Ducati.org forum banner

see how to set the belt tension with no tools!!!

111K views 34 replies 18 participants last post by  DaveS 
#1 · (Edited by Moderator)
see how to set the belt tension with no special tools!!!

Finally, after a couple of months since I had this idea, I made a report so you guys can follow and do it yourselves!!!!

The principle is to put each cyl @ tdc end of the compression stroke when the cams are @ rest and somehow check the slack on the belt......this is how you can do it yourself without any fancy tools and be pretty close to factory specs:

First, remove everything needed to get to the belt covers and remove them, then remove the coils and spark plugs.

Second, align the green dot in the layshaft pulley with the mark in the clutch cover, put a screwdriver through the spark plug hole and rotate the crank (bike on 6th gear, the rear tire in the air) back and forth to make sure you are @ tdc, the cam's slots should be aligned with the head surface to make sure they're @ rest and you are @ tdc on the final of the compression stroke.

Third, mark the cam's pulleys to know your front cyl cam timing alignment marks, it'll look like this.......(layshaft pulley is already marked from factory, the cam's pulleys I did without).......

see pics 1, 2, 3.

now you can remove the belt to do your valve adjustment by unloosing the 12mm nut on the eccentric adjuster and removing it or readjust the belt (will show you later)

Fourth, find tdc on the rear cyl with the screwdriver through the spark plug hole but make sure you are @ the end of the comp stroke (cams @ rest, cams slots will look exactly like on the front cyl, cam's slots align with the head surface, shaft pulley will be approx 1/4 of turn away from green dot), mark them, this will be your rear cyl cam timing alignment marks.......it'll look like this...

see pics 4, 5, and 6.

now you can remove the belt to do your valve adjustment by unloosing the 12mm nut on the eccentric adjuster and removing it or readjusting the belt.

Fifth, to adjust the belt we'd need a hertz reader (should be @ 100hz, not lower than 90) so I figured if I can adjust it with the proper tool and then measure somehow the slack on the belt It'd be pretty accurate, so after trying different methods I came up with the perfect tool, a 10 mm wrench!!!!!

So the idea is to centre the tool in the middle and push up and down on the belt to check the tension, It'll look like this........

see pics 7, 8, and 9.

After adjusting the belts with the wrench method I rechecked them with the Hz reader..........

see pic 10.

Auto part Engine Vehicle Car Automotive engine part


Auto part Wheel Tire Automotive tire Rim


Auto part Engine Automotive engine part Clutch Tire


Auto part Vehicle Engine Fuel line Car


Auto part Engine Gear Tire Automotive wheel system


Auto part Motor vehicle Vehicle Engine Car


Auto part Tire Automotive tire Wheel Vehicle


Auto part Vehicle Automotive engine part Engine Steering part


Auto part Tire Wheel Automotive wheel system Brake


Electronic device Technology Gadget Electronics Portable communications device


Sixth, if you removed the belts you'll have to do the rear cyl first, align the timing marks, then put the belt, then the adjuster and the 12mm nut with its washer, then put a big flat screwdriver between the nut of the adjuster and the ID ring to rotate it and tension the belt and gently tighten the nut (this will allow you to have some tension and readjust it as needed).
Once you get to the desired tension then torque the nut to 26NM but hold the adjuster @ the same time (it might twist and you'll have to start over with the adjustment).

Before you put it all back together rotate the engine slow by hand to make sure you didn't miss the alignment marks, it should turn over smoothly without any problems.

Seventh put it back together and go for a test ride!!!!!!!!!!!!!!
 
See less See more
10
#4 ·
This is a nice idea, and we all want to achieve the 100 Hz setting without spending 100s of $$.
Many of us have been setting belt tension for years, by feel. Not too loose, not too tight.
The variable in the method you describe is just how much pressure is exerted against the belt to achieve the 10mm distortion.
Part of the process is having a tool which holds the eccentric pulley whilst you tighten the nut. I made such a tool, with 2 x 3.8mm pins set 10mm apart, and an offset handle to clear the airbox (on the rear head).
The other part of the process is checking them frequently - particularly after fitting new belts, or doing trackdays.
I guess the question is, how close to the 100Hz setting were you able to achieve with the 'spanner method'?
 
#5 ·
This is a nice idea, and we all want to achieve the 100 Hz setting without spending 100s of $$.
Many of us have been setting belt tension for years, by feel. Not too loose, not too tight.
The variable in the method you describe is just how much pressure is exerted against the belt to achieve the 10mm distortion.

+1 My thoughts as well

I have a question though Devimau,
Is 100 mhz measured at the midpoint between the cam shafts? According to the service manual, 110mhz is recommended at the midpoint between the crankshaft and belt tensioner. I'm just wondering why different spot was picked as a measuring point.

To me it seems the longest section of the belt (red circle) would yield the most accurate result.
Or is it the way you guys been doing for a while on Ducatis?
 

Attachments

#6 · (Edited)
Subbie79: the mathesis or dds have an optical reader and gates has a mic and those signals have to be picked up were you pointed the red circle on the horizontal cyl and between the layshaft puley (is not the crankshaft) and the fixed roller on the vertical.....however..... 100 hz with those readers equals to a flex of the belt plus and minus 5mm (that's why I used a 10mm wrench as a center point) @ were you pointed the green circle.
And yes, I've been doing it for quite a while, I think you missunderstood.....
Pat, the flex is pretty much all the way you can, there's a point were the belt just doesn't flex anymore, me and others who tried it read within 3hz from each other.
I remember the 6mm allen method on the 2v motors but It never worked for me, this one I've been monitoring it for a few adjustments now and is quite accurate, always within 2 or 3hz for me.
Ducati makes a tool to hold the tensioner while you set the tension but I find it easier to star with low tension and tighten the 12mm nut snug and work my way up with a long flat and skiny screw driver on one of the holes for the tool and works quite well, it makes it way easier.
 
#7 ·
Thanks for going to the trouble of putting all this up Devimau. Very helpful.
I was thinking of lashing out on the Riva freqency meter, as the two I'd downloaded (even with a $50 uni-directional microphone) didn't give meaningful readings.
I still like my tensioner 2-pin tool for setting, and may make a fixed tool to replace the 10mm spanner (what you call "wrench") for best accuracy.
Thanks again for sharing. Within 2-3 Hz is incredible - that accuracy is difficult to achieve even WITH the frequency meter.
 
#10 ·
FYI,

I have used Visual Analyser with a cheap mike on my laptop giving reliable readings. VA is a free app. Just google Visual Analyser.

Cheers
I have Visual Analyser downloaded also, but get a constantly changing readout, even when all around is quiet.
That's why I bought the uni-directional mike - but I still couldn't get a meaningful reading.
Might go to plan B and make a 10mm 'gap tool' like Devimau's.
I still set my belts by feel, and have had no problems.
I believe the service interval on these is too long - they'll always be loose when you get to them if you leave them for the recommended interval.
That's all about matching Honda et al on servicing, not optimal maintenance.
 
#11 ·
BZR,
It was a while since I did mine so I cant recall the
specifics as far as reliability is concerned. I took
multiple measurements and they all clustered
around the same Hz so they were at least repeatable.

The difference between say 95 and 100 Hz was a very
small movement of the tensioning roller and yes with
the analyser you do not get a constant 100 hz when
the belt is plucked. You get a range with the 100hz
being the "main" frequency.

I also set mine using a
spacer for the deflection of the belt and then
tested with the meter. The repeatability was
also "good enough" using the criteria that it
should not be over 100Hz so as not to overtighten
(the cylinder length increases increasing belt tension)
and therefore stress the belt. Ducati is quite tolerant
of the belt being looser but not tighter.

Frankly, the change in Hz is so sensitive to tension
roller movement I just cant see the mechanics in
the workshop devoting the time to get them perfect. It
seems a bit of overkill, remembering that belts have
been used reliably in machines for a long time without
a problem and without setting by using a Htz meter.

Ive agree with Pat1098 that setting them by feel
so they are firm (ie for me 4mm deflection at one
finger pressure between the cam rollers) is probably OK
- it certainly was with my 749.

BTW on my 749 the belt tension prior to adjustment
was way below 100hz and different tension on each
belt. After adjusting to the correct tension (with new belts)
I admit to not being able to discern any difference
at all in the engine performance, fuel economy etc. Ive
done app 7000km since then without any problem.

I have attached the photo of the belt deflection using
a 12mm drill bit wedged between the top casing and
the belt. The gap between the top casing and unflexed
belt is app 8.5mm givng a 3.5 - 4mm flex which was
app 100hz.

Hope this helps.
 

Attachments

#12 ·
A bump on belt tightening

There is likely to be a lot of this happening about the place soon as a result of the present eco conditions, however in this case it relates to my 999.
The belts were fitted by a noted technician not in favour of using modern technology, but that's another matter.
Having just changed the oil and filter 4000kms later, thought I'd check the belt tensions. I'm a long way from anywhere and a novice Ducati-wise, so the best I could do was do some research and utilize my guitar tuner.
I considered the belts to be lose - and they were, but at the same time I had no yardstick with which to do a comparison. Finally in this regard I had success by having the tuner mounted close by, gradually applying tension to the belt, and plucking it at the same time. This raises 2 issues I'd be grateful for clarification on from a learned technician please.
When I released tension on the eccentric nut, and then commenced to turn it in an anti-clockwise direction to tighten the belt, the nut went through 1/2 turn or more before that started to happen. That occurred on both belts. my thought was that the increased tension would have commenced immediately I started to turn the spanner in the a-c direction. Is what happened to be expected?
The other issue is the re-use of the fuji (?) nut.
Chris Kelly in his youtube video on the subject suggests min re-use. In my case esp with the first belt, where I was feeling my way, the nut was getting nipped up and released numerous times until I found where I was at. No doubt it would be beneficial to replace it - yeh, easier said than done, but I wonder in this context, if re-use actually means tensioning it to the required spec?

I concur with the comments made in the previous post to this, that to find and tension to that exact spot is very difficult. I was amused by the way Chris seemed to quickly move on from measuring the desired tension, to tightening up in the video. My belts although tighter that they were, are both still a little below the mark. If I put a drill bit in between the case and the belt next to a cam roller that causes just a slight deflection on the belt, all as in tune.:)
 
#13 ·
Hi Rockit - and thanks for the PM.

Firstly, if you feel your belts are loose, you are going to improve things by taking them up, so long as they're not too tight.

And there's the rub. The previous post with the 5 mm up and down deflection (between the cam pulleys at TDC on the firing stroke) - 10 mm in total - is probably a reasonable guide. (In fact I made a tool to mount that idea on the centre thread post on the cam cover).

But you'll need to know where your cam pulleys are in relation to their timing if you don't have the top covers off.

I had limited success with the Visual Analyzer software as I said, even with a $50 microphone.

No substitute for having adjusted so many Ducati belts since 1980 (and knowing they were all OK).

As to replacing the locknut on the adjusting roller - if you nip it up it should be fine. I replaced mine with Ti ones, but I used the stockers for a few adjustments first.

The rear one is in danger of getting marked anyway, the way it is obscured by the frame, so you'll probably replace them eventually. Won't cost much - if you don't go for the Ti ones..

Flanged 8mm x 1.25 pitch nyloc nuts should find them at a bolt supplier if you go that way, but they usually have a $10 minimum. Time for another tin of CRC..

:D
 
#15 ·
Good stuff, I'm waiting for my new belts to arrive in the mail.
 
#16 ·
Firstly - no iphone for me, too technologically sophisticated for me, I'm still at the learning about Ducati level.
Thanks for the response Pat. I would like to squeeze just a bit more out of you on the eccentric nut tensioning. I've re-traced my steps. My Ducati Service and Repair Manual on CD (written in Eng. and It. - looks official??) seems to make no mention of the nut tensioning direction, which under the circumstances could be a serious omission. The Ducatitech youtube video - and it is Chris Keeley not Kelly of CA Cycleworks, uses anti clockwise. There is also a thread - "Timing belt eaten alive" on Ducati MS stating the same thing, and what the consequences can be from doing it wrong. I have read the thread when it first came up, but can't reproduce the link here as I'm not a member there, and can't get it back up.
I'm just concerned because the initial impression I got on mine was that I was rotating the wheel from a position where it had been tensioned clockwise, to one of anti c w. - it seemed to swing through such a large arc. Then, the bike has travelled 4k kms without any trouble, so whatever was done in the first instance seems to have been ok. Of course in my case, this apparent excessive movement, might just have been the result of initially releasing tension, then having the belt move the wheel to a position of least resistance from where I then started rotating it a.c.
All my work was done with the cogs in their marked positions, so things should be right in that regard. Although with this doubt - Thomas being my middle name, I'll remove the plugs and use a depth measurer to confirm TDC, and go over it all tomorrow. One of these days I'll get the bloody thing out of my shed and be able to move onto the next job.
 
#17 ·
No worries Rockit. You refer to the direction of "nut tensioning" - I presume you are referring to the direction of rotation of the actual eccentric roller centre.
And yes - you rotate the eccentric anti-clockwise, so the two holes you engage rotate above the nut.
If you rotate this eccentric the wrong way the belts may touch between the rising and descending runs. (And it would be hard to engage the adjusting 'spanner').
As for finding TDC - remember there are two TDCs (being a four stroke). Only every 2nd TDC (compression stroke) sees the valves inactive to set tension ideally. On the 'other' TDC - between exhaust and intake cycle - both valves are slightly active.
 
#18 ·
Gee, this is becoming quite a saga. I could mention Blue Hills, not many will have heard of that, plus it hints at my age.
This is the link to the youtube video to which I have referred. This particular part is the No2 of a series of 3.

Pat, I understand that you have made a special tool to utilize 2 holes. This I don't understand. Perhaps the holes are specific to your bike and not the 999? The video displays my setup. One thing though, I don't understand how he can get the jaws of a shifter on the nut. It is about flush with the surface of the pulley. I tried - no way. I was able to use a 27mm ring spanner that worked ok. Ha! problem was the size seems to be 26mm?? and is the reason I couldn't optimally tension the belt. Must be Ducati specific, the step up that I've always encountered is 24 then 27.
No desire to get the job finished just now. We have a low pressure weather system working here at present resulting in much rain. If I opened the shed doors it would be blowing straight in. Anyone in the market for some pure fresh rain water? Plenty about here, tanks are over-flowing and have been for weeks. 800m of gravel road out the front is slush and full of holes. Will have to be happy with my DR for a while.
 
#19 ·
  • Like
Reactions: WiscoDuc
#20 ·
I'm having a hell of a time with the vertical cylinder. I'll sett tension at 100 hz, rotate the engine and it'll be at 120 hz. Rotate it again and I'm at 85 hz. The horizontal cylinder is staying constant. I'm about to take the average and leave it be.

It feels like it's at the same tension as the horizontal cylinder.
 
#21 ·
From what I've written in my posts, it is evident that I am no authority on this subject. The thought that comes to my mind regarding the vertical cylinder, is to wonder if you have changed your TDC as is shown in the video I mentioned. There is (was ?) also a good current thread on this on the other site. There it advises to turn the motor forward in the usual direction of motion, 270 degrees rather than back 90 degrees as the video suggests.
 
#22 ·
I was turning it forward 1 full turn from TDC. I messed with it long enough to get it set finally.
 
#23 ·
One more question, I'm about to do the 15k service my own,
Let's say I get the cam belt lock tool,
Do I still need the engine turning tool to rotate the engine back 90* like the vid said?
Thank you very much for tips and hints.
Also, along with my other thread, has anyone done a valve adjustment themselves?
Could you help me out with hints and tips on this? Cause I think the task is hard and require an engine drop, I assumed. And my passion could possibly outweigh my talents on doing this valve adjustment myself
 
#26 ·
hello, i had a question before i start on this the coming weekend. I was a little confused on TDC on the vertical belts? Im looking at the cams and the "T" is different( i mean upsidedown) compared to ducatitechs vid. Reading the comments on youtube others have also noticed this as well? Wanted to confirm the pic @ TDC at comp?

Thanks ahead!!!
 
#27 ·
Make yourself aware of the fundamentals of a four-stroke engine. Every SECOND TDC has the same cam timing.
That is - four 'strokes' (up and down movements of the piston) - which equals two revolutions of the crankshaft for a complete cycle.
This alternating function at TDC is facilitated by the cams running at half engine speed, so the 'other' TDC is different.
Hope this helps you get a grasp of it.

:)
 
#30 ·
#29 ·
Of course how far you have to turn the wheel to rotate the engine varies according to which gear you have engaged. Half a turn of the cams will equal equal one revolution of the crankshaft.

And to the enquiry about valve adjustment - no, you do not need to drop the engine. But there is a fair bit of dismantling involved. Preferable to remove all the front fairings, tank and airbox, to give full access.

The oil radiator can just hang down, and the coolant radiator can be positioned around to the LHS of the bike without draining it. Then removing the cambelt covers and top cam covers will give you access for the valve job.

Anyone attempting the valve adjustment for the first time should know that there are two oil drain back holes, which are big enough to swallow a valve-retaining clip, particularly on the rear head (where they point upwards).

They should be blocked before any clips are removed..
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top