Click for JetForums Click for Ocean Alexander Click for Cross Click for Furuno Click for Nordhavn

Centaflex PTO coupling

Discussion in 'Stabs, Tabs & Gyros' started by TahoeJohn, Dec 28, 2020.

You need to be registered and signed in to view this content.
  1. TahoeJohn

    TahoeJohn Member

    Joined:
    Aug 8, 2020
    Messages:
    155
    Location:
    Tahoe VIsta
    I've noticed a fair bit of vibration on the port main engine and am now looking at the front of that engine where it drives a hydraulic pump (which powers the thrusters + the windlass).

    On the front of the crankshaft is a Centiflex coupling which drives a Spicer flexible driveshaft (about 10" long) that in turn drives a clutch in front of the hydraulic pump. This Centiflex coupling has one bolt missing, and I believe it's an M20 bolt, so not a small one. In addition to this missing bolt, there's a fair bit of slop in the coupling, even near one of the bolts that's still there.

    I took a short video of the setup and movement; hopefully this works:
    https://vimeo.com/495100482

    This type of coupling is completely new to me, and I'm trying to determine what's best to do with it. Should I just find a replacement for the missing bolt and tighten everything securely, or do these Centiflex couplings have a limited lifetime and need to be replaced periodically? It's just under 20 years old.

    What about the Spicer driveshaft -- maybe get it rebuilt with new U-joints?
  2. d_meister

    d_meister Senior Member

    Joined:
    Mar 4, 2010
    Messages:
    468
    Location:
    La Conner, WA.
    Try one of the other bolts in that hole and torque it to specs for the type and size bolt to be sure the hole isn't stripped. If that goes ok, take the other two bolts out and check them for wear or damage, as you would the first one. Then, use one as a sample to purchase the correct size and grade of all 4 bolts. With the coupling off, look at the metal bits in the rubber where the bolts go through to the flange and check that the holes aren't elongated and the metal and rubber isn't damaged. I would order a new one and keep that as a spare. That slop in your video doesn't belong and will not be there when the coupling is properly mounted. I would use Loctite on the threads, since it's come loose once, already.
    The U-joints should rotate on their two axis' smoothly, and there should be no play in the assembly when twisted. I would mount one coupling in a vise and grab the other with my giant Cannelocks and load the joints while testing smooth operation. It's worth noting that the range of motion isn't great on this application, unlike the rear end of a car that travels up and down quite a bit, so it's not likely to be excessively worn. Best thing is take it to a driveshaft shop and let them recommend a course of action. The driveshaft is correctly phased, now, so don't allow someone to disassemble it unless there are companion marks to reassemble it correctly. A driveshaft specialist would know that.
    Now you can search for that 4th bolt. It's in that engine room somewhere!
    Metric Recommended Bolt Torque
  3. TahoeJohn

    TahoeJohn Member

    Joined:
    Aug 8, 2020
    Messages:
    155
    Location:
    Tahoe VIsta
    Thank you - very helpful. I'll need to get a 17mm hex socket and then follow your suggestions. The ABT Trac manual lists the torque at 300 ft-lbs, vs the 480 Nm specs you shared, equivalent to about 350 ft-lbs. In any case, they need to be tight!

    The next step is to figure out why there is slop under one of the still-present bolts. I'm also seeing if I can get a new rubber element, just for peace of mind.

    I like the idea of taking the Spicer unit to a driveshaft shop and get their recommendation.

    Thanks again!
  4. d_meister

    d_meister Senior Member

    Joined:
    Mar 4, 2010
    Messages:
    468
    Location:
    La Conner, WA.
    The bolt that won't go down tight may be damaged, but it could also be that the threads run out before any clamping pressure is achieved. The shank length may be longer than it should be. I can't post images, for some reason, but the anatomy of a bolt is HERE. I just want to be sure I've got the right terminology. Turns out "shoulder" is wrong :)
  5. TahoeJohn

    TahoeJohn Member

    Joined:
    Aug 8, 2020
    Messages:
    155
    Location:
    Tahoe VIsta
    Update: I got a 17mm hex socket with a 1/2" breaker bar and tried to get the bolts out. I managed to get all four of the radial bolts out but only one of the axial bolts. The problem is that the engine spins... anyone know a good way to stop a Cat 3412 from rotating?

    I did go get a Milwaukee 120V impact wrench, but it's only rated at 300 ft-lbs and these bolts were torqued to at least that, plus the 19 years of being stuck in there.

    Any ideas?

    Also, it appears that the missing bolt didn't just spin out, but rather broke off just below the surface of the flange on the crankshaft. Once I get the rubber element off, I'll need to go to work on extracting the broken part. Good fun.

    A couple pictures, here's the center hub loose:
    hub out.jpg

    And these M20 x 65mm bolts are pretty big. I don't have small hands.
    bolts.jpg
  6. DOCKMASTER

    DOCKMASTER Senior Member

    Joined:
    Feb 26, 2012
    Messages:
    1,396
    Location:
    Ketchikan, Alaska
    Either get yourself a real impact like a Dewalt 20v or similar or put something in one of those radial bolt holes that sticks out so you can block it off and keep from spinning. Will a good quality 1/2” extension fit in one of those holes? Then use some wood blocks to lock against a stringer or something solid.
  7. TahoeJohn

    TahoeJohn Member

    Joined:
    Aug 8, 2020
    Messages:
    155
    Location:
    Tahoe VIsta
    I actually looked at that DeWalt 20V but the one I saw was only 150 ft-lbs.

    However, your idea of a pipe or 1/2" extension in one of the empty bolt holes is a great idea! Since I'll be replacing this rubber element anyway, I don't care if the hole gets bungled up. That and a cheater pipe on the breaker bar should do it. Thanks!!!
  8. DOCKMASTER

    DOCKMASTER Senior Member

    Joined:
    Feb 26, 2012
    Messages:
    1,396
    Location:
    Ketchikan, Alaska
  9. d_meister

    d_meister Senior Member

    Joined:
    Mar 4, 2010
    Messages:
    468
    Location:
    La Conner, WA.
    It's a real testament to the Loctite on the bolts. For one bolt to break, and the coupling being loose as shown in the video, I'm wondering if the bolts bottomed out without achieving clamping force.
  10. NYCAP123

    NYCAP123 Senior Member

    Joined:
    Mar 14, 2008
    Messages:
    11,205
    Location:
    Long Island, NY
    Penetrating oil
  11. Capt J

    Capt J Senior Member

    Joined:
    Jul 11, 2005
    Messages:
    14,432
    Location:
    Fort Lauderdale
    Heat breaks down locktite ( from a torch), but be VERY careful as not to create a fire.
  12. NYCAP123

    NYCAP123 Senior Member

    Joined:
    Mar 14, 2008
    Messages:
    11,205
    Location:
    Long Island, NY
  13. TahoeJohn

    TahoeJohn Member

    Joined:
    Aug 8, 2020
    Messages:
    155
    Location:
    Tahoe VIsta
    This is very good to know. Once I get to the point of extracting the broken bolt, this is probably going to be key... thanks!
  14. NYCAP123

    NYCAP123 Senior Member

    Joined:
    Mar 14, 2008
    Messages:
    11,205
    Location:
    Long Island, NY
    Broken bolt? You can try an easy-out, but most likely you'll have to drill it out.
  15. TahoeJohn

    TahoeJohn Member

    Joined:
    Aug 8, 2020
    Messages:
    155
    Location:
    Tahoe VIsta
    Success. Dockmaster's suggestion of a pipe worked like a charm; most the work was just filing down a 1/2" steel pipe until it fit snugly into one of the radial holes in the rubber element. That, plus a 30" long, 3/4" extension pipe on my breaker bar, had the remaining two bolts out in about 10 minutes.
    1.jpg

    There is obviously a significant amount of wear on this whole assembly. This 1/4" thick brass disk sits behind inner coupling, plus one of the rubber element bolt holes is similarly ovaled-out:
    2.jpg

    The broken bolt inside the flange at the 9 o'clock position:
    3.jpg

    The great news is that the flange, to which the rubber element bolts, is removable so I can just take it down to a machine shop, have it cleaned up, get them to remove that bolt and chase the other threads, and then it will be time to reinstall everything.

    Everything looks solid on the crank itself:
    4.jpg

    Thanks for all the thoughts and suggestions!
  16. TahoeJohn

    TahoeJohn Member

    Joined:
    Aug 8, 2020
    Messages:
    155
    Location:
    Tahoe VIsta
    I now have a new Centaflex rubber element (with 8 new bolts), have rebuilt the Spicer driveline, and have cleaned up the other parts. The last missing piece is this bronze (?) disk/spacer that was sandwiched between the engine flange and the inner PTO coupling: stackup.png


    As posted previously, it's significantly worn / ovaled and can't be reused, so I'm trying to source a replacement. Centa (now Rexnord) doesn't make this; ABT Trac (who made the overall thruster system) doesn't, either. I'm checking with Horizon to see if it's something they manufactured and added, but haven't heard back yet.

    I don't understand what the purpose of that disk is. The inner coupling will be very firmly attached to the rubber element and sees little axial force (really zero). I'm tempted to reassemble this without that disk in place, but would welcome any thoughts or insights into what it is and why it was there.
    Last edited: Jan 13, 2021
  17. d_meister

    d_meister Senior Member

    Joined:
    Mar 4, 2010
    Messages:
    468
    Location:
    La Conner, WA.
    Could it be that the bolts holding the coupling are too long? It looks as if there is little meat for the bolts to hold onto in the base of the flange that the bushing mounts too. I would screw the bolts in until they bottom out, then measure the length exposed, and then compare to the corresponding length of the journal where the bolt fits inside the coupling. It seems to me that the entire situation resulted from a lack of clamping pressure achieved by the bolts (SEE VIDEO) The people that assembled this utilized the most immediate solution available to them to solve the mounting problem, since there was likely a machine shop on site. Another solution would be shorten the bolts, but it would be necessary to trim to fit so that the maximum number of threads are in the flange.
    The method of attaching the coupling should be studs, so the maximum amount of material in the threaded mounting plate would result, but the problem is the recess where the head of the Allen cap screw fits. Reviewing the video, it appears another solution could be to use studs and nuts with spacers in the Allen head cavities. It doesn't appear that nuts and washers would interfere with the driveshaft mounting (?). Watching the video, it appears certain that the bolts were bottomed out without clamping the coupling, especially when considering the amount of force was required to loosen the bolts. Loctite is strong, but not long breaker-bar strong.
  18. TahoeJohn

    TahoeJohn Member

    Joined:
    Aug 8, 2020
    Messages:
    155
    Location:
    Tahoe VIsta
    Thanks for the thoughts. The 8 bolts that attach the rubber element to both couplings are supplied by Centa, so I'm fairly confident that they are the correct length. There is nearly an inch of threads in both the axial holes of the crankshaft flange and the radial holes into the inner coupling, so I believe that there is adequate space for the bolts to not bottom out.

    To clarify my earlier picture, this bronze spacer fit entirely inside the rubber element, so it's not intended to compensate for bolts being too long nor space the rubber element away from the crankshaft flange:

    stackup2.png

    I think the slop shown in the video resulted from the bolts once being tight but the rubber element wearing out after 19 years of use.

    Centa specifies (in their technical drawings) a spacing of 8mm between the pink and white couplings above. One theory might be that Horizon felt that this was so important that they machined an 8mm thick spacer. But with four M20 bolts connecting the rubber element to this center coupling, there just isn't any play.

    One thing that surprised me initially is that the inner diameter of the new rubber element is actually about 30mm larger than the diameter of the inner coupling; after you snug all those four radial bolts down, the rubber element is significantly pre-loaded:

    b.png
  19. Capt Cole

    Capt Cole Member

    Joined:
    Jan 24, 2017
    Messages:
    120
    Location:
    Pittsburg CA
    What a great thread. Clear and concise documentation of the issue and multiple helpful suggestions. This is exactly why I routinely monitor this site. Compliments to all that contributed.
  20. d_meister

    d_meister Senior Member

    Joined:
    Mar 4, 2010
    Messages:
    468
    Location:
    La Conner, WA.
    It appears that you have the Centaflex A design, Type 3
    Type A.jpg
    which has only rubber where the attachments are made. I was assuming that these are made like others I'm familiar with that had actual metal lugs molded into the rubber that would allow the bolts to be torqued down against metal.

    The type 3-S has smooth sleeves to drive the coupling, and I would hope there are special shoulder bolts for the driven hub,
    TypeA_3S.jpg
    If this is the first failure since 2001, then the application appears to be adequate. That must be some tough rubber to have broken a bolt, which started this whole thing!