Skip to main content

Mk 3 Vibration at low revs

Forums

My Mk 3 Commando suffers very unpleasant vibration below 3000 rpm. I have heard also of other Mk 3 owners having similar problems.

The bike was completely rebuilt from the crankshaft upwards several years ago and was fitted with new isolastic mounts. The bike is very refined above 3000 rpm but as the speed drops the vibration levels become very unpleasant. I have no experience of earlier models but I am told they do not suffer like this.

Observing the engine as it runs up through the rev range it is clear that the unit goes into resonance on the mounting system at about 2600 rpm (43 Hz first order). This would appear to be a very high frequency for resonance of an engine mounting system giving a clear reason why vibration does not become acceptable until 3000 rpm is reached.

Do other owners of Mk 3's suffer similarly? Is it that the replacement isolastics mounts are harder then the original factory units? Can anyone identify the resonant frequency of the engine mounting on earlier bikes?

I would add that the adjustments to the head steady spring or the shim clearances have no effect.

Can anyone help?

Bob Southall

 

Permalink

I found by changing the isolastic adjustment, mostly at the front ,I was able to gain some control over where the smooth range is on my bike. (Around 2,000 rpm upwards)  About 25 years ago I added a Norvil Stainless head steady, this had the effect of slightly increasing the vibes, but not by much.

I decided to leave things where they were. The ride becomes ultra smooth over 3,000 rpm, even with the new head steady arrangement. (There is less capacity for engine movement over the original 'mechano' like fish plate))

The ride stays smooth to 5,000 rpm, after that I don't know. Too many covid cops about!

Bike is a 76 Registered Mk III. (Left the factory in June 75) 

Permalink

Definitely not right; a Mk3 should be sooo smooth when correctly setup with the sprung head steady set just right, the best of all Commandos. There were complaints a couple of years ago that new Isolastics rubbers are/were far too stiff and made the ride harsher than the originals - I'll try to find a link on that for you.

Permalink

In any mechanical system the natural frequency is proportional to the square root of the ratio of stiffness to mass. That always makes it difficult to alter. To halve the frequency needs 4 times the mass...which is obviously impossible. Or one quarter the spring stiffness, which is difficult because, although rubbers do vary, they probably don't vary as much as that. So hard rubbers will have some effect but probably not to cause a massive change. Are they running out of space to move? Or are they the wrong dimensions?

If it's only a few hundred rpm, then it could more easily be the rubbers being too stiff.

Permalink

It has been discussed on here before Bob, so if you search there will be some more information; sorry I cannot point directly to it.

There seems to be agreement that replacement isolastics in the past have been too stiff. But I do recall someone having success by drilling holes in the rubber to weaken them.

If it helps, I will say that I purchased new isolastics from ANIL in early 2020 and they work just fine. No idea about other supply sources. I would expect that they have tried to get the rubber stiffness to match the original specification. I recall reading that Norton had to keep reducing the stiffness during development until the vibration became acceptable.

Permalink

I’m no Commando expert and have only owned my 1976 MK3 Interstate for a few months and covered 600 miles or so but this is how I’ve found mine to behave. I find my bike runs absolutely smoothly whilst constantly accelerating up to 4k rpm. I have not been past that mark as I don’t want to stress the engine as it hasn’t covered many miles since it was rebuilt. When keeping a constant speed then it needs to be running at 2.5k rpm at which point it is very smooth, any less and it begins to vibrate. I find keeping a constant 30mph is only achievable in 2nd gear as using 3rd gear the revs drop below 2.5k rpm. At around 35mph I can keep the bike running along very smoothly in 3rd. Can’t really remember what speed it begins running smoothly in top gear but I think it’s around 45mph and definitely 2.5k rpm. From what I have read and been told this is exactly how the MK3 with standard gearing and well set-up isolastic’s should behave. As stated, I’m no expert so please feel free to correct me if I’m wrong. 

Permalink

The factory fitted iso rubbers were not all the same hardness, it varied with each revision with the MK3 being hard and possibly the hardest, no idea for the reason, possibly durability. The aftermarket followed suit and so all replacements from the end of the factory onwards overtime became the same hardness as the MK3 or even harder. ANIL a few years ago finally cottoned on that this was the cause of the complaints of extra vibration after the fitting of new iso rubbers. They went back to the drawings and started to make them to one of the softer earlier versions and included the MK3 vernier rubbers in the exercise. You need to work out from receipts or date of restoration etc which iso rubbers are fitted, if they are not the soft ANIL version then they need changing or you can soften them by drilling holes in the sides. My MK2a with original rubbers shakes worst at idle and the vibrations reduce in amplitude with rising revs until 2500rpm when they disappear completely.

https://atlanticgreen.com/images/frontiso.jpg

Permalink

Found this discussion on the subject: https://www.accessnorton.com/NortonCommando/hard-isolastic-rubbers.16265/#post-236222

Sounds like it's been sorted out for the current Andover offering.

I would add to John's comment, above, about the Mk3. In 1978 I was able to ride a new Mk3 with 500 miles on the clock, straight after it's first service, and it was sublime. Even the electric start worked! My ride at the time was an ex-police Mk2A Interpol so a new bike was always going to feel good, but that properly setup, and new, Mk3 remains my benchmark for what is possible.

Permalink

Mk3 the hardest?

Quote: "The factory fitted iso rubbers were not all the same hardness, it varied with each revision with the MK3 being hard and possibly the hardest, ..."

John, an interesting comment, are there facts or evidence for this?

I bought a new Mk3 in 1976 and that engine really shook around at low revs. My new ANIL isolastics are possibly stiffer than the old ones, but impossible to compare over 50 years perhaps.

Permalink

John, an interesting comment, are there facts or evidence for this?

See the graph from Atlantic Green included in the post.

Permalink

Had this problem myself, buying replacement mk3 iso from a famous auction site!

Followed advice of  Andres Weingartner posted on 6.7.20. Drill 4 x 5mm holes in each side, only had to do front on mine to stop the mirror shaking and my fingers going numb.

Original post is on site under forums / commando if you search back.

Regards Martin.

Permalink

Lots of really helpful info here and the issue appears to be much as I feared.

I worked as an NVH and structural analysis consultant to the automotive industry for many years including on the design of engine mounts. 4 cylinder car mountings had vertical mode natural frequencies below the 2nd order out of balance of the engine at idle. It was not as simple as that because many modes in the 6 degrees of freedom could couple if not well spaced on frequency and cause problems.

The Commando mounts due to the plastic guide shims remove many of the degrees of freedom leaving perhaps only vertical bounce and the forward nodding couple. The pure bounce mode ought to be set below the first order frequency of the engine at idle (16.6hz at 1000 rpm). The engine mount stiffness would have to be 4 times design to have pushed this frequency up to 2600rpm (43hz) on my machine which seams unlikely. The bounce mode will have much more acting mass/inertia than the couple which may therefore be a higher frequency mode and may be the problem. This could account for one respondent fixing the problem by softening only the front mount which would have a big impact on the nodding couple but this is only conjecture.

It would be nice to have a few accelerometers and analysis kit to see what exactly is going on. I am however long retired and no longer have access.

Interestingly it was common for engineers to drill holes in engine mounts in development to assess the effect of lowering stiffness. It was generally referred to as "Black and Deckering"

I will be doing some drilling (after much dismantling)

Many thanks to all who contributed.

 

Bob Southall

 

Permalink

I read an article many years ago about the Isolastic system design (written by Tony Foale I think). The idea is that the engine rotates about the rear mount and moves up and down on the front with little fore and aft movement. The article discussed why the rear mount was not in the optimum position and how crankshaft balance factor affected the engine movement. Can't remember the detail but it is clear that any restriction on the front Isolastic mount, whether due to overly stiff bushes, over-tight clearances or a bent frame will have a very noticeable effect on vibration, as will an out-of-balance crank .

Permalink

Stan's comments are very interesting. It would be really interesting to find the article but there is certainly some logic an what is said.

Given the control of the shims there are only 3 degrees of freedom for the engine. Vertical, fore-aft and rotation about the crank axis. Purely by conjecture I would estimate that in the fore-aft plane the position of the centre of gravity would be somewhat behind the crank axis. This is based on the mass of clutch and gearbox behind the engine balanced a little by the rotated position of the engine putting barrels and heads in front. The rear engine mount is also further behind the crank axis than the front mount is in front of it.

Given these 2 conditions the vertical out of balance at the crankshaft axis acting through the c. of g. and the mounts would then tend to give a component of fore aft rotation as well as bounce.

Again by conjecture in the vertical plane the centre of gravity would be a somewhat above crank centre line due to the mass of the heads and barrels. The fore aft out of balance at crank centre line, created by the balance weights, acting on this c. of g. and also with a high rear mount would then also create a fore-aft couple as well as a fore-aft force. Both of course being 90 degrees out of phase with the vertical.

The level of fore-aft out of balance compared to vertical is of course dependant on the balance factor. The more the vertical is balanced the greater the fore-aft force becomes.

This all appears to tie up with the article quoted. It also supports the theory that the natural frequency of the rotation mode will have a significant effect on vibration on the frame. If this natural frequency is above idle, given the significant rotational force input, you might expect this to create a problem as the engine speed passes through the resonance.

I may of course just be making far too many wild assumptions but I will drill my front mounts. When I get round to it I will report back.


Norton Owners Club Website by 2Toucans