8
$\begingroup$

According to maintenance manual, we usually carry out engine wash (CFM56) with pure water as scheduled task. Pure water is injected into the engine from the LP compressor inlet while the engine is cranked.

I wonder if the engine performance would improve significantly after having a wash. What is cleaned from the engine gas path? is it carbon residue or something else.

As it is said that the water should be injected at 1 o'clock position on the LPC (aft view), why does it need to be at that specific position? does it clean certain components at that position? Thanks for your explanation.

| improve this question | |
$\endgroup$
  • 1
    $\begingroup$ That reminds me my glider trainer that told me washing the wings helps not loosing one point of L/D ratio $\endgroup$ – Manu H Nov 27 '19 at 16:57
8
$\begingroup$

A performance increase after a compressor wash is expected to happen, but it does not return to the original performance as there are more degradation mechanisms at play. This answer explains that organic build-up causes a change in airfoil shape causing the recoverable performance degradation. Another source may be deposition of salts and dissolved constituents to deposit on the blades as a result of ingesting salt water from which the water evaporates in the hotter stages of the compressor. E.g. erosion is a non-recoverable degradation (for a compressor wash). This is clearly shown in the figure below.

enter image description here

From: Modern Gas Turbine Systems, High Efficiency, Low Emission, Fuel Flexible Power Generation, Woodhead Publishing Series in Energy

I wonder if the engine performance would improve significantly after having a wash.

The performance increase is considered significant enough, else we would not go through all the trouble of performing the procedure. Fuel costs is one of the major direct operating costs, reducing fuel costs by washing the engine doesn't only save money, it also reduces the amount of used fuel (nice green image for the airline).

As it is said that the water should be injected at 1 o'clock position on the LPC (aft view), why does it need to be at that specific position?

This could be done to prevent build-up of water in the lower section; now it has to travel half the engine down, by the time it reaches that location the water will be transported aft by the churning blades. The water is therefore better distributed, which is confirmed by this practical answer.

| improve this answer | |
$\endgroup$
  • $\begingroup$ Could you please include what exactly the performance factor is? I'm thinking that the engine thrust should still be somewhat similar if not identical even if the engines are dirty because of the engine control computers, or at least on engines that are EPR controlled. If the EPR and thus the thrust are not as they should be the FADEC would just inject more fuel. So engine thrust would be the same just at a cost of more fuel burn. Is this what the performance factor describes? $\endgroup$ – Jan Nov 30 '19 at 20:35
  • 1
    $\begingroup$ @Jan Frequently the exhaust gas temperature is a direct measure. Engines are EPR or N (fan spool speed) controlled. Indeed, fouling means that more fuel needs to be added, to maintain thrust, but that will make the engine deteriorate faster, so you want to clean it once in a while to prevent accelerated damage accumulation. $\endgroup$ – 0scar Nov 30 '19 at 23:42
5
$\begingroup$

The compressor blades accumulate a coating of organic material from the atmosphere on the blades, which in the later stages, where the temperatures get above the material's flash point, may be mostly carbon from cooked organic material like pollen and bugs (the same sort of blackened accumulated crud you'll see on the butterfly of a late stage compressor bleed shutoff valve when you remove it). It reduces the efficiency of the compressor since the blade contours are altered in subtle ways by the crud coating. The water bath is just to erode that stuff off. It's basically the same thing as cleaning bugs and dirt off wing leading edges.

Not sure about the reason for the clocking instruction. Either to clean something like a probe at that position, or conversely, (and more likely by my guess) to avoid excessive water impingement on a probe or sensor on the other side.

| improve this answer | |
$\endgroup$
1
$\begingroup$

I work within a engine test facility. During a compressor wash process we inject water and detergent. This also depends on how much work have been carried out during the overhaul on the engine. The positions are given only to provide an even flow of water through the engine. Sadly I have no experience with live aircraft, so I can not comment when their on wing.

| improve this answer | |
$\endgroup$
  • $\begingroup$ I wonder how long it takes to perform engine wash and what crank the engine in test facility. As we use pneumatic starter, which only allowed to run in few minutes, therefore, we cannot wash for too long. $\endgroup$ – dingdingdong Dec 10 '19 at 23:41
  • $\begingroup$ Hi. Same in our test facility, we turn the engine pneumatically so we still have to adhere to starter limits. Each wash cycle take 3-5 minute depending on engine types. $\endgroup$ – kyrun Harris Dec 11 '19 at 18:57
  • $\begingroup$ Thanks. I also want to know if only compressor would benefit from engine wash. How about combustion chamber and turbine? do we necessarily perform seperate turbine or combustion chamber wash? $\endgroup$ – dingdingdong Dec 12 '19 at 1:33
  • $\begingroup$ Yes it would benefits the combustion chamber and turbine, once the VBV’s are scheduled closed the water will penetrate the whole core. Core wash at the test facility take around 2 hours to complete. Do you schedule VBV’s when completing a core wash on the wing? How to you motor the engine? Thanks $\endgroup$ – kyrun Harris Dec 12 '19 at 5:49
  • 1
    $\begingroup$ Yes, we do close the VBV when washing engine. We use its starter which pneumatically powered by bleed air from Apu. It also takes around 2 hours to wash, however, mostly for preparation and close-up. The main process takes only 30mins each engine. Cheers $\endgroup$ – dingdingdong Dec 12 '19 at 13:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.