There are three components that define RISK. Most of your comments address only 1 of those, the failure rate. But that's the one you use to justify dismissing the parts count in a Soob. Yes, it's a little tough to nail the fail rate on the rotary engine. I think it's a LOT tough to nail down in ANY aviation application - Lyc, Soob, V6, whatever. The other two components have equal weight. What EFFECT the failed component will have on flight. How likely to notice the defect before flight. It's not too difficult to assess those two components accurately. Those two are the no-brainers. We don't need a lot of schooling to figure out that a single CAS is a show stopper, water or oil pump is a serious but manageable situation and Oil Pressure Switch a minor annoyance. So if we have the discipline to use these methods, we've got most of the risk equation nailed. I'm getting lost here. How does mastering the no-brainers but still largely clueless on failure rate get MOST of the equation nailed? Therefore, we don't have to be precise in defining the fail rate. We just have to do our best to be moderately objective. Which brings us right back where we started. Consequence and pre-flight detection are no-brainers. We can guess at the failure rates without seriously compromising the process. Our only remaining obstacle is to become moderately objective and we now have you to impart that skill to us. Can I be forgiven if I suggest that we already have one or two moderately objective souls amoungst us, and further, that we pretty much know who they are?

But you know what this is essentially about? If we use our natural method What might this "natural method" look like? Would mine look a lot like or a lot different from yours or Cletus Johnson's? of determining risk, we make substantial judgement errors. You know

what's worse, we don't have any feedback to tell us how wrong we are. Now I'm REALLY getting lost. Up until now, I thought we were (individually and collectively) up to our ass in feedback. Folks helpless to stop telling us how wrong we are. BOY! Was I ever reading this list wrong! We tend to put more emphasis on items from recent memory, like anecdotes we heard recently. We attach significance to emotional appeals. I'm thinkin' that part will vary a lot from one individual to another. For my own part, I've been drawing lessons from the blunders (real or imagined - but often fatal) of my brethren for upwards of forty years. I like to think there are a lot of rational, level headed individuals on this list. If there is a shortage of anything, I'd say it is of hysteria. It's difficult to realize these truths. The Japanese analyzed design judgement errors, and found if they followed the FMEA approach, there were fewer failures. No doubt. But getting back to the issue at hand, what did their data look like? Were they working with two legs of the stool out of three (no data)? Or did they have ALL their shit together?

Here's an excellent example. I've seen people describe their reasoning

for going rotary. Often I would hear: "It only has three moving parts,

therefore it's safer". I suspect that 80% of you believed that. That is a gross error in perspective. No maybe, it's a huge distortion. But then again, maybe it's not. Maybe there's something important there. If it really IS a huge distortion, perhaps there's some readily available, compelling evidence to support that notion. It's a

wonderful theory, it has a component of truth in it, but totally fails

the tests for significance. This part (only my personal opinion) smells like one of those sweeping, unsupported generalities that my daddy warned me about. If a logical theory is not significant, you dump the theory. But first, you'd do well to have some really compelling evidence to support its insignificance.

When you assess risk, at some point you have to look at the whole

picture. Converting a rotary engine is extremely risky because there are

so many creative activities you are doing. As opposed to converting to a ??? (Soob? V6? Inline-4? - you pick one) engine that is devoid of/has dramatically fewer creative activities? Each has many potential

oversights, so the oversights occur. You have lot's of evidence that

describes this. The ratio of dead sticks, the number of alarm conditions, whatever you choose. But you need the courage to face the truth. It's not easy. Now I won't argue that perhaps Soob or V6 has fewer dead-sticks per installation than rotarys (largely because I don't have the data to support such a position). Can we entertain the notion that some of this might involve: a) we (the rotary community) are operating much closer to the rated peak performance of our engines than, say, your Soob; b) we are eschewing critical factory components for what we feel are [potentially - key word here] superior, more robust custom components? I can understand that such activity would increase risk initially. The (widely held, I think) hope is that these efforts will produce a more robust, reliable fuel control (for example) that will advance the state of the art down the road. The goodness or folly of that notion rests with the individual IMO. Stipulating the added risk, I must point out that the significance of such risk (not by any means quattified at this stage) must be determined by the risk taker - not by som

Now if we can take VERY effective action on those risk items, the ones we have all the failure data on? Or the ones we can inutit are obviously risky? then we have a power plant that has some real potential. Many of it's failure modes are gradual, thus reducing the EFFECT of the failure. Basically giving us more time to land safely. In theory, the rotary is less prone to torsional risks...although there are indications can you share some of those? that too fails the significance test. Unless we were to include Lycs with all their crankshaft ADs, or Ivo props, or gawd knows what we'd find if the Soobs and V6s were run at or above rated rpm. We get the satisfaction of being different and thumbing our noses at nay sayers. Any auto conversion qualifies here. We contribute to what can very likely be a viable alternative that exceeds Lycoming performance on all levels. And here too.

So this guy that spent all those years looking into the root causes of

failures, hopes he'll eventually be able to explain these things well

enough, provide enough facts, that a life or two will be saved. I suppose that sounds corny, but that's my view. My view is that the pivotal aspect of his investigation was reliable failure data. Seems to me we can all pretty much agree that the rest is pretty intuitive and not at all difficult to master (even without taking night courses).

On Mon, 06 Jun 2005 17:27:06 -0500 Jim Sower <canarder@frontiernet.net>