Pitt944T

Hi folks

So i did read Clarkes and used the search feature but this one seems to escape me.

If I set tension to 40lbs at TDC then roll back to 10 degrees before TDC the belt seems loose (when measured at the 10 degrees BTDC position).

If I set the tension to 40lbs after rolling back counter clockwise by 10 degrees BTDC and Then turnover the engine again back to TDC the tension seems high, like 70 lbs (although I can still twist the belt through 90 degrees by hand)

I am using a crikit tool to measure tension, yes its imperfect but I actually get very consistent readings. The online guides seem to rely on a more dynamic "max value" style gauge (like the arnworx or porsche tool) which the crikit isnt.

SO...

Measure at TDC? or after rolling back to 10 degrees before TDC?

I thank you.

Van

I'd probably err on the loose side...

Here's a video explaining how I do it:

fejjj

Van,

Your videos are very well done and most helpful.

Thank you for taking the time and effort to produce these!

Pitt944T

Thanks! was that video taken with the engine turned backward briefly to before TDC or after turning it by hand only in the direction of usual rotation?

Pitt944T

Looks like maybe the factory manual has the answer. It never talks of testing the the tension directly following forward rotation. Always rotate back 10 degrees then measure

Van

When you use the factory tension gauge, you always turn the engine backwards a little bit to put the "slack" in the long run, where you'll be measuring it.

If the belt is too loose, it might slip on the water pump, which will, of course, not give you good cooling, but it will also cause friction to heat up the belt and likely cause it to fail.

If the belt is too tight, it puts excess side loads on the roller bearings and the water pump shaft, which is likely to cause premature failure in one of those things.

Tiger03447

Great video, Van! More things to ponder!

odonnell

What do you guys think of revolving the engine (by hand) several times before checking the final tension, so the tension is evenly distributed across the belt?

It's because of the tools that tell you to turn the crank backward to put slack on a certain length... I remember when I first got this car and did the belt, that got me into trouble, and it was a meaningless reading from the tool because I wasn't sure how much slack was enough.

Now when I do a belt, on any car, I like to install it and set a preliminary tension. Then rotate the crank 4-5 times, and assess the tension, repeating until it's satisfactory. Is that a valid approach with the 944 timing belt tools available?

Van

Absolutely - it should always be rotated a few times. The primarily reason it to make sure the belt is running on the proper (natural) spot on the pulleys. Otherwise there will be small amounts of lateral deflection that are causing the belt to be shorter. And if you tensioned it this way, it will appear loose after a few revolutions.

Pitt944T

Can you define "got me into trouble"? I admit to being confused over this. I absolutely would turn by hand a couple of times after installing the belt but should we then turn back before we measure. The factory manual (when using the factory tool) says a categoric "yes". The factory manual always measures after turning counter clockwise 10 degrees for both initial setting and tension check.

I am curious why we wouldn't follow that factory procedure.

odonnell

I remember I had an older version of the Arnnworx tool... the instructions said to turn it back as well, I cannot remember the amount. I ended up (apparently) putting too much slack on the measured length, and hence over-tightening it. No damage, but the tensioner roller was very noisy so I replaced it with a new one and had a well-rated indie shop set the tension with their factory tool.

The last couple times I've just done it by feel (updating tension at 1000 and 15000 miles along with TBs on other cars), and check it every couple months to make sure nothing crazy is happening. For my next belt change I want to get a new Arnnworx tool, though.

bonus12

Van's video provides multiple methods to ascertain and set the correct tension. My first time setting tension, I was troubled because the tension seemed to be changing, or inconsistent. That's how I learned to set the tension once, spin the motor a few times, set it again, spin the motor, etc. The tension would be unevenly distributed on both "sides," if you will," of the belt's rotation path.
It's important to use as many methods as are helpful to measure tension, and important to measure tension after distributing it by rotating the motor.
I know a lot of people use the tool, but there's nothing like honing your instincts. In some respects, a tool may only deceive.

mikey_audiogeek

Not quite true. However not wrong either. With a 944 engine with spring tensioner you set the tension without rotating back 10 degrees.
However whenever you measure the tension using the factory tool on any 944 engine, you need to clock the engine back 10 degrees to put the slack on the correct part of the belt.

Info here:
https://rennlist.com/forums/924-931-...l#post12287879
https://rennlist.com/forums/924-931-...l#post10804511


Cheers,
Mike

Lizard944

Van's video is exactly 9:44 long. Coincidence? I don't think so...