasp on the line:
  • density of jet fuel 0,785
  • density of Kathon 1,05
Which are the actual correct units to be used for diluting the additive according to AMM and the spec-sheet? Molecular, mass, or volume?

DaveReidUK 27th Apr 2020 15:46


Originally Posted by FlightDetent (Post 10765050)
Perhaps mg/kg or ml/m^3, anyone?

Are you suggesting that fuelling in cubic metres could catch on ?

FlightDetent 27th Apr 2020 20:20


Originally Posted by DaveReidUK (Post 10765085)
Are you suggesting that fuelling in cubic metres could catch on ?

Ok, call it kilo-liters if m3 sounds overly continental for the islands. 8-)

Two morals of the story:
  • mg (mass) / l (volume) is perhaps fine for a winery (density 1,005 - marginal error). For jet A1 (0,785) and Kathon (1,05) mixing volumetric and mass units gives you a nice compound error 26% even if you get the multipliers right and/or avoid the trap of mis-reading a comma in-lieu of a dot for the decimal divider.
  • If there's ml one side, for PPM to work instinctively i.e. avoiding the order-of-magnitude error, you need to have kilo-liters (m^3) on the other. Well pointed that's not the case, hence the argument of PPM being a stupid choice of a field-deployable unit.

Question stands. Which units - weight or volume - are used in the original Kathon recipe of the AMM?


cashash 27th Apr 2020 21:01


Originally Posted by FlightDetent (Post 10765276)
Question stands. Which units - weight or volume - are used in the original Kathon recipe of the AMM?


The answer is 'volume' - I did notice though that if you are using Biobor then the units are 'weight'

https://cimg1.ibsrv.net/gimg/pprune....53c7fe880e.png


cashash 27th Apr 2020 21:05

As an aside if you use a metered injection rig to carry out the addition of the Kathon, the AMM provides a handy calculation chart to get the correct amount.



https://cimg0.ibsrv.net/gimg/pprune....0e6a030d52.png


All times are GMT. The time now is 02:52.


Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.