So I really need some advice from the experts out there.
Has anyone ever heard of a post mod bolt shearing before?

Yes you are not the first

What could have caused it to shear - over torqued, fault in bolt metal????

Benelli supply that bolt with a shortened head which will weaken it, although yours broke further up the shaft, so it was probably a re-used bolt or it had a fault.
Replace it with a standard R8.8 M10 x 1.25 x 25 set screw from your local bolt shop and use the OEM Belville washer. Don't be tempted to add a flat washer. That defeats the Belville washer and will cause it to loosen.

Can I knock the shaft out from the alternator side with a 13mm bar, leaving the z25 gear in place?

Yes, that's the only way to get it out.

Should I replace the z25 shaft with new if I can get the remain of the bolt out?

If it isn't damaged you should re-use it

Finally the worrying question -should the main clutch basket gear have any play - it rocks slightly like when you find wheel bearings goosed? If it shouldn't rock - what is the likely cause?

The gear can have some play in the rotating direction if the cush springs aren't tight, but if you are talking about movement perpendicular to rotation that should be barely perceptible. I have a perfectly good '03 basket that I just checked and all I can see is an ever so slight movement of oil between the basket and primary gear. If yours has more than that the rivets (three) may be loose. If that is the case, I don't know how to tighten them.
Colin has had one apart, so he may be able to advise.

Will definitely need the alternator cush drive gear seen in the picture - wonder if they are readily available?

I believe so. Talk to Maniac or Bennets.

 

Thanks for the quick response and advice Engenia.

Waiting for a clutch holding tool to arrive that I ordered of fleebay. Then plan to get the clutch off and pull the z25 shaft out completely to examine and remove the broken bolt. Hopefully will be able to recover it and just fit a replacement bolt as you advised.

Will keep the thread running with any info I think will help.

 

You don't need the tool if you select top gear and jam the rear brake on.

 

So managed to strip the clutch down tonight. Basket is post mod and all secure with no damage. Found that the axial play in the clutch is caused by the play between the needle roller bearings and the 'spacer' sleeve that the bearings run on. No signs of damage or wear?

Found the following thread which seams to indicate axial play as normal? No difference in axial play After replacing the bearing's with new.

http://www.benelliforum.com/forum/tornado-tre/10361-revised-clutch-basket-bearings-uh-oh.html

Can anyone else confirm if this axial play is normal.

 

Alternator drive failure on 2013 TRE K

I bought a 16,500km 2013 TRE K a month ago. A horrible clatter at low revs got the better of me so I proceeded to investigate. I found both alternator drive coupling halves were loose on their shafts, despite the alternator nut and Z25 shaft bolt being held in place by the Loctite. With help from this forum I ordered new bits from Maniac Motors and proceeded to consider the root cause of the alternator drive failure. After reading everything on this forum and finding via the Triumph forum that those triples have the same problem, I have resolved to do a couple of things differently on reassembly. As I see it there are two issues: 1/. Spline clearance: Inertia effects subject the alternator drive to torque reversals on engine compression and firing cycles. If the centre bolt generated enough friction to avoid slipping under these reversals there would be no need for a spline, but it doesn't, so there is! It stands to reason therefore that the splines should have an interference fit, or at least zero clearance when clamped in position, to prohibit slipping when the torque reverses. If the splines on my new bits fit with even the slightest clearance I will Loctite them (and worry about disassembly only if and when the need arises). 2/. No Free Bolt Shank: A bolt is in effect a spring that clamps faces together. It works best if it has some free length that can be stretched as the bolt is tensioned. That provides tolerance in bolt tension and ensures tension is maintained under dimension changes due to temperature variation. On the Benelli almost the entire bolt is bound by the thread inside the Z25 shaft. There is no unrestrained bolt shank free to elongate under tension, so there is hardly any margin between under-tension and over-tension. Benelli introduced the Belville washer in an attempt to overcome this issue. That has proven inadequate by the failure on my bike and the bolt failure that is the original subject of this thread. In my opinion it would be better if Benelli counter-bored the Z25 shaft for some distance at the outer end, before the thread starts, to provide some unrestrained bolt length. Rather than risk messing-up a new shaft I propose to achieve the same result by turning the thread off a new bolt for some distance under the head. This will allow unrestrained elongation. I think a 30 mm bolt length will work, so I propose 15 mm unthreaded shank, leaving 15 mm thread on the end of a M10 x 1.25 mm pitch Class 10.9 bolt (with the head machined down to 5 mm high). I proposed to use 60 Nm torque, which should be about 1/3 turn from snug, as recommended for this size and class of bolt. I will run a trial on the old shaft first to confirm the thread can take it. I proceed in anticipation that my alternator drive will finally run quietly and be found tight on subsequent checks. Anyone who reckons I am barking up the wrong tree is welcome to explain the flaws in my reasoning.

 

2/. No Free Bolt Shank: A bolt is in effect a spring that clamps faces together. It works best if it has some free length that can be stretched as the bolt is tensioned. That provides tolerance in bolt tension and ensures tension is maintained under dimension changes due to temperature variation. On the Benelli almost the entire bolt is bound by the thread inside the Z25 shaft. There is no unrestrained bolt shank free to elongate under tension, so there is hardly any margin between under-tension and over-tension. Benelli introduced the Belville washer in an attempt to overcome this issue. That has proven inadequate by the failure on my bike and the bolt failure that is the original subject of this thread. In my opinion it would be better if Benelli counter-bored the Z25 shaft for some distance at the outer end, before the thread starts, to provide some unrestrained bolt length. Rather than risk messing-up a new shaft I propose to achieve the same result by turning the thread off a new bolt for some distance under the head. This will allow unrestrained elongation. I think a 30 mm bolt length will work, so I propose 15 mm unthreaded shank, leaving 15 mm thread on the end of a M10 x 1.25 mm pitch Class 10.9 bolt (with the head machined down to 5 mm high). I proposed to use 60 Nm torque, which should be about 1/3 turn from snug, as recommended for this size and class of bolt. I will run a trial on the old shaft first to confirm the thread can take it. I proceed in anticipation that my alternator drive will finally run quietly and be found tight on subsequent checks. Anyone who reckons I am barking up the wrong tree is welcome to explain the flaws in my reasoning.

I thought long and hard about exactly this when I put my RS back together. The primary problem IMO is torsional vibration in the Z25 shaft, which can be felt at 4-5000 RPM with the engine loaded even with all the alternator drive parts in good condition. I try to avoid loading the engine in this RPM range for this reason.

The further the bolt thread is engaged into the shaft past the alternator coupling, the greater the torsional vibration the bolt will have to cope with; effectively the shaft will be twisting more towards it's middle and will be pretty much static at end where the bolt head sits. This suggests to me that a longer (fully threaded) bolt would likely make things worse, though with an unthreaded, slightly reduced shank the bolt would be more free to twist with the shaft rather than trying to resist it, as well as being better able to maintain clamping force.

It's a shame there isn't space to fit an overrun clutch as this would help significantly to cut down on torsional vibration.

 

If you have trouble getting the alternator cush-drive rubbers you can try buying Kawasaki ZX750N rubbers (Kawa part number 92075-1790). They fit and work just fine.

 

Yep. Thanks for the tip. I have already been down that road but I decided to get Triumph rubbers from the UK - the price was the same and for little extra the o-ring was included. I do like the look of the Kawasaki rubbers though. They appear to have extra ribs to ensure a tight fit. That is important to avoid the slack that results in impact and sees them hammered to bits. Maybe I will use them at the next service. I now wish there was an alternative to Benelli gaskets. The price of those is outrageous (valve cover and clutch cover). I did try to find gaskets from a Chinese supplier but they only seem to offer bits for the smaller bikes.

 

You raise an interesting point Mike, thanks for troubling to respond. My underlying assumption was that the torsional strain in the Z25 shaft is insignificant. You are correct though. However minor, the phenomenon does exist. I hope that friction and Loctite have sufficient strength to absorb the torque developed in the unrestrained bolt length. If my modification fails I will declare your concern vindicated.

Yes, an overrun clutch would avoid the torque reversals, as the car industry discovered. Perhaps with some determination one could be incorporated using a housing spacer and longer bolts but that is a bigger project than I am contemplating. The rubber element coupling would have to be retained to absorb the torsional spikes. The type of spring damper used in the later automotive alternator decoupler pulleys might be a more compact and lasting solution but selection of the correct characteristics would require dynamic analysis my grey, threadbare head is unwilling to consider.

Part of the Benelli problem is that the Denso rubber element coupling design is not ideal. In industry it is understood that the rubber elements must be in compression at all times so that they never lose contact with the coupling blades, as this causes wear and impact. The elements are therefore made larger than the coupling cavity. They must be assembled by inserting half the elements then applying more than the rated drive torque to open-up the remaining cavities and enable insertion of the remaining elements. In the motorcycle situation that would make assembly impossible, thus we have an imperfect compromise prone to failure. However I do not fully understand why apparently the Denso coupling on Benellis have a chronic problem, on Triumph triples an occasional problem and on Kawasakis they last the life of the bike. While four cylinder engines have smaller torque variations at higher frequency that result in lower dynamic loads, I suspect there are subtle differences in the detail designs that could make an interesting investigation.

P.S. It has just occurred to me that one of the aforementioned subtle design differences I think I see, from coupling offers on eBay, is a shorter bolt on some Kawasakis, maybe only half as long as on my TRE K. It could well be that you are correct and I am barking up the wrong tree. A shorter bolt would only engage at the end of the shaft within the coupling spline where the stiffness of that component would prevent significant torsional strain. My bolt is 25mm long and goes into the Z25 shaft 10mm beyond the coupling. The looseness of my coupling and the broken bolt that started this thread could both be explained by rotation in service resulting from the Z25 shaft twisting back and forth. However further examination of the photo at the start of the thread shows the Belville washer had been bearing only on the outer edge, indicating there was not much bolt tension. The failure would therefore likely have been by fatigue due to stress cycles imposed as the coupling wobbled about. A bolt is supposed to be tensioned more than the applied load so it never sees load cycles hence will not fatigue. Interesting question though - perhaps I should buy another TRE K and fit a very short bolt for comparison with my long bolt proposal!

 

Y However I do not fully understand why apparently the Denso coupling on Benellis have a chronic problem, on Triumph triples an occasional problem and on Kawasakis they last the life of the bike. While four cylinder engines have smaller torque variations at higher frequency that result in lower dynamic loads, I suspect there are subtle differences in the detail designs that could make an interesting investigation.

Again I suspect this is due to excessive levels of torsional vibration when the engine is under load. After all the Z25 shafts are known to break, particularly the older ones with the oil hole and I suspect this down to very high cyclic torque reversals when the alternator drive train hits it's resonance point.

The primary difference with the Kawaskai engine is that it has four cylinders which give torque reversals from the crank only twice per revolution since all pistons reach TDC/BDC at the same time. A three cylinder engine with 120degree crank creates much higher frequency torsional vibration components, even if the amplitudes are a little lower. The difference between the Benelli and the Triumph is clearly more subtle since the designs are fairly similar. It may be as simple as the resonant frequency of the alternator drivetrain is shifted to less used engine RPM range.

 

I haven't put the Kawasaki rubbers to a long term test yet, but here is my experience so far: I checked the standard rubbers after 3,000 miles when I was doing the Z25 upgrade and they were showing signs of breaking down. I couldn't get hold of Benelli rubbers at that time and, as I wanted to get the bike back on the road I refitted the originals. At the 6,000 mile service they were completely shot. I thought it an unacceptable risk to continue fitting standard rubbers if they were likely to to fail so quickly and looked for an alternative. Somewhere on this forum there was a thread mentioning the Kawasaki rubbers and I bought a set. I rode with them for 3,000 miles, inspected them while I was checking the Z25 bolt torque and they looked perfect. Pretty soon I'll be checking them again at 6,000. If they are OK again then I won't bother checking them for a while. As I said, no long term test results, but to date and bearing in mind that this is a test sample of one, I think they are better than the ones which were fitted at the factory.

 

Thanks Kevin, that is great news!
It supports an inkling that the Benelli coupling elements might be plastic and the Kawasaki elements rubber.
Plastic is very different to rubber. It is cheap to mould but has two characteristics that make it unsuitable for shock absorption: 1/. Its spring constant increases with the rate of strain. 2/. It has little or no natural damping. It therefore does not soften the torque impulses that can overstress the drive when they try to instantly accelerate a large inertia like the alternator rotor. Plastic transmits shock like a rigid element. I visualise it acting almost like glass, hammering the drive and breaking itself into bits.
On the other hand rubber has a pretty constant spring rate regardless of the strain rate and a natural hysteresis that slows down (dampens) the rebound. This makes rubber very suitable as the compression element in a coupling (when sized correctly). It softens the torque impulses on compression and firing cycles, accelerating the alternator rotor more gently. That avoids the torque peaks that hammer the drive components, loosen splines and cause metal fatigue.

Did you notice a reduction in handlebar vibration with the Kwakka rubbers?

 

A couple of other things, unlike B1gneil I found that there was no slack in the Kawasaki rubbers, in fact they were really quite snug, which was reassuring. Earlier in the thread you were bemoaning the cost of the gaskets for the primary drive and camber. I agree, fifty odd quid is a lot to pay for a primary drive gasket, however the aluminium composite ones are definitely reusable. The gaskets seem to come in two flavours, one made from a fibrous material which tends to break very easily and is essentially a use once gasket, but Bennetts sold me one which has an aluminium core coated another side with a softer material. I've used it three times now with a very light swear of Threebond 1215 and it is perfectly leak-tight. Similarly the cambox gasket is rubber (nitrile?) and is reusable. I use a smear of the same sealant on the semicircular parts on the cam-chain side and have had good results. I assume that if the valve seat are in good enough nick the chamber is vented well enough and there is very little pressure in that volume

 

Thanks again Kevin, what you have told me is what I hoped. You have made me a happy man with confidence that my money was well spent and this will be reliable bike when the new alternator drive is installed using Kawasaki coupling elements. I also hope that, as I read somewhere in this forum, elimination of the harshness in the alternator drive will stop the nasty vibration of the right handlebar that numbs my hand. I considered leaving the gaskets off my parts order and reusing the originals but given the high shipping cost I decided to be safe and include them. Your advice gives me hope that this will be an infrequent cost while Nellie and I grow old together. I love Nellie's note, torque and handling so I see no reason to part as long as I can be confident I will not have to walk home. I invested in a top box to carry luggage securely and plan to have the ECU remapped on a dyno to eliminate Nellie's crankiness at low revs. For a total cost of about $10k I should have an excellent bike for my purposes.:clap:

 

Difference between urethane and buna-n alternator coupling elements

If I am right about the materials used for Benelli couplings vs Kawasaki as mentioned in Post # 16, the following information pinched from the Lovejoy site is relevant. Neither material is perfect but Buna-N is the only choice for shock absorption, although it must be replaced when it has taken a set leaving clearance in the coupling. Note the bits in bold:

1. NBR (Nitrile Butadiene Rubber) a.k.a. Buna-N -- is the standard and most economical material for jaw coupling spiders. It offers the best combination of temperature and chemical resistance, misalignment, and damping ability. Rubber has the best resiliency in bouncing back from deformations that occur in cyclic or heavy shock loads. This is the only material suitable for reciprocating engine applications.

Most sizes of NBR spiders are 80A-shore hardness and are black in color. Temperature range is -40°F (-40°C) to 212°F (100°C). Also referred to as "SOX" by some manufacturers. This material will allow the Jaw coupling to experience a torsional wind-up at full torque load of 4°-10°, depending on the coupling size.

Another attribute of natural rubber products used in compression is that they take a permanent "set" or loss of volume after just a short time in operation. This does not become a performance problem until the spider thickness is anything less than 75% of its original size, at which point it should be replaced. This limits their selection in motion control/precision applications since increased free-play in the coupling results from the "set". Shelf life of natural rubber elastomers is 5 years.

2. URETHANE has a 1.5 times greater torque capacity than NBR due to its greater compressive strength (either 40D or 55D shore hard ness is used) as well as better abrasion/wear characteristics. It holds up better to environmental conditions such as ozone, ultraviolet, and some oils-chemicals versus the NBR. It is limited to -30°F (-34°C) to 160°F (71°C) temperatures however, and should not be used in heavy cyclic or start/stop applications since the damping ability is limited. ..... Urethane spiders ... offer a shelf life of 5 years.

 

Triumph rubbers too small

Triumph rubbers arrived from UK. They smell like rubber so I presume they are Buna-N, which is good. They are unsuitable for Benelli unfortunately, as they are narrower and longer than required, to suit a smaller diameter, deeper coupling. That results in radial play and probable axial thrust in the Benelli coupling. I await Kawasaki rubbers from Partzilla, USA, the only supplier I found with new stock and reasonable shipping charges.

 

Interesting thread. Something I hadn't considered when I took my alternator off 18 months ago after buying my bike, which I havent managed to run in yet ! ( only 440 miles so far )

As my Alternator drive 'rubbers' are now coming up for 14 years old presumably they should be changed in any case as the shelf life has been far exceeded.

So, can anyone say what the original Benelli items are made of for sure ? plastic or rubber ?

Thinking of ordering some Kawasaki ones to fit at service time despite the fact mine ' looked all right ' when I took them out.

 

"If it ain't broke don't fix it"? Possible causes & Loctite

I like the principle "if it ain't broke don't fix it". On the other hand my car still runs fine when the fuel light comes on but I learnt the hard way to take the simple step of refilling before the situation becomes more complicated.

Assuming the history of your alternator drive is unknown I would examine it carefully. If everything is tight and the "rubbers" show no wear I would leave them alone to avoid the risk of introducing a problem. On the other hand if there is the slightest slack anywhere I would fix it to avoid the inevitable trouble. Case in point: on my bike the evidence indicates the drive half coupling had come loose and flogged back and forth. As the slack grew it hammered the driven (alternator) half coupling with increasing severity until that also came loose. While the play in the coupling on the alternator shaft is only slight, I face the likelihood that the new coupling will be a loose fit on the alternator shaft due to wear. This consequential damage could have been avoided if the loose Z25 half coupling was addressed early. (To minimise the financial pain I procured a used Triumph alternator in the hope I can use the armature in the Benelli if necessary.)

One of the steps in Root Cause Analysis is to identify possible causes. Benelli alternator drive failure apparently occurs on some bikes but not others. Possible causes I can identify are: Variation in component fits due to machining tolerances, variation in material used for the "rubbers", variation in assembly practices including bolt tension and retaining compound application, variations in operating conditions (eg. rev range, temperature). Any additions anyone?

Another small development: Since Australia does not have the "High Temperature Sealer" specified in the service manual for application to the splines I referred the problem to Loctite technical support. I was advised to use Loctite 620 without primer and leave it to cure for three days before starting the engine, to ensure it develops full strength. Information was provided showing the use of Loctite Primer gives a faster cure but with lower strength. Hence I will use brake cleaner to degrease the parts and dry them completely before assembly.

 

I like the principle "if it ain't broke don't fix it". On the other hand my car still runs fine when the fuel light comes on but I learnt the hard way to take the simple step of refilling before the situation becomes more complicated.

Assuming the history of your alternator drive is unknown I would examine it carefully. If everything is tight and the "rubbers" show no wear I would leave them alone to avoid the risk of introducing a problem. On the other hand if there is the slightest slack anywhere I would fix it to avoid the inevitable trouble. Case in point: on my bike the evidence indicates the drive half coupling had come loose and flogged back and forth. As the slack grew it hammered the driven (alternator) half coupling with increasing severity until that also came loose. While the play in the coupling on the alternator shaft is only slight, I face the likelihood that the new coupling will be a loose fit on the alternator shaft due to wear. This consequential damage could have been avoided if the loose Z25 half coupling was addressed early. (To minimise the financial pain I procured a used Triumph alternator in the hope I can use the armature in the Benelli if necessary.)

One of the steps in Root Cause Analysis is to identify possible causes. Benelli alternator drive failure apparently occurs on some bikes but not others. Possible causes I can identify are: Variation in component fits due to machining tolerances, variation in material used for the "rubbers", variation in assembly practices including bolt tension and retaining compound application, variations in operating conditions (eg. rev range, temperature). Any additions anyone?

Another small development: Since Australia does not have the "High Temperature Sealer" specified in the service manual for application to the splines I referred the problem to Loctite technical support. I was advised to use Loctite 620 without primer and leave it to cure for three days before starting the engine, to ensure it develops full strength. Information was provided showing the use of Loctite Primer gives a faster cure but with lower strength. Hence I will use brake cleaner to degrease the parts and dry them completely before assembly.

I respect your viewpoint Rolf, as you seem to have a lot of knowledge based on engineering experience, and I am happy to leave my alternator alone for the time being, but will have a good look at everything again when I have got as far as a 3000 mile service, or if any unusual noises begin ! I usually prefer the maxim ' prevention is better than cure ' but have also been known to follow ' if it ain't broke don't fix it ', sometimes though I have reached the one which starts ' Oh bugger, I should have done........... '

 

New parts arrived - Kwakka rubbers fit perfectly!

My Benelli parts and Kawasaki alternator drive rubbers have arrived.

The Kawasaki rubbers (Part No. 920751790) look and smell like rubber, not plastic, and they are a good tight fit in the coupling. Each segment is compressed about 1mm when the coupling halves are pressed together. The only risk I foresee running these rubbers in the Benelli is that they may become loose in the coupling with use. If and when that happens I will replace them to avoid trouble. Backlash, oscillation and impact would result in accelerated deterioration and could loosen the coupling halves on their shafts once again.

If or when the rubbers will become loose in the coupling is the big mystery. Surely they cannot be worse than the original "rubbers" that were loose and very visibly worn after about 3 years and 18,000 km. As I ranted earlier; I am very optimistic that the Kawasaki rubbers will run happily for years.

Maniac motors supplied a Class 10.9 bolt. That has the advantage it can provide more tension than the original Class 8.8. However I will probably use my modified Class 10.9 bolt that has 12mm of unthreaded shank under the head. As stated in an earlier post, the intention is to allow the unthreaded section to elongate so it can help maintain tension. History may eventually reveal that the modification is either a good or bad idea!

On the new Z25 shaft both the gear and coupling are a clearance fit, hence the need for retaining compound. I am disappointed that this is necessary. I have more faith in mechanical interference than magic potions to prevent movement and fretting in any drive subject to torque reversals.

The Z25 shaft & gear have no match marks and the shaft has one oil hole. I understand from another thread the match marks were discontinued.

And yes Kevin, the clutch cover gasket is the reusable aluminium and rubber flavour, so the financial pain has eased!

 

Kwakka rubbers in - inspection method requiring enlightenment & sensitivity

I fitted the new drive half coupling and Kawasaki rubbers today. To give the retaining compound time to cure I will not start the engine till the end of the week, for participation in the Friday run with the local gang of motorcycling retirees.

The Kawasaki 920751790 rubbers in the new coupling:

I reused the old Z25 shaft because the backlash of the new coupling on the new shaft was not much different to the old shaft. Likewise I reused the old male half coupling on the alternator. Hopefully, with new, gentle rubbers, the bolt tension and retaining compound will prevent coupling movement on the splines.
Trying to avoid going back to this thing ever again, I broke the rule that says "change only one variable at a time" by using a modified Class 10.9 bolt to hold the coupling on the Z25 shaft:

Applied torque was 60Nm on threads that were cleaned and unlubricated, except for the retaining compound. The bolt was snugged then the new cone washer was centred by prising with a screw driver, before full torque was applied. Some 40mm square tube slotted at one end was used to restrain the coupling while the bolt was tensioned. It was also used to hold the alternator shaft while 45Nm was applied to the nut.

The coupling rubbers were lubricated with a smear of lithium grease before aligning the coupling blades with the gaps between the rubbers and pressing the alternator home by hand (it went-in fairly easily after a bit of wriggling and messing about).

Through all this I pondered Grahams issue from Post 21. I realise my use of the term "if it aint broke don't fix it" will be misinterpreted. To clarify, I don't advocate waiting for catastrophic failure but I also I prefer to the avoid the strategies of disassembly for inspection and replacing parts that are still servicable. Faults can be introduced this way. A better strategy is non-invasive Condition Monitoring; resorting to invasive intervention only when required.

How do you monitor the condition of a Benelli alternator coupling? I encountered two issues:
1/. Coupling rubber backlash can only be found with the coupling assembled. Graham said his rubbers "looked all right" but I initially thought mine looked pretty good too. It was only after removing the coupling halves from their shafts and assembling them with the old rubbers that I found the rubbers were loose in the coupling. They were worn on the circumference. That increased the gap between segments, allowing backlash.
2/. The bolt can be tight while the coupling is loose. The Benelli service schedule says the coupling bolt should be checked at every service. That check is flawed. My bolt was tight when I checked it but I found looseness and backlash in the half coupling behind the bolt.
It seems therefore, the best way to assess the condition of the alternator drive is to check for backlash while it is fully assembled.

This is a suggestion that may suit gentlefolk with the capability to undertake a delicate check that could damage their alternator if they get it wrong. If you believe you have the necessary enlightenment and sensitivity; read on:
Through the ribs of the alternator casing, near the slip ring end, there is a pressed metal fan. This fan can be used to check for drive coupling backlash. It requires very carefully inserting a small screwdriver from below and behind to gently press a fan blade in one direction, then the other.

This way I confirmed that my alternator drive now has no backlash. The only movement is a slight deflection allowed by the flexibility of the new rubbers.
If any backlash is detected the rubbers should be replaced as soon as possible, to avoid oscillation, impact forces and loosening of the coupling halves on their splines.
The method is only for those who can distinguish the fan blades, which are difficult to spot, from the windings and other alternator innards. Best disconnect the battery during this check, to avoid possible calamity.

On Friday I hope to find-out whether or not my alternator drive is still tight after perhaps a 300km run, cruising at around 3500-4000 rpm. I also hope my hand is no longer numbed by an awful vibration in the right handle bar.

 

Coupling still tight with Kawasaki rubbers, after 340km.

Nellie took me for a 340km run today, after Wednesday's exercise fitting a new female half coupling to Z25, refitting the male half coupling to the alternator and installing Kawasaki rubbers. I had been advised to leave the retaining compound to cure for 72 hours before starting the engine. My patience lasted only 36 hours. Never the less, the clatter from the alternator is gone. The vibration of the right handlebar was greatly diminished and no-longer numbed my hand. Slight vibration was still detectable at 3500rpm by gripping the handlebar tightly. It was just enough to tell me the bike was alive and was no-longer a problem. After the bike cooled-down I checked the alternator backlash with the risky, screwdriver in the fan method, described in the previous post. There was no discernible backlash. The Kawasaki rubber installation appears to be success. Time and operation will tell whether or not the remedy lasts but it is looking very promising.

A warning if you attempt the screw driver in the fan check - use a torch so you have enough light to see inside the alternator. Rock the rear wheel while the bike is in gear, to make the fan blades move so you can identify them. Please do not poke the wrong bit or use force, as it might damage your alternator.

 

Not so great news. I checked the Kawasaki rubbers I used and have found damage similar to that on the Benelli ones after 3,000 miles. There is no vibration and the rubbers are still a good snug fit, but it looks like I'll have to change them again after another 3,000 miles.

 

Not so great news. I checked the Kawasaki rubbers I used and have found damage similar to that on the Benelli ones after 3,000 miles. There is no vibration and the rubbers are still a good snug fit, but it looks like I'll have to change them again after another 3,000 miles.

View attachment 23977

Darn! But thanks for the info Kevin. Mine still has no backlash after 4000km but I had best remove the alternator after tomorrow's run to check for damage to the rubbers.

 

Kawasaki Rubbers failing

Eventually checked the Kawasaki rubbers after 5,000 km. They are still doing the job, the coupling halves are still tight and there is no backlash in the coupling but the rubbers are tearing, like Kevin's, where the edge of the coupling blades bears on them. It looks like the rubbers have been trying to extrude through the gap between the coupling halves. The tearing is only on the drive faces, which I take to show they can't handle the force of firing impulses working against the alternator rotating inertia and drive torque. I guess it's back to Maniac Motors for their elements and replacement at every service. The only thing I can see that might make a difference is to take a bit off the faces of the alternator so that the coupling halves engage deeper, leaving less gap for the rubbers to squeeze through.....but I am reluctant to risk messing-up a good alternator.

 

I can't remember how sharp they are to begin with, but would rounding the edges of the coupling blades be worthwhile?

 

Sharp corners and gaps.

I removed the sharp edges before initial assembly Mike. Perhaps it would help to add a spacer behind the drive half coupling to reduce the gap between vane tips and face of the driven half coupling. First step might be get an accurate measurement of the existing gap with the aid of plasticine. The damage on the rubbers in my bike suggests the gap is over 2mm. Maybe 0.5 millimetre would be more appropriate. My bike would have more gap than original due to wear after running for an unknown time with both coupling halves loose on the splines.

 

I have replaced the Kawasaki rubbers with Benelli parts in the alternator coupling on my TRE K 1130. My suggestion in an earlier post that the Benelli rubbers might be plastic appears to be incorrect. They appear to be some sort of synthetic rubber, although they lack the rubbery smell of the Kawasaki rubbers. Examination of the used coupling rubbers shows they compress and harden in service and polish with wear. After use the Kawasaki rubbers look like plastic in the same way the used Benelli items did. The Benelli rubbers are bigger than the Kawasaki items so there will be less space inside the coupling to accommodate deflection, which I presume results in the coupling having more torsional stiffness and hopefully slower breakdown of the rubbers. However it also means it will not absorb the torque spikes as well and will put higher loads on the other rotating parts of the alternator drive. My coupling has stayed tight for 6000 km since I refitted it with retaining compound, a modified bolt, Maniac Motors heavier cone washer and the Kawasaki rubbers. I hope it will continue to do so with the Benelli rubbers. This evening with the new alternator rubbers and a new clutch housing the bike ran quieter than I have ever heard it. The mechanical cacophony that had become so familiar seems to have diminished dramatically. If it starts to clatter again in the future I will have to assume something is awry.