Yes Daniel, you did spy on me correctly, and thanks for your flattering finding via Google!
You, Larry and others have brought up excellent view points or examples on what engine ignition system to be adapted on their own KRs, and I agree it's a personal choice simply because the home-build experiment aircraft "industry" is quite a different beast, and there are too many unique factors and differences on how their planes are actually fabricated, even though everyone seems to follow the "same plan" or purchased the same kit or materials... In other words, the generic knowledge on safety or redundancy configurations of system design normally adapted by the commercial aerospace or nuclear industries (heavily regulated by government agencies) may or may not be applicable to the home-build world, and this is due to many factors ranging from cost, supplier and brand to availability, spacial constraints or even tooling and personal preference considerations! You are also quite accurate on the reliability characteristics of electronic components, of which the early high rate of failures are due to infant mortality (burn-in period) caused by manufacturing process or material defects. As Luis mentioned that the E-ignition modules he installed seem to exhibit fairly low failure rate, however unless the sample size of failure or anomaly data is sufficiently large, the confidence level of any calculated reliability or risk results on these e-modules can't be good enough, as you elaborated. The industrial trend toward digitized engineering in product or system design has largely driven by cost reduction and relentless seeking for profitability by industry stakeholders or business owners in my opinion. The use of FPGAs and software control on complex system designs have simplified the work load of design engineers in a significant way, but unfortunately at a cost of increased risk transferred to product users or system operators. The big risk trap is that, very few design engineers (now dominated by EE graduates) fully understand or well trained in the area of "assurance engineering". For instance, there is generally lack of understanding of severe ESD/EMD/EMC impact on performances of critical electronic components.... Precisely for reasons as discussed, we need to wait or watch cautiously when decide to utilize any new and fully digitized electronic/chip-based products in any of our safety critical systems. Believe or not, human understanding and predictive assessment techniques and math models on mechanical components are built upon over 200 years of experiences, and it's therefore far more robust than our handling on the beasts of semiconductors, microelectronics and the ever increased complexity of software codes attached for the flow control of sensitive flux of "intangible" particles (we call as "electric current"), and there are simply inconceivable number of triggering events, such as environmental stressors (moisture, thermal, ESD/EMD/EMC, radiation, lightening, aging, to name a few) which can suddenly ruin the day without any signs of warning....! v/r, Dr. Hsu On Fri, Jul 8, 2022, 5:26 PM Daniel Branstrom via KRnet < krnet@list.krnet.org> wrote: > I happened to Google Dr. Feng Hsu. If I've got the right person, his > credentials are impressive in the field of high tech risk assessment. I > would also like to add that the other people who have contributed to the > discussion are also not lightweights. I'm probably in the lightweight > category, but I've been around experimental aviation for about 50 years. > > Flying, to me, is a matter of risk assessment. Or, as the book about Jimmy > Doolittle put it, "the calculated risk". In terms of different ignition > systems, all have their advantages and disadvantages. > > As I understand it, the discussion has been mostly about ignitions for VWs > and engines designed for aircraft. Many have their roots based in the > 1930's and subsequent years, when 80 octane was readily available. With > 100LL, lead fouling has become a real problem for many piston powered > aircraft based on earlier designs. > > On Corvairs, designed in the late 50's/early 60's when fuel with high lead > content was used, IIRC, William Wynne claims that lead fouling is not a > problem on them, but that the ignition systems, coils, condensers, and > electronic circuits, are the weak points. As a consequence, his system only > has one ignition plug per cylinder, but two separate ignition systems with > their own coils that feed into a coil switcher before the distributor and > plugs. > > IIRC, His system IS powered electrically separate from the unit itself, > unlike mags, which generate their own power. He points out (pun intended) > that Kettering (points, coil) will continue to function at a much lower > voltage than for electronic ignition. The Kettering types also give warning > that they are about to fail. When the voltage to them falls off, their > performance falls off, especially in starting, whereas an electronic > ignition doesn't give warning with low voltage, they just quit. > > Many Corvair flyers have a second battery that just powers the ignition > for several hours if power is lost from the main battery or from the > generator or alternator. > > William now sells a dual system for Corvair engines with an electronic > ignition as the 'primary' one, with a Kettering system (like the Corvair > automobiles have) for a backup. As a backup, there is little pitting or > electrical wear on the Kettering points because they aren't powering the > plugs when the electronic system is on. Yes, they are both on the same > shaft from the engine, and he builds in a heavier bearing on the > distributor shaft, as I recall, but there's been no problems that I know of > with distributor shaft failures. There is also no engine coupling to fail. > > Correct me if I'm wrong, but in the early years of e-mag use, there were > several reported failures. Apparently, they are more reliable now. Their > self-generating power puts them in a separate category from regular mags > and totally electronic ignition systems. > > I've heard it reported that with electronic systems in general that if > they're going to fail, they do it early in the use cycle (less than 25 > hours). As people in the experimental aircraft world, we have to be careful > who we follow because the sample size for other than magneto systems is > relatively small, so that may skew any statistics. IMO, events are > important, but there may not be enough to make them statistically relevant. > > I had a friend who built a KR-1 with a single mag. The thing that caused > him to do a rebuild before even 50 hours were up was that he got a > Brazilian, instead of a German-made cam for his VW. He built it according > to plans, and he found the CG was way aft. > > What you choose to be your ignition system is a matter of personal choice. > I appreciate the reasoning expressed by the various people on this forum, > and that there's been on flaming or disparaging in the discussion. > > It behooves us to do a thorough inspection and run-up before we take to > the air, and to monitor the performance of our engines as much as possible. > They do seem to talk to us, if we 'listen'. > > Dan Branstrom > On 7/8/2022 9:56 AM, Dr. Feng Hsu via KRnet wrote: > > Good morning all, > > I enjoyed reading all the excellent stories, failure event descriptions, > arguments of pros & cons and thoughts (from John, Chris and Luis etc.) > pertaining the debate of safety on dual mags, dual E-ignition or one meg > plus one E-ignition redundant aircraft engine systems.... Well, I am glad > to see this discussion actually got us into a classical technical debate on > pros & cons in safety between conventional engineering vs digital > engineering, which has been going on for several decades across all the so > called "high integrity and high reliability" industries, such as nuclear, > aerospace, chemical and medical ....A famous topic debate on this issue in > the aerospace or aviation industry for example is the debate on "flying by > wire"!! Frankly, I am on the camp of opposing the concept of "flying by > wire" within the safety/reliability & risk assessment academic community, > and this is simply because the folks who have been advocating the concept > of "flying by wire" are the "extremist" in engineering design... For the > same reason, I am a none believer of a truly autonomous and safe unmanned > cars & trucks or ground-based transportation system based on AI technology > due to theoretically obstacles or limitations in coding highly reliable and > safe software packages based on human brains. The chilling fact is that > over 75% catastrophic accidents from all high-integrity industries were > contributed by software and human errors combined! Guess what, the single > most "unreliable component" in our man-machine systems (such as aviation) > is the human brain itself! > > Obviously, my views on this topic is taking a balance between the two > approaches and avoid going for both extreme ends of the argument! In > other words, I believe an one mag (mechanical) plus one E-ignition > redundancy makes a lot of sense in risk reduction based on theorems & > reliability engineering principles. I understand what's in the mind of Luis > on dual "independent e-ignition" component and DC power channels... The > issue here is that the two e-ignition channels are not truly independent as > you would believed! Yes, it looks independent in a physical sense but it is > inherently connected or identical due to the design and manufacturing > processes! Although I have not done a model based risk assessment myself to > compare the trade-offs in risk & reliability of these two ignition > configurations, but I am confident (based on expert judgement) that a meg > plus e-ignition configuration would highly likely to be the winner in terms > of safety risk concerns... I am sure other people in the aviation industry > have done a rigorous model based risk assignment comparison on this matter. > > Just to offer some food for thoughts for folks here who might be > interested in this topic debate...! > > Best of luck! > > Dr. Hsu > > On Fri, Jul 8, 2022, 8:07 AM Luis Claudio via KRnet <krnet@list.krnet.org> > wrote: > >> I think these discussions are healthy and welcoming. I read them and go >> "hmmm" then I make my own decisions. I have been flying since 1966 when I >> took my first ride as a civil air patrol cadet in High School in a PA-22 >> 108HP. As a student at Embry-Riddle, I was an understudy of Dr. Richie >> (RIP) a true pioneer in failure analysis (FMEA) of everything >> aviation-related. I understand failure modes, and most importantly how >> to mitigate the risk for each independent system or collectively (through >> risk priority number= severity x occurrence x detection...) I am also >> very familiar with E-Mags and other electronic ignitions which is why after >> careful consideration I chose a dual SDS ignition system. It wasn't by >> "gosh or by golly" that I arrived at my conclusion. Considering that my >> initial choice of ignition system was a Dual Bendix D3000 mag with a single >> failure mode (the main coupling), I asked what else is out there... and >> here is what I considered >> >> 1. A dual mag setup such as the aforementioned dual mag - Four points, >> four capacitors, one coupling = Total 5 points of failure minimum >> 2. An independent mag with an electronic ignition driven by the timing >> gears - Minimum of four points of failure (points, capacitor, two couplings) >> 3. An "E-Mag like" with an electronic ignition driven by the timing gears >> - Engine couplings = two points of failure (minimum) >> 4. Two independent electronic ignitions with no mechanical moving parts - >> dual independent battery backups - Component failure >> 5. Just screw it and go sailing... >> >> My most predictable component failure is the spark plugs. So moving >> forward I then wrote a proactive maintenance schedule in my POH and >> annotated it in my conditional inspection form to reduce the risk of >> component failure. Additionally, during my selection process, I >> prioritized the failure modes of each system, mechanical and electronic... >> from highest RPN to lowest RPN noting that it’s worth emphasizing that it’s >> nearly impossible to address every potential failure. Instead, I focused on >> addressing the potential failures that would most jeopardize the safe >> outcome of my flights and I chose from analysis and not from bells and >> whistles... just my rambling thoughts here trying to justify that $50K I >> spent on getting educated at Embry-Riddle back in the day... keep building >> >> Luis R Claudio, KR2S N8981S >> On Friday, July 8, 2022 at 06:23:13 AM CDT, victor taylor via KRnet < >> krnet@list.krnet.org> wrote: >> >> >> To answer Chris yes I do have a flying KR2 with a single electronic >> ignition though I was actually talking to Dan Diehl yesterday about adding >> a magneto for a backup. >> I do contract work for Velocity Aircraft as a pilot. We love electronic >> ignitions and put them on every airplane. They are the way to go and in the >> past have gone 100% electronic ignition but over time have had enough >> failures that we went back to having one mag. I also test fly new airplanes >> for M-Square who builds the Zenith CH-750 SLSA. I actually have had two >> electronic ignition failures with brand new aircraft there in the last >> couple of years. M-Square is still committed to 100% electronic at this >> point. Just to be clear I’m not knocking electronic ignitions nor their >> reliability. If you have dual electronic ignitions though you likely are >> relying on the same electrical system to run both systems. Up until four >> months ago I had never had a magneto failure in my 40 years of flying but >> recently I lost a magneto while ferrying a Grumman AA1A. Fortunately the >> other one got me to the nearest airport where a safe landing was made. >> There is a reason why aircraft manufacturers such as Velocity have reversed >> their positions on going 100% electronic. And that reason is failures in >> the past and accidents such as the one in Kissimmee. >> When Teladyne Continental first built their full FADEC engine system it >> kept experiencing total ignition failures in flight. After over a year of >> flight testing with multiple in flight failures yet not a single one in the >> test cell they finally got to the bottom of it. The problems was found to >> be the frequency that the US government uses to communicate with submarines >> was interfering with the FADAC system. The fix was simple by shielding the >> system. >> One of the functions of this group is to make the KR’s safer. That’s done >> by discussions and experience. Mostly bad experiences and I’ve personally >> had those bad experiences as a professional test pilot. All of us in this >> group look at canopy latches a little closer today than we did a year ago. >> Controls have gotten balanced, fifth bearings have been added to engines, >> fuel tanks are being moved to the wings etc. You would think that over the >> years we’ve fixed every possible failure point. Have we? >> It’s our duty in this group to be devils advocate at times and to point >> out past failures of similar systems. >> >> Victor >> >> > On Jul 8, 2022, at 00:24, John Gotschall via KRnet < >> krnet@list.krnet.org> wrote: >> > >> > >> > Hi all, >> > >> > Quite a bit of commentary about risk, system failure etc. >> > >> > So I am convinced everything will fail sooner than later. I make my >> living on failing machines, so I may appear a pessimist about machines, or >> an optimist about future employment fixing them. On the upside there is so >> much failed stuff I get paid more than just several times every day since >> 1988 to set those broken machines straight. >> > >> > I find it particularly interesting to see the failures that come in >> groups or waves. I experienced such an odd failure scenerio in aviation >> recently. >> > >> > I bought an experimental flying boat, a volmer home built, OMG, what a >> pretty unit! And a blast to fly! A cub can't touch a volmer for fun. >> Sadly I broke that and am still working to fix it. >> > >> > Anyway there is an annual seaplane fly in at NW Idaho, and I and a >> friend went in the volmer. What a blast! the mountain crossing, middle of >> nowhere overnight camping, etc. However the alternator quit near Idaho and >> we made our way there and home by never turning on the battery switch the >> whole time, except to run the starter motor, and then to cross under the >> Seattle class B. Needed adsb working for that. >> > >> > I found the High current alternator output wire had failed by vibration >> fatigue and had simply broken off that big alternator output post. Simple >> to fix, and I replaced the entire wire with welding cable, better for >> vibration. No big deal. >> > >> > Then I broke the Volmer (another story), and needed another flying boat >> so bought a lake LA-4. We found it in Arkansas and flew it home to the >> Seattle area. >> > >> > The same seaplane fly in came up in Idaho this year, and we went again >> this time in the lake. A complete blast for all the same reasons, but on >> the way home the alternator quit outputting current! This time the plane >> has heavy electrical laods that quickly consumed and flattened the battery >> in short order (less than 40 minutes) on the final leg home. instead of >> panic we just turned west to get out from under the class B, shut all the >> electrical off and used the hand hydraulic backup pump to operate the gear >> and flaps. Made the uneventful landing at home in the backyard and went >> digging for the fault. >> > >> > Dang if it wasn't the SAME DAMN WIRE that broke in the Volmer! On the >> second occurrence of the same trip. Except this time the wire pulled out of >> a bad crimp rather than broke off. Pulled out of a bad crimp on a >> certified factory built plane. hmmm. >> > >> > Well, both the Volmer and the Lake are dual mag lycomings. An o290 and >> o360. >> > >> > So I will say mags are a great choice, from experience. In both cases >> mags brought us all the way home. My kr has one mag and one electric >> ignition. It'll probably make it home without an electrical system. >> > >> > cheers! >> > >> > >> > jg >> > >> > >> > >> > -- >> > KRnet mailing list >> > KRnet@list.krnet.org >> > https://list.krnet.org/mailman/listinfo/krnet >> >> >> -- >> KRnet mailing list >> KRnet@list.krnet.org >> https://list.krnet.org/mailman/listinfo/krnet >> -- >> KRnet mailing list >> KRnet@list.krnet.org >> https://list.krnet.org/mailman/listinfo/krnet >> > > -- > KRnet mailing list > KRnet@list.krnet.org > https://list.krnet.org/mailman/listinfo/krnet >
-- KRnet mailing list KRnet@list.krnet.org https://list.krnet.org/mailman/listinfo/krnet
