Timing belt failure log MTBF

[erikv11]

Y'all are not paying much attention to the thread ... OP only wants to know about failures.

The goal of this thread is to get > 30 logs of timing belt failures with mileage to see if we can get an idea of how long these parts last.

Entry format

First line: year, model, engine, year replaced, mileage replaced,

Second line - mechanical commentary on what failed, if anything

[polskamafia mjl]

Using replacements before failure data obviously gives a lower bound to MTBF vs actual MTBF, but for the purpose of this study, that will suffice.

It sounds like OP is also interested in replacement w/o failures.

[abscate]

Clarification - all data is useful - even ...my car has run for 80k on original timing belt - never replaced, never failed. Such a data point will pull down the MTBF lower bound, but thats ok.

[instarx]

Clarification - all data is useful - even ...my car has run for 80k on original timing belt - never replaced, never failed. Such a data point will pull down the MTBF lower bound, but thats ok.

Of course you can include change intervals before failure, but if you do your results won't say anything at all about MTBF. All you will have is is mean time between belt changes.

[backfoot100]

OP's requested data on my TB changes.

'98 S70 T5. TB tensioner is hyd.
5/03; 40K, purchased from original owners
11/03; 50K, 1st TB. I mistakenly thought these were 50K intervals. My bad.
6/06; 121K, 2nd TB
2/08; 179K, 3rd TB, pulleys, tensioner, water pump. This change was only a 60K interval again. I was starting to freak out that the OEM pulleys, water pump and tensioner were still in there and figured discretion was the better part of valor.
2/10; 250K, 4th TB, pulleys, tensioner
4/12; 324K, 5th TB, pulleys, tensioner, water pump

Currently at 357K. Every TB changed looked virtually unworn. Started changing tensioner and pulleys at every change just for piece of mind. Water pumps every other change. Always reading about the horror stories of catastrophic failures which is the reason for this thread right? I like to err on the side of caution.

[abscate]

Always reading about the horror stories of catastrophic failures which is the reason for this thread right?

Exactly. For someone in your shoes, keeping car indefinitely, your err on the conservative side. You also learn to do it yourself and save 4000 USD in the bargain.

Getting a lower bound and standard deviation on the MTBF is useful to the person who buys a 100k Volvo project car, and wants another 20k out of it, for example. WIth this info, you can gauge how much of a risk you are taking running existing TB.

[JimBee]

The predictive validity of this analysis will be suspect due to huge random variation in other factors related to how hard a car was driven, maintenance items, whether it was used for towing, possibly seal leakage and maybe belt manufacturer. Also, as others have pointed out, how long it was out of service could affect longevity (failure data could help validate or discredit that theory).
It might be quite easy to control for some of those factors simply by asking for a little more information and ranking answers on a scale of 1 - 5 for the respective factors—then of course including all that data in the research model. Those scaled data would still be subjective but could be somewhat helpful in making the predictive model a little more factual, rather than loosely conjectural.

[rspi]

JanuWan: I promise you, after 155,000 miles, everything wasn't in perfect condition. I'd be afraid to try to push rollers past 140,000. I seriously doubt they will go 210,000. Those things had to be loose.

JimBee: I don't think it matters how hard you drive these cars. The belt, rollers and water pump will go 70,000 under any legal circumstance. Even if you are able to cruise at 110+ mph daily the rollers will make noise before failing.

If people would use their senses and listen to their car complain and do something about it before it fails, and follow maintenance schedules for stuff that clearly has time change intervals, a broken timing belt would be a very rare occurrence. Matter of fact, I'm sure there are more cars blowing head gaskets than loosing timing belts. Only because the owners are playing "engine roulette" by adding coolant daily/weekly. I pull every timing belt cover off every Volvo I see in the JY and I'd say that less than 15% of them are there due to timing belt failures.

[hansenjg]

If you are still looking for data, here is mine for my 2000 C70 Ragtop. Bought in 2005 with only 20K on it. Changed OE belt at 125K miles in 2010. Also replaced tensioner and water pump at that time. Should not have replaced pump I think. 65,000 miles later (after a fairly long time of losing just a little coolant constantly) the water pump seized up and took the timing belt off. Engine destroyed. When I opened it up to see what I could do with it, the timing belt was still in pretty good condition except for the cuts and scrapes from when it got thrown off. Don't use cheap water pumps and don't replace them unless necessary!

[mecheng]

Which statistical software are you using? Are you using Weibull statistical analysis?


I can't give you a failure point but:
My 850 had the belt changed at 70,000 miles and it looked as close to perfect as possible. Only the timing belt was changed. The car has 100,000 miles and everything is perfect: rollers, tensioner, WP.