Modern Vespa

I started my Colt conversion experiment 2009.8.10 with 19.000 km, now 2010.8.22 one Year and at nearly 35.000 km so 16.000 km (10.000 miles) later it is time to see how it worked out.



If you are new to my blog, here a short recap for you.
I wanted to replace my rollers with the Dr Pulley sliders but on my way to find out if I had a 6 or 8 roller setup I had to take the variator apart. Since in the discussion groups there was always mentioned a lower weight for better acceleration I wondered what would happen if I would remove 2 rollers from my 8 roller setup. I was so amazed from the performance boost, that I wanted to know what happen if I would remove 4 rollers. Then I got into measuring and testing the 8,6 and 4 roller setups for acceleration, RPMs and millage.

What the 6 or 4 roller conversions are doing, is moving you in a more optimum power band.



So for the final conclusion, did the Colt conversion was a 100% success?

Short answer is yes - without any reservations.

So to put the result in one spiffy sentence:

The Colt conversion it the ZERO money mod what brings you 30% better acceleration, eliminates the paint shaker and the strangled sealion, makes your bike feel like it has a sports bike engine, installs a fat grin on your face every time you open the throttle, has no ill effect on your gas bill, transmission belt or engine, gives you the extra edge in passing cars and you can do it yourself right now.

for the complete results you are welcome to have a look at my blog
http://malaguti-spidermax.blogspot.com/

You seem to have done the homework, and the testing, and pulled off a pretty interesting scenario for some of us to give it a try. Next time I am into the non-Costa 500, I am curious enough to give it a go.
Your resourcefulness given where you live is impressive, but I suppose living off the main continents must make you that way--I would never think to MAKE rollers for a Variator!

He didn't Rob - he simply took half of them out. So instead of 8 sliders he only has 4. Interesting. I dunno if I could live with it leaping right on up there to that RPM band tho...

If you read his Blog, he was at least trying to make rollers, I do not recall if he succeeded. I have decided mine is running fast enough to suit me of late.

Indeed, a very nice blog he keeps.

My `04 Aprilia 500GT scooter has 4 rollers and the newer versions went from 6 to 8 rollers. It seems only the newer scooters like mine have unhappy riders versus I'm happy with my original setup on startups. At the same time I've yet to ride a newer version of my scooter so don't really know if there's a difference from a dead stop.

The Scarabeos are different now of course--(we knew each other back on Scarabeo riders forum on Yahoo)--I miss my 500gt sometimes, mainly on long trips! Good for you --keeping yours going.

So, if I'm doing my math correctly, this would be the same as using 11 gram weights in an MP3 500, right?

Mark

very interesting...thanks for posting

Actually, probably not - the acceleration curve is described by a calculus formula. It's the sum of an infinite number of instantaneous offsets (which is going to be a function of both centripetal force and the physical size of the weight) times mass, so while the average response may be the same as a lighter weight, the acceleration profile would likely be different.

Why not three (I mean in principle - I appreciate they have to be symmetrical to balance the load)? There must be a reason for so many rollers - perhaps more rollers reduces individual wear, so four rollers will wear out faster? (and/or induce more wear on the races?)

By the way Gran Canaria, you have been most impressive with your first "new member" posts here and around Modern Vespa, very informative and thought-provoking. Thank you!

That is one great website and the pictures are the best I've seen posted. I find the lesser roller idea mind boggling. Keep up the good work.

thanks 4 ur kind words guys

the thing with the weights is one part the other part is on how many rollers you distribute it.

I thought that having the same weight I had on 4 rollers (85 gram)



with 8 roller (woody / steel combination) would have the same effect but I was wrong there.


This setup let the engine spin much to high and changing 4 single steel insert rollers to double steel insert with 95 gram total did not work either.

So it needs a certain minimum weight per roller to actually work properly.

It takes me now 10 minutes to take the transmission apart and back together, I have done it so far more than 15 times sometimes twice a day.

These are all the tools you need to do this job, plus a Phillips head screw driver to the the silver plastic cover off (4 screws).
Thankfully the Malaguti engineers did not design any tupperware around the transmission.
In case of the Spidermax you just need to remove 2 hex screws to get the pillion foot rest out of the way.


and a boltcutter to spread the driven pulley for easy inserting the belt


So I can report no ill effect not using the proper tool to block the driving pulley when getting the nut off doing it the quick and dirty way.

For the rear nut I just use my home made parking brake to keep the rear wheel from spinning.




see:http://malaguti-spidermax.blogspot.com/2008/10/hold-it-right-there-mister.html

Like that bolt cutter method, Gran.

was the closest thing to hand to compress the spring without putting strain on the belt

I just pull it back with my fingers and work the belt in to the groove.
Crimping those cables all day has paid off

You make that sound so easy. Perhaps you could provide a little more detail.

Many thanks

I grab the clutch/rear pulley assembly from the clutch side, and just squeeze the pulley towards the clutch and push the belt into the groove with a finger or two on each hand. Wrapping something around the middle of the belt to squeeze the belt together makes it a littler easier. I use a tie wrap.

Gotcha......thanks

My hand strength is not quite as strong. I put the clutch on the ground, put a foot in the middle and pull up with both hands. Learned this at demo at Amerivespa in Chattanooga.

Thanks for the detailed information, have enjoyed reading your blog.

I'm going to try the colt modification on my MP3 500 in the next couple of weeks a the 6000m belt change.

Have you noticed any other engine or transmission wear which mint be attributed to the steeper rev/acceleration curve ? Any greater consumption of lubrication?

I'm in the Seattle area and commute by freeway frequently, so I'm curious to test the results with long/straight stretches at 70+ mph.

For the knowledgeable mechanics and engineers out there: can you project any specific, if improbable, catastrophic failure modes for this modification? Could the removal of 4 rollers in any way stress the components of the variator or housing?

"For the knowledgeable mechanics and engineers out there: can you project any specific, if improbable, catastrophic failure modes for this modification? Could the removal of 4 rollers in any way stress the components of the variator or housing?"

Good! I've been waiting to hear this.

Thanks for your comments, and I will definitely post my results after I've completed the modification. The difference in dry weight between the Malaguti & the MP3 500 seems to be about 40kg, it will be interesting to see if this makes any significant difference in the performance.

My question about failure was based in ignorance, and my inability to interpret the engineering: do the rollers themselves bear a meaningful amount of stress as part of their normal functioning & physical arrangement; if so, would the removal of 4 (since we know 6 works) rollers too much increase the load on the rollers? It seems no, based upon your results.

Conversely, do the two plates in contact with the rollers need support from the 6 or 8 roller positions in order to prevent any kind of deflection or deformation of the plates over time? Again, an ignorant question as I'm unable to figure out the force distributed between the plates & rollers as the rollers drive outwards. If the force is relatively small it would seem this kind of failure is unlikely then!

Thanks for pointing out about the acceleration curves etc., and that it could be an issue with a cold engine. After I complete the mod I'll be sure to warm it up before pushing it too hard.

Quaere: with the different acceleration curve, would re-mapping of the fuel injection help, hurt, or be indifferent? Does the acceleration curve have any implications for the exhaust, e.g., would it result in greater benefit from the installation of one of the aftermarket exhausts?

The cold engine issue is good mechanical sympathy for whatever engine in whatever state of tune. It's not nice to put it under high load when cold.

Regarding good warmup to operating temperature of an engine, it's better to set off on your journey with light engine load until it is up to temp, than to idle excessively, and then set off on journey under load.

Idling doesn't bring it up to temp as quickly, and I've read an engineer's description of what strangely greater valvetrain wear occurs at idle while cold, less than light load cold operation.

There is also the issue that separate engine oil from transmission hub oil means thinking your engine is at operating temp (and that is water only, oil still takes a lot longer to come to temp, and certainly not while cold start idling), you still have cold oil in the trans hub.

Remember to take it all with a pinch of salt, something has to be really wrong before any of this negligible stuff matters... and the hazards on the road from other vehicles or rider error is far greater probability of ruining the engine/scooter before any negligible wear and tear items will.

"The difference in dry weight between the Malaguti & the MP3 500 seems to be about 40kg"
mp3 Dry Weight (lbs/kg) 244/ 538
spidermax Dry weight: 204.0 kg (449.7 pounds)
you are right there
but when you compare a 90 kg rider with a 60 kg rider it might even out a bit
my test setup is around 310 kg for bike, add ons and rider.

"so, would the removal of 4 (since we know 6 works) rollers too much increase the load on the rollers? "

first fact, the rollers don´t roll, the pressure and the friction to high for that. they get dragged out by centrifugal force and that is the reason why you never have an even wear profile around the rollers but one or two spots where a bit of material is lost. ( tip rotate 90 degrees and you have a new set of rollers - tried and proofed that)



this is a roller after 18000 km in the 8 roller setup.
it looks much worse than it is.
the plastic did loose less than 15% of its thicknes in the roller above on that spot
the same roller is still working after now 37000 km with no ill effect.


"Conversely, do the two plates in contact with the rollers need support from the 6 or 8 roller positions in order to prevent any kind of deflection or deformation of the plates over time?"

No, before I decided to test the radical Colt conversion I had a close look at the craftsmanship of the whole transmission.
When I was a teen I learned at VW in Germany for 3 and a half years to create and shape metal parts for cars.
So i have a pretty good idea what materials can cope with what kind of stress.

The pulleys are so sturdy you can roll over them with a big truck and they won´t even notice, also they would work without deformation with only 2 rollers.

"Again, an ignorant question as I'm unable to figure out the force distributed between the plates & rollers as the rollers drive outwards. If the force is relatively small it would seem this kind of failure is unlikely then!"

see above

(at 37000 km)

"Thanks for pointing out about the acceleration curves etc., and that it could be an issue with a cold engine. After I complete the mod I'll be sure to warm it up before pushing it too hard."

The cooling water cycle at the malaguti is set up in a way that in the cold start phase just the water in the engine block is circulated.
when it gets hot enough the first radiator line opens, if that gets too hot the second radiator line opens, after that the cooling fan takes care.
with this setup the engine has reached optimum temp in less than one minute.

Joe, a thousand thanks! You have provided a real education for me on this topic, and it underscores the good that these message boards can provide. You have been most generous with time and information, particularly in light of my many questions born of ignorance.

I'll post again after this weekend once the modification is complete.

Danke sehr / merci bien / mille grazie,

Paul

WHOOOOOOOO-EEEEEEEEEE!!!!!!!

Completed Joe's Colt Mod. Very, very happy. Have not had opportunity to measure the difference, but it is significant by subjective measures. Here's my initial, highly subjective take on it. Bear in mind that I replaced belt & (4) rollers at the same time, so part of the perceived difference may be attributable to the new belt & rollers.

At least a full second off of 0-60mph
Accelerating uphill radically better
Acceleration response to WOT above 50mph better
WOT from full stop brings the tach up to 7k+ very quickly, and then smoothly back down to 4-5 if on the flat under 50mph
Noise, vibration, smoothness all comparable to 8-roller setup, given the understandable pitch change for higher revs. I would say even smoother in the 3-5k band.
Acceleration drops off more abruptly upon releasing throttle
Upon coasting to full stop, a steeper and slightly delayed drop down to idle RPMs.

In short, still a very smooth machine with noticeably more punch on the acceleration. This mod is dangerously fun on twisty-hilly road.

Thank you again Joe for trail-blazing!!

Paul

P-mac,
Have you noticed any difference near the top end? Specifically at freeway speeds such as the 75~80mph plus range? I mostly ride in town but on occasion hit the freeway on my scooter.

Haven't had an opportunity to drive over 65-70 for more than a couple seconds for concern of safety conditions and those who enforce them. I'll be writing here again after I've had some freeway time at 65-80 for any meaningful distance. Taking a road trip on the blue highways of central western washington coast this thursday thru Saturday and will have much more opportunity to observe the behavior over time, speed, conditions and distance.

you are welcome Paul
it sure is a big difference ain´t it?

you will loose a bit in max speed with 8 rollers the 8000 rpm limiter kicks in at 169 km/h


with 4 rollers it is at 160 km/h when the rpm limiter kicks in

I gladly accept this penalty since our speed limit here is 120, and even if i could race with 169 (real 160 according to the gps) i would do this mod in an instant again.

p-mac,
I would love to see any updates on your 4 roller experience. This may be my next mods.

I asked paul a few days the same question, how he likes the Colt conversion.
Here is what he wrote to me:

VERY!!!

Have put 500+ miles on it in widely varying conditions & will provide more feedback when I have time to write in detail.

No regrets at all. Has provoked some questions (e.g. "why does this do that?") but has not created any worries or concerns...just really big smiles!

Thanks,

Paul

Thanks Gran Canaria. I am going to try your colt conversion. I just need to decide if I am brave enough to try it myself vs going to the dealer. Leaning to doing it myself. It is a good reason to buy new tools.

Cheap electric impact wrench is the best new tool I've bought, and all thanks to the CVT mods. Wish I had it years ago, would have avoided lots of rounded nuts. I also bought the Powerbuilt digital torque adapter, but it didn't arrive in time, so I used the electric impact wrench to do it up again.

i 100% agree
it is 5 min work with the impact wrench



copy from my blog...

Hello !
I've read all articles on your great blog but didn't find how did you improvised variator lock tool ?
I'm desperately seeking that tool for my Fuoco but unfortunately version for 400/500 models is quite rare.
Thank you !
Andrej


I never needed it.
The impact wrench is all you need and to hold the rear brake when you remove or tighten the 21 mm nut.

So far I opened and closed the transmission more than 30 times with no ill effect. I also never used / needed a torque wrench in my 30 something years of working with vehicles.
A trick is to mark before you loosen something, so when you are tighten again and the marks align you see that it is screwed tight enough.

Since making the colt modification I've put 1,000 miles on, through longer trips over 300 miles, short runs, straightaway, curvy, hilly, stop/go commuting, 75mph cruising, and all weather types except hard snow (& the light snow wasn't intentional).

In short, my assessment is WHEEEEEEE!!!!! (still).

Sightly longer assessment: for all speeds+conditions where RPM >2500 this mod is a no-cost no-brainer; if you spend most of your riding time in stop and go traffic under 20MPH, you probably don't want to make this mod.  Everyone else: go for it!

Long assessment:  
This mod is for you if you want to have a dramatic increase in performance without committing to a $200-300 part.  The RPMs remain constantly in the 5-6 area almost constantly, and my top speed seems limited to about 83-85MPH.  I'm really not comfortable cruising faster than 80 so that's fine with me.  The optimal torque/power band is evident especially riding in steep curves.  My objective scientific assessment of the net improvement under these kind of riding conditions is WOOOOOO-HOOOOOO!!!! (said with the facial expression and intonation of a surprised macaque to get the full effect).

As the mod's inventor pointed out a possible risk is the strong temptation to exceed the limits imposed by statutory authorities or by good judgement.  That said I would rather that they be the limiting factor, not my ride.

In the course of trying this mod I think I've figured out why (as have many others here as well I'm sure) Piaggio would put such a compromise transmission on the MP3: smoothness.  The stock transmission, while by all reports is no J Costa, it is smooth in it's transitions from lowest revs up to the 5k range.

 The colt mod, while in all other respects absolutely delightful, has an abrupt disengagement point somewhere as the RPMs fall below 3K and before they reach 2.5K - thus my tongue in cheek caveat to people who have stop and go commutes at slow speed.

A related (I think) tendency is for occasional stalling when the engine is cold, as I don't think the system is tuned/designed to handle the abrupt 3-2.5K drop.  This disappears after there's two or three bars on the temp gauge.

Again, thanks for this thread, the ideas and information, and most of all, the massive fun!

Paul

It is now 4 years later since my initial post here.
My Spidermax has now close to 75.000 km and over 55.000 km with my 4 instead of 8 roller mod.
Since the original Piaggio rollers changed quality for the worse and tend to destroy them self in the 4 roller setup I switched to four 21 gram Dr. Pulley sliders.
These were reduced to 19 gram and working flawless since February 2012 in combination with a 2.5 mm washer.




If you want to reead more, have a look at my blog
http://malaguti-spidermax.blogspot.com/