Shutting off the fuel in a Seneca I

[avbug]

Like I said, a lot have people have made the same mistake, some of them fooled by a similiar experiment to what you did. Complacency.

When debris moves to block something during the takeoff, or a partially opened valve causes the failure...you'll wind up in the same boat as a great many others who have made the same mistake.

Nobody ever believes until it happens to them. Too bad.

 

[Timebuilder]

When debris moves to block something during the takeoff, or a partially opened valve causes the failure...you'll wind up in the same boat as a great many others who have made the same mistake.

No mistake, my friend.

"Test, don't guess." It is an axiom that is tried and true.

If you are having a fuel problem, such as inadequate flow due to a partially blocked valve, you will find out about that current condition during the runup in this airplane.

If a blockage occurs after the runup, it happens long after the fuel selectors have been returned to the "on" position, and after the testing required by the checklist. Then, the strongest possiblity might be assymetric thrust on the takeoff roll. If the failure happens at rotation, I will abort the takeoff. It's all training, and the willingness to follow the training.

At that point, I have the same likelihood of having a failure as any other twin, and I have proven that the crossfeed is capable of operation if I need to use it following a real inflight engine failure. I would be in big trouble if I had not beeen in the habit of following the checklist testing of the valves, and at the time I needed to use them, found them failing to operate. I think that is a far bigger problem than having a failure during runup.

Apparently, so does Piper.

 

[avbug]

Piper places the information in the normal proceedure to cover themselves. There is no need, nor requirement, that it be done during taxi. A checklist is not a "do" list. It's nothing more than a list to ensure that you've covered everything. Great lattitude is possible in the method by which it's conducted.

"Test, don't guess." It is an axiom that is tried and true.

So what you're really saying, then, is that it's okay, because it hasn't happened to you. Yet.

It's happened to many others. Just not you. Yet.

Good logic.

Or perhaps you're just saying that because it didn't do it one time, it won't do it any other time. Also good logic. Similiar logic is found in dehorning all the cows each time the chicken coop burns down.

 

[Timebuilder]

Piper places the information in the normal proceedure to cover themselves. There is no need, nor requirement, that it be done during taxi. A checklist is not a "do" list. It's nothing more than a list to ensure that you've covered everything. Great lattitude is possible in the method by which it's conducted.

According to the checklist for this airplane, the valves are to be placed in the "x-feed" position during the taxi. It is very plain about it. The runup portion of the list makes equally plain that the valves are to be returned to "on" for the runup portion.

Every Fed and DPE I know, which is inclusive of the one's that flew with both myself and students in the airplane in question regard every checklist item as a "requirement." In fact, it was a fed that suggested the idea of returning to "on" before reaching the runup area in order to ensure proper operation of the fuel system.

So what you're really saying, then, is that it's okay, because it hasn't happened to you. Yet.

No, not at all. "Test, don't guess" is an axiom of mechanical repair. If you have tested a device or system, instead of guessing that the system will perform properly when called upon, you have a far better chance of having that system or device perform as intended when it is needed. For example, no one would fail to take a compression reading on a cylinder during an inspection for fear of disloging some dirt or grit during the removal or reinstallation process.

What gets me is you know this.


Logic? This is the logic of the preflight, the logic of the runup, the logic of the annual inspection, the basis of quality control theory, and the foundation of the self-checks that computers use before loading the operating system.

To not test these valves would be illogical.

Or perhaps you're just saying that because it didn't do it one time, it won't do it any other time. Also good logic. Similiar logic is found in dehorning all the cows each time the chicken coop burns down.

I'm saying that this check is a requirement of the operation of the aircraft, and that regular adherance to this procedure helps to keep the moving parts of these valves clean and lubricated with fuel, and decreases the chance of having a problem with these valves when needed to perform in flight.

Besides, we sold off most of the chickens and I burned down the coops personally, when I was 11.

It was on my father's checklist.

 

[avbug]

No, not at all. "Test, don't guess" is an axiom of mechanical repair.

As a working aircraft mechanic and inspector, I suppose I've missed that, over the years. Thanks for bringing it to my attention. Axiom, is it?

Placing the fuel valves in the crossfeed position during taxi was never on any checklist in the airplane when I flew the Seneca.

If you have tested a device or system, instead of guessing that the system will perform properly when called upon, you have a far better chance of having that system or device perform as intended when it is needed. For example, no one would fail to take a compression reading on a cylinder during an inspection for fear of disloging some dirt or grit during the removal or reinstallation process.

Really. Tell that to the crew of Alaska Flight 261. While there's no doubt as to the misuse of grease on the jackscrew that failed, there's also no doubt that overuse and overtesting lead to the failure. The same may be said for the failure of many components that have only so many cycles or repititions in them.

As for making a bizarre connection between moving a fuel selector prior to the most critical phase of flight (takeoff), and performing a compression test during a routine mechanical inspection, it's nonsensical and stupid. One can kill you, the other can save your life. Dislodging degris or releasing debris that has been stopped at the selector can do harm, whereas debris isn't an issue in any respect during a compression test. To say it's an apples and oranges comparison would be to make an understatement. Not even in the same ballpark.

Logic? This is the logic of the preflight, the logic of the runup, the logic of the annual inspection, the basis of quality control theory, and the foundation of the self-checks that computers use before loading the operating system.

Again, a nonsensical statement.

Moving selector valves, especially shutoff valves, or switching to a fresh tank before takeoff, is dangerous, stupid, and gets people killed every year. Even examiners. Being an examiner doesn't mean one's brilliant, intelligent, or even much of an authority. It means one can administer a practical test and determine that a student has met a minimum particular practical standard. End of story.

 

[great cornholio]

I know someone that almost died from their instructor shutting off the fuel valve on the seneca 1 in the pattern. Happened a few years ago at Flagger Co airport in FL. Just north of DAB. It was the Phil Air crash. Anyways the owner of the school was checking out 2 new MEIs so they could teach in the seneca. Sometime in the pattern Phil turned off the fuel selector and of course the engine failed. Somehow they got slow and they also failed the right engine with right hand traffic pattern. So they got low and slow and then tried to feather the prop before hitting power lines...but too late did a Vmc roll into the ground and caught on fire. The 2 guys up front werent too hurt although Phil almost lost a foot, the MEI in the back was burned up pretty bad and spent months in the hospital. I guess my point is get a new MEI man. I taught in the PA 34 for over 200 hours. I never shut the fuel selector off below 3000 feet. Throttle only till then. With throttle the airplane feels like its on one engine but youve got an out if you need it. And our checklist called for checking the fuel valve in cross feed for one min before we left the parking spot.

 

[Timebuilder]

I know someone that almost died from their instructor shutting off the fuel valve on the seneca 1 in the pattern.

That was a BIG mistake that never should have happened.

I never shut the fuel selector off below 3000 feet. Throttle only till then. With throttle the airplane feels like its on one engine but youve got an out if you need it. And our checklist called for checking the fuel valve in cross feed for one min before we left the parking spot.

Good. You can expect this regime to serve you well.

 

[Timebuilder]

As a working aircraft mechanic and inspector, I suppose I've missed that, over the years. Thanks for bringing it to my attention. Axiom, is it?

The vast majority of mechanics don't work on airplanes. Happy to share that axiom with you. I was bitten on the keister several times before I began to regard testing as an axiom.

Placing the fuel valves in the crossfeed position during taxi was never on any checklist in the airplane when I flew the Seneca.

Okay...

It was on both Seneca I checklists I used, in two examples of the airplane.

Apparently, it was on "great cornholio" 's checklist.

And this afternoon, I found it also on an aftermarket checklist for the same airplane in a pilot shop while purchasing some charts.

That's three unrelated instances of the same operation during the taxi phase for this airplane.

Really. Tell that to the crew of Alaska Flight 261. While there's no doubt as to the misuse of grease on the jackscrew that failed, there's also no doubt that overuse and overtesting lead to the failure. The same may be said for the failure of many components that have only so many cycles or repititions in them.

Apparently, this was a problem that was unrelated to "testing", and that wear resulted form use. Without causing massive thread drift, I think I can say that I know you (and respect you) enough to say that you might agree with me that anything a human designs might be subject to flaws. Maintenance procedures are often changed as a response for our inablility to design "perfect" machines, as was the case with the jack screws.

As for making a bizarre connection between moving a fuel selector prior to the most critical phase of flight (takeoff), and performing a compression test during a routine mechanical inspection, it's nonsensical and stupid. One can kill you, the other can save your life.

Apparently, those nonsensical guys at Piper see this in a different light. Checking compression on a cylinder is a good idea to allow you to have a reasonable expectation of service from an engine.

Similarly, the engineers who designed this aircraft want us to prevent problems by use, not over use in the checking and assurance of the function of the crossfeed fuel system before every flight. Their preflight routine makes the fuel valves just one more in an entire series of moving parts that must work in all phases of flight, critical and non critical.

Now here is an area where I bet we agree.

Say that a pilot never uses the crossfeed check during the taxi, and has never had to use it in flight. Not for training, not for a real engine failure. What will happen when someone DOES make the check? Maybe exactly what you are suggesting; a piece of packing or dirt might be dislodged and a possible engine failure could result. Maybe this would happen on takeoff, and maybe, just maybe, the valve would fail completely if asked to perform its crossfeed function in a real emergency.

Personally, I would rather be working with a device that has a history of operation, demonstrated repeatedly, than to rely on a device that I have not tested based on a fear of failure and against the recommendation of the manufacturer.

Dislodging degris or releasing debris that has been stopped at the selector can do harm, whereas debris isn't an issue in any respect during a compression test

You must work on some very clean engines. Not me.

Anytime you disturb a spark plug, things that have been outside of the combustion chamber gain an opportunity to enter. The reward of checking the compression and the plug electrodes, maybe even the opportunity to use a scope to have a look inside the cylinders far outweighs the risk of dirt or debris entering that cylinder. It's a good trade, one I'm sure you'd agree with.

Similarly, a valve which is operated regularly and within its tolerances and design limits, which isn't really in question with a hand operated fuel valve, it seems to me, has little chance of creating a problem from anything being dislodged. A valve which is ALWAYS in one position certainly can build up sediment or dirt in that position, and moving that valve could indeed cause that dirt to be dislodged.

Maybe that's why Piper wants that valve moved and checked for proper operation. At least, that's my impression. I'm sure not an A&P, and I don't pretend to have a working knowlege of the differences in design of magnetos made by Bedix or its competitors, or any of a hundred topics that I have no business knowing about. This topic, however seems easily within the realm of my experiences with engines and plumbing.

Remember the hazardous attitudes for pilots? Maybe this antidote is pertinent here: "follow the rules, they are usually right."

Logic? This is the logic of the preflight, the logic of the runup, the logic of the annual inspection, the basis of quality control theory, and the foundation of the self-checks that computers use before loading the operating system.

Again, a nonsensical statement.

No, my friend. An illustration of how devices require checks for correct operation. Piper has determined what they think is a prudent check for their equipment. I agree with their assesment.

Moving selector valves, especially shutoff valves, or switching to a fresh tank before takeoff, is dangerous, stupid, and gets people killed every year. Even examiners. Being an examiner doesn't mean one's brilliant, intelligent, or even much of an authority. It means one can administer a practical test and determine that a student has met a minimum particular practical standard. End of story.

I don't think anyone is nominating examiners for MENSA membership, or even talking about using a "fresh" tank.

Now I grant you, as above, that moving a valve that is never moved as a part of normal ops might be a mistake. This particular valve, the one we are discussing, gets moved before every flight, if done according to the checklist. I know of NO, not even one, instance of this check causing a problem in the Seneca.

Now, we can talk about the necesssity of switching fuel tanks in the normal operation of other Pipers, including all of the singles I've flown, and the main and aux tank valves in the Navajo, if you like. None of those planes have much utility at all if you can't move the fuel valve. In fact, not moving the fuel valve in the singles HAS killed pilots.

An interesting irony, isn't it? That would be the end of the story, would it not?

Follow your checklists kids, and do it every day.

 

[avbug]

The vast majority of mechanics don't work on airplanes.

The vast majority of mechanics don't work on airplanes? Do we work on submarines? Handguns? Reticulated Orgasmatrons? What?

As an aircraft mechanic, I work on aircraft. Go figure.

Apparently, this was a problem that was unrelated to "testing", and that wear resulted form use.

Actually, it really was related to testing and running checks. Had they put the airplane on the ground instead of screwing with it, they'd be alive, pure and simple. They tested it to death. Literally.

Maintenance procedures are often changed as a response for our inablility to design "perfect" machines, as was the case with the jack screws.

This was not the case with the jackscrew. There was nothing wrong with the maintenance program; the work card had been falsified (as is often the case, unfortunately), and the wrong grease had been used. If the work had been done properly, there would never have been an issue. The proceedure was fine. Company issues and personnel issues caused the failure.

Apparently, those nonsensical guys at Piper see this in a different light. Checking compression on a cylinder is a good idea to allow you to have a reasonable expectation of service from an engine.

Oh, so very wrong. No mechanic with any understanding or experience believes a compression test says anything about engine life, because it doesn't. General standards are provided stipulating that 75% of the input pressure should remain during a compression test, but that provides NO insight into the life of an engine. Oil analysis does, but compression testing does not; compression testing varies with engine temperature, with the user, and with the test set. Even varies with the same test set and user, and even with the same relative engine temperature. Laymen and the uninformed see it as some great panacea of engine information; the compression test means almost nothing.

Maybe exactly what you are suggesting; a piece of packing or dirt might be dislodged and a possible engine failure could result. Maybe this would happen on takeoff, and maybe, just maybe, the valve would fail completely if asked to perform its crossfeed function in a real emergency.

Absolutely not...not unless you attempt to take off with the crossfeed in use. Is there a place on your checklist where you are directed to do so, or is it prohibited? Case in point. You won't have an engine failure on takeoff related to moving the selector valve during taxi...unless you move the selector valve during taxi. Exactly my point; thanks for making it.

Personally, I would rather be working with a device that has a history of operation, demonstrated repeatedly, than to rely on a device that I have not tested based on a fear of failure and against the recommendation of the manufacturer.

Perhaps so. But with more experience, you'll learn that some components have limited life in them. Published, or not. They'll only work through so many cycles and tests. Running exhaustive tests may mean that you're going to have component failure in use. Think about it; it will either fail when you're testing it, or using it in practice. If you test it more than you ever use it, you're probably going to cause it to fail eventually during testing...but if not, then you've put yourself in the position of wearing it out in testing, and not having it when you really need it. Sometimes it's far better to just leave alone until you need it.

 

[Timebuilder]

As an aircraft mechanic, I work on aircraft. Go figure.

That's right, you do. Most mechanics are not aircraft mechanics. Most work on other things, but the commonalities are the basic physics and knowlege base of internal combustion engines. Mechanics are more alike than they are different, IMHO.

Go figure?

Actually, it really was related to testing and running checks

As you later said:

There was nothing wrong with the maintenance program; the work card had been falsified (as is often the case, unfortunately), and the wrong grease had been used. If the work had been done properly, there would never have been an issue. The proceedure was fine. Company issues and personnel issues caused the failure.

The company "issue" was the use of a "procedure" of falsification. Of course, falsification was not the "official" procedure, but it was the one being used, apparently. The de facto procedure, so to speak.

Other maintenance procedures which are modified on aircraft in general happen because we are fallible humans, and we are unable to make a "perfect" device, or even to service that device perfectly, 100% of the time. All of the information on the CD-roms that mechanics use point to this design ability deficiency on the part of humans, the thousands of AD's, the modified procedures, the limited service life values. Think Parker Hannifin. They get a lot of blame when pilots can't fly partial panel, or haven't a good standby electric gyro on board. We are fallible humans, always looking to blame someone else for our faults.

So, we have a jackscrew that was not even being maintained to accepted standards, nor was the approved lubricant being used. If your argument is that the valve could be worn out by this moving and checking, as was the improperly maintianed jackscrew, that might be possible in a very extreme case. I'd imagine most cases are far from extreme. I'd move that valve rather gingerly, actually, because it was expected to function on almost every training flight. It had done so (functioned well) for several years, and since it is a checklist item, I expect it to do so for years to come. The mechanics who work on the aircraft are generally pretty sharp, and they no doubt check that valve themselves.

No mechanic with any understanding or experience believes a compression test says anything about engine life, because it doesn't.

It does not say EVERYTHING about engine life, but it does speak to the condition of the engine, which is a large component of how we might judge the egine's "life." You cannot return an engine to service if the compression is not up to snuff. How many cylinders and and other parts such as valves and rings, are serviced or replaced every day due to loss of proper compression? Hundreds? Thousands? Why? Because compression is an essential element of an engine in good condition. A condition that lends itself to service life. If wear or damage makes compression too low, you have effectively run out of service life. If it is getting lower over time, you can see a trend that means you will soon be out of service life.

General standards are provided stipulating that 75% of the input pressure should remain during a compression test, but that provides NO insight into the life of an engine.

If the engine compression does not meet the standard, can it be considered for continued service, ie: life? No, it cannot. It is then deficient, and cannot be used in that condition. A part that cannot be used can certainly be considered "dead," or at the end of its life. It can resurrected, yes, but until then, it's over. The Easter theme is not intentional, for those about to accuse me of thread drift.

Laymen and the uninformed see it as some great panacea of engine information; the compression test means almost nothing.

A "panacea?" I don't know ANY mechanics that see it that way. In fact, even when I had tens of thousands of dollars worth of specialized electronic equipment at my disposal, I never though of anything, not any tool, as a panacea. Okay, you were being scarcastic. That's cool.

At it's most simplistic, you can count on compression decreasing as the service life is spanned, and it is a reliable tool whether you use a leak down set or a classic compression gauge. When troubleshooting an engine performance problem, for example, a reasonable level of compression is considered fundamental by all manufacturers of engines. The purpose of the engine is to provide power, whether you are talking about a G-V engine or a Briggs and Stratton atop your backyard lawnmower. Without compression, you have no ability to produce power using current technology.

Oil analysis is a wonderful tool for judging engine wear, including bearings, rings, almost anything that rubs against something else with oil. In liquid cooled engines, the oil lab routinely looks for evidence of coolant. At one point, I had a fleet to maintain from Ford Tempos to LN9000 dumps and IH TT's, and every kind of implement and bulldozer, loader, and backhoe in between. Oil analysis was a teriific tool. It no doubt saved us a lot of time and trouble, to say nothing of tens of thousands of dollars.

Absolutely not...not unless you attempt to take off with the crossfeed in use. Is there a place on your checklist where you are directed to do so, or is it prohibited? Case in point. You won't have an engine failure on takeoff related to moving the selector valve during taxi...unless you move the selector valve during taxi. Exactly my point; thanks for making it.

I read this a couple of times and I'm not sure of what you might have intended to say. I can observe that the valve being returned to "on" before even beginning the runup IS a checklist item, and that all of the items are intended to be performed as required.

Are you saying that following the checklist might lead to an inadvertent takeoff in crossfeed? Perhaps, but not if you are exercising due dilligence.

But with more experience, you'll learn that some components have limited life in them. Published, or not. They'll only work through so many cycles and tests. Running exhaustive tests may mean that you're going to have component failure in use. Think about it; it will either fail when you're testing it, or using it in practice. If you test it more than you ever use it, you're probably going to cause it to fail eventually during testing...but if not, then you've put yourself in the position of wearing it out in testing, and not having it when you really need it. Sometimes it's far better to just leave alone until you need it.

Good. I had no problem at all with that.

In fact, I agree with your principle. There is, however, a corollary. That is that something which is not checked or exercised, like a muscle or joint, or a valve, cannot be considered to provide reliable service when called to operate. You're a jumper, aren't you? Isn't there some sort of limit date on a packed chute, beyond which it must be opened and repacked before it can be used? It seems I read someone mentioning this here at some point. One day I'll take the jump course and go up. Some of the posters here no doubt would rather I skipped the use of the parachute. In November, they may volunteer to jump without one.

At any rate, I have to have some level of faith in the record of use of this valve in the many Senecas still in service, along with the example of Piper being able to design and manufacture a reliable valve in the singles (not that electric thing that fails in the Aerostar) as a basis for following the checklist and an expectation of a reasonable number of life cycles for the valve. If this moving during taxi has caused these valves to fail in the past few decades, I haven't seen a report that suggests that is the case, or an AD on the valve that suggests that Piper has found this valve to hold a design defect.

Nothing is perfect, except for, well, you know.