Centaflex coupling fault / design flaw

[ianj99]

Whilst being fitted by the yard's engineer, the Centaflex-M coupling (sold under the Quietlife DB127 brand) split along one of the clamping bolt holes as you can see from the photo. There are hairline fractures of varying lengths visible adjacent to the other bolt hole locations. Even if the crack and fractures had not occurred during the initial fitting, I'm sure the coupling's reliability in service would be marginal and potentially life threatening if it was to fail at the 'wrong moment'.

I would appreciate comments as to the design because it seems to me and to the engineer and other experienced yard staff, that the holes are far too close to the edge. Either this is a design flaw or a manufacturing defect however Centa are dragging their heels over deciding whether to replace it despite a promise to have done so first thing on monday morning.

Photo 1 shows the cracks - the one on the left is full length and through into the bolt hole.
Photo 2 shows the bolts holes are not central in the clamp flange.
The coupling is the clamp fixing type and the clamp flange tapered and fits in the tapered bore and tightened by means of the 6 bolts.

Ian

 

[Talulah]

I would send an email with photos to centaservice@centa.de.
http://www.centa.info/?show=service&c=#

 

[ianj99]

I would send an email with photos to centaservice@centa.de.
http://www.centa.info/?show=service&c=#

Thanks. I will, though the supplier has already had the photos and passed them to Centa almost a week ago. They were promised a decision yesterday am by Centa (presumably the UK office)

I am amazed that a global company cannot reach a decision over a sub £200 (retail cost) coupling. It just beggars belief and puts anyone off dealing with them...
Ian

 

[Poignard]

I have had one of those couplings in my boat for more than 10 years without any problems and I would be very interested to know what the manufacturer has to say, so please keep us informed.

What's the background to this story? The rust suggest it's not a new coupling.

 

[pvb]

Whilst being fitted by the yard's engineer, the Centaflex-M coupling (sold under the Quietlife DB127 brand) split along one of the clamping bolt holes as you can see from the photo. There are hairline fractures of varying lengths visible adjacent to the other bolt hole locations. Even if the crack and fractures had not occurred during the initial fitting, I'm sure the coupling's reliability in service would be marginal and potentially life threatening if it was to fail at the 'wrong moment'.

I'm confused by your pics, because they don't look like the bits of a new coupling. Was it brand new or second-hand?

There are specific maximum torque recommendations for the clamping bolts - were these followed by the yard's engineer?

 

[ianj99]

I'm confused by your pics, because they don't look like the bits of a new coupling. Was it brand new or second-hand?

There are specific maximum torque recommendations for the clamping bolts - were these followed by the yard's engineer?

It is an new coupling but a year old as I've not had the opportunity to have it fitted until 2 weeks ago. (The supplier has a record of the sale so its not second hand.)
It has been stored on board for the intervening period.
I asked the yard to fit it whilst also fitting a Kiwi prop and a 7year time served engineer attempted to fit it.
No documentation was supplied with the coupling as regards tightening torque.

It maybe the case that the smaller bore versions have plenty of metal between thread hole and bore, but this was the largest available as my shaft is 1.5" and there is no reason that I can see for the holes not to have been drilled on a larger pitch circle.

I am not asking Centa for any admission of a design flaw or manufacturing defect, just to authorise a refund or replacement.

If you would be happy to rely on less than 1mm of steel between thread holes and bore, then fine, but as I said, experienced marine engineers are dubious to say the least.
Ian

 

[pvb]

No documentation was supplied with the coupling as regards tightening torque.

Ah! Maybe overtightening was the problem then.

 

[ianj99]

Centa have agreed to replace it for which I thank them.

It had of course occurred to me that the bolts were overtightened and this would then have left me in dispute with the yard.
I will compare the replacement with the original but I am reassured by the comments from those who have been using one for years without problems so hopefully the replacement will give long and trouble free service.
Ian

 

[eagleswing]

our centa M13 is 5 years old , & in fine shape..

we have the centa M13 fitted to a 30mm (small amount less than 1.25 inches) propshaft which was original equipment installed by the manufacturer on a nuaticat 321 2005. when we removed the centa to replace our propshaft, the centa coupling was inspected. i recall that the 6 retaining bolt holes , and the 3 jackbolt holes, were quite a distance away from the inner edge of the coupling (unlike the coupling shown in OP's photos...) i recall having checked the proper torque for the 6 retaining bolts, as supplied to me by a centa engineer, and it was surprisingly low . i also doublechecked the torque for the centa against that recommended for the same size vetus coupler-- it was the same range, only off by a pound or 2. it certainly looks like OP's retaining bolt holes were spaced much too close to the inner surface. even significant overtorquing of the bolts bolts should not have resulted in the double stress fractures OP saw. i'd be interested to hear the explanation for this one...do keep us posted.

s/v eagleswing

 

[pvb]

Centa have agreed to replace it for which I thank them.

It had of course occurred to me that the bolts were overtightened and this would then have left me in dispute with the yard.
I will compare the replacement with the original but I am reassured by the comments from those who have been using one for years without problems so hopefully the replacement will give long and trouble free service.
Ian

Good outcome. Make sure you get proper max torque settings for the clamping bolts and that the yard uses them. As eagleswing says, the max torque is surprisingly low - for your M127 it's only around 23Nm.

 

[Poignard]

The engineer ought to have appreciated that this coupling relies for its effectiveness on the tapered collet gripping the shaft more tightly as it is drawn into its socket and thereby compressed, and that it should not have been tightened until something broke but that there would be a torque setting for the bolts which he could, and should, have ascertained. I don't have the torque wrench settings to hand at the moment but they are not very great.

I hope he also knows that the clamping bolts should be torqued up in stages, i.e. following a sequence so that the collet is drawn in evenly. Also that the shaft and inside face of the collet should be thoroughly degreased and the special grease supplied by the manufacturer should be applied to the outside (tapered) face of it.

Another thing to make sure of is that the collet is the right size for the shaft, ie don't use a metric size collet for an imperial size shaft

Don't place undue reliance on 'experience'. More damage is caused by people who think they have nothing more to learn than by those who know they don't know much and take the trouble to learn.

Finally, I think it's a good idea to paint the torque settings on the inside of the hull near the coupling so that the next man to install it has no excuse for overtightening the bolts.

 

[eagleswing]

centa design change, post 2008...

out of curiosity i went to the centa website. they are proclaiming that in mid-08 they have had a 'design change... allowing a reduced diameter.. at the same power.' the ad does not say what diameter was reduced but from OP's pictures it surely looks like there is a reduced diameter allowing the retaining bolt holes to be perilously close to the inner edge of the coupler inner surface..

so much for what one can learn from an ad....

i guess we are all OK if ours is pre-08 ???

 

[doug748]

out of curiosity i went to the centa website. they are proclaiming that in mid-08 they have had a 'design change... allowing a reduced diameter.. at the same power.' the ad does not say what diameter was reduced but from OP's pictures it surely looks like there is a reduced diameter allowing the retaining bolt holes to be perilously close to the inner edge of the coupler inner surface..

so much for what one can learn from an ad....

i guess we are all OK if ours is pre-08 ???

Well spotted.

Cynics have been known to say that makers downsize everything till it breaks, and then make it a bit more robust.

 

[pvb]

out of curiosity i went to the centa website. they are proclaiming that in mid-08 they have had a 'design change... allowing a reduced diameter.. at the same power.' the ad does not say what diameter was reduced but from OP's pictures it surely looks like there is a reduced diameter allowing the retaining bolt holes to be perilously close to the inner edge of the coupler inner surface..

so much for what one can learn from an ad....

i guess we are all OK if ours is pre-08 ???

The "reduced diameter" refers to the outer housing which bolts to the gearbox, not to the inner part which clamps on to the propshaft.

If you look at the diagrams of the 2 versions (original and revised), you'll see the clamping arrangement is basically unchanged.

 

[Avocet]

I think the holes are far too close to the edge and are "asking for it" - torque setting or no torque setting! If the visible edges of the holes are that close to the edge, the outermost tips of the tapped threads will be even closer. Still if they've agree to replace it, that's an end to the matter.

As for the off-centre holes in the other photos, that's harder to comment on - they possible use that mating part on more than one model, each of which has different pitch circle diameters, so maybe that's not so much of a problem.

 

[pvb]

I think the holes are far too close to the edge and are "asking for it" - torque setting or no torque setting! If the visible edges of the holes are that close to the edge, the outermost tips of the tapped threads will be even closer. Still if they've agree to replace it, that's an end to the matter.

Remember that these holes are around a tapered section, so the thickness of steel around the hole will increase progressively towards the bottom of the threaded hole. It's very possible that the splits were caused by the inner cone-shaped insert "bursting" the outer cylinder apart as a result of overtightening the clamp bolts.

 

[Poignard]

As for the off-centre holes in the other photos, that's harder to comment on - they possible use that mating part on more than one model, each of which has different pitch circle diameters, so maybe that's not so much of a problem.

If you mean the three holes with plastic bungs in them, they are tapped holes for the jacking bolts that are needed to extract the collet.

 

[Neil_Y]

Good answer, we supply the same type of clamp from Norway, and they rely on the taper to clamp tight with very little torque on the bolts.

 

[doug748]

Good result.
Do let us know if they offer their explanation for the failure.

 

[Neil_Y]

This looks similar to the clamping unit used for Hydradrive and marine joint, if this works I've taken some pictures to show that the bolt holes are closer to the edge at one end due to the taper. This unit has two components the Inside ring has a taper that sits against a taper in an external ring, when both parts are pulled together the inside clamps down on the shaft at the same time as the outside expands into the cup of the coupling.