Belt Tightness

Quick way to check is to take the cover off completely and find the longest unsupported run of belt (can't remember off hand which side) and try and twist it. If you can't get it past about 45 degrees using 'moderate' finger pressure, then you're fine.

Much more than it's too loose and if there's no way you can twist it to 45 degrees without really trying, it's too tight.

Ideally, I confirm this using the 'Haynes' technique and put a spring balance on the completely slackened belt and pull it to approx 10lbs, tightening the tensioner simultaneously. Then double check as above.

If you really want to be anal, then a 6mm Allen key should pass relatively easily between the belt and the tensioner on the longest run on the vertical belt, but not a 7mm one.

For the horizontal belt, a 5mm key should pass between the belt and tensioner but not a 6mm one. The conventional wisdom being that the rear cylinder head expands more due to the lack of cooling air flow. Though to be honest, I've done both using the 5mn Allen key technique and never had an issue in the air cooled Dukes.

I tend to check using all three techniques as it's just as quick and gives you peace of mind.

 

If they are used belts they should be a little looser. Think it should be about 10% less tension iirc.

 

It seems that some owners recommend having the vertical cylinder belt slightly looser than the horiz cylinder, as the barrel and head expansion is greater.

 

Yes, good point.
If you tension used belts to new belt spec, you will be over-tightening them as new belt tension takes into account initial stretch that will already have taken place.

 

from recent experience i would say this is correct.
run it up to 40C see how much play there is then.

 

It's air cooled. How and where do you measure the 40*?
But yes, good idea to run it up to temp

 

of the dash dude.
i just fitted exe's belts at the week end using the 5mm Allan key method. i used a 5mm drift. by the time i rode it home the belts where howling. i rechecked when cold, all seemed fine tho i had slight free rotation on the horizontal cyl but non on the vertical. i ran it back up to 40c with the covers off, the slack had gone from front, the rear was like a piano wire. i let it cool down, readjusted the rear belt, gave it about 2-3deg of free movement at the cam shaft and ran it up again, perfecto..
chatting to exe about it apparently they (exes) don't stretch as much.
btw, i am v.happy with his belts. i have fitted gazillions over the years. they look and feel top quality.
so they do.
brw.

 

Maybe the 900s have a dash readout but my '96 750 didn't.
In service tension is 90 hz
Check it out with a phone app (Gates Carbon) and microphone is probably simplest.
I don't know the in-service Allen key spec and haven't seen any reference to it. Does anyone else? @Arquebus??

 

you slip the Allan key between the belt and the idler. 5mm should slip through 6mm shouldn't.
90hz but from what of the 4 runs on each cylinder? are they all the same length?

 

The other way is do what I sometimes did for a check, having first used the Allen key method.

A fishing weight scales with the moveable dial marked at 10 lbs - add a loop or similar 'device' (I used a teardrop shaped thin stainless plate with a hole for the scales hook; and a hole for the pulley stub) to the scales hook which can be hooked onto the adjustable pulley stub and pull against the scales spring with one hand - when it gets to 10, tighten the allen bolt.

 

Is that 10lb for in service or new though?
The op is checking his belts for in service tension, not new fit and is there an Allen key measurement for in service tension??

 

The Ducati workshop manual uses the same method with their special tool (effectively the same thing as I used) but they don't differentiate between new and used belts - they just say it is good practice to change the belts each service.

IMO it doesn't matter unless of course the belts are well used - IE, two years; when you would change them anyway. However a Ducati Italy technician advised me that the belts don't really need changing until three years old.

I didn't get any issues with belt adjustment at around 18 months old - but as I was a low mileage rider I rarely checked them until I changed them at two years old. Actually, my old belts that were removed, were in such good condition they could probably be used again.

 

I did see a Ducati service bulletin quoted some time ago that stated that ideally the inservice tension should be 90 hz but there was a lower limit which was acceptable, I think it may have been 80 hz.
I think the belts are expected to loosen fairly quickly after installation and then pretty much maintain that tension. I don't think they are expected to keep losing tension at a linear rate over time

 

This may sound a bit prehistoric, but I'm not realy in favour of belts, whether for driving the camshafts or the rear wheel.

I much prefer the good old chain (and shaft drive) even though the rear wheel etc can get pretty filthy. (chain or gears / pushrods to drive the cams).

This is because IMO the internal combustion engine is still fundamentally a basic old technology despite all the 'improvements' over the years.

This doesn't mean I don't like the internal combustion engine; but when I look at the matter of belts there are probably only two supposedly accurate ways of getting them right.

One is using a simple spring tool to either check tension by pulling on the belt, or by pushing against it to check the deflection at a certain load.

The other is using a PC or similar so it can be tested by sound.

Both are a little bit silly and advanced for checking something supposedly high-tech fitted to a prehistoric device.

Especially compared with chain adjustment - just a push with a finger to check the deflection and if you don't get them quite right, it either doesn't matter or you will hear / feel it before failure actually happens.