EGT/CHT vs A/F sensor for Engine Leaning

[normancanderson]

So here is an excellent example of a simple question getting completely off topic - The intent of the original query was to determine if the modern A/F ratio meters provided any benefit for use in aviation applications. There was never any intent to imply that developing an automated mixture control system was viable for most homebuilders.
WLIU makes a good point about air cooled engines not running at stoichiometric (14.7:1) ratios. But the ironic thing is that I found most of the useful discussion on the application of the A/F ratio meter on the aircooled VW racing engine forums. They agree - the 'best' ratio for engine performance AND lifespan is around 12:1 (a.k.a. - 'rich of peak'); racers have never generally been bothered about fuel economy. I was hoping to get some useful feedback on the relative merits of using only EGT for mixture control vs direct A/F monitoring or a combination of the two. It seems this simple question has hit a sensitive point for many contributing to this forum - but that off-topic discussion has served to distract from the original intent.

[JCosh]

I currently have about seventy hours on my GlaStar since converting it to EFI via a kit sold by Flyefii.com I have a Lyco 360 in normal cruise CHTs are 300-350, EGTs are 1200-1250 and the AF is approx 13.5 -1. I like the EFI since I previosly flew behind a Subaru, and was not interested in going back to changing the mixture for altitude/temperature changes. Also wanted an engine that would restart instantly after a hot shutdown, which I'm sure you know is not the forte of an IO 360. Just my two cents.
Juliette

[normancanderson]

To expand on JCosh comments - there has existed EFI for aircooled automotive engines for over two decades - with no burned pistons, scoured cylinder walls, warped heads, etc.. Since my application (a modified version of an aircooled VW engine) currently uses only non-mixture controllable carburetors, it just seems like either a mixture controllable carb (or better yet, EFI), supported by an A/F mixture sensor would provide much more accurate control over the engine than just EGT peaking. Faster response times, too. I can see the value of also having EGT and CHT to make sure that the mixture setting isn't doing any harm, but those sensors then become more like oil temp or fuel pressure - something to warn the operator of problems, not something you use to prevent them. Or am I missing something?

[Mike M]

So here is an excellent example of a simple question getting completely off topic - The intent of the original query was to determine if the modern A/F ratio meters provided any benefit for use in aviation applications. There was never any intent to imply that developing an automated mixture control system was viable for most homebuilders....that off-topic discussion has served to distract from the original intent.

KA-CHING!!

i installed a Bosch O2 sensor for an '86 Buick (it was economical) running a simple bar graph indicator to show relative mixture settings on my O320 Lyc-powered homebuilt. "calibrated" it by leaning to the setting i wanted to hold and noting the reading. from then on, i could quick-set the mixture to the same point reliably and repeatably as long as i used unleaded fuel. then i burned a few tanks of leaded fuel. that changed the sensitivity gradually until it quit showing the actual ratio. so. if you want to use an A/F ratio meter, i recommend you find one that does not use an automotive O2 sensor from an '86 Buick to sense the ratio if you plan to use leaded fuel.

your mileage may vary.

[Mike M]

In order to make use of the technology that you are talking about, you will have to have an engine control computer FAA certified.

not for an experimental aircraft.

[steveinindy]

there has existed EFI for aircooled automotive engines for over two decades

The question I have in response to that is how hard is it to "tune" an EFI to a particular engine? Wouldn't it require extensive tweaking which would provide a higher risk for engine damage than simply going with the tried and true methods that are used to establish the TBO intervals anyhow?

I can see the value of also having EGT and CHT to make sure that the mixture setting isn't doing any harm,

So are you suggesting that we have a redundant system to allow the pilot to override the EFI if they settings are doing harm? Because unless that is done, you have three choice: land ASAP, risk engine failure or shut the engine down. The latter is obviously not a good choice in anything but the most extreme situations. Option 1 isn't always available unless we're willing to chance a forced landing or never leave the traffic pattern.

but those sensors then become more like oil temp or fuel pressure - something to warn the operator of problems, not something you use to prevent them. Or am I missing something?

Paying attention to them is how you prevent a problem. I'm probably one of the biggest advocates for EFI and FADEC but I also think we have to be cautious before getting in a rush to try to assign a more complicated solution to the problem of engine failure. The problem/question isn't can EFI or FADEC be safely applied to aircraft engines (because the answer is yes) but rather can we safely do it as a post-production add-on to the multitude of engines (aircraft and otherwise) that power experimental aircraft? The technology is there but at the same time we have to remember that not all of us are mechanically or electronically astute as we probably should be for such a task. I say this because I happen to be one of those people when it comes to anything firewall forward other than fire suppression. There's a very good reason why I'm going with something off the shelf for the powerplant and recognizing my own limits- which a lot of other homebuilders likely share- is the best way to keep myself and my passengers out of trouble.

In other words, Michael Crichton got it right in The Lost World when he was talking about the character Dr. Thorne (a Stanford engineering professor) who liked to point out to his students that to engineer you have to understand psychology because while your calculations and everything may be perfect, as soon as you put those plans into the hands of others, they will screw it up all too often. It's not something we like to think about or admit but if we want to grow as a community and as individuals, we must have this sort of introspection inherent in everything we do.

[normancanderson]

cdrmuetzel makes a good point about 'poisoning' the O2 sensor. This is also a concern for the automotive racing community, since they tend to use leaded fuels for high octane applications (like drag racing). Here is a quote from one of those forums "I use the innovate LM-1 with the Bosch sensor and we raced with it for 5 years, on VP C14 fuel (114 octane, leaded) , never had it alter its AFR reading. Engine is tuned well with tan on plugs and it runs at 12.8. Always powered it up, and I replaced it after 5 years for something to do, not because it was necessary, and the new sensor just read the same AFR as the old one, so the old one went into the box as a spare. So it is possible to obtain the time I said. And what I said about the 50 the 100 hours is what Bosch specify." So maybe the Buick sensor was more susceptable to contamination, or the engine was running too rich for that sensor. Definitely something to be addressed in such an application.

In response to steveinindy's comments - Q1- O2 sensors generate a voltage that is used by the EFI controller to regulate the A/F mixture. This voltage can be adjusted to provide control over the actual ratio (essentially tricking the EFI controller into giving the operator the ratio desired). Q2- The concept, whether using EFI or a carburetor, would be to provide mixture control to the pilot to get the desired engine conditions, just like mixture control and EGT are used today. Q3 - I think a lot of folks on this thread are over-thinking the question - O2 sensing can be used with a mixture control system of any kind to more accurately set the A/F ratio than using EGT. No fancy EFI or FADEC is required. This is just another sensor to help with engine management, and the original question was whether or not anyone had tried it. It sounds like a few have , with varied results. The biggest issue having been the leaded fuel potentially altering the readings.

Thank you everyone for your inputs, I think I have a grasp of the potential issues to be addressed. Finally, I have to admit that I'm truly amazed that, on a forum like this, there would be comments saying things like 'don't try something new, because others might try also it and screw things up'. To me, trying something new is what the spirit of the EAA is all about. Otherwise, this would just be the Cessna maintenance forum. Granted, you need to do your homework, and do your the best to mitigate risks. Can you imagine what life would be like if the Wright brothers heeded such advice?

[Mike M]

...comments saying things like 'don't try something new, because others might try also it and screw things up'.

if what i said sounded like that, it's not what i meant. what i was trying to say was, "don't duplicate a system i used without knowing its shortcomings." i've noted that when somebody asks a simple question on this forum, folks with no actual hands-on experience in that particular area will toss a wet blanket on the whole idea. some of us just ignore them. i'll check out that other sensor, thanks.

[steveinindy]

Finally, I have to admit that I'm truly amazed that, on a forum like this, there would be comments saying things like 'don't try something new, because others might try also it and screw things up'.

The problem is that you have to remember that not all EAA members have the requisite brain power and/or experience to go experimenting. A lack of one or both of those- especially in concert without either the restraining influence of the community or the common sense to know when you're in over your head- is a potentially lethal issue when someone starts getting away from strictly following a set of well-tested plans or a kit.

If you think I'm being hyperbolic, I have two words for you: Daniel Lloyd.

The "experimental" moniker for the majority of us is and probably should remain more a reflection the certification category of our aircraft more than an implicit encouragement to try new and untested concepts especially given how many of us plan to haul our family and friends around in our creations. Given that a lot of folks in the experimental community seem to treat the "test flight" period as more of a "get it over with" hurdle rather than a true testing and learning experience once they realize that the aircraft is off the ground and has something that passes for control, I can't in good faith argue that we need to be pushing those same folks to try new things without a lot more support and oversight than we can currently offer. It's better to be sitting around dreaming about possible improvements that are beyond our personal abilities than for our friends to be standing around a freshly closed grave questioning "what the hell was he thinking?".

o me, trying something new is what the spirit of the EAA is all about.

I would tend to agree with the above caveats.

Granted, you need to do your homework, and do your the best to mitigate risks.

Here's my thoughts on that:
1. Most pilots, unless they do test flying or aerobatic flying for a living, tend to kind of ruin their pants and freeze when they either encounter a stall or a sudden engine failure.
2. Most pilots' stick and rudder skills (my own included) leave much to be desired especially when it comes to low speed flight.

How do you suppose we mitigate these risks in the few pilots/builders who are confident enough in themselves to go forward with a serious engine modification? The issue is that people who become pilots tend to be more self-confident in their own abilities than your average person. We tend to be very poor judges of our own abilities and tend to rate ourselves superior to our peers at a rate greater than the general population. Those of us who decide to build seem (I say "seem" purposefully sense I've never seen any psychological cohort study of builders vs. non-builders and am going solely upon my own admittedly anecdotal evidence) to be more self-confident than the rest of the pilot community because of the determination it takes to spend years putting together a craft to bear us into the world of flight.

The problem is that- as a general rule- the more self-confident a person gets, the more driven they are towards a particular goal, the more at risk they are for "tunnel vision" or a selection bias towards those things that will lead them to that goal. It's the same underlying issue with "get-home-itis", scud running and all the other incredibly questionable pilot behaviors we see time and again in NTSB reports. It's also why, while the credit usually goes to one person (such as a Burt Rutan, a Ed Heidemann, a Kelly Johnson to name a few), almost to the point of statistically absurdity, they all had a team around them to keep them grounded or to otherwise provide feedback. Great ideas seldom- contrary to the way it's oft portrayed- spring from a lone genius who is working against all his detractors. Hell, sometimes the person who gets the credit didn't have anything really to do with the practical application of the idea (Alexander Fleming and penicillin anyone?)

Some of the best advice I was ever given came when a professor of mine invited me to lunch with a friend of his who was in town to visit. We get there and I realize that I'm about to have lunch with James Watson- as in one of the guys who figured out the double helix structure of DNA- and I found myself able to ask him what the best advice he had for a young researcher was. It came in the form of two points (paraphrasing here since I was too awestruck to think to write the actual words down):
1. Never be the smartest person in a room. If you think you are, you won't try as hard.
2. You will always learn more from those who disagree with you than those who will tell you that you have it all figured out. Anyone on your team who can't give you at least one reason the current idea won't work should be considered for termination because they are not obviously not giving it their full attention.

Can you imagine what life would be like if the Wright brothers heeded such advice?

I really hate that sort of logic because they are generally strawman arguments. There's a pretty big difference between the exceedingly cautious way the Wrights approached their testing (which explains to a far greater degree their success and survival than a simply pig-headed "you can't stop me" attitude) and development and the often "throw stuff at the wall until something sticks and assume that's the best solution" approach that a more than a few homebuilders currently use. As I said before, the best advice often comes from those with whom you disagree.

[Frank Giger]

My thoughts:

If one is knowledgeable enough about what an additional gauge actually means and can use it in a useful manner, I can't see how it would hurt - though I do wonder if the effort here is going to measureably improve performance or longevity of engine life. I'm the guy Steve is referring to when speaking about guys that don't need another gauge and wouldn't be able to interprete its data properly.



Secondly, the statement that pilots and homebuilders "tend to rate ourselves superior to our peers at a rate greater than the general population" is just plain silly as it a defining characteristic of our nature as aviators - because it's true!