Replacing the chain tensioner

[BrianG]

I agree but in this case I was trying to make the erratic running better when all along it was a software problem.

 

Please explain................

[motoguzznix]

The Valtech/Stucchi type 'Blade' tensioner is in some ways a lot more robust than the Guzzi item. Yes, I know that some people don't like 'em and I know that some people have reported breakages of both blade and spring. I've had one broken spring on dozens I've installed and generally prefer them to the later model Guzzi item but both are a HUGE improvement over the original 'Foot' type tensioner from the early chain drive models. I think like most things both have advantages, both have disadvantages. When the chain or tensioner becomes a problem on my Griso I'll probably install a Valtech type as I think these work better at their true task of tensioning the chain but I certainly have no intention of doing it pre-emptively.

Pete

 

I totally agree with your statements on the chain tensioners.

 

Some years ago a friend of mine installed a P3 cam together with the standard Guzzi spring loaded cam tensioner in his Cali 2. The P3 cam causes very brute valve accelerations and has almost no ramp on it. He noticed a rattling noise from the timing chain at approx. 3000 rpm.

When the tensioner was replaced by a stucchi device, the noise was gone.

 

My conclusions: the standard tensioner is good for the standard cams that cause very modest valve train accelerations. Even there it is on its limit (only at idle) when you consider the problem stated by ratchet. An engine with a hot cam shold have the stucchi tensioner in it.

 

On my own engine I would not have replaced the stock tensioner as there was no indication for a malfunction. but the engine was apart at 9000 km because I had to balance the crank due to the severe engine vibrations. So I installed the stucchi tensioner when reassembling and noticed the idle is completely steady now.

 

1. At steady idle (cold), the flywheel mark is steady as she goes.

 

2. When the motor is warm at idle (evidently providing less drag throughout the valve train due to lower viscosity of the oil), the flywheel mark jumps erratically, off and on, alternately by 4, 5, or 6, maybe as much as 7 mm. <_>

 

Don't know wot "acceptable" scatter is, but I reckon this ain't good.

 

3. Over 3K RPM, it steadies out again, as Dan predicted, (evidently due to the increased drag of the valve train with RPM). When RPMs are on the rise responding to throttle, it's steady as a rock -- exactly as it behaves on the road.

 

Logically, to me this indicates at least some probability -- even liklihood -- of a wonky tensioner.

Ratchet

 

Thank you for performing the measurement with the timing light. This is really enlightening and generates a fact out of the theory.

 

Let me add some words to Petes statement.

The reason for that scatter is the oscillating rotation of crankshaft and camshaft.

 

The crank is accelerated when the mixture burns and is decelerated when the pistons come to TDC and BDC because the pistons have to change direction which costs energy.

 

If you take a look at the camshaft along the axis, you will notice the concentration of cams on one side of the shaft, whereas the opposite side is free of cams for a lot of camshaft degrees. This is why the camshaft of a 90° V-engine is very sensitive to oscillate during rotation and needs accurate tension of the chain to prevent the oscillation.

 

When the cam begins to open the valve, the chain tension increases as the valve spring forces have to be overcome. Every time when the cam rotates over the max lobe, the follower wants to accelerate the camshaft via the valve spring forces and so reduces cam tension. The first three lobes follow very close to each other but after the last lobe there is enough time for every play the valve train likes to do. Especially at idle speed the cam has so much time to decelerate that the scattering can be made visible with the timing light.

 

The cranks oscillating rotation is now the driving force for the cam that itself oscillates and the combination of the two results in an oscillating movement that I can't imagine in my brain, but some mathematics should be able to do this. I do not care since the stucchi tensioner is able to cope with this.

[mike wilson]

The reason for that scatter is the oscillating rotation of crankshaft and camshaft.

 

The crank is accelerated when the mixture burns and is decelerated when the pistons come to TDC and BDC because the pistons have to change direction which costs energy.

 

If you take a look at the camshaft along the axis, you will notice the concentration of cams on one side of the shaft, whereas the opposite side is free of cams for a lot of camshaft degrees. This is why the camshaft of a 90° V-engine is very sensitive to oscillate during rotation and needs accurate tension of the chain to prevent the oscillation.

 

When the cam begins to open the valve, the chain tension increases as the valve spring forces have to be overcome. Every time when the cam rotates over the max lobe, the follower wants to accelerate the camshaft via the valve spring forces and so reduces cam tension. The first three lobes follow very close to each other but after the last lobe there is enough time for every play the valve train likes to do. Especially at idle speed the cam has so much time to decelerate that the scattering can be made visible with the timing light.

 

The cranks oscillating rotation is now the driving force for the cam that itself oscillates and the combination of the two results in an oscillating movement that I can't imagine in my brain, but some mathematics should be able to do this. I do not care since the stucchi tensioner is able to cope with this.

 

This is also true of other engine configurations (such as single cylinder machines) and, interestingly, there is a method of distributing the wear when using gears to drive the cams. Velocette, for one, used it on many models.

 

http://dliengineering.com/vibman/huntingtoothgearsets.htm

 

Unfortunately, this option is not available on aftermarket sets for Guzzis, so the same teeth have to carry the stress and wear revolution after revolution.

[Guest ratchethack]

Ratchet

 

Thank you for performing the measurement with the timing light. This is really enlightening and generates a fact out of the theory.

Ernst, giving credit where due, it was Docc who came up with the enlightening suggestion. However, I still can't believe that not only hadn't I thought of doing this myself, but since I did it, surprisingly enough to me, there doesn't seem to've been any interest from anyone else here in doing this quick, simple, and what I consider to be a very directly "incriminating" test!

 

I suspect that not everyone gets the same degree of scatter, due to a wide range of contributing factors that would tend to throw variance into the mix. In my case, as mentioned previously, I reckon that mix is currently conspiring to make my "scatter" as significant as it is.

. . .

If you take a look at the camshaft along the axis, you will notice the concentration of cams on one side of the shaft, whereas the opposite side is free of cams for a lot of camshaft degrees. This is why the camshaft of a 90° V-engine is very sensitive to oscillate during rotation and needs accurate tension of the chain to prevent the oscillation.

 

When the cam begins to open the valve, the chain tension increases as the valve spring forces have to be overcome. Every time when the cam rotates over the max lobe, the follower wants to accelerate the camshaft via the valve spring forces and so reduces cam tension. The first three lobes follow very close to each other but after the last lobe there is enough time for every play the valve train likes to do. Especially at idle speed the cam has so much time to decelerate that the scattering can be made visible with the timing light.

 

The cranks oscillating rotation is now the driving force for the cam that itself oscillates and the combination of the two results in an oscillating movement that I can't imagine in my brain, but some mathematics should be able to do this. I do not care since the stucchi tensioner is able to cope with this.

Ernst, thank YOU for this enlightening observation.

 

I've probably seen a few Guzzi camshafts here and there in shops and maybe in photo's (having no need to inspect my own to date), but never noticed the orientation of the lobes mostly all on one side. Of course, exactly as you point out, this orientation would combine with the natural crank oscillations of the 90 degree single-pin motor to tend to create exactly the kinds of whipping and jumping of the cam chain (against the relatively weak counter forces of the tensioner) that the scatter I measured identifies at idle -- as well as at off-throttle situations on the road!

 

As I mentioned earlier, I b'lieve Raz is really on to something here with this thread. Your observations are shoring this up.

 

Wot's easy to imagine is that all of this has likely been very well understood (somewhere) for 40 years, since this motor hasn't changed in any significant way relative to wot we're talking about here since 1967!

 

Wot I think this means, Gents, is that all we've really "discovered" here is a bit of understanding that was probably never even "lost" -- it's been "out there" all along, just buried to our own understanding beneath the surface and unknown to er, shall I put at least a few of my fellow Guzzisti on this thread in the same category with meself and call us "perpetual students" of the Guzzi big block?!

 

Again, as I noted previously, one "obvious" solution to cam chain control has been timing gears. Again, let's snuff that one for reasons that ARE pretty universally understood, and just not go there.

 

With many decades of development behind it, I've gotta hope that there's at least one aftermarket tensioner that does a much better job than the stock tensioner (or at least the one I'm using ) for "stock" state-of-tune motors!

 

Which one to choose?!

 

Pete prefers the "Valtek type". Parting question -- Wot makes it a "Valtek type", and other than Valtek, who makes 'em?

 

A related Q: Has no one ever successfully modified a tensioner to gain better chain control??

 

Great thread and one I hope continues to develop.

 

As some point very soon I'll be re-tensioning my camchain -- one way or another. Would much prefer to do it on more of an informed basis and less of an "experimental" basis!

 

Raz, you seem to've elected y'erself to the guinea pig position. How's it goin', my friend?

[docc]

 

 

Wot's easy to imagine is that all of this has likely been very well understood (somewhere) for 40 years, since this motor hasn't changed in any significant way relative to wot we're talking about here since 1967!

 

 

Although the cams have certainly changed. I remember in Field's book Wittner discussing how much went into the V11 cam development with Crane. Yet I have no idea how it compares to other cams even the Sport 1100 or 1100Sporti.

 

I'm going to try to get the strobe on mine today. Should be especially good since mine idles well and doesn't flame out . . .

 

Watching the gear teeth through the window they jump 2-3 mm every 6-10 seconds. No change as it came up to temperature. I'm pretty sure it was fully warmed up after touching my elbow to exhaust. Certainly no 7mm jumps as that would shift the tooth from bottom to top in the window. Ratch, are your marks jumping around constantly or periodically ?

[Guest ratchethack]

Watching the gear teeth through the window they jump 2-3 mm every 6-10 seconds.

Docc, this is the kind of read that would lead me to b'lieve I'm in a situation that could stand considerable improvement. Would like to see results of others -- but would urge to make sure to get it up to operating temp!

 

. . .Ratch, are your marks jumping around constantly or periodically ?

The best term I can think of is "erratic". No periodicity to it wotsoever. That is, with engine warm, the mark will be constant for 1 or 2, or as many as 5, 6, or 7 seconds, occasionally longer, then it will jump. Also, no regularity as to how far it will jump when it jumps, could be anything up to ~7 mm. Far less frequently, it will fail to light the strobe 'cause it dropped a signal or even 2 successive signals. Again, with the engine cold, on acceleration, or held at 3K RPM for example, it's constant -- no jumping or dropped signal.

 

NOTE: FWIW -- Without a distinct mark on the flywheel itself, I can't imagine how anyone could get a credible read on "scatter" by watching the flywheel teeth, but that's just me. Seems to me the teeth would appear to overlap in and out o' phase with the strobe and confuse the reading?!?! In my case, there happened to be a mark on the flywheel in front of the gear teeth at the firing signal point for the right-hand cylinder. I think this mark may actually have been an accidental thing (it looks like a paint dot, but it's irregular in shape and not ideally where you'd expect a timing mark to appear ), but it was distinct enough and in perfect location to track for the purposes of the strobe with confidence. This is wot I used as my reference for the strobe. It would be easy enough to put a mark on the flyhweel with a marker -- that is, if you knew ahead o' time where to put it.

[John A]

Please explain................

It was a very early 15M ECU and it was before any dealers that were within 800 miles had any software to change it with.It sure ran crappy and there was no information to be had anywhere on them.I tweaked everything that was possible to tweak.I also thought maybe the cam chain was too loose.Long story short:FIM UltiMap and Doug Lofgren who has been doing carb and fuel injection since before he was weaned.All new fuel and ignition maps,on the dyno .I had to scramble to get all the engine mods done before the scheduled dyno time.Dual plug heads,1100 Sport throttle bodies,cam and all that.Plus I had to make sure all my tweaks were exactly as they should be.I set the phase sensor to the minimum spec,.6mm and the maps were made with that.Lord knows it's hard enough just to keep the damn things running right and it takes a different mindset to be a high performance mechanic,the more you squeeze out of the machine the more you have to tend to it.

[raz]

Raz, you seem to've elected y'erself to the guinea pig position. How's it goin', my friend?

I'm desperately trying to convince myself doing a cam chain job in the one-wall carport while it's raining is a Bad Idea™. I have the option to take it inside my living room, where I had it all winter, but GF tend to dislike that. Also, my first try buying a strobe light resulted in zilch. Not many need those nowadays...

 

The timing chest cover can be removed with sump full of oil, right? Oil level should be way below that.

[Guest ratchethack]

Hey Raz. Wot better time to do the chain tensioner than when it's raining??

 

Any more encouragement I can throw on the pile? -- Just say the word. I'll come up with something.

 

WRT oil level, I just took a gander, since I have this same consideration comin' up meself. The dipstick puts the oil level a few inches below the bottom of the timing chest, as you've already noticed. However, oil gets in there SOMEHOW, and IMHO it would make good engineering sense not to allow it to remain pooled up in there, especially when there's a gasket joint around the bottom -- that'd be a perpetual leak just begging to happen. . . So I reckon it must get in there by a steady supply as long as the engine is running for a wild ride 'round the sprockets, and then drains back into the sump through a hole in the aft side of the timing chest at the bottom.

 

WRT the timing light -- may I suggest find a neighborhood Old Geez who tunes his own "vintage" bucket o' bolts and see if you can borrow his.

 

Good luck. The entire Guzzi planet -- or at least me -- will be watching for your results!

[docc]

Ratch, did you compare the strobe signal from side to side? I wonder if it could be dropping the signal on one side only?

 

Also, a wiggle test might be appropriate where you have an assistant watch the strobe and you wiggle every wire and connector you can get to for the timing pick-up, coils etc. looking for any effect on the signal.

[Guest ratchethack]

Ratch, did you compare the strobe signal from side to side? I wonder if it could be dropping the signal on one side only?

Docc, it's an interesting question. I didn't check it on the other side and I'm not sure wot it could tell me of any significance or value if the dropped signals were on one or both sides.

 

But o' course, it's SO easily done, and now I've just gotta know. . .

 

Will advise. [. . .sigh. . .]

Also, a wiggle test might be appropriate where you have an assistant watch the strobe and you wiggle every wire and connector you can get to for the timing pick-up, coils etc. looking for any effect on the signal.

Since the bike runs perfectly under load, at higher RPM's, and acceleration without as much as a hint of hesitiation or flat spot by "seat-o'-the-pants dyno", and exhibits no other anomalies of any kind -- let alone a dropped signal under such circumstances (not ever, not even once), I've got no reason to suspect a faulty connector or ground path, etc.

 

So I'm not likely to be too keen on futzing around with connections that're eliminated from suspicion.

 

Many thanks f'er the additional thoughts, though.

 

Now I've got one more thing on my Saturday list. . .

[raz]

I just remembered I have one more fact to indicate we are on the right track: When using Cliff's My16M, the chain whip is not by far as self-amplifying as with the original ECU. Probably MyECU deals with borderline signals from the pickup much better (I think at some times, like when cranking, it reads the signal into an ADC instead of using the digital signal from discrete circuitry). So the sudden will to change rotation doesn't happen, for instance.

 

I'm currently installing a roperplate. Pure therapy. Weather looks promising so I'll probably have a go at the chain tomorrow. I tried buying a strobe light again today at what I thought was a logical store to buy one. They didn't even know what it was . This makes we want one even more (before they start costing huge money at pro stores) so I will try again tomorrow at a proper car parts store. Hopefully they are open Sundays, I think they are. When I'm done with this I will store the strobe light together with my vinyl record player for use after WW3

[Guest ratchethack]

Ratch, did you compare the strobe signal from side to side? I wonder if it could be dropping the signal on one side only?

FYI -- Just put the strobe on the LHS high tension lead and took a read. Exact same symptoms as with RHS cylinder. There's no timing mark on the flywheel here either, but I found a recognizable pattern of clutch debris on the flywheel as a point of reference that gave me a clear read.

 

Wot d'you make of it, Chief?

 

And wot if I got different scatter and dropped signal reads right and left? D'you think this would be reasonable or even possible?

 

Assuming inadequate chain control (as I am now with some degree of partially informed expectation, if not certainty) I can see how it might tend to drop a signal more on one cylinder than the other due to the interaction of the 90 degree single-pin layout with the cam lobe orientation as Ernst pointed out. Perhaps scatter could tend to show up on one side more than the other for the same reasons. . . ? If I think about this any more I might strain something. . .

 

Enquiring minds (well, you know. . .)

 

NOTE: It's considerably warmer today than it was yesterday, when I read the scatter on the RHS cylinder, and the bike was quite a bit warmer from a longer ride. It didn't take long before I could tell the engine temp was running up pretty high as I was getting a read. The temp rise at the time seemed to decrease both the frequency and degree of scatter. I reckon both the temp sensors, (air intake and cylinder head) acting independently, have quite an influence on the ECU as they go thru their cycles.

 

On a related note, In case anyone reading missed this, Todd Eagan at GuzziTech is now offering a Dynotech VDST that actually looks pretty handy.

 

http://www.guzzitech.com/store/TR-VDSTS.html

 

Just wot I need -- another Guzzi toy. . .

[docc]

 

Wot d'you make of it, Chief?

 

 

 

Actually, I'm pretty sure yer fooked. It's the same as when you rebuild yer forks and they squeak like yer kid's swingset.

 

Whatever's going on with your bike is the same affliction Nose2Bleed has. We tried swapping his coils (fixed the brackets in the process) and it still coughed and missed when hot. Could it be the pick-up itself is faulty? Seems a damn sight easier than pulling the timing chest and all the sprockets . . .

 

I thought the 'wiggle test' worth a try. I learned this trying to sort a turbo Merkur (Sierra).

 

The Brooklands publication "How to Prepare the Sierra for Motorsport" declares in the troubleshooting section for electrical systems: (and I paraphrase with reference to five years watching folks sort their V11s),

 

"The most common electrical faults in [Moto Guzzis] are caused by wiring faults i.e. breakage of wires or connectors, electrical shorting, poor or intermittent connections, or contamination by water or mud, rather than by component failure."

 

While there is no substitute for meticulously sorting your harness I must admit my Merkur (Sierra) was plagued by a component failure in its fuel management system. It ran so well when I found it , I slipped the clutch and blew a head gasket!

[Guest ratchethack]

Actually, I'm pretty sure yer fooked. . . .

Whatever's going on with your bike is the same affliction Nose2Bleed has. We tried swapping his coils (fixed the brackets in the process) and it still coughed and missed when hot. Could it be the pick-up itself is faulty? Seems a damn sight easier than pulling the timing chest and all the sprockets . . .

Um, thanks for the reply Docc, but I beg you differ. . .

 

With all due respect. . .

 

As mentioned many times previously, I'd tested, cleaned, and eliminated the timing sensor as a problem long ago. Again, I have no symptoms at all under acceleration, under throttle, or at RPMs over ~3K.

 

I'm not particularly interested in wot's a damn sight easier than pulling the timing chest, as long as there's no logic wotsoever behind chasing after stuff lacking any indications of problems -- particularly when electrical connections are so easily ruled out, as I've mentioned. Life is far to short, n'est-ce pas?

 

As I b'lieve was noted previously, I reckon I ain't anywhere near "fooked" by any reasonable interpretation, my friend. As Pete observed, timing chain slop and resultant signal scatter such as I've ID'd with the strobe (thanks again to your suggestion) seems to be not only tolerable by many -- up to 200K miles in some cases, as Pete has mentioned -- but evidently not even noticeable by many. . .

 

I'm not having any more of it, however. And I'm determined by any and all means at my disposal to improve the aforementioned. I b'llieve at least Raz and myself are on the same beam here, so to speak.

 

I reckon expectations for both of us at this point are high for considerable improvement, based on our current understanding of the physics at work here, and I also expect that significant returns on effort expended are more'n likely around the corner.

 

Unless I completely miss my best analysis so far, this thread has just begun, my friend. . .