BW 350 PSRU Failure, please share the word!

Discussion in 'Chevy' started by Andreas K, Aug 19, 2016.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Aug 19, 2016 #1

    Andreas K

    Andreas K

    Andreas K

    New Member

    Joined:
    Aug 19, 2016
    Messages:
    4
    Likes Received:
    5
    Location:
    Beautiful Wisconsin
    Fellow Pilots,

    I’d like to share the gearbox failure that I recently experienced in my Vans RV-10.

    https://drive.google.com/open?id=0B_...0RTM280TzAyaDg

    https://drive.google.com/open?id=0B_...mlmVjJWdEVibUk


    My setup:

    Airframe: Vans RV-10
    Engine: GM LS 7 (approx. 400 hp, 800 ft/lbs of torque @ 2700 prop RPM, WOT @ SL)
    PSRU: BW 350 (Bud Warren, Geared Drives), Cast Aluminum Case, 1:1.75 reduction ratio
    Propeller: Whirlwind Aviation RV-10 composite prop, 80 inch, 46 lbs

    I was in phase one at around 33 hrs. when I experienced a total thrust failure shortly after takeoff. As it turns out the prop shaft of my PSRU broke and caused a disconnect between the engine and the prop.
    In my third flight a couple month back I made a 3 G turn for 20 seconds as part of my flight testing.

    Several conversations with the manufacturer of the BW350a and BW350b at AUTOPSRUS revealed that this turn caused the gearbox case to flex and misalign the supporting bearings. The prop shaft bent slightly and possibly cracked at this time.
    I could fly another 30 hrs. before the shaft finally broke. The shaft is splined at the area of breakage and the core of the shaft is 1.25 inches thick (compare that to a Lycoming main bearing).

    I believe a combination of the 3 G turn and torsional vibrations caused fatigue and subsequently shaft failure.
    I never used more than 75% of the available power and flew the airplane like an airliner (minimum G’s) except for that one turn.

    I am reaching out to you in order to spread the word. Maybe you know somebody who is flying or building an airplane with this PSRU. I destroyed my gearbox because I was unaware that a 3 G turn could do such harm.

    Please remember, that I am not an engineer and every setup is different. I CANNOT provide a G-limit that is safe.

    I can only suggest to STAY AWAY FROM ANY G’S and STAY OUT OF TURBULENCE

    Many Happy Landings

    Andreas K
     
  2. Aug 19, 2016 #2

    BBerson

    BBerson

    BBerson

    Well-Known Member HBA Supporter

    Joined:
    Dec 16, 2007
    Messages:
    12,185
    Likes Received:
    2,394
    Location:
    Port Townsend WA
    Photo looks like a broken quill shaft to me. But I have no knowledge of this unit or if it has one.
     
  3. Aug 19, 2016 #3

    cheapracer

    cheapracer

    cheapracer

    Well-Known Member

    Joined:
    Sep 8, 2013
    Messages:
    5,469
    Likes Received:
    3,823
    Location:
    Australian
    One of the downsides of social media displayed right here.
     
    steveair2 and BoKu like this.
  4. Aug 19, 2016 #4

    Turd Ferguson

    Turd Ferguson

    Turd Ferguson

    Well-Known Member

    Joined:
    Mar 14, 2008
    Messages:
    4,829
    Likes Received:
    1,745
    Location:
    Upper midwest in a house
    not a very convincing autopsy.
     
    BoKu likes this.
  5. Aug 19, 2016 #5

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,792
    Location:
    Calgary, Alberta, Canada
    There have been other failures with these gearboxes. Another in the smaller 200Z unit was a gear failure, possibly again TV related on a 6 cylinder EZ30. The other was in another LS powered RV10, oil seal failure and a forced landing, fortunately with no injuries. The original oil seal retention method was not so good and several people had the same problem.

    The new owner has addressed many of the known issues with the BW design but the fact remains, there was never any TV studies done to my knowledge and you never know what's lurking without doing that.

    Until a gearbox design has several examples going past 500 hours each with no problems, you just never know.
     
    don january likes this.
  6. Aug 19, 2016 #6

    don january

    don january

    don january

    Well-Known Member

    Joined:
    Feb 11, 2015
    Messages:
    2,664
    Likes Received:
    1,097
    Location:
    Midwest
    It's sad the plane suffered damage, but I'm glad the pilot is not hurt. It's hard to put reins on that many ponies
     
    Chris In Marshfield likes this.
  7. Aug 19, 2016 #7

    clanon

    clanon

    clanon

    Well-Known Member

    Joined:
    Jun 6, 2007
    Messages:
    1,101
    Likes Received:
    217
  8. Aug 19, 2016 #8

    Chris In Marshfield

    Chris In Marshfield

    Chris In Marshfield

    Well-Known Member HBA Supporter

    Joined:
    Jan 29, 2008
    Messages:
    1,230
    Likes Received:
    1,006
    Location:
    Germantown, WI USA
    BoKu likes this.
  9. Aug 19, 2016 #9

    RJW

    RJW

    RJW

    Well-Known Member

    Joined:
    Feb 9, 2011
    Messages:
    704
    Likes Received:
    201
    Location:
    Wisconsin and Kansas
    The failure could have resulted from torsional vibration. But ordinary loads from a 400HP motor and a heavy prop are enough to destroy these boxes. They use automotive (competition) quick-change gears. The gear teeth are sturdy enough. The machined, 10-spline, 1.25” shafts however are not sufficient to handle the loads in high-power, heavy prop applications. I’ve been studying gearboxes based on quick-change gears for some time. My analysis has led me to believe that quick-change gears, in an arrangement similar to that in Bud Warren’s box, should be limited to about 250HP and use a light wooden or composite prop.

    Also, Bud was marketing these boxes as good to somewhere around 450HP. The new owner calls them BW350 boxes and limits them to 350HP IIRC. But even this I think is optimistic.

    Rob
     
    clanon and cheapracer like this.
  10. Aug 19, 2016 #10

    Turd Ferguson

    Turd Ferguson

    Turd Ferguson

    Well-Known Member

    Joined:
    Mar 14, 2008
    Messages:
    4,829
    Likes Received:
    1,745
    Location:
    Upper midwest in a house
    So would a TV failure be sudden and catastrophic or would there be cracking and slow(er) propagating failures that lead up to the final failure? Could a tear down and inspection revealed this was coming?
     
  11. Aug 19, 2016 #11

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,792
    Location:
    Calgary, Alberta, Canada
    The failure is sudden but there is almost always something cracking over some time before the failure. TV can impose loads 10-30 times over the peak torque output of the engine. If you have this happening in a constant operational range, no gearbox which would be a reasonable weight in an aircraft can withstand those forces for long.

    As Dan H stated on VAF, the shaft design does not follow best practice and a 1.25 inch shaft for this level of torque and the usually heavy C/S props, is likely inadequate.

    I tell people, any time you fly with a geared auto conversion, it's truly experimental as they've rarely been professionally developed or adequately tested. Eyeball engineering does not always work out well in this field. Basing your purchase on outside looks, cool factor, nice powder coated and anodized parts does not always turn out well. It's what's inside all those nice CNC'd cases that counts.

    To be fair, I believe the high time user has something over 500 hours on his gearbox in another RV10. As far as I know, he's not had any serious issues to date. However TV is insideous- a different engine or prop combination often changes the TV signature dramatically so what worked in a similar case might be unreliable when changes to engine torque or propeller inertia occur.
     
    Last edited: Aug 19, 2016
    Midniteoyl, BJC and BoKu like this.
  12. Aug 19, 2016 #12

    BoKu

    BoKu

    BoKu

    Well-Known Member HBA Supporter

    Joined:
    Aug 16, 2013
    Messages:
    2,348
    Likes Received:
    2,982
    Location:
    Western US
    There's some good material in the entire VAF thread on this topic:

    BW 350 PSRU Failure, please share the word! - VAF Forums

    A couple of notes and speulationisms:

    * @Billski, I think we would really benefit from your input, based on what's here and in the VAF thread.

    * I'm among those keeping a big ol' grain of salt handy regarding any statement that these boxes not being designed to handle a 3g load factor. If it's in their published specs, well, I'm wrong. Hopefully someone on their side with have their say on it one way or the other.

    * The shaft for which Dan Horton links a picture is like the poster child for a torsional stiffness gradient concentration. It sure would be interesting to get this thing into a real FEA system and see where it goes red, and at what torque.

    [warning: off-topic sermon ahead]

    Getting philosophically wider afield, designing highly loaded power transmission equipment is not rocket science, but only until it is. At issue is that pretty much anybody with a Boston Gear catalog can eventually arrive at a set of gears and shafts and bearings that move 400hp between two RPMs. It only gets really difficult when you go to do it with less than a couple hundred pounds of iron. That's when the stresses and strains start to approach the levels at which you must expect fatigue. The trouble, of course, is knowing where to expect it. That's when some experience really shines.

    During WWII, engineers all over the world got a pretty good handle on what it takes to move these great gobs of horsepower out of a reciprocating engine and into a propeller, and they did it without FEA. But it did take a lot of math and physics and engineering, and it also took a lot of broken engines and gearboxes. Probably the biggest thing we learned was that it is a thorny problem not particularly amenable to universal solutions, and that every different engine is a different challenge. But with the free world at stake, we invested what it took.

    What's interesting is noting that in the wake of WWII, after we'd developed all this geared engine technology at great cost, geared aircraft engine development pretty much fell off a cliff. The big guys were all using turbines inside about 20 years, and in the General Aviation realm of 100hp to 300hp the vast majority of reciprocating engines remained direct drive motors that accommodate a (relatively) low output speed by throwing lots of displacement at the problem.

    Of course, the usual mistake is seeing how an O-360 gets only about 1/2 hp per cubic inch and assuming that it must be a very inefficient solution. Which might be true if for whatever reason you have a fetish for hp/displacement. But airplanes don't care about that. They only really care about hp per installed mass, hp per installed volume, and specific fuel consumption. And factoring in their reliability and robustness, the O-360 does pretty darn well at those things.

    Thanks, Bob K.
     
    Terrh, Midniteoyl, Glider and 6 others like this.
  13. Aug 19, 2016 #13

    moto

    moto

    moto

    Member

    Joined:
    Oct 25, 2008
    Messages:
    7
    Likes Received:
    1
    Location:
    Queensland, Australia
    This is a sad reminder of a local accident In that case, Terry Kronk's 80% scale, LS-powered P-51 broke the same shaft as described here, with the loss of both the aircraft and Terry. I have pictures of the broken shaft. The redrive had only just returned from Bud Warren.

    There does not seem to have been enough testing of these units for them to be making the claims they do...

    Marty




     
  14. Aug 19, 2016 #14

    rv6ejguy

    rv6ejguy

    rv6ejguy

    Well-Known Member

    Joined:
    Jun 26, 2012
    Messages:
    3,749
    Likes Received:
    2,792
    Location:
    Calgary, Alberta, Canada
    The clutch design on these drives which decouples the engine from prop during startup does little to alleviate high amplitude TV further up the rpm range. I've never been a fan of this design or any other which uses a clutch inside unless it's a breakaway clutch designed to protect the engine in the event of a prop strike.

    TV is inadequately dealt with on most Experimental drives. I'm not saying that's what caused these failures but it's caused lots of problems in lots of drives.

    Many vendors also don't understand that they need to do plenty of testing before release and that they should insist that the same prop and engine combination used during that testing is used by every customer of their drive. The testing is fairly meaningless otherwise.
     
    Last edited: Aug 20, 2016
  15. Aug 20, 2016 #15

    Andreas K

    Andreas K

    Andreas K

    New Member

    Joined:
    Aug 19, 2016
    Messages:
    4
    Likes Received:
    5
    Location:
    Beautiful Wisconsin
    Please do me a favor and look for published G limits. As of 8/10/2016 I couldn't find them. This is the reason for my post, to raise awareness of the fact.

    I know a gentleman who successfully flies with the cast aluminum gearbox, a 2 bladed Harzell prop and a LS1 engine but he flies as smooth as he can and hasn't pulled any G's. He has 150+ hours on his RV-10.

    I heard of an accident where a pilot made a high speed fly-by and a "high" G pull-up at the end of a runway which destroyed the gearbox. Could have been Terry Kronk's P51.

    I assumed that "high" G starts at 5 or 6 (with my fighter pilot background). Big mistake.

    I only pulled 3 G's with my setup and lost all power at 300 ft on takeoff 30 hrs. later.

    The statement, that my 3 G turn damaged my gearbox and the analysis about the flexing case comes from the manufacturer of the current BW350b gearbox. I have 2 independant witnesses for that.


    Blue skies

    Andreas K
     
    Glider likes this.
  16. Aug 20, 2016 #16

    TFF

    TFF

    TFF

    Well-Known Member

    Joined:
    Apr 28, 2010
    Messages:
    11,721
    Likes Received:
    3,320
    Location:
    Memphis, TN
    Hi G is only how high the component is rated. 3 g is not a lot but it is more than traveling planes like the RV10 normally fly as passengers are usually not fans of Gs. It is possible in turbulent air straight and level. Is there any bending or misalignment possible? It is where the hobby of engines comes in vs Lycoming. Do you have a picture of the whole drive on the engine?
     
  17. Aug 20, 2016 #17

    BJC

    BJC

    BJC

    Well-Known Member HBA Supporter

    Joined:
    Oct 7, 2013
    Messages:
    9,714
    Likes Received:
    6,505
    Location:
    97FL, Florida, USA
    I never would have critical equipment that self-distructs at 3 g in a relatively fast cruising airplane.


    BJC
     
  18. Aug 20, 2016 #18

    Vigilant1

    Vigilant1

    Vigilant1

    Well-Known Member Lifetime Supporter

    Joined:
    Jan 24, 2011
    Messages:
    4,153
    Likes Received:
    1,882
    Location:
    US
    Does this pass the sniff test? We're talking about 3Gs and a total weight on the output bearing of, what, 40 lbs? So a 120 lb load pulling straight down on the prop shaft during the duration of this turn caused a misalignment of the quill shaft/bearings and a fracture? What does the manufacturer claim the maximum G loading >should< be, given the weight of various props/any prop extensions?

    I'm not an engineer, but some sort of fatigue related to TV seems a lot more likely to me.
     
    Last edited: Aug 20, 2016
  19. Aug 20, 2016 #19

    Autodidact

    Autodidact

    Autodidact

    Well-Known Member

    Joined:
    Oct 21, 2009
    Messages:
    4,513
    Likes Received:
    799
    Location:
    Oklahoma
    It isn't the loading from the gs, it is the turn rate required to maintain those gs - that causes a bending moment due to gyroscopic precession that can be very high depending on the rate of the turn
     
  20. Aug 20, 2016 #20

    BBerson

    BBerson

    BBerson

    Well-Known Member HBA Supporter

    Joined:
    Dec 16, 2007
    Messages:
    12,185
    Likes Received:
    2,394
    Location:
    Port Townsend WA
    The rate of turn is probably greater in a rapid pull-up. The system should be designed for maneuvering.
     

Share This Page



arrow_white