VansAirForceForums  
Home > VansAirForceForums

- POSTING RULES
- Donate yearly (please).
- Advertise in here!

- Today's Posts | Insert Pics

  #1  
Old 03-30-2008, 01:27 AM
gmcjetpilot's Avatar
gmcjetpilot gmcjetpilot is offline
 
Join Date: Jan 2005
Posts: 4,280
Default Lycoming or Alternative Which is MORE reliable?

In another thread "statistics" where brought up. It was implied Lycomings are falling apart and alternative engines are bullet proof, at least the bottom end. So I decided to look my self. The goal was to filter out as much as I could to find the number for LOSS OF POWER, for each engine type, Lyc, Subaru, Mazda, Chevy, Ford. The Mazda, Ford and Chevy are fairly easy to search since there are so few accidents. Subaru had the highest total accidents of the Alternative Engines, due to fleet size I assume.


Using the NTSB data I tried to find the answer. Of course fleet size and hours of utilization is hard to pin down. This is how I attacked it.

For the alternative engines I searched for the engine name and variations. Than I went through each probable cause to determine if the engine was the cause and what part.

I eliminated any Undetermined, WX, Pilot Error or Airframe issues, so if the math does not add up that is why.

For the Lycoming I could not just type that in. So I searched the reason or probable cause field with key words such as: Crankshaft, Connecting Rod, Rod Cap, Bearing - failure, fatigue and fracture. Than I went through the results and eliminated other engines and as with the alternative engines, undetermined, WX, pilot error and airframe. The data base seems to go back to 1985, although I searched back to 1970. So assume this is 22 years worth. If it was not in the database I did not included it. No claim of perfection her just a good broad swipe.

Accidents due to internal Engine Failure

Lycoming (all)
32 total accidents
17 crankshaft failure, fatigue & Fracture, 1 experimental
15 rod, rod cap, rod bolt, failure, fatigue & Fracture, 2 experimental

[Note: maintence or improper engine assembly, e.g., under torquing rod bolts was included. There where at least 6 due to improper maintenance or manufacture defect. 3 failures in Experimental's.]

Breaking the "32" down to just typical "RV" engines failure/fatigue/fracture:
360s 4 crank & 4 rod failure, fractures and fatigue - 8 total
320's 1 crank & 1 rod failure, fractures and fatigue - 2 total
235's 0 crank & 1 rod failure, fracture and fatigue - 1 total


Subaru
42 accidents
16 pilot error
8 Undetermined
18 total ancillary & mechanical failures:
12 Ancillary
1 Vapor lock
1 Cooling lines
1 Waste gate servo
2 Fuel / Pump failure
7 electical / ignition

6 mechanical
1 PSRU failure*
1 Bearing failure
1 Belt timing gear
2 valve train or exhaust valve
1 Connecting rod failure over temp
*Consider PSRU as part of the engine.

Chevrolet / General Motors
13 total accidents and (1) major engine failure, (2) fuel related and (1) reduction drive.
The following Chevy or GM engines used: 5-V8, 6-V6, 1-4 cyl, 1-corvair


Mazda
Total 7 accidents: (1) Cooling, (1) Lube System, (1) exhaust/prop, (3) Carb/Fuel sys


Volkswagen
36 total accidents: (2) cranks, (1) lube system, (3) Ignition, (3) Carb/fuel


Ford
Total 8 accidents: (1) Cool, (1) Lube external, (1) Ignition, (1) Fuel (carb ice)
(2) Model-A's, (6) V6

************************************************** *******

Totals - accidents due to core engine failure
Lyc 32 (11 for 235, 320 and 360's only)
Subaru 6
Chevy 1
Mazda 1
VW 3
Ford 0

The above ignored electrical, cooling, ignition, turbo, timing and fuel issues. Of the alternative engines, adding the electrical, cooling, fuel and misc, you would have this.

Total accident due to loss of power core + ancillary
Subaru 18
Chevy 4
Mazda 6
Ford 4
VW 9

The higher number for the Subaru & VW is no doubt related to a larger fleet size. It is clear that the ancillary items on alternative engines, electrical, cooling, fuel and misc cause more loss of engine power than rods and cranks. The VW had almost as many accidents as the Subaru, but most where undetermined, pilot or other.

This is not a PERFECT cut at it, only a broad brush approach. I could have missed other "core" causes of loss of power, but a good faith effort was made.

Of all the Lyc powered Experimental planes, there where only (3) hard engine failures.

I did not search for fuel issues, starvation, carb ice, vapor lock and so on with the Lyc but there are 100's of fuel related accidents, but I did not include fuel starvation or carb ice with any of the above.

The fleet size of the Lycoming and Continental is many times larger, 200 times (?) more than the alternative engine fleet, especially in the high performance fixed wing plane category.

There are 66 times more registered Lyc 320/360's than Subaru engines with the FAA and only 2 times the failures. HOURS FLOWN? I can only imagine that the Lyc flys more but that is hard to prove.

Do all the other Lyc's count in the statistics: 235, 290, 480, 540, 720 and radials count as well? There are near 65,000 to 80,000 Lycs on the books (I think) with the FAA, not including turbines. How many are flying I don't know, but the 80 core failures for all certified engines (Lyc, TCM, Frank, P&W, Wright.....) over approx 70 years is about one a year.

Many of the Subarus are the lower power E85's used on Gyrocopters, but you don't read about all engine failures in the NTSB database, since many just Gyro down; however many did make the list of 42 accidents for Subaru powered "planes".

It is safe to say Lycoming has no (or few) cooling lines, belt or electrical causes of power loss in the data base. Belts took down at least one or two Alt engine planes, since they drive "internal" timing or cams. I did not check the Lyc but belts tend not to be critical as in engines with a belt driven valve train.

Conclusion: Draw your own, But I don't see a systemic problem with Lycoming or Continental reciprocating lower end or valve train. Both Continental and Lycoming went through a Crank production QC debacle at one point in time. You can see from the data the Continental crank problem on the C210's, early 1990's, and on the Lyc, late 90's, when their crank issue appeared. There are a handful of crank failures due to manufacture defects. If not for this transient QC issue, there would be about 5 less crank failures. Some of the above crank/rod failures where due to prop strikes or poor maintenance clearly, a cause and affect. I did not remove them from the count. The lesson is proper maintence and overhaul.

Although my search only went to 1985 it included all the planes in service made before than till now. If you parse the data even further you will find may be a dozen real holly failures out of the blue or due to manufacture failures for ALL aircraft engines. That ain't bad. Against the fleet size and time in service (since 1960) that's acceptable.

Fleet size is hard to pin down but from a "FAA registered" data base got the following: Lyc 320/360's - 40,000 (at least 65k for all Lyc horz recips), Subaru - 609, Ford - 121, Mazda - 4 (sounds low), Volkswagen - 1,235, Chevy (none listed).
__________________
George
Raleigh, NC Area
RV-4, RV-7, ATP, CFII, MEI, 737/757/767

2020 Dues Paid

Last edited by gmcjetpilot : 03-30-2008 at 01:54 PM.
Reply With Quote
  #2  
Old 03-30-2008, 02:52 AM
rv8ch's Avatar
rv8ch rv8ch is offline
 
Join Date: Feb 2005
Location: LSGY
Posts: 3,133
Default accidents vs. failures

Hi George,

Interesting and very valid analysis of the FAA data. One observation I would like to throw out there is that we all know that there are lots of engine failures that do not end up in the database.

Because of this, one could argue that we really don't have accurate numbers.

One thing is still clear to me - by far the main cause of accidents is the judgment of the pilot flying the aircraft.
__________________
Mickey Coggins
http://rv8.ch
"Hello, world!"
Reply With Quote
  #3  
Old 03-30-2008, 06:26 AM
David-aviator David-aviator is offline
 
Join Date: Feb 2005
Location: Chesterfield, Missouri
Posts: 4,514
Default

Interesting read, George. If anything it supports a conclusion that simple is better when considering a mechanical contraption like a general aviation aircraft engine. There are many old engines out there doing their thing very well year after year.

I am intrigued by and drawn to the Subaru engine because of its incredible design success in an auto but do wish it were a bit more KISS in an airplane. There are a gazillion electrons flowing to keep it humming and I keep hoping they all know where to go and what to do.

The difference from a magneto to the typical electronic ignition system in modern auto engines is fascinating. So far it has proven quite reliable - at least for me. The other systems needed to keep it running are getting more and more reliable also, so we just keep on doing it to to fix what is wrong and get it better. Same can be said for Lycoming over the years.
__________________
RV-12 Build Helper
RV-7A...Sold #70374
The RV-8...Sold #83261
I'm in, dues paid 2019 This place is worth it!

Last edited by David-aviator : 03-30-2008 at 08:23 PM.
Reply With Quote
  #4  
Old 03-30-2008, 07:06 AM
TSwezey TSwezey is offline
 
Join Date: Oct 2005
Location: Savannah, GA
Posts: 1,849
Default

Thanks for data George. As David stated KISS seems to be the key to reliable. Alt engines are not nearly as KISS as the Lyco. Will they be made 100 % bulletproof probably not. The best we can do is try to cover every angle that could possibly go wrong. If there are any failures find the problem and fix it. There are several potential areas of failure in my engine: Loss of electrical power, loss of coolant, loss of belt (not for timing but for the water pump), PSRU failure and computer failure. Only testing will truly evaluate how reliable these systems are. I plan to fly very conservatively for quite awhile.
__________________
Todd
N110TD
RV-10 Vesta V8 LS2/BMA EFIS/One formerly flying at 3J1 Hobbs stopped at 150 hours
Savannah, GA and Ridgeland, SC
Reply With Quote
  #5  
Old 03-30-2008, 10:11 AM
rv6ejguy's Avatar
rv6ejguy rv6ejguy is offline
 
Join Date: Mar 2005
Location: Calgary, Canada
Posts: 5,729
Default

Wow George, you must have spent a few hours crunching that data! Very interesting.

I'm not quite clear on what you searched under , all or experimental but I spent about an hour counting 72 catastrophic failures in only 5 years by typing in only "connecting rod", searching all types and sifting the causal conclusions. I only counted the valid ones. I was surprised how many others were caused by improper assembly or bad maintenance- maybe 50%!.

My numbers don't agree with your's at all and at a quick glance again today, I only had to go back to Jan. 2006 to find 17 catastrophic failures caused by straight mechanical reasons. There were many others where an unexplained loss of oil during flight caused rod failures. I did not count any radial engine rod failures here. Failures were rods, cranks and valves.

There were 14 crank failures alone in the late '90s alone in the first Lycoming crankshaft debacle. I know of 3 catastrophic failures ( 2 broken rods and a jug separating from a cracked case- 2 Lycomings and 1 Continental) just on aircraft operating from my airport in the last 4 years. This is a very small sample and shows that there are a lot more than one of these happening per year fleet wide, in North America.

We can't blame the engine for failing due to overheating, lack of oil or poor maintenance/ overhaul procedures. As I sifted the stats, these came up again and again. This seems little different than the race car engine world. Some people just should not assemble or work on engines even if they appear or think they are qualified. Clearly some of the A&Ps doing work on engines are plain scary.

There are many ways to sift the data but I do agree with your basic conclusion that you are more likely to have a power loss condition using auto power than a Lycoming.

In 2008 so far, we've had 4 power loss accidents in RVs, 3 Lycoming and 1 Subaru. They do happen, stay up on your forced landing skills. The recent stat mentioned on VAF shows that a majority lose control after power loss. Keep the nose pointed down and don't turn back to the field unless you have a lot of altitude.
__________________

Ross Farnham, Calgary, Alberta
Turbo Subaru EJ22, SDS EFI, Marcotte M-300, IVO, Shorai- RV6A C-GVZX flying from CYBW since 2003- 441.0 hrs. on the Hobbs,
RV10 95% built- Sold 2016
http://www.sdsefi.com/aircraft.html
http://sdsefi.com/cpi2.htm



Last edited by rv6ejguy : 03-30-2008 at 12:00 PM.
Reply With Quote
  #6  
Old 03-30-2008, 10:42 AM
Jconard's Avatar
Jconard Jconard is offline
 
Join Date: Apr 2005
Location: Twin Cities
Posts: 438
Default

Ross,

Georges numbers would not include the Canadian stuff at all because the NTSB does not investigate those.

I am willing to assume that unreported accidents occure proportionally, hence eliminating that as a variable.

Not sure if georges numbers were restricted to the 320/360, but based on fleet size, I think they were, I believe the power outs you are describing are with the big sixes, based on your previous posts.
__________________
John C Conard
J.D.
Citabria 7KCAB (Former)
RV-7 Flying
Reply With Quote
  #7  
Old 03-30-2008, 11:52 AM
rv6ejguy's Avatar
rv6ejguy rv6ejguy is offline
 
Join Date: Mar 2005
Location: Calgary, Canada
Posts: 5,729
Default

Actually there are a number of crashes in the NTSB database which occur on foreign soil and no not even involve US registered aircraft. Not sure why that is. In fact, I stumbled on one yesterday that happened operating from our base where 3 were killed and a rod failure was the cause.

There are a number of other well publicized accidents where catastrophic failure was the root cause in other countries. The double engine failure on a Navajo in Australia a few years back comes to mind: http://www.warmkessel.com/jr/flying/td/jd/57.jsp

I too would agree that there are many other engine stoppages which go unreported and don't lead to an accident. This means there are a lot more happening than the NTSB stats show- all the more reason to be sharp with your engine out skills.

I'd also agree that there appear to be far more problems with high hp turbocharged six cylinder engines than 320/360 Lycomings.
__________________

Ross Farnham, Calgary, Alberta
Turbo Subaru EJ22, SDS EFI, Marcotte M-300, IVO, Shorai- RV6A C-GVZX flying from CYBW since 2003- 441.0 hrs. on the Hobbs,
RV10 95% built- Sold 2016
http://www.sdsefi.com/aircraft.html
http://sdsefi.com/cpi2.htm



Last edited by rv6ejguy : 03-30-2008 at 11:58 AM.
Reply With Quote
  #8  
Old 03-30-2008, 01:18 PM
TSwezey TSwezey is offline
 
Join Date: Oct 2005
Location: Savannah, GA
Posts: 1,849
Default

I believe the NTSB reports include any American made airplane regardless of its registry.
__________________
Todd
N110TD
RV-10 Vesta V8 LS2/BMA EFIS/One formerly flying at 3J1 Hobbs stopped at 150 hours
Savannah, GA and Ridgeland, SC
Reply With Quote
  #9  
Old 03-30-2008, 01:52 PM
gmcjetpilot's Avatar
gmcjetpilot gmcjetpilot is offline
 
Join Date: Jan 2005
Posts: 4,280
Default For Just 235's, 320's and 360's

Quote:
Originally Posted by Jconard View Post
Ross, Not sure if georges numbers were restricted to the 320/360, but based on fleet size, I think they were, I believe the power outs you are describing are with the big sixes, based on your previous posts.
Ahaa the usual brain trust responds, ha ha. Yea I am glad you enjoyed it. It did take time and I don't "own" it and feel free to add, comment or criticize. One piece of wisdom I'll parrot "It's only as strong as its weakest link." Clearly from the stats that weak link is the pilot, sorry to say. That includes just making the decision to take off with known issues. Clearly as pilots and builders we have more responsibility or I like to say control over our fate and our passengers fate.

No, the Lyc numbers where for ALL Lycs from 50hp (O-145) to 450hp Lycoming TIGO-541

But it is easy to break out the 235's, 320's and 360's. Breaking down the 32 total for all Lycs to typical RV Lycs:

360s - 4 crank & 4 rod failure, fractures and fatigue - 8 total
320's - 1 crank & 1 rod failure, fractures and fatigue - 2 total
235's - 0 crank & 1 rod failure, fracture and fatigue - 1 total
.
Total 11 out of the 32 for small disp 4 cyl Lycs
The stats are far from perfect. You have to than look at each one and evaluate the cause, such as prop strike, manufacturer defect, maintenance or the "surprise". The latter act of fate is more rare.

You could do some PhD thesis on NTSB statistics. There are holes in the data and accidents not reported for sure. For a broad brush I think it shows a few things.

One thing that jumps out is electricity is critical to planes that need it for the engine to turn. Not being sarcastic, but true right. There where 7 pwr loss events for the Subaru due to electrical wiring including one ECU going wacky. The other engines, VW, Ford, Chevy, Mazda all suffered at least one or more electrically related loss of power events (recorded). I suspect as more Lyc flyers go to dual electronic or FADEC, we'll see Lycs having electrical loss of power accidents.

Zero failures across the board would be nice but unlikely. For many reasons, regardless of the engine, zero failures is not likely. Safe to say, if starting with a good part, installed, flown and maintained properly your "statistics" improve drastically, provided you don't do a stupid pilot trick, which no pilot is immune from either.
++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++++++++++++++++++++++

BLAST FROM THE PAST? (pics from PrimeMover.Org)

You want water cooled? 1,200 HP flat 12 cyl O-1230

1,200HP not enough, two SO-1230's put into a "H" configuration thru a gear box for 2,300HP XH-2470

Still not enough liquid cooled power? How about the XR-7755,
From 7,755 cubic inches (bore of 6.375", stroke of 6.75") and 7,050 lbs it gave 5,000 horsepower in 1944; 7,000 horsepower was the target. Nine liquid cooled in-line four cylinder engines about a common crankshaft. Two contra-rotating prop shafts. The camshafts each consisted of two sets of lobes. One set of lobes for takeoff, the other for economy cruise. The camshafts were shifted axially to switch lobe sets. 580 gallons of fuel per hour at takeoff power, BSFC of 0.43 at cruise. It was aimed for the B-36 but political pressures caused the B-36 to be fitted with the Pratt & Whitney R-4360 instead. Smithsonian's Garber facility has the last survivor. PIC and one more Picture PIC me in goggles (kidding not even born, but cool pic huh)

If there where as many GA planes as cars today or jets come later, you wounder what recip engines would have be developed for planes. It's fair to say from the highly experimental XR-7755 to the first O-360's and soon after 540's over a decade later, they knew how to make complicated engines. In the 1930's Lycoming was making supercharged water cooled V-12's.

A new light weight version of the Continental O-200 going back to the 1947 (C-90) is new again. It will be put into Cessna's first new GA plane, their C-162 LSA Sky catcher (is that a good name?). Can't they come up with something better (name and engine)? Apparently not. May be this still is a good configuration, flat 4 air-cooled direct drive?

What happened to CoolJugs? Last I heard they had dyno and manufacturing vendor issues. But is it the inertia of air cooled engines that restricts new technology, or is it that new technology (current state) has not earned its way onto planes? Clearly going from air-cooled to liquid means finding a place for the radiator'(s). If there was some skin technology that would be good, but pressure loss from running long small tubes plus weight would be a challenge. Combing the heat exchanger and engine in one saves space and weight (ie air cooling). Of the liquid cooled engines there where 3 accidents from three different engines. It's not the core of the engine but its a factor like electrical.

Lycs in a way, indirectly have been taken down by cooling, by pilots overheating the engine and causing failures, even later failures from residual damage. If not causing outright failure, a Lycs life span is reduced by two things: inactivity and overheating. I think a car's reputaion is in part due to being driven daily and not overheated.

Clearly a Lyc is for trained pilots not "mini van soccer mom" (no offense) simple, turn the key and drive it. Some see the future aircraft engine as turn the key and drive, go to jiffy lube once twice a year, simple. Even jets need some care. Jets now have "FADEC" that protects the engine during start and operations. Early jets could be turned into melted junk very easy if the pilot did not watch temps like a hawk. Electronics can help, but total dependence is another issue. Jet engines can go into a fail-safe mode with less protection (ie "Jurassic" mode) but still run. Jets once started run on their own if supplied fuel. They can even suction feed with their own mechanical pump at reduced power. There use to be cables running out to the engine, now just wires? Progress or opening door to more or different problems? To me farm tractor technology has its charm. I love electronic gadgets, but I know they run on smoke and mirrors. If you ever burn up electronics what comes out? Smoke. Therefore smoke is the power that runs electronics. I know how electronics work, but I know how a push pull cable work even better.

May be they (Lyc) learned stuff in the 40's. It is clear before jets the P&W radials where and still are working well. Into the 60's most regional airlines had pistons and some are still winning races at Reno. I am amazed. Personally when it comes to furniture, I like modern, danish, not antiques. Actually antiques creep me out a little. But antique engines I love? Go figure.... Pick your poison Gents.
__________________
George
Raleigh, NC Area
RV-4, RV-7, ATP, CFII, MEI, 737/757/767

2020 Dues Paid

Last edited by gmcjetpilot : 03-30-2008 at 03:34 PM.
Reply With Quote
  #10  
Old 03-30-2008, 03:21 PM
rv6ejguy's Avatar
rv6ejguy rv6ejguy is offline
 
Join Date: Mar 2005
Location: Calgary, Canada
Posts: 5,729
Default

Quote:
Originally Posted by gmcjetpilot View Post
Ahaa the usual brain trust responds, ha ha. Yea I am glad you enjoyed it. It did take time and I don't "own" it and feel free to add, comment or criticize. One piece of wisdom I'll parrot "It's only as strong as its weakest link." Clearly from the stats that weak link is the pilot, sorry to say. That includes just making the decision to take off with known issues. Clearly as pilots and builders we have more responsibility or I like to say control over our fate and our passengers fate.


You could do some PhD thesis on NTSB statistics. There are holes in the data and accidents not reported for sure. For a broad brush I think it shows a few things. One is electrical is critical to planes that need it for the engine to turn. Not being sarcastic, but true right. There where 7 pwr loss events for the Subaru due to electrical wiring including one ECU going wacky. The other engines, VW, Ford, Chevy, Mazda all suffered at least one or more electrically related power loss events. I suspect as more Lyc flyers go to dual electronic or FADEC, we'll see some Lycs having electrical loss of power accidents.

Zero failures across the board would be nice but unlikely. For many reasons, regardless of the engine, zero failures is not likely. Safe to say, if starting with a good part, installed, flown and maintained properly your "statistics" improve drastically, provided you don't do a stupid pilot trick, which no pilot is immune from either.[/indent]
I agree with the electrical thing being the most likely cause to bring an auto conversion down and I stated this yesterday. My accident was caused by a bad Denso clone alternator and my inattention to monitor the charging system- so pilot error really. Second was my improper evaluation/ design of the electrical system, thinking that if the alternator failed, I'd notice it and land so why did I need a backup battery?

Lessons learned-

1. A low voltage light on the panel is useless in direct sunlight.

2. Install a backup battery just is case everything else goes wrong.

3. Install an aural warning device for low voltage.

4. Use only genuine Denso parts.

4. Pay attention, dummy.

I think we already saw a dual EI failure last year on a Lycoming RV didn't we? Luckily it turned out ok and does not appear in the NTSB database.
__________________

Ross Farnham, Calgary, Alberta
Turbo Subaru EJ22, SDS EFI, Marcotte M-300, IVO, Shorai- RV6A C-GVZX flying from CYBW since 2003- 441.0 hrs. on the Hobbs,
RV10 95% built- Sold 2016
http://www.sdsefi.com/aircraft.html
http://sdsefi.com/cpi2.htm


Reply With Quote
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -6. The time now is 02:28 PM.


The VAFForums come to you courtesy Delta Romeo, LLC. By viewing and participating in them you agree to build your plane using standardized methods and practices and to fly it safely and in accordance with the laws governing the country you are located in.