What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Eggenfellner Engines - Technical Only

Status
Not open for further replies.
Convince me

Whoa Nellie.....

Speaking for myself, this is not a case of a professional engineer picking on the weak. (snip...)

Three, the only agenda is to shine light in the dark places. We may find gold, or we may find cockroaches, or we may find both....but I didn't set out to find just one in particular.

Agree 100% with Dan. I haven't seen anyone trying to insult Jan. I have seen some difficult questions. These are exactly the questions I'd want good answers for before buying Sube.

I'm just the demographic Jan should sell to, by the way. I'm a fairly early adopter of new technologies, and I'm competent enough in the relative disciplines to install a "alternative" engine without hand-holding. I'm going to need an engine in a few years (I hope!) and have no bias against Subaru. However, just as I know exactly what I'm getting design-wise with Lycoming, I'd expect to know exactly what I'm getting design-wise with any alternative engine before I plunk down 20 - 25K.

It was mentioned that Lycoming wouldn't be responsive to questions about the crankshaft recalls. I found all I needed to know, including metallurgical reports and expert testimony, with a few minutes worth of googling.

T.J.
 
As you now know, the exact testing requested was performed by some of the "big guys" in the industry. Namely MT propellers, Germany. They sell certificated propellers to million $ airplanes and are experts at this very thing. Every STC need detailed information of this sort, before the propeller can be bolted and flown on a certificated airplane. As a company selling experimental engines, we do not have to do any of this but we still did.
Jan
 
Test results from industry standard tests are not intellectual property, nor are they trade secrets.

Any organization may choose to publish or not publish whatever it wants, but just to clarify since we are throwing out legal terms of art here, nothing which is being requested would be considered a "trade secret" or "intellectual property" Those terms would typically apply to unique designs or processes which are not standard, nor widely practiced, and which give a company a competitive advantage.

The fact that a company does or does not perform certain testing, and the results of that testing would not be shielded or protected as trade secrets or intellectual property in any proceeding between a customer and a vendor for example, unless the vendor could establish that the tests themselves were unique enough that revealing the results of the test would reveal the methodology, which would reveal a design that was unique.

As I understand it the claim is that industry standard tests have been performed, by a third party (MT), the follow up question is simply for the results of those tests so that claims and designs can be vetted. In addition, folks are interested in where the resonance falls in the spectrum, which frankly I find fascinating.

A clip from an email, which offers a single conclusory statement that loads are lower than direct drive engines is of little help, as I understand the question here. I think folks are looking at the system before the propshaft as well as whatever loads ultimately lead to the prop. I think from reading the whitepapers, that simply measuring prop loads/vibration actually tells you very little about what is going on, resonance wise, in the rest of the system.

Seems like releasing the data for such vetting would be a competitive advantage especially since such hardcore verification is unavailable from the less prolific alternative vendors.

I do not think it is insulting at all to follow up on generalized answers with more direct pin point questions clearly designed to get at the specifics.

I wonder if there is just a "failure to communicate" it seems that Jan wants to discuss "how well these things run", and the nerds (term of endearment) want the detailed information supporting the claims about "how well these things run".

But it seems that the questions are appropriate, given the nature and stated purpose of this thread. I for one am excited at the prospect of hearing and reading the analysis. The information shared thus far on torsional resonance and such has been really educational for me. I don't think anyone has insulted Jan.

But I wonder if, given the different goals between Jan, and the folks asking the technical questions, if this thread will meet its stated goal. It certainly seems to be a great opportunity to go beyond marketing and enter the realm of truly educating the market, with facts, on the Egg package.

I am not Jan, but I think such openess would pay huge dividends for years to come, in terms of new customers, more installations, and more successful installations....I would have to imagine that it would help both Jan and the industry as a whole. JMHO

I hope the discussion continues and the facts come out.
 
Jan, please forgive a request for clarification.

The specific question was "Did you apply a strain gauge to the propshaft and record vibratory torque? Your response was "Yes we did."

However, in a post a few minutes later you refer to an MT-conducted propeller test as "the exact testing performed".

Very large difference between the two. Could you clarify? You conducted a test per the above question and MT did a prop survey?
 
Supplying test data is reasonable

........we have decided that the information requested is for company use only. We are not willing to share testing that cost thousands of $ and countless hours to generate. We are happy to share general information but that is not where this is going.

You still have the opportunity to use this thread for what I thought it was about. Namely, discussing how our engines go together and why they work as well as they do. General information about ECU's, cooling, exhaust, reduction drives, engine mounting, etc. Is available from me. Company collected data, used to make our engines work, is not available. Like you know, it is all in the details. We are keeping the details.

Representatives from Honda discussed harmonics with their engine project with me recently. Very interesting and fun.

Jan

Jan,

I don't want to question your integrity, but neither do I agree with your assertion about questions and test data.

I'm considering an engine change in 5-to-6 years. Considering the lead time on your product, that is not too far in the future. Options range from an IO360, some TBD diesel, your product, etc. Information about the color of valve covers is interesting, but, as a potential user of your product I have a desire for more substantial information.

I'm an EE in the semiconductor industry. We supply product test data of almost every form imaginable to our customers so that they will know how to use it, what to expect in terms of performance and operating limitations, and how it compares to the competition. On top of that, many customers will request and get customized testing on issues they are concerned about.

I'm trying, unsuccessfully, to imagine denying a request for information relating directly to a potential failure or operating life of a product. Taking additional data, which costs money, is negotiable. Sharing existing data is just a given. Know-how, in terms of how you achieve that performance is a different story. I can see how sharing that would be a problem.

While I'm writing, I drive a 2004 WRX STi. I really like that motor and use all 300 horses as often as possible. I've got close to 40k on it, with no problems. That experience has caused me to be naturally drawn to your product and I have been periodically checking your web site for updates and information.

The power, reliability, simplicity of operation, and economy of the modern auto engine are all a draw.

As someone with limited wrenching skills & experience, however, it seems like your product is too experimental for me. I do not get the feeling that I can bolt in your motor and go.

That doesn't mean the motor is bad for others, but what ever motor I choose is going to need to be very turn-key in terms of installation and maintenance.

Also, there is going to have to be data that makes me feel like I'm not going to become a glider pilot with one of my grandchildren in the right seat. That data can be in the form of test data or in the form of successful flying hours (like the lyco), but I've got to have it.
 
Steve:

I am not an engineer, I am not a engine mechanic, I am a salesman: however, I have assisted in installing an IO-540, and have just completed installing an Eggenfellner E-6TI on my RV-10. From experience, I can tell you that the Egg is much easier. Just about everything is already installed. I installed the engine with a hoist, by myself, and have now just about completed the wiring, and will do a test run, probably next week.

With the Lyco, we had to do baffling, oil cooler, starter, alternator, etc etc; all this is done on the Egg.

I guess you can make a case for needing all this test data, if you want, and have the engineering slant to analyze it, however, I went with the data I got from the satisfied customers I talked to, and the excellent reputation that Subaru engines enjoy.

My last airplane was a Piper Arrow, and that engine was a maintainance nightmare. It contributed to my decision to go alternative, and made me question the belief by many on this list that the air cooled Lyco is the panacea they think it is.

In summary, I am sure you can find many reasons to decide to go Lyco, but don't use installation difficulty as one of them.
 
Helpful, but.......

Steve:

I am not an engineer, I am not a engine mechanic, I am a salesman:......

In summary, I am sure you can find many reasons to decide to go Lyco, but don't use installation difficulty as one of them.

That's a helpful data point, but your career choice doesn't hinder you from being a great mechanic.

I don't think being an EE helps my wrenching at all. It probably does make me predisposed to liking lots of data.

I have a general uneasiness with withholding TEST data.

Another pilot made this comment: " If you called up Ford, GM, or Toyota and asked for their test data they would laugh you out the door"

Since the cost of the motor and the cost of the cars are about the same, its a somewhat valid comparison. So:

If Jan makes samples of his motor available for testing by several aviation magazines + adds a significant warranty + local dealers who can solve product problems, then we'll have an apples to apples comparison.

I still have 5-to-6 years to make my choice, so I haven't closed the subi out as an option. I'm just saying that withholding test data doesn't warm me to the idea at all.

Open exchange of information builds trust.

AND - I am not hostile toward Jan or his product. I think its a cool product that may have specific advantages. Maybe I'll be flying behind one eventually.

I'm just being open about my thought process in terms of considering it as an alternative.
 
I took a few minutes to follow up on Bill's comment regarding Powersport's torsional guy, to wit:

<<I'm rotary oriented and got to be pretty friendly with the guy that did the newest Powersport drive. He did his PHD thesis on the vibration studies of the PowerSport rotary and drive combo. He wouldn't reveal the info to outsiders either.>>

His name is Steve Weinzierl. The Powersport analysis was for a Master's. He was with Polaris, consulted for Powersport and is apparently with Schrick Inc today, working on lightweight geared aviation diesels:

http://ftp.uspto.gov/web/patents/patog/week12/OG/html/1316-3/US07191742-20070320.html

If he was in fact a hired consultant to Powersport, he lips would be sealed by contract, a normal practice in consulting.

However, that hardly means information is not available. I direct you to a few pages of the Powersport website:

1. Complete Solidworks drawings of the engine and PSRU:

http://www.powersportaviation.com/Home/Reduction drive/Reduction drive.htm

2. Photos of torsional measurement using the encoder method;

http://www.powersportaviation.com/Home/Testing/Testing.htm

3. A Campbell diagram; 1st and 2nd natural frequencies plotted with a full series of engine orders, ie, the forcing frequencies. They did indeed use the "very stiff" approach; F1 about 340 hz:

http://www.powersportaviation.com/Critical2.xls

I guess ya'll get the point.

On a personal note, I really like this Weinzierl fellow. I offer two quick snips from web search results.

First, he has the respect of Kevin Cameron, my all time favorite gearhead:

<<Steve Weinzierl had that case of the prototype triples that Fuji had built for Polaris a number of years ago. They all broke cranks in testing (20 of them, and not cheap either!) and the Polaris people were just throwing up their hands. Steve, in his methodical way, got out the thick books on crankshaft math simulation and plowed through it all - approximating the flywheel discs as separate masses and working up all the shafts as torsion springs. And the math showed about what they had - a bad maximum at 8500. He added 2-mm to one or two shafts, parts were made, and on test everything was sweet and calm. KC >>

....and a bit from a Forbes Magazine article on Victory Motorcycles (a division of Polaris):

<<Testing the cooling system fell to development manager Steve Weinzierl.
Working on a tight budget, Weinzierl strapped a Czech-built Velorex sidecar onto a prototype Victory bike and, with colleagues, took it to Death Valley, Calif., for worst-case cooling trials. For this kind of testing, one guy rides the bike, which is studded with temperature-sensing thermocouples like a patient prepped for an electrocardiogram. Wires from the thermocouples lead to the sidecar, where a second guy records the temperatures.

One day Weinzierl decided to use the instruments on the Victory bike to record the oil-sump temperature of a Ducati Monster, which he sees as the benchmark for this sort of cooling system. "We drilled a hole in the Monster's oil plug and put in a thermocouple," he recalls. "It was 121 degrees in the shade. I rode the Ducati, and while we were rolling I'd pull over to about ten inches from the Victory and hand the guy in the sidecar the wires from the thermocouple. This was at speeds up to 90 mph. It wasn't entirely legal, but we got the data." >>

My kind of guy. <g>

Ya'll lighten up and let's talk tech.
 
I took a few minutes to follow up on Bill's comment regarding Powersport's torsional guy, to wit:

<<I'm rotary oriented and got to be pretty friendly with the guy that did the newest Powersport drive. He did his PHD thesis on the vibration studies of the PowerSport rotary and drive combo. He wouldn't reveal the info to outsiders either.>>

His name is Steve Weinzierl. The Powersport analysis was for a Master's. He was with Polaris, consulted for Powersport and is apparently with Schrick Inc today, working on lightweight geared aviation diesels:

http://ftp.uspto.gov/web/patents/patog/week12/OG/html/1316-3/US07191742-20070320.html

Correct about his masters, but like so many people good in the T/V world he was sucked up by the automotive world so fast it was like he vanished. I believe he is now a VP at Schrick USA and was responsible for the crank and crankcase of the VW W-12 and by association the Bugatti Veron W-16! hardly typical. I bet he is going crazy in a management position!


If he was in fact a hired consultant to Powersport, he lips would be sealed by contract, a normal practice in consulting.

However, that hardly means information is not available. I direct you to a few pages of the Powersport website:

1. Complete Solidworks drawings of the engine and PSRU:

http://www.powersportaviation.com/Home/Reduction drive/Reduction drive.htm

Hardly unique, since Jan has shown disassembled Gen-3 reduction drives.

2. Photos of torsional measurement using the encoder method;

http://www.powersportaviation.com/Home/Testing/Testing.htm

3. A Campbell diagram; 1st and 2nd natural frequencies plotted with a full series of engine orders, ie, the forcing frequencies. They did indeed use the "very stiff" approach; F1 about 340 hz:

http://www.powersportaviation.com/Critical2.xls

Very unusual which is why I contacted him in the first place. The rotary is unique in the ability to use the "very stiff" setup due to the super strong nature of the straight through e-shaft. Additionally he was more than a consultant at Powersport. They licensed him the ok to build all engines >300 HP as a separate company. To bad that for all that Powersport is now all but a non-entity. Not Steves fault, but my point is that disclosure of test results won't pay the bills for you.

I guess ya'll get the point.

On a personal note, I really like this Weinzierl fellow.

<snip> So do I. Bill

My kind of guy. <g>

Ya'll lighten up and let's talk tech.

We face several technical and legal troubles with producing parts for aircraft. A super small pool of buyers and a legal profession waiting buzzard-like for anybody to mess up. I'd love to have that F1 data on all the new engine packages, but I'd wager that half (at least) would give you a blank stare if you asked them what the F1 frequency of their system is! The people who know are charging a bunch for all their parts. Check out the prices for some of the EPI stuff Ross linked too earlier. Dan and Mike I'm really on your side, but when we have a manufacturer gutsy enough to even come on the forum I'm supprised and pleased. I'd go on but it's really off-topic in this tech-thread.
Back to lurking.
Bill Jepson
 
The specific question was "Did you apply a strain gauge to the propshaft and record vibratory torque?

I said yes before. Very interesting process. Scary to see the the same resultd when done on a pusher aircraft with a direct drive 360 engine. It is a wonder it can stay together.

Jan
 
Possible fatigue / crack starter

Yes, we are learning this. The front seal is riding on stainless to avoid rust over the years. The steel is soft and can get scratched between the manufacturing process and installation. I hope we cured it by taping each seal surface for protection, until installation. Also, we went from a 10 mm wide to a 12 mm wide seal and so far all is good.
Jan

Jan,
I've dealt with seal-on-shaft difficulties over the years. The seal will always cut into the shaft. This could be especially problematic in an application like yours because this seal groove will become a stress riser and a likely candidate for fatigue failure. One way to eliminate the problem is to use a seal with a bore of, say 1/8" larger diameter than the shaft, then shrink a 1/16" thick sleeve around the shaft. Thus, if the seal does wear a groove into the sleeve, and that forms a fatigue crack in the sleeve, the crack will stop when it is cracked through to the ID of the sleeve. The crack will not propagate down into the shaft itself. This is especially nice if you like to keep the prop attached to the airplane. :)
 
I looked at the seal change out instructions in the link.

Very nicely written instructions, good photo usage.

And the use of a screw to "jack" the seal out was a new tip for me.

Thanks for the link.
 
Jan, Scary? Why?

I said yes before. Very interesting process. Scary to see the the same result when done on a pusher aircraft with a direct drive 360 engine. It is a wonder it can stay together. Jan
Jan, Lyc 360 pusher's are scary. My friend flys this config, with a fixed wood/glass prop. I figured harmonics on the prop end where OK, because its wood. Also I assumed the engine was OK as well since there are 100's if not thousands of LongEZ's, Cozy's, E-racer's, Velocity's Lyc 320/360/540 pushers flying around for a few decades now. I have heard of wood props coming apart but no engine/crank issue, attributed or associated with the "pusher" configuration.

Sorry to ask a non-Egg question, but why do you say scary? Why do you say. "It is a wonder it can stay together"? This concerns me. It would upset me greatly if something happened to my friend and I did not say something to him. :confused:
 
Last edited:
Jan,
I've dealt with seal-on-shaft difficulties over the years. The seal will always cut into the shaft. This could be especially problematic in an application like yours because this seal groove will become a stress riser and a likely candidate for fatigue failure. One way to eliminate the problem is to use a seal with a bore of, say 1/8" larger diameter than the shaft, then shrink a 1/16" thick sleeve around the shaft. Thus, if the seal does wear a groove into the sleeve, and that forms a fatigue crack in the sleeve, the crack will stop when it is cracked through to the ID of the sleeve. The crack will not propagate down into the shaft itself. This is especially nice if you like to keep the prop attached to the airplane.

The design is very similar to what you describe. The seal is riding on a larger diameter than the output shaft for that very reason.

Jan
 
I have heard of wood props coming apart but no engine/crank issue, attributed or associated with the "pusher" configuration.

I had 2 Vari-Eze's (O-200 and 2.2 Subaru) myself and loved to fly in them. I was just refering to data I have seen on direct drive crankshafts, driving propellers in general. (The report ai saw was on a 360 pusher) Like spagetti in motion. Designed to handle the job for sure and very impressive designs.
 
<<I was just refering to data I have seen on direct drive crankshafts, driving propellers in general.>>

Perhaps you could offer a link or bibliography reference for such data? Like George, I would love to see it.

Fair trade; here's an illustration taken from the SAE Journal, Vol 47 No1 (50 or 60 years ago), measured crankshaft vibration amplitudes for a small geared Lycoming (CAA TC 210), with and without 2nd and 4th order pendulum absorbers.

 
I had 2 Vari-Eze's (O-200 and 2.2 Subaru) myself and loved to fly in them. I was just refering to data I have seen on direct drive crankshafts, driving propellers in general. (The report ai saw was on a 360 pusher) Like spagetti in motion. Designed to handle the job for sure and very impressive designs.

The problem with the pusher is the prop slams into very turbulent airflow, twice per revolution, imposing significant stresses on both the prop and the crankshaft. This is what caused the failure of the prop on the Rutan Voyager aircraft within the first couple of hours of operation. A composite prop was installed, which functioned for the remainder of the program.

This is one of the many design compromises of pusher aircraft which have severely limited it's success in light aircraft design.
 
<<I was just refering to data I have seen on direct drive crankshafts, driving propellers in general.>>

Perhaps you could offer a link or bibliography reference for such data? Like George, I would love to see it.

Fair trade; here's an illustration taken from the SAE Journal, Vol 47 No1 (50 or 60 years ago), measured crankshaft vibration amplitudes for a small geared Lycoming (CAA TC 210), with and without 2nd and 4th order pendulum absorbers.


Dan,

Since most geared, production engines employ pendulum dampers, could the absence of such dampers on the Subaru be causing the problems they are having?
 
The problem with the pusher is the prop slams into very turbulent airflow, twice per revolution, imposing significant stresses on both the prop and the crankshaft. This is what caused the failure of the prop on the Rutan Voyager aircraft within the first couple of hours of operation. A composite prop was installed, which functioned for the remainder of the program.

This is one of the many design compromises of pusher aircraft which have severely limited it's success in light aircraft design.

[OFF TOPIC]

I thought Voyager eventually wound up using a conventional metal CS prop on the aft engine?
 
<<Dan, Since most geared, production engines employ pendulum dampers, could the absence of such dampers on the Subaru be causing the problems they are having?>>

No. It merely dictates another approach. Pendulum absorbers were the hot new technology of the time.

<<The problem with the pusher is the prop slams into very turbulent airflow, twice per revolution, imposing significant stresses on both the prop and the crankshaft. >>

Again no; "slam" suggests the wrong picture; the airflow disturbance is merely another source for a forcing frequency, which can be a variation in shaft torque, or a variation in load, or both. The forcing frequency itself is not necessarily large in terms of shaft torque variation, but as usual, if it matches a natural frequency the system resonates and things can go straight to heck very quickly. "Pusher propeller" problems of this nature are among the most well-studied in the torsional vibration field because they're among the oldest; consider the blade of a ship propeller passing through the wake of the sternpost.

Did you know a propulsion system torsional vibration analysis is required before Lloyds will bind insurance coverage for a ship? Ships with no steerage tend to go on the rocks, much like airplanes with no propulsion tend to return to earth.

BTW, it is not just a pusher problem; the same issue can arise with tractor twins due to airfow along the fuselage (order = # of blades), or because the prop disk was placed very close to the leading edge of the wing (2nd order for a 2-blade, 6th order for a 3-blade).

I digress.....returning now to Egg -> technical.
 
Last edited:
Today we further simplified the Fuel injection system of the engines by eliminating the need for the throttle position sensor (TPS). We have kept it until now, for good measure, but enough testing has shown it to be of no practical value with the engine in an airplane. It will save 3 wires and one pin on our newly designed engine harness and firewall Cannon plug.

We no longer use O2 sensors in the system but a bung is still provided for those that would like to add a wide-band fuel / air ratio meter to the panel.

I have now completed the design of a pusher configuration cooling system for the engines and will test it in the defiant this summer. I successful, we can provide pusher engine packages where the cooling has been implemented into the package.

Jan


Jan
 
The following diagram is taken from a classic SAE paper, Karl Lurenbaum's
"Vibration of Crankshaft-Propeller Systems". Mr Lurenbaum plots the measured 1st and 2nd mode natural frequencies for several experimental metal propellers, as well as the engine's exciting orders and the engine's own 1st crankshaft natural frequency. Remember, when an exciting frequency matches (intersects) a natural frequency, the component(s) with that natural frequency resonate and the vibratory amplitude can get very large. On this plot, every point of intersection with an engine order means potential trouble.

Note the tight grouping of 1st mode natural fequencies for many of the subject propellers beginning around 4500 vib/min and rising to about 5250 vib/min at 2500 RPM. (Centrifugal force stiffens the blade and raises blade frequency as you spin faster.) The first mode can be visualized as an arc-shaped bending of the entire blade.

The 2nd mode blade frequencies are spread out quite a lot. The 2nd mode vibration can be visualized as the blade forming something like a very weak question mark.

We also see the crankshaft's 1st torsional mode at about 11,000 vibrations per minute (183 hz).

The exciting engine orders are the forcing frequencies. These (2, 4, 6, etc) are for a 4-cyl 4-stroke. The 2nd order is firing frequency and the most powerful.

The life or death propeller question is the amplitude of vibration at each intersection. There may be elaborate computer models good enough for prediction of accurate blade vibration amplitudes. However, there are so many variables that Hartzell, MT, etc still find it necessary to stick strain gauges all over the blade and fly with telemetry.

The engine manufacturer is generally responsible for determining crankshaft vibratory modes. In this case the crank could be excited by the 6th, 8th, and 10th orders.

The Schwartz No 32018 would be a really bad choice. It would resonate in the first mode at both 950 and (much worse) 2250 RPM. It is a real mess in the 2nd mode; note the intersection of 2nd mode blade frequency, crankshaft 1st mode torsional frequency and the 6th order exciting frequency at about 1840 RPM. At best you could expect a prohibited RPM range.

Cutting edge stuff? Actually, no. This paper was published in SAE Transactions Vol 39, No 6, Dec 1936. Lurenbaum was a researcher at Deutschen Versuchsanstalt fur Luftfahrt, the German state aviation research institute. He presented his paper at the SAE's semi-annual meeting in White Sulphur Springs WV in June 1936. We were a little behind in such things at the time, so Mr. Lurenbaum's visit was fortunate given the events of 1939 and after.

Funny how we could be given such detailed data by a soon-to-be enemy nation, but 72 years later we can't get the slightest peek from a vendor <g>

 
The Bf-109B had a Schwarz propeller, the chart you posted may well have data for that application. It was about that time the 109 was being tested for war in Spain. The early versions had a fixed pitch Schwarz wood prop.

MT, another German company with roots back to those days, does extensive tests with their props to certify them for various applications. But I have not seen any data from those tests. It probably is filed with the FAA but the average customer does not see it. Maybe they decided sharing all that data in 1936 was not such a good idea after all. :)
 
Max temp on the unit was 134F, I hope it isn't running too cool. :) Coolant temp was on the thermostat at 189 and oil was 180 with the cooler blocked off.

Customer flight tests show that the G3 drive is running below engine oil and coolant temps. That is a good sign for long life. This was on a cool day.

Jan
 
<<But I have not seen any data from those (propeller vibration survey)tests.>>

Exactly. I doubt anyone outside MT has seen the data from the recent survey on a customer's airplane, including Jan. I'm not so sure the prop people share all their information with conventional engine manufacturers either, although the three teams (engine, propeller and FAA) would be working together in the course of a certified installation and would talk shop in the process:

http://rgl.faa.gov/Regulatory_and_Guidance_Library/rgAdvisoryCircular.nsf/0/4cfce91224a85b8586256acd00707ebb/$FILE/ac20-66a.pdf

BTW David, you mentioned the customer's name in another thread. What was it again? And is his installation a 6-cyl or 4-cyl?
 
Last edited:
Vee have vays of making Das Props

MT, another German company with roots back to those days, does extensive tests with their props to certify them for various applications. :)
David since you correct me (and you are usually right) :D one clarification, MT-Propeller was founded in 1981 by Gerd Muehlbauer, whose prop experience prop design goes back to 1968. But you are right the ideas he uses (natural composite ie wood) go back the the 1930's, pre-WWII Germany.

The engine tends to drive the prop vibration not the other way around. However prop does affect direct drive engines at some "orders" and of course would affect a PSRU. The PSRU as a middle man has a hard act to perform. It has to deal with both the unique harmonics of the engine and prop and it's own. As Jan pointed out, prop balance is important (1st order).

A direct drive crank is a massive thing. The going in position to fight all the pounding from combustion, rods pushing and prop harmonics & inertia, is just make it stout and rigid. If stresses are low, regardless of harmonics it will last. Excitation at higher orders also tend to be less energetic, so they can be tollerated more than lower order harmonics (sometimes). If not, you see "pendulum absorbers" or dampers at critical orders (frequencies). On Lycs you see 3rd, 5th, 6th & 8th order dampers typically. These orders are mostly associated with the torque, inertia and pounding the crank takes from the cyl pressures (via the rods). None of the orders that typically get crank dampers are prop related, unless you have a three blade prop. Some engines don't need any damper because the crank stresses are low enough to tolerate. The direct drive crank has to tolerate 1/2 and 1-1/2 order vibs by brute strength. That design philosophy works for PSRU's, "when in doubt make it stout".

Below are some significant "Orders" of vibration and the source of that vibration on direct drive engines. (bold = where prop acts on engine or vise verse)

Vibration Orders (from Sacramento Sky Ranch)
--------------------------------------------------------------------------------
1/2 order
-minor component, single prop blade aerodynamics.
-major component, Gas pressure (torque).
-ranked relative cylinder influence: Continental 1,2,3,4,5,6; Lycoming 5,6,3,4,1,2.
-notes: Amplitude depends on mount stiffness, crankshaft torsional stiffness, cylinder position and gas pressure.
--------------------------------------------------------------------------------
1st order
-minor component, slight amounts of inertia and dead weight. Torque increases with order.
-major component, rotating imbalance.
-related orders 2,4,5,7,8. Higher orders are torque orders.
-notes: Amplitude increases with engine speed. Movement of propeller from rotating imbalance can cause 2nd order vibration on 2-blade propeller or 3rd order vibration on 3-blade propeller
--------------------------------------------------------------------------------
1-1/2 order
-minor component, none.
-major component, gas pressure (torque).
-related orders at 40% power, 1-1/2 and 1/2 order have approximately same amplitude. At higher power settings half order is increasingly stronger than 1-1/2 order.
-ranked relative cylinder influence: Lycoming 5,6,3,4,1,2, Continental 1,2,3,4,5,6,
-note: All 1/2 orders are gas pressure. may be more accurate assessment of gas pressure variation than 1/2 order. Amplitude also depends on mount stiffness, crankshaft torsional stiffness, cylinder position and gas pressure.
--------------------------------------------------------------------------------
2nd order
-minor component, propeller dynamics on 2-blade propeller. Rotating balance. Connecting rod couple (slight - subtracts from inertia) and propeller phase angle.
-major component, inertia.
-related orders 1,4,5,6.
-cylinder influence, none.
-note two blade propeller phase angle in relationship to forward crank pin may have slight influence. Inertia forces increase with engine speed.
--------------------------------------------------------------------------------
3rd order
-minor component, gas pressure (counteracts inertia). Phase angle of 3 blade propeller. Aerodynamic excitation of 3 blade propeller.
-major component, inertia. Third order is a major order.
-related orders 6,9.
-cylinder influence, none.
-note: Amplitude increases with rpm. Amplitude decreases with gas pressure. Crankshaft resonant frequency just above red-line rpm. Propeller phase angle in relationship to phase angle of 3rd order harmonic may increase, decrease, or have a neutral affect on 3rd order amplitude. Inertia forces increase with engine speed.
--------------------------------------------------------------------------------
4th order
-minor component, inertia 1/3 of gas pressure and counteracts. Slight connecting rod couple adds to gas pressure.
-major component, gas pressure (torque)
-related orders 1,2,5,7,8.
-cylinder influence Continental 1,2,3,4,5,6, Lycoming 5,6,3,4,1,2.
-note: Amplitude increases with manifold pressure on engines without 4th order counterweights (pendulum absorbers).
--------------------------------------------------------------------------------
4-1/2 order
-minor component, none.
-major component, gas pressure (torque).
-related orders 1-1/2, 7-1/2,
-cylinder influence, may vary if engine has 4-1/2 order counterweight (pendulum absorber): Continental: 1,2,3,4,5,6, Lycoming: 5,6,3,4,1,2.
-note Counterweights (pendulum absorbers) may be used to absorb 4-1/2 order torques. Amplitude increases with manifold pressure on engines without 4-1/2 order counterweights (pendulum absorbers).
4-1/2 order vibrations are especially hard on propellers.
--------------------------------------------------------------------------------
5th order
-minor component, 8% inertia.
-major component, gas pressure (torque)
-related orders 1,2,4,6.
-cylinder influence, may vary if engine has 5th order counterweights (pendulum absorbers): Continental 1,2,3,4,5,6, Lycoming 5,6,3,4,1,2.
--------------------------------------------------------------------------------
6th order
-minor, component, less than 1% inertia, less than 1/2 of 1% connecting rod couple.
-major component, gas pressure (torque).
-related orders, 3,6,9,12.
-cylinder influence, none.
-note: Counterweights (pendulum absorbers) absorb 6th order torque. Vibration measurements may vary depending on manifold pressure and engine rpm. Especially with high manifold pressure at resonant rpm. (This is why some engines have limits on high MAP with very low RPM, in a certain range.)

The above is for direct drive, but it does give us a clue what the PSRU is sitting in the middle of. The PSRU has its own harmonics or major orders as well. This all has to match the Prop and engine. Bolting up a new prop or engine to a PSRU is not trivial. There may be new "interactions" between vibration modes that may not be desirable. Also turbo or supercharging can have affects on all components.
 
Last edited:
Does this apply to the STI?

Today we further simplified the Fuel injection system of the engines by eliminating the need for the throttle position sensor (TPS). We have kept it until now, for good measure, but enough testing has shown it to be of no practical value with the engine in an airplane. It will save 3 wires and one pin on our newly designed engine harness and firewall Cannon plug....
Hi Jan,

Does this apply to my STI engine as well, or is this only for engines using the new ECU?

Thanks,
Mickey
 
<<As Jan pointed out, prop balance is important (1st order). >>

Sure it is, but the forces generated are translational, not torsional.

<<The PSRU as a middle man has a hard act to perform. It has to deal with both the unique harmonics of the engine and prop and it's own......The PSRU has its own harmonics or major orders.>>

Sorry George, no, it does not. "Harmonics" refer to the forcing frequencies, the many inputs to overall vibratory shaft torque variation. "Order" is merely the number of times per revolution that a particular harmonic introduces a torque variation. The PSRU itself has no way to force a vibratory shaft torque variation.
 
David since you correct me (and you are usually right) :D one clarification, MT-Propeller was founded in 1981 by Gerd Muehlbauer, whose prop experience prop design goes back to 1968. But you are right the ideas he uses (natural composite ie wood) go back the the 1930's, pre-WWII Germany.

Not this time, George - you are correct. MT does not go back to WWII but the technology does.

Your post is way over my head with regard to the various orders of vibration. I do know that vibration was the cause with the Comet and the Lockheed Electra disintegrating in flight after production and entering service. Perhaps those events motivated research to tell us what is known about vibration today. The later jets had no such problem.

With regard to the MT-PSRU-Subaru combination, the vibration source is the prop. I say that because it can be smoothed out by dynamically balancing it. So what order of vibration is that?
 
What?

DanH: Sure it is, but the forces generated are translational, not torsional.
A PSRU transfers torsion does it not. "Translation force"? Is that gear box rattle of helical gears which have a thrust component? I did not know a gear would resonate. I thought the shaft it was on would resonate in bending and torsion as well. A gear box with balanced forces, like a planetary spur gear would solve radial and thrust forces, right? All components, case, gear and shafts should be very stiff in torsion & bending. You got me.

<<The PSRU as a middle man has a hard act to perform. It has to deal with both the unique harmonics of the engine and prop and it's own......The PSRU has its own harmonics or major orders.>>
DanH: Sorry George, no, it does not. "Harmonics" refer to the forcing frequencies, the many inputs to overall vibratory shaft torque variation..
Well I took vibration analysis in mechanical engineering school. I did graduate and went on to grad school, so if its OK with you, lets keep this mistake our little secret, other wise they will come and take by diplomas away. :D

Sorry for the sloppy language. Lets see if I can say this right.
-Harmonics is the natural frequency all structures have, which has the tendency to oscillate the structure at maximum amplitude, also called it's resonance or resonate freq. It's not really the forcing function. Harmonics has nothing to do with "forcing function", which is a mathematical term to shape a boundary constraint and simplify vibration problems (laplace transforms).
-Frequency is the number times a structure oscillates per second. Order is the number of oscillations per "event' like one RPM.
-Bottom line, why we care about all these oscillations? It can fatigue structure to failure or the oscillations can get so large ultimate loads can be exceeded and failure occurs that way.



"The PSRU itself has no way to force a vibratory shaft torque variation."

Not sure what that means. PSRU transfers torque from the engine to the prop. I believe it might not be critical but there is torsion or shaft twist at some frequency, it just may not be critical. Everything has a natural freq or node. Bottom line is to say out of that resonate freq. The "issue" is the prop, PSRU, engine and engine case, engine mount and airframe is a vibratory system and they all have different harmonics. It's all connected and where our little helper comes in, dampening.

I think this is the "HOT TOPIC", what kind of dampening can solve a very complicated problem of three separates systems connected, all with different harmonics or natural freq. Well I want to say if the basic match of components are bad dampening is not going to save you. You have to be close. Clearly Lyc needed to add helpers from some applications.

I know the prop affects the engine when directly connected.
I know the engine affects the prop when directly connected.
I know when a PSRU is in the middle, the prop and engine affects the PSRU.
I know the PSRU does not isolate the prop and engine from each other completely.*
(* unless extensively dampened or isolated, which is often too complicated, expensive or heavy)

Now what is critical and what is minor? That is where the engineer comes in. You can only hope to simplify and minimize the affects of harmonic (such as avoiding some RPMS). Even a GE CFM6 or RR RB211 jet engine have harmonics. You hear it every time you start. It's just a fact of life. The best bet is minimize the affect with clever design.

Harmonics has been the down fall of many structures and systems designed by professional engineers. It's not easy. Test test test.
 
Last edited:
Not Just Balance Anymore

Not this time, George - you are correct. MT does not go back to WWII but the technology does.

Your post is way over my head with regard to the various orders of vibration. I do know that vibration was the cause with the Comet and the Lockheed Electra disintegrating in flight after production and entering service. Perhaps those events motivated research to tell us what is known about vibration today. The later jets had no such problem.

With regard to the MT-PSRU-Subaru combination, the vibration source is the prop. I say that because it can be smoothed out by dynamically balancing it. So what order of vibration is that?

Wow David, that's a great insight into the issues here, their understanding,
and the solutions. Since Dan and George are currently engaged in a "credential dispute", maybe I can help.

What YOU are talking about is prop balance. Making the blades equal in weight so as to reduce radial imbalance and vibration. A fairly easy solution if you have balancing equipment.

What this thread has been addressing is not "prop balance", rather harmonic vibration. This comes about when the power pulses of the engine feed into the reduction shafting and gearing, and the elasticity of the system either cancels out or intesifies the pulse in a sort of cumulative way. If the system is designed wrong, the harmonic vibrations are intesified exponentially and cause the destructive failure of the gearbox, ie, Gen 1 and 2.

Now, I am not an engineer, but I can speak PILOT to you, and if you think radial balance is the only issue these gearboxs face, you should get on the phone tomorrow with Jan for a heart to heart.
 
Last edited:
....Now, I am not an engineer, but I can speak PILOT to you, and if you think radial balance is the only issue these gearboxs face, you should get on the phone tomorrow with Jan for a heart to heart.

I know prop vibration is not the only issue but one that has been emphasized because of a concern it is bad for the PSRU, so we balance the prop.

I also know harmonic vibration is a different beast but know nothing about designing it out of an airplane or controlling it. It would seem, as smooth as the 6 cylinder Subaru is, there is no danger of harmonic vibration originating with it but that is wishful thinking and I know it. For sure it is OK in an auto, but add a gear box and prop, it gets complicated as has been stated.

The fact that MT found the combination to their liking, as far as their prop is concerned, is of some comfort. Plus there are quite a few hours on the system with the MT and Quiti props at just about every rpm with the 2.5 and H6 engines. I am not privy to what testing Jan did (or Van's with Lycoming installations) but I am sure he did some work in this area.
 
I am not privy to what testing Jan did (or Van's with Lycoming installations) but I am sure he did some work in this area.

We did and also, as much as real world testing is dismissed in this thread, that is what first and foremost what we believe in. Be assured there are always reduction drive units with more time, heavier loading and in much worse test conditions than those out flying.

Also, Dave, do not dismiss your own feelings for what "seems" right while flying your airplane. If your reduction drive is running cool, smooth and not making metal, you are doing your part.

The elements in favor of our system has also been ignored and barely mentioned. There is a huge benefit to lighter propellers, heavy flywheels, springs, oversized gears and shafts, short shafts, small engine cylinders, many cylinders, fuel injection, etc., to keep all this working safely together. The forces discussed are always there, our job is to minimize thir magnitude.


Jan
 
>>....the forces generated are translational, not torsional.<<

An out-of-balance propeller can be modeled as a rotating disk on a shaft. If we introduce an additional mass so as to unbalance the rotor, the resulting radial force F translates from the Y axis to the X axis to the -Y axis to the -X axis and back to the Y axis to complete the cycle:



The vibratory motion in either axis may indeed be graphed as a classic harmonic function....



....the peaks positive and negative representing the displacement within that axis.

However, that detail is of no interest in a discussion of torsional mechanics; the rotating force F imparts no change on shaft angular velocity or torque. The effect is to bend the shaft, not twist it.

As for the ability of the PSRU itself to generate harmonic (or periodic) variations in angular velocity or torque....well, maybe you could list a specific in-context example. Offhand I can think of two, but didn't list them previously because neither are relevant to a discussion of the Egg system; standing waves in a belt PSRU and gear mesh in high speed turbine systems. Can you add a relevant example?
 
<<I know prop vibration is not the only issue but one that has been emphasized because of a concern it is bad for the PSRU, so we balance the prop.>>

I noticed the dynamic balance requirement in the Egg literature. Jan, could you please share the specific physical reason for the requirement? If a customer is so foolish as to ignore the requirement, what will be the result?

David, regarding the MT prop survey, you mentioned the customer's name in another thread. What was it again?
 
DanH David said:
Andy Parish flew to Deland for the vibration survey with MT engineers. I am not sure if it was I who mentioned his name but it was Andy Parish. He is a retired USAF F-15 instructor and flies the RV-7A Subaru H6 he built.
 
We did and also, as much as real world testing is dismissed in this thread, that is what first and foremost what we believe in.

The elements in favor of our system has also been ignored and barely mentioned. There is a huge benefit to lighter propellers, heavy flywheels, springs, oversized gears and shafts, short shafts, small engine cylinders, many cylinders, fuel injection, etc., to keep all this working safely together. The forces discussed are always there, our job is to minimize their magnitude.


Jan

In fact, this is what most OEMs in the auto industry use today. While these have been extensively validated, they have also ended up with similar configurations- heavy flywheels, spring or elastomeric dampers/ absorbers, big gears and stiff shafts in their drive trains. Both the GM LS6/ Corvette and BMW setups I have here in the shop use these components.

While instrumented testing should be part of these developments, there is no substitute for getting flight time on these systems. Sometimes all the theory in the world and even vibration testing will not uncover all potential faults in a design- exactly why the OEMs also do extensive real world testing.

We learn and improve from experience on any given system. 15 years and thousands of flight hours experience in this area should not be discounted.
Potential customers are more likely to ask how many flight hours vs. failures there have been when considering a PSRU/ engine package than looking at a bunch of data they don't understand.

Jan has made steady improvements to his packages as power has crept up from the early days of the atmo EJ25s. Mistakes have been made (as we have all made and even big companies like Textron make) but the latest packages are better than ever- on almost every front. This is a huge task, offering FF packages like this. Only those who have done it can appreciate the amount of time and effort that goes into such a task.
 
<<Potential customers are more likely to ask how many flight hours vs. failures there have been when considering a PSRU/ engine package than looking at a bunch of data they don't understand.>>

Ok, let's find out. It is really a marketing information issue, so I've opened a separate thread with a poll.

Please note the poll asks participants to consider the questions in the context of all vendors, not just Egg. The poll result may help all vendors with their marketing decisions.

Returning to technical; Jan, how about those "propeller balance required" questions?
 
I noticed the dynamic balance requirement in the Egg literature. Jan, could you please share the specific physical reason for the requirement? If a customer is so foolish as to ignore the requirement, what will be the result?

The entire airplane and everything in it will fatigue from prop vibration. Bearing life in the PSRU is also reduced.

Jan
 
Thanks but I can't think of any examples

>>....the forces generated are translational, not torsional.<<

Ahaaa got it, gyroscopic precision, I totally forgot, right hand rule, Doha! If the planes maneuvering (angular velocities) in yaw and pitch is not too aggressive, than translational force should not be significant, as long as thrust bearing is preloaded from prop thrust. Staying out of certain operational power & maneuvering regions is goodness. Determining those is the trick. It could be an easily avoided condition that is doing damage. Avoid that condition, gear box last indefinitely. Not unlike stay out RPM's for metal props on Lycs. If there is no practical impact on normal operations its a fair compromise. Ultimate aerobatics might be out for PSRU's (unless they're designed for it). Besides temperature, a permanent accelerometer on the PSRU would be nice. If vibrations are too high, investigation is needed. It could be prop, engine or PSRU. Freq and amplitude might determine source of vibs. Vibration from the prop (as Jan says) is added stress that is not needed or desirable. In other words "Super Position" of loads, they add in the vibration world. You need to avoid potential bad vibs where you can. Prop balance is one of them and it's easy. I balanced my RV4 prop three times to get it perfect. Although it was a lyc/hartz combo it made a difference.

One BIG issue with certified gear boxes in planes from my experience, both piston and turbine, is you should keep a preload or thrust on the prop to "pre load" the thrust bearing. On turbines the prop controller and engine torque are talking to each other, with a device called NTS or "negative torque sensing". NTS is in part for engine out, auto-feather and in part to protect the gear box. NTS rears its ugly head if you reduce power to min and descend too fast, so you DON'T do that. Back driving the engine through the prop, ie low power high rate/speed descent, is a no-no in some geared piston Continentals. Part of this is BACK-LASH slapping between the gears, which can cause impact damage. Bottom line even certified PSRU's have operational limits (some limits are official and some are good practice). Than you say limit backlash to zero? Well that's expensive, impossible and creates other problems. Gear boxes are part engineering, part art.

As for the ability of the PSRU itself to generate harmonic (or periodic) variations in angular velocity or torque....well, maybe you could list a specific in-context example. (Well I recall the long prop drive shaft on the BD5 had this issue? May be that was bending only or bending torsion couple?) Offhand I can think of two, but didn't list them previously because neither are relevant to a discussion of the Egg system; standing waves in a belt PSRU and gear mesh in high speed turbine systems. Can you add a relevant example? (I doubt it, ha ha, I'll do sum thunking on it.)
Thanks Dan for explaining it to me, I understand your perspective. I'm way rusty on this stuff. Keep on brain storming, interesting stuff. Cheers
 
Last edited:
George,
<<Ahaaa got it, gyroscopic precision>>

Ordinary centrifugal force.

In fairness, I may have confused you with the use of the word "translational". Translational usually refers to motion of a mass where all particles in the mass maintain their relationship to each other within the X-Y-Z reference. In this example only the particle at the center of rotation has true translational motion.

I know you studied this stuff forty years ago. As an informal student I've been studying it on and off for the last ten years or so.....and I suspect neither of us can remember what we had for breakfast yesterday <g>
 
Last edited:
Sometimes it is interesting to run some numbers just to satisfy curiousity. Here I got curious about the forces on a front bearing due to an out-of-balance propeller.

First we need a fair yardstick for the worst out-of-balance we're likely to find in the field. 1.0 in/sec is the accepted maximum vibratory velocity at the beginning of a dynamic balance session. If the first run shows more than 1.0 in/sec, you're supposed to remove the prop and do a static balance. Thus 1.0 in/sec is the absolute most we should expect with a new prop; they did a static balance at assembly.

At 2500 RPM (41.666 hz) and 1.0 in/sec vibration, acceleration in G is 0.68.

The measuring device is mounted on the nose of the engine, the mass being accelerated. Pick whatever number you like. I'll use 250 lbs here to purposely be excessive; the actual value would require some study of engine mass CG and mounting restraints.

250 lbs x .68 = 170 lbs radial force at the propshaft, worst case.

A good dynamic balance will reduce velocity to 0.1 or less, so the radial force would decrease 10x to 17 lbs or less.

So, dynamic balancing is clearly worthwhile. However, even the maximum likely unbalance may not be a major culprit in propshaft bearing failure; the load is low compared to other routine bearing loads; side thrust due to gears, the gyro load, or even axial thrust.
 
Last edited:
Not as technical as this thread has become, but my buddy flew his 9A for the 1st time after installing the Gen3 gearbox. He has the 2.5l Subbie. He reports he used to have a climb out of 1,000 - 1,200 FPM now he has 2,000! It is cold here and that may have some affect, but he is really pleased with the new found performance. Speed tests will come today after he installs his wheel pants.

Another happy Subbie / Egg customer.
 
Good to hear, but...

That would indicate an improvement of anywhere from 66% to 100% in performance. Since the power loss through a gearbox is typically negligible (assuming a few percent at the most), such an improvement could mean one of two things:
  1. The prior gear ratio was such that the prop was operating well outside its efficient range of RPM (which would mean the setup was significantly wrong and needed correcting anyway), or...
  2. The improvement in performance is a result of the engine producing more power and the wings producing more lift (i.e. the cold weather you discussed).

Glad to hear your friend is happy, but to be a useful (quantitative) data point, all such performance figures would need to have extraneous variables removed (same w/b, engine settings, density altitude, a/c configuration...flaps, fairings, etc).

Thanks for sharing. It's certainly useful from the standpoint of knowing that another satisfied customer is out there. :)
 
This thread has been very informative but I have a few questions.

As I see it, there at least five ways to connect a prop to the power source:

1) Direct Drive
2) Gear Drive
3) Belt Drive
4) Shaft Drive (Quill shaft)
5) Viscous coupled shaft (Including air drive)

I've tried to list the "High Level" Pro's and Con's of each, what am I missing? Are there other options?

Direct Drive:
When talking auto engine conversions, Direct Drive is usually not an option due to the high RPM's generated. (There are some exceptions, mostly to do with replica aircraft.)
Pro: Simple and easy
Con: Bearing life is a problem as these engines are generally not designed for the prop loading.

Gear drive:
Pro: Get's the RPM right in the required range (RPM selection)
Con: All the stuff you guys have been talking about (Loading, weight, cooling, etc.)

Belt Drive:
Pro: RPM selection, light, simple, and absorbers, power pulses, light weight when compared to gears, no cooling oil required.)
Con: Belts break.

Shaft Drive:
Pro: ???
Con: Quill shafts are expensive and usually only work where this a long disance between the power source and the prop.

Viscous coupled shaft (Including air drive)
Pro: No physical connection between the power source and the prop
Cons: Cost
 
Status
Not open for further replies.
Back
Top