Timing Belt Tensioner


Tensioner spring, main body, and adjusting bolt (sitting in the main body).

Examine the tensioner body and look at the shape of the slot in it. The adjusting bolt fits in this slot. With the bolt loosened, the tensioner can then move. Tightening the adjusting bolt moves the tensioner body to the left.

I replaced the timing belt and its tensioner on my 1991 Civic LX in July 2007 at 197k miles. Everything went well except for setting the tension. I followed the factory service manual steps, but I did not understand what the steps I took were doing.
On my first attempt to set the tension, the belt was too slack. When I started the car a muffled rat-a-tat-tat noise, like rubber on plastic, came from under the hood. Putting my hand on the TB cover as the car ran, I could feel that the noise was in synchonization with vibrations on the TB cover. I took all apart again, rotated the crankshaft by hand, and watched how un-smooth the motion of the TB was. This is when I realized what the rat-a-tat-tat noise was: It was the TB hitting the TB covers. I concluded I had set the tension wrong.
I studied the tensioner more and experimented with it installed and TB covers off, trying to understand the forces involved so I could set it properly. Eventually I learned from Gator88 at honda-tech.com that the bigger hole on the periphery of the tensioner housing (and opposite the hole to which the tensioner spring attaches) must mount on the peg on the engine block. Big doh on me. This peg is a pivot point for the tensioner. One should look for this peg, since so far I have not seen a factory service manual that emphasizes it. Googling for information on installing the tensioner turned up only one quasi-functioning site (jdmcivic.com) that talks about the "big hole" and the "little hole" on the tensioner. I suppose the "big hole" having a function should have been obvious, but this went right by me for quite awhile. For me, this reinforced the maxim to always question each design feature.
Once I had the tensioner properly mounted I was able to get the tension set properly using the manual's directions to (1) loosen its adjusting bolt a half-turn or so, (2) rotate the crankshaft so that the camshaft rotates by three teeth [here I also watched for tension on the forward side of the timing belt], (3) tighten, and then torque to spec, the adjusting bolt. I then rotated the crankshaft fully a few times and watched for smooth turning of the belt over the sprockets' teeth. Some observations on this adjustment process:
  • The factory service manual states that the tensioner "is spring loaded to apply proper tension to the belt automatically after making the adjustment [given in the manual]." I think what this statement means is that, during the adjustment process, the spring applies just the right amount of force to locate the main body of the tensioner so that it puts the proper tension on the timing belt. Once the adjusting bolt is torqued down, though, the spring and even the peg are irrelevant. The aforementioned "auto adjustment" takes place only for an instant while one is following the adjustment process. Since the belt likely stretches over its lifetime, this argues for maybe re-adjusting the timing belt periodically during its six year/90k mile life.
  • I suppose the spring stays in tension at all times, though this is a fixed tension, since the adjusting bolt is torqued down. So this argues for changing the spring every other TB change or so. If the adjusting bolt somehow comes loose, it also seems that the spring and pivot point ensure some fail-safe operation until the operator can hear enough noise to get off the road, assuming s/he knows enough to do so.
  • Does it matter which way the hooks at the ends of the tensioner spring are oriented? It seems to me the difference would be small, and thus there is no warning about this in the manual.
  • Getting the belt on is easier with (a) mounting it first on the larger camshaft sprocket, then focusing on the smaller crankshaft sprocket; and (b) the tensioner spring off its upper mounting point. A coat hanger bent into a hook to hold the end of the spring and pull it onto the upper mounting point worked really well.
  • I could only get a torque wrench in for final torquing of the tensioner by going in from underneath, with the crankshaft pulley off.
Site http://www.tegger.com/hondafaq/misc/tensioner.pdf shows a pdf drawing of the tensioner from a 1991 Acura service manual. (Thanks to Tegger for this drawing and his discussion on this matter.) Watching the aft side of the TB go from slack to reasonably tense when tightening the adjusting bolt seems to be very important. Checking for smooth operation by rotating the crankshaft a few times also seems important, or at least a potential telltale as to whether one has messed up.

Back to Table of Contents