×
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!
  • Students Click Here

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Jobs

Crankshaft/timing belt pulley interface
6

Crankshaft/timing belt pulley interface

Crankshaft/timing belt pulley interface

(OP)
I am interested in understanding the reasons for failure of a woodruff key joining a crankshaft to the timing belt pulley. This has happened twice on the same engine and I would like to find a reliable solution.

The engine in question is a (late 80's) Japanese 1.3 litre inline four with dohc and direct hydraulic tappets.

The key originally wore asymetrically as if it were taking an abnormal load in only one direction of rotation, that is it was changing into a Z-shape as sketched below. There was corresponding wear on the pulley and shaft where they contacted the worn parts of the key and all corners were rounded since the key was rocking in its seat (with the ensuing cam/ignition timing variation). The pulley inner diameter was enlarged and the pulley could wobble on the shaft but the shaft was not worn significantly. The accessories pulley which bolts onto this pulley was not significantly damaged.

CODE

 ________
|        /
|       /
|      /_
|_       |
  /      |
 /       |
/________|

The shaft keyway was enlarged using EDM, a new key was ground and a good pulley fitted.

After less than 20000km we've had to remove the timing belt and the same problem has happened. The key is not as worn as previously but the pulley ID is up by ~0.7mm. Shaft OD is effectively unchanged since flipping the pulley over onto the undamaged side it is tight(the pulley edge protruded slightly more than the shaft). The bolt retaining the pulley onto the shaft was loose and threads were damaged).

When trying to find the cause a local machinist asked about AC but the car doesn't have it. The accessories pulley seems to be a damper with a rubber ring - could this be the culprit? There was an issue with the alternator being slightly misaligned from the drive pulley but this was addressed at the time a new key was fitted.

Did anyone here encounter anything similar? Any suggestions to avoid another failure in a year's worth of driving?

I would try to use threadlock on the shaft-pulley interface but I am not sure whether it is appropriately strong...

I have seen similar damage on two other engines of roughly similar age - an inline three from the same manufacturer (but not the same engine family) and another inline three from another Japanese manufacturer.

N

RE: Crankshaft/timing belt pulley interface

I'm no expert but I'm gonna guess the three cylinder crank harmonics are the root cause of the problem. A press fit along with a good hard key might reduce the movement? Loctite designed for cylinder sleeves might help? If the damper is bad or the incorrect one for the engine, it could exacerbate or be the entire issue.  

RE: Crankshaft/timing belt pulley interface

If the accessory pulley/ damper or any damper has been removed, replace it.

I'd use Loctite to repair a loosened press fit, but not a super- strong grade; probably PST since it's in my toolbox anyway.  You don't need great strength because of the large area involved; you just don't want the pieces moving relative to each other.  For that reason I'd bed the key in Loctite, also.

 

Mike Halloran
Pembroke Pines, FL, USA

RE: Crankshaft/timing belt pulley interface

(OP)
Yes it could be harmonics, but on the other hand the repeat offender is a four cylinder which should run much smoother.

Damper and timing pulley are both stock, that's why I'm perplexed.

What's a good grade of loctite for such an application?

N

RE: Crankshaft/timing belt pulley interface

Sorry I misread and didn't see it to be a 4 cyl. which has slightly better harmonics but it too can still be the issue?

I'm not sure what Loctite number they use these days for cylinder press-fit sleeves but I'm sure Loctite has that info. on their website. Be advised that this type of Loctite is designed to be a permanent installation, thus higher strength. The Loctite website should advise on the different strength fluids available and if heat is required for removal as is the case with most permanent chemical locking fluids.

RE: Crankshaft/timing belt pulley interface

The crank snout needs to be in good condition, the hub needs to have a good tight interference fit and the key ways need to be good, straight and true.

The key needs to be hand filed to a gentle tap fit. The surfaces need to be clean and all burs need to be removed.

The pulls and hub need to be concentric and in balance.

Regards
Pat
See FAQ731-376: Eng-Tips.com Forum Policies for tips on use of eng-tips by professional engineers &
http://eng-tips.com/market.cfm
for site rules
 

RE: Crankshaft/timing belt pulley interface

Do you know for certain that the crank pulley end bolt was properly torqued? This is a common failure mode as a result of too low torque on the crank pulley bolt.

Another Japanese engine I know well requires 217 foot lbs of torque (factory spec) on the end bolt to keep the pulley from rocking on the keyway.

I know of several instances of do-it-yourselfers changing the timing belt and not re-torquing the pulley bolt down to the required torque.. The rocking pulley damages the crankshaft to the point the only cost effective solution is a new short block.  

I'm sure you are already aware of the critical function the harmonic damper pulley plays in damping crankshaft torsional vibration and why the pulley to crank connection is critical and carries considerable torsional stress.

RE: Crankshaft/timing belt pulley interface

2
Suzuki, MAzda, or ??
What is the torque spec for the crank bolt?

Does either of these look familiar?
http://i9.photobucket.com/albums/a53/whitfimb/ZuIFS/Crank1.jpg
http://carphotos.cardomain.com/ride_images/1/1185/4241/2962120001_large.jpg

Sadly Insufficient crank bolt torque 30,000 km ago is likely behind it all.

A bozo non-mechanic (he's real good, he works on Corvettes and he races motorcycles) replaced my daughter's Integra's timing belt, apparently without consulting a manual.  He got the exhaust cam timing off a tooth ( 3 times), so it ran terrible with retarded cam and ignition timing.  I checked things out after maybe a hundred miles, found the crank bolt maybe 45 lb-ft, and  the crank key was dented several thousandths as had the timing sprocket keyseat.

A few weekends ago I changed the timing belt in Volvo 245 I bought several years ago, but only recently started driving.
The crank bolt spun right out the second the impact wrench hit it.  The crank key had lightly pounded the damper keyseat so I flat filed all the hub and cam sprocket faces. They needed it. The bolt torque spec is some large number, plus an additional 1/6 turn.

In the old days some dampers and timing gears/sprockets were an interference fit on the crank snout to prevent the deadly micromotions that a key, really ANY key, is simply not man enough to resist.  NOwadays it is common for easy-to-assembl slip fit components to rely completely on heavy crank bolt torque to apply immense axial clamping to hold things still against the persistent and insistent squirming of the cam drive and damper and etc. If that end of the crank was "stock" with no added great fat pulleys or added driven components I'd say at the last timing belt or crank seal change there was not enough torque applied to  the bolt.  Perhaps in addition a few faces of the components needed to be dressed flat and thus shook a few salty sprinkles on the cyanide cupcake. Insufficient torque by itself is enough to allow (cause?) the start of a awful cascade of damage to perfect keys, crank, and hub in a few days. Slightly insufficient torque may take a year.  The alternating torsional and radial loads of the cam drive will jiggle and wiggle the sprocket/gear, the key will be unfairly asked to transmit torque without benefit of the powerful frictional forces on the mating faces, and fretting and pounding will always make a mess of everything real quick. The second time all the faces were damaged so in service even if the torque was correct they were quickly buffed down or embedded and sank together a few thousandths of an inch in use and the bolt preload and precious clamping was lost again.  

There are websites that describe patching up damaged components with various compounds like Loctite. Loctite on a fastener on guard securing a metallic joint may keep the fastener from "loosening" but if the bolt's clamping force is not adequate and adequately maintained the battle is quickly lost. (And the proper torque would make Locite un-necessary).

RE: Crankshaft/timing belt pulley interface

It's not just the key way that gets damaged.. the diameter of the end section of the crank starts getting worn down and out of round to the point the pulley wobbles on the crank.   

RE: Crankshaft/timing belt pulley interface

I'm with DanEE, Tmoose and Greg on this one.
The most likely cause of key /keyway damage is insufficient torque on the crankshaft nut/bolt.

I've seen this on a Fiat 1 litre FIRE and a Land Rover 200TDi - both down to not enough torque. Can't remember waht the Fiat requirement was, but the LR was about 240ft.lb. and it was nowhere near that when I dismantled it.

Bill

RE: Crankshaft/timing belt pulley interface

(OP)
Thanks to everybody for your advice and opinions.

I agree with the consensus that insufficient torque was the culprit.

I'll check if there's any damage to the crank but I'm a bit wary of machining it again - the keyway was already enlarged from 5 to 7 mm a year ago. And yes it costs a bit less than a short block! :)

I'll have a new key made and if necessary face the surface where the pulley rests. Unfortunately the slight wear to the crank won't enable an interference fit but I hope some loctite as a filler will help. Then lots of torque!

Just for the record the engine is a Suzuki G13b and the other cases I've seen were a Suzuki F8 and a Subaru EF10. All three were in bad shape - the Subaru was pretty much as the photos Tmoose posted.

RE: Crankshaft/timing belt pulley interface

I agree, the bolt must be tight and the surfaces it pulls against must be flat and true.

My experience is with large blower drive hubs on snouts not really designed to carry them.

In these cases, a really tight bolt is not enough. Even with the next size up bolt long enough to reach into the front main journal area, they still work loose if ANYTHING else allows any movement.

I have seen the edge of the key way pushed up to make a bur, but I never saw a snout worn down. Maybe that is because of very regular mainteance.

If the interference was to lose, we would opt for a slightly smaller hub bore and dress up the snout wit a file and stone. Still a lot less cost than a new short block.

If the pulley has been lose long enough to significantly wear the snout, I guess a costing exercise would be necessary as it would be crank out to repair the snout.   

Regards
Pat
See FAQ731-376: Eng-Tips.com Forum Policies for tips on use of eng-tips by professional engineers &
http://eng-tips.com/market.cfm
for site rules
 

RE: Crankshaft/timing belt pulley interface

(OP)
I got my hands on the service manual which mentions a torque of just 80 ft lb for the bolt. It seems quite low compared to what people here have mentioned. I'll check the condition of the threads on the crank (those on the bolt are damaged) and if they're OK I'll risk tightening a bit more than that.

Pat, how do you define "too loose" as describing the fit? I would assume that if it has to be tapped lightly to slide onto the snout then it will not rock significantly unless the bolt is loose. Is that right? The owner (my brother) ordered a new pulley so within a week or so I can confirm the fit.

Regards,
N

RE: Crankshaft/timing belt pulley interface

The problem with these designs is that a bolt isn't designed to prevent rotation of the parts or it shouldn't be if it is. The joint should be in rotational shear. So the hub should be splined, keyed, doweled, etc. A pressed on hub is a poor substitute but better than a slide fit.

Loctite will help some as will tightening the bolt tighter IF the threads and bolt can stand increased tightening torque without failure. Determining if the threads should be dry or lubed is VERY important in this instance as it can change the clamp force by 30% or more between the damper and crank.

RE: Crankshaft/timing belt pulley interface

On a blower hub, at least 0.001" interference is required. It needs to be pulled on with the bolt. It takes a reasonable torque just to pull the hub on even with the hub and snout interface covered with anti seize.

 

Regards
Pat
See FAQ731-376: Eng-Tips.com Forum Policies for tips on use of eng-tips by professional engineers &
http://eng-tips.com/market.cfm
for site rules
 

RE: Crankshaft/timing belt pulley interface

haven't been here for a while but I can tell you how we used to repair this problem.
firstly as stated prev the key is only there to postion the pulley, pulley should be a transition fit and torqued, the best way to go now is to replace the crank as to repair it properly will require removal anyway and a used crank shouldn't cost the earth (measure and polish 1st)
If crank is out you can knurl and grind the snout it should be a nominal metric size  and make up a stepped key to suit the 7mm keyway in the crank and 5mm in the new pulley remember the keyway is worn away from rotation direction and the step needs to match this, you will need to face the shoulder and use a radiused wheel if required also make sure the pulley bore ''overhangs''the snout so that the bolt face or washer torques on to the pulley not the snout.Don't use loctite as in the end its only plastic and will soon compress.If the thread is worn you may be able to helicoil but there could be a problem with the tap cutting into the key way.    

RE: Crankshaft/timing belt pulley interface

As an example of reliance on bolt torque, the VW Audi 2.5 I5 TDI does not even have a proper key - just a notch for orientation in the crank and matching lump on the pulley/harmonic damper assembly.  And the torque....? 200Nm + 270º..... and that is tight.  It's a big bolt with a fine thread.  It does seem to work, but it makes changing the timing belt an even bigger challenge.

Nick

RE: Crankshaft/timing belt pulley interface

You don't need to take the crank pulley off a TDI engine in order to change the timing belt.

You are supposed to take the cam pulley off. That one has no keyway at all - it's just a taper fit with a center bolt to lock the taper. It's usually not too hard to knock that pulley loose.

RE: Crankshaft/timing belt pulley interface

Brian,

That is quite true for the 4 cyl TDI.  Unfortunately not true on the I5 (although part about the cam sprocket is correct).  Wish it was....!  Don't know about the V6 TDI - I've not worked on one.

Nick

RE: Crankshaft/timing belt pulley interface

The large bolt cranked tight is a Hail Mary cost savings approach... <LOL>

RE: Crankshaft/timing belt pulley interface

With torsionals on a 3 or 4 cyl. engine that may NOT be enough... <LOL>

RE: Crankshaft/timing belt pulley interface

For the priest that heard my confessions when I was still young enough to be gullible enough, three Hail Marys and one Our Father was always enough for any sins, crimes or whatever else I managed to get up to.

Regards
Pat
See FAQ731-376: Eng-Tips.com Forum Policies for tips on use of eng-tips by professional engineers &
http://eng-tips.com/market.cfm
for site rules
 

RE: Crankshaft/timing belt pulley interface

If he only knew what you are up to now... <LOL>

RE: Crankshaft/timing belt pulley interface

With all the talk of "over-torque" "bigger bolt" and "interference fit"...please tell me that none of you has ever installed a flywheel on a Mini !

Rod

RE: Crankshaft/timing belt pulley interface

Oh yeah..... they can be aggrevating little monkeys!  Getting them off the taper being the main challenge.  It's been a while since I worked on a Mini - can't say I'm missing 'em though!

Nick

RE: Crankshaft/timing belt pulley interface

(OP)
They are definitely unwilling to let go of the crankshaft! Had a beaten up Metro a couple of yers back, bought with a compression leak and managed to run (hard!) for a few months. I was going to change the block since I had a spare but had to give up because of the flywheel.

On the original topic, the car is running again. I cleaned up the edges of the keyway, had a key and a new washer made up, refreshed the tapped hole in the snout. That was one hell of a job with the engine in place - the car was pretty far from home. We assembled with permanent loctite between shaft and pulley and mineral oil on the bolt threads and tightened to "this feels a lot" (I need to buy a torque wrench, I know!)

I'll check the torque when the new timing belt arrives in a couple of weeks and can work somewhere more convenient.

Thanks for all the help!

RE: Crankshaft/timing belt pulley interface

Where do you suppose the timing belt pulley ended up compared to the cranksh'ft TDC ?

RE: Crankshaft/timing belt pulley interface

(OP)
What do you mean by that?

RE: Crankshaft/timing belt pulley interface

Close enough.

RE: Crankshaft/timing belt pulley interface

(OP)
I have no idea what the term SFA means, but if tmoose meant "how far is camshaft timing from nominal" I haven't measured anything.

When the keyway was enlarged a year and a half ago the timing pulleys ended up offset a few degrees - I'd guess a few (less than 5) crankshaft degrees advanced with respect to nominal. this was because the key was stepped an extra 2mm on the trailing side, but the keyway was enlarged a little less than 2mm on the trailing side with the difference on the leading side.

RE: Crankshaft/timing belt pulley interface

hereza link to some Results of shifting a stock 4 cyl Volvo cam a bit either way from the popular, but to my mind optimistic and technically hollow (not measured) "straight up."

http://forums.turbobricks.com/showthread.php?t=209648
http://www.ovtuners.org/gallery2/v/members/klr142_001/MY244/dynocharts/?g2_page=1

On a 1.25 inch diameter crank snout 1 degree ~ 0.011"

If it runs well I probably would not care, but I like to think I'd make an effort to make measurements to estimate what I'd done, in case I did not like what happened.  I check cam timing to some extent with every timing belt change, rather than relying on either the accuracy of the factory machining or sprocket etc markings, or my skill at aligning them.

 

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources