Peugeot FE53 series 3 1994 Scoot

[Florida Antique]

If you get some 11 gram rollers and decide you need 10's you can always alternate 9 and 11 gram rollers and effectively end up with 10 gram rollers. Another thing you might consider is if the bike is shifting at the right point but the clutch is engaging at too high an RPM, you may be able to fix that by putting in lighter clutch springs. Changing the rollers will just change the shift point causing to start shifting before the clutch engages, the lighter springs will allow it to engage at a lower RPM but still allow it to shift in the middle of the powerband.

[fe53series3]

May 11, 2015 17:36:27 GMT -5 Florida Antique said:

If you get some 11 gram rollers and decide you need 10's you can always alternate 9 and 11 gram rollers and effectively end up with 10 gram rollers. Another thing you might consider is if the bike is shifting at the right point but the clutch is engaging at too high an RPM, you may be able to fix that by putting in lighter clutch springs. Changing the rollers will just change the shift point causing to start shifting before the clutch engages, the lighter springs will allow it to engage at a lower RPM but still allow it to shift in the middle of the powerband.

I see what you mean...... but in all honesty, I don't have a real grip on the fundamentals.

It sounds a bit like the mix of idle circuit, idle jet, needle, and cutaway.
IE. more ways to skin a cat..... but each setup offering a slightly different outcome.

I CAN get the clutch gripping earlier (with lighter springs).
But from what I saw, running the motor with the cover off....
The first thing that makes the belt spin up, are the rollers.

I take it that you are suggesting that the clutch should be gripping earlier - just before the belt makes it's first main move.
Whereas at the moment, I'm relying upon the rollers to fling out earlier, to spin up the belt, to cause the clutch to grab.
IE. with an earlier grabbing clutch, I wouldn't need additional weight.

Is that a good understanding?

From memory, it would appear to be like this..... because I remember looking at the front pulley, while opening the throttle....
... the rear wheel spun, only when the front pulley made a definitive move.

However..... that test was done with badly fitted, extremely flat rollers.
I need to look again tomorrow.

But can we leave the clutch till later?

That question highlights where my brain is struggling.

A certain weight flings out, spins up the belt, causing the clutch to grab (as a result of c.forces).
That much is clear.

The clutch could grab earlier in the spin up.
OR
The weights could fling out earlier, to spin up the belt sooner, causing the clutch to grab earlier.

Okay.... that's it.... those are the two scenarios.

Would you (or anybody) like to have a stab at explaining what comes next?

The thing is.... you talk of shifting (gear), but it's a variomatic....... that's where I'm lost - even though I do notice a change in gearing.

Feedback

Here's how it currently rides:

Gear (phase) 1: Engages @ 4.6k and speed increases at a linear level, for low to low-mid speed running.
Gear (phase) 2: Whilst tootling along.... whack open the throttle...... the engine immediately revs to 7k.... bike accelerates but gearing definitely goes higher (steadily)..... revs stay close to the same, but the bike continues to accelerate (I can feel the bike speed catching up with the RPM).
Gear (phase) 3: After speed has caught up with the initial 7k burst.....then speed progressively rises with the revs.

Am I interpreting the rider feedback correctly?
At what point do we say 'the gearing has shifted'?
Is it phase 2, or phase 3?

.... and how do the roller weights affect this gearing.

Otherwise, apart from explanation.... the only other way I might understand this, is when I've changed the weights.
At the moment, I've merely replaced knackered 9g rollers, for 'round' 9g rollers.
Perhaps if I go much heavier, the result will not be so successful.

I'm a bit lost
 

[Florida Antique]

I will see if I can help. The front pulley (variator) drives the rear pulley (driven pulley or torque driver) which drives the clutch. The clutch will engage at the same RPM (not engine RPM, clutch RPM)each time. Currently idling the engine will not spin the clutch fast enough to engage it. As you say the engine RPM has to be around 5000 RPM to engage the clutch. Since we do not know what the clutch RPM is I have to assume that at 5K RPM the front pulley is almost fully retracted. By changing the rollers, you made it easier for the front pulley to contract slightly causing the back pulley to spin faster which made the clutch engage 400 (engine)rpm sooner. Adding weight to the rollers will cause this to happen even sooner so the variator will collapse at an even lower RPM. The problem with this is that the point at which the variator collapses entirely will also happen at a lower RPM. If it happens low enough that the engine is not in its powerband, it will not reach top speed because the horsepower at lower RPM ranges will not be able to overcome the increased resistance of the higher ratio. In other words,if the engine is currently in it's powerband when the variator is collapsing, putting larger weights in it may cause it to collapse too quickly.
If you want the clutch to engage at a lower RPM then you can change the springs in the clutch so it will engage while spinning slower. It should engage after the variator has collapsed on the belt but before the driven (back)pulley starts to expand. Then you will be able to use the entire range of the variable ratio that this is designed for. The ultimate goal is that the RPM of the engine reaches the powerband and then the driven pulley starts to expand and then when it reaches maximum expansion the scooter will travel faster with higher RPM.
Given the example above the variator is collapsing and the driven pulley is expanding during phase 2. Heavier rollers will cause that to happen at a lower RPM and lighter rollers- higher RPM. By the way what the rollers actually do is over come the resistance of the contra spring in the driven pulley.
All I was saying in my earlier post was that if the engine seems to be in its powerband when is in phase 2 then you may need weaker clutch springs and not heavier rollers.
On my Aero, it has 16 gram rollers in the clutch. That was because they had heavy grease in the clutch from the factory. The grease is unnecessary and so I took it out. Now that the rollers do not have to overcome the increased resistance of the grease, the clutch still engages where it always did but as I accelerate, the driven pully expands way too early (about 2500 RPM). This causes the acceleration to be very slow and you can hear the engine laboring to increase the speed. It eventually gets there but I know it is hard on the engine. I have read on other forums that people who remove the grease from the Honda clutches have to drop the weights down to 11 or 12 grams to allow the engine to get to the powerband before it starts to shift. In my case, I feel that changing the roller weights will accomplish what I want to do. In your case I am not sure. Increasing the roller weight may cause it to shift outside of the powerband like mine is doing now. In that case the only was to make the clutch engage earlier is with lighter springs and keep the existing roller weight. Hope this helps a little,

[niz76]

fe53, I think you have a better understanding of the CVT than you're giving yourself credit for
From the way you describe your scoot, it seems you may already have a decent CVT tune- but I could be wrong. Most folks- I think- tend to take off at full throttle. If you whack the throttle from a standstill, at what RPM does the clutch engage (scoot start moving)?
Basically, you want the clutch to engage while the belt is still "little in the front, and large in the rear." Ideally, you'd like to bang right into the powerband and stay there while the scoot accelerates (say 8k for example)- the pipe plays a big part in what that # is. (I can't remember what pipe you're running)
This setup might feel "too racey" for some- it's really up to your personal preference... Kinda sounds like your current CVT setup is good for your "style" of driving.
Basically you just want to take advantage of the best "1st gear" i.e. Have the scoot start moving (clutch engagement) while the belt is still "small" in the front...

[fe53series3]

Many thanks to Florida Antique & niz76

I hadn't fully 'clicked on', that the front pulley phase, is distinctly different from the rear pulley phase.
This was missing from my understanding.
I understood, in an overview way, how it works, but the effort put in by Florida Antique, delivered the missing detail, exactly where the Devil exists.

I believe I now understand when you talk about 'the shift'.

There are two gearing systems:
The front pulley increasing in diameter (as a result of engine RPM), and the rear pulley decreasing in diameter (as a result of rear wheel RPM).

The clutch should grab before the rear pulley begins to shrink.
Also
The front diameter should only increase slightly..... to get the clutch to grab...... in this way, the full range of roller movement is available for changing gear.

If the clutch is grabbing too late..... the front diameter must increase more than the ideal, so wasting available movement.

So the tests should be done with the CVT exposed.

I need to watch the front pulley belt for movement, as the revs rise.
If it needs high RPM before it moves, then the weights are too light.

If the front pulley moves at say 3.5k........ I would then want the clutch to grab, to launch the bike, and maintain drive down to 3.5k.
Any lower than 3.5k..... I would lose drive.

(At the moment I lose drive at 4.6k...... meaning the engine is screaming, just in order to maintain movement - at junctions I sound like a communist era Tupolev winding up the motor just in order pull away from a junction.)

In principle; that deals with the front pulley and clutch.

The second test will be to see if the rear puller is decreasing in diameter too quickly.
I presume that this is a function of the spring.
If it has gone weak, then the rear pulley will decrease/come apart too quickly.

I guess the only thing I can look for, is to see the wheel spinning, but no movement on the rear pulley.

Summary
We want maximum available movement of the front pulley, once the bike is moving.
Given this is correct.
We want the clutch to grab at an acceptable launch RPM.
Given this is correct.
We want the rear pulley to start moving after the clutch has grabbed, when the bike is moving at an appropriate speed.
IE. the rear pulley movement should be dependent upon rear wheel RPM or 'road speed'.

Tuning the CVT
A stronger spring will delay the moment when the rear pulley moves...... meaning that the bike must be going faster.
Therefore a stronger spring should increase the top speed of the bike, but only if the engine has power to drive it to its full movement range.
If the rear pulley never gets to its smallest diameter...... that available gearing will have been wasted (and acceleration potential will not have been reached).
If the rear pulley gets to its smallest diameter to quickly, the rising gear ratio will be to quick, and acceleration will be reduced.

Similarly...... the front pulley should achieve its largest diameter at peak power RPM.

With front and rear gearing correct:
The front gear will be at its highest at peak power.
The rear wheel gearing will be at its highest at peak power.

If the bike is over geared..... the gearing can be lowered in two ways:
1. Restricting the diameter increase of the front pulley, with lighter weights.
2. Restricting the diameter decrease of the rear pulley, with a stronger spring

Therefore I need an engine that is powerful enough, to gain maximum front pulley diameter, and minimum rear pulley diameter.
This would give the maximum top speed, whilst utilising the full range of gearing that is available.

As this is currently unrealistic (niz see page three for pic of standard exhaust)...... it would appear that this is the reasoning behind tuning the CVT to suit the power of the motor, and gain good useable acceleration, rather than looking at achieving maximum speed.

Is all this correct?

 


 

[niz76]

Seems like you've got a pretty decent handle on it! Brent has done tons of homework on the subject of CVT tuning and documented it beautifully in video. The clutch springs and contra/main spring are typically available by color code/RPM designation- i.e. blue/1000 RPM, Yellow/1500 RPM, and Red/2000 RPM. It's usually suggested to stick with either stock or 1k/blue contra (large main spring) as the heavier/Higher RPM contra usually just makes the clutch harder to work with (harder to squeeze open with your hands). Maybe a scenario where someone travels on a lot of hills might want a 1500 or 2000 RPM contra to force the belt back to a "lower gear" for hill climbing- but usually stock or 1k contra/main spring is used.
With that piece of the puzzle filled in, that leaves the smaller clutch springs and roller weights to play with. Again, the clutch springs are typically available in the same color coding/RPM designations- blue/1k, Yellow, 1.5k, Red 2k (this is the typical color code but some manufacturers like Malossi for example may use different colors like white or green etc.). The RPM designations on the clutch springs symbolize how many more RPM's than stock (approximately) the clutch would engage at. For example, if your stock clutch springs engage at 4000 RPM, then switching to the Blue/1k clutch springs would mean now the clutch engages at 5000 RPM. Yellow 1.5k springs would engage at 5500 etc. So, 1000 "more", 1500 "more", 2000 "more".... Then the last piece of the puzzle is roller weights!
An example of a pretty popular CVT setup is: Blue 1k Contra spring/ yellow 1.5k clutch springs/ and 6 gram roller weights. <---- Of course this is just an example of a typical "tuned" setup. You might like your CVT setup a little differently depending on your driving style/needs.

[Florida Antique]

The front pulley increasing in diameter (as a result of engine RPM), and the rear pulley decreasing in diameter (as a result of rear wheel RPM).



Read more: 49ccscoot.proboards.com/thread/11963/peugeot-fe53-series-1994-scoot?page=5&scrollTo=175183#ixzz3ZwWKIOjX

Well not quite, The reason that the rear pulley expands is that the front pulley is compressing forcing tension on the belt which causes the contra spring in the rear pulley to compress expanding the rear pulley sheaves effectively making the diameter of the rear pulley smaller. If you can get that to happen while the engine is in the powerband portion of it's RPM range then you will get maximum acceleration. As far as the powerband is concerned, with a stock exhaust, the engine designers want the widest powerband available. So in your case with the stock exhaust, variator tuning is not as critical as probably anything from 4000 rpm to 8000 RPM will be in your powerband so that's where you want the shift to happen. Making the clutch grab at a lower rpm will help you with low speed riding so you don't have to wind up the engine to get the bike to move. With aftermarket pipes, they make additional horsepower by creating a resonance point where the engine creates maximum efficiency over a much narrower RPM range. Outside of that narrow range the engine horsepower is much lower than with the stock pipe. In that case, variator tuning becomes much more critical since you only want it to shift in that narrower RPM range to give maximum acceleration. I don't know if you were into motocross or have ever seen a race but the older 2 stroke motocross bikes were actually designed to create 35 or 40 HP from a 125cc engine but only over a RPM range that was maybe 1000 RPMs wide. So the engine would rev to 15,000 RPM but would only produce that horsepower from 10,000 to 11,000 RPM and the rider would keep it in that range using the multi (6) speed transmission. With the scooters, we have to try to keep in in that range by using the changes in the belt drive since we do not have a multi speed transmission to shift. In your case, since we are looking at a mostly stock engine with a stock exhaust, I would think that there may not be much you can do to optimize the transmission other than fixing whatever was broke (or flat spotted). You are correct in thinking that the CVT is designed to give the scooter maximum acceleration if it is working correctly and the belt is traveling the entire range. Top speed increases can only then be achieved by increasing the engine RPM or by changing the rear end gear ratio.

[fe53series3]

For myself, my starting position is to gain drive at lower revs; and gain the ability to drive slowly.

I tested it today, and was just able to maintain a slowish speed at 4.6k....... any slight movement of the throttle, and it shoots forwards.

So I was riding down a 'community garden' path (with people in their gardens), trying to go as slow as possible, yet the engine is screaming away.

I just don't need this amount of power being expended, just to keep me at 10mph.

I'm wondering if this is a sign that the rear pulley spring is too weak for the current clutch setup, because, no sooner is the bike moving, it is also accelerating.

.... and because the engine is already revving at 4.6k, if the rear pulley diameter decreases, the bike just shoots off.

It is very hard to control.

If I can get the clutch to bite at a lower RPM, then I should have low speed controllable drive.

Anyway, we'll do the test with the cover off, to see at what revs the front pulley moves.

[Elektrey]

Do as I do and just give it the beans and the coast and then WOT for a few seconds then coast again. helps you go slow.

these scooters can balance really well at low speeds so you don't always need throttle.

[niz76]

Do you think the previous owner may have installed aftermarket clutch springs? I've got at least 3 sets of stock clutch springs here I'd be more than happy to mail but not sure it's realistic to send them from "across the pond"? (Florida, USA).
Heavier roller weights may also put you in a more desirable slow speed RPM...

[Elektrey]

I can send you some of my stock clutch springs if you pay for shipping, they engaged at 4500 for me.

really the rpm that the clutch engages is ONLY counting when your not moving once you start moving you can go below that rpm and still have the clutch engage.

I'm kind of having a hard time understanding your problem, I'm able to give just a tiny bit of throttle and have my scooter accelerate very slowly if needed.

Maybe you are too rich/lean in your midrange and it just wants to jump to WOT when you give it partial throttle? Thats what was happening to me when I was rich.

[fe53series3]

Niz and Elektrey, thanks for those kind offers.

Let's first have a look at the test data that I've just this minute acquired (clutch assessment, I think, must come later). 

Visual Inspection of Pulley Movement Against RPM (cover off)

The first thing to say, is that it is extremely hard to control the revs, when the RPM hits around 3.8k.
It hits a powerband, and surges ahead, as efficiency improves...... and there is no load to keep it in check.

It may be that the air intake speed, increases to a point where the needle starts to deliver full flow, for that taper position.

But, by gaining the knack, it is just possible to back off a touch, and hold the rpm.

We found:
There is a very small initial movement at around 4.2k, but no compression of the belt.
(the belt is 0.7mm down.... well within operational tolerance, but this still must have an effect.)

At 4.4k belt expansion occurs.
At 4.6k clutch grip occurs.

Clearly RPM readings are very hairy; but this data does tell the story.

Therefore, at this stage, the problem lies with the roller weights.

Vis a vis Elektrey's comment
"I'm kind of having a hard time understanding your problem, I'm able to give just a tiny bit of throttle and have my scooter accelerate very slowly if needed."
         Read more: 49ccscoot.proboards.com/post/new/11963#ixzz3ZxBvb8Vs

Regardless of clutch grab point...... definitively, with the current 9g rollers; there can be no drive at all, until 4.4k engine RPM.

With the PHBN carb mod, the engine has way sufficient power to launch the bike at a lower RPM than this.
In fact the low down power is entirely wasted.

I think that 'launch at 3k' is definitely possible AND I then wouldn't have this problem of 'hard to control revving'.

I could launch at 3k..... the engine would then be under load..... the carburation progression would be in play..... and once moving, the motor would slip into the initial stages of the powerband, and I would be able to access all that acceleration, when the mass of the bike and rider is already moving. 

With momentum established.... Newtonian mechanics indicate that acceleration will be much easier to achieve.

Granted..... We'd then have to configure the gearing to suit..... but it might just all fall out in the wash.

If we can get the front pulley to engage at 3k....... this would then spin up the rear pulley earlier, causing the clutch to grab earlier.

We could then (at that point) judge the point of clutch grab, because engine RPM would be 1k lower...... but due to belt expansion, the clutch would still spin up, forcing it to grab, at what will be a lower engine RPM.

Which will be fine, because we'd quickly go from 3k to 4k.
In fact....... if the CVT performs as it is doing now...... if the throttle is wound back, we'd go immediately from 3k to 5k...... and we'd just be entering the powerband, and getting the most from the carb mod.

Anyway..... I'm prepared to be wrong about this.
What I'm sure about is that, at the moment, I'm wasting available power, that should be being used to get the bike and rider moving.
I'd like to access the power (that is currently being wasted on launch), when I twist open the throttle at crawling speed.

That's when I want the kick in the pants!
... and when I'm already moving, that kick should be so much greater!

Obviously...... comments and opinions welcome re the Newtonian thinking.
But it's gonna be a hard sell, to tell me that I can't get the carburation right, to enable a powerful launch at 3k.

...... but genuinely..... my ears are open.



 








  






 

[Elektrey]

while on the stand it will rev a lot faster than when your on the road because theres no load.

It only engages at 4.4k RPM because of the rear pulley spinning, if your in a higher gear then 4.4k RPM will spin the rear pulley even faster and the clutch will still be engaged.

you should be able to give it partial throttle to rev it slower up to 4.4k and then once there be able to control acceleration as you see fit. Once you start moving the clutch will be engaged no matter what unless you slow down to below the clutch engagement speed which is whatever speed 4.4k RPM at the lowest gearing you have is (which is honestly not much probably a few mph).

It's hard to understand whats going on, I'd love to take a video and show you how I handle slower speeds on my scooter but i'm waiting on some parts atm.

It would help if you took a video so I can know exactly what you are talking about, there is a lot to read and a lot of terms lol.

[fe53series3]

May 12, 2015 16:27:49 GMT -5 Elektrey said:

while on the stand it will rev a lot faster than when your on the road because theres no load.

To be fair..... there is no load until the clutch engages...... therefore.... prior to clutch engagement.... rev control is pretty hairy, because the load doesn't kick in until 4.6k.
Tickover is at about 1400rpm so we have + 3.2k of revs to manage, before any load.
Worse..... as we approach 4k the revs go wild, as we come on to full taper, and engine efficiency.
The clutch grabs, and we launch at a startling pace...... very good fun, but barely rideable as a road bike.

It only engages at 4.4k RPM because of the rear pulley spinning, if your in a higher gear then 4.4k RPM will spin the rear pulley even faster and the clutch will still be engaged.

Yes.... agreed..... that is the objective.... to get the rear pulley spinning earlier, so that the clutch will engage earlier.

you should be able to give it partial throttle to rev it slower up to 4.4k and then once there be able to control acceleration as you see fit. Once you start moving the clutch will be engaged no matter what unless you slow down to below the clutch engagement speed which is whatever speed 4.4k RPM at the lowest gearing you have is (which is honestly not much probably a few mph).

No... it is very hard to control the throttle as we get to 4k, because the revs surge.
Throttle opening is controllable to about 3.8k..... thereafter the revs take off without even further opening of the throttle.
The reason is normal.
Engine revs increase, efficiency is met, and the revs self fuel the bike, by increasingly sucking fuel through the carb.
Without load, a perfect air mix is irrelevant..... the revs simply soar; and the bike takes off.

Now under load, throttling back immediately causes a drop in revs LOL typically to below 4.6k, meaning all drive is lost.

As I said earlier.... it IS possible to ride on the cusp of 4.6k...... but it requires total concentration, and the bike is still motoring along at about 12mph.
A touch more throttle, and the bike surges ahead..... any less and all drive is lost.

It's hard to understand whats going on, I'd love to take a video and show you how I handle slower speeds on my scooter but i'm waiting on some parts atm.

It would help if you took a video so I can know exactly what you are talking about, there is a lot to read and a lot of terms lol.

Taking a web cam video on the move is not that easy to set up.
While I acknowledge.... a video would bring home the bacon........ the simple fact is that the bike has no drive below 4.6k, and yet at 4.6k the throttle is very hairy, and the bike launches like a rocket.

This much, we can envisage without recourse to video.

What I believe is that.... if launch occurred below 3.8k.... throttle control would be improved.
My personal belief, as yet untested, is that I could launch at a controllable 3k, and easily wind the throttle and find the powerband...... taking full advantage of the fact that I'm already moving, and taking full advantage of the available powerband.

Re the "lot to read, and lots of terms"...... I appreciate that.
But don't forget...... as I write the report, I'm creating a record that can be accessed...... in principle by me, but also by others.
.... I'm sharing my growth in understanding of the bike.
The detail is genuine...... and only with that detail, can WE judge change, when another mod is made.
I fully appreciate that a vid would highlight launch probs and low running probs....... but I don't have a headcam..... and so we've gotta accept that the launch and slow running problems are as stated.

I see no other option to 'adding weight to the rollers.
I'm not a fan of this, but there is no other option.
We must get the front pulley engaging at a lower engine RPM.

Anything below 3.8k should offer a great education.

[Elektrey]

"the simple fact is that the bike has no drive below 4.6k, and yet at 4.6k the throttle is very hairy, and the bike launches like a rocket."
I get that, I'll try to make a video to address that once my parts come in (hopefully tommorow)

realistically a launch closer to your powerband will get you going faster.
the idea is that you let the engine rev up to the powerband without any load (so it does it quicker) and then you dump the clutch so the clutch starts spinning at the RPM at the powerband.

If you launch at 3k then you will have to take 3k to 4.4k with the load of yourself on the bike, instead of using the clutch shoes to transfer that energy. hopefully that makes sense.

I can't figure out why you can't control the throttle that well at low speeds. If you have the throttle slide almost all the way closer even if your in the powerband you wont be allowing alot of air/gas into the engine and won't be making much power. What is your idle RPM at?