Changing CVT behaviour

Dedicated to the CVT...

Moderators: Dan J, Diesel Dave, Crazymanneil, Stuart

Post Reply
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Changing CVT behaviour

Post by alexanderfoti »

HI all

On my superdream diesel (D850 Kubota and Aixam CVT transimission). It is quite sluggish between 10-20 mph. After that it takes off with a suprising amount of acceleration.

You seem to have to give it lots of revs to get it to accelerate quickly from 10-25 and even then its not that fast.

It would seem then, that I need to get it to shift UP slightly earlier. Can I do this by increasing the weights slightly, or decreasing the spring force? (weights would be easier?)

If weights, what sort of weight am I looking at? Something pretty small I imagine?
BertTrack
I luv the smell of Diesel...
Posts: 334
Joined: Tue Jun 05, 2012 3:53 am
Location: Netherlands

Re: Changing CVT behaviour

Post by BertTrack »

Keep in mind that lowering the spring force could lead to the belt slipping.

Sounds like slipping could be the issue. With what i know now i'd first check your secondary side spring though. I mean it has enough grunt to take off and above 20mph so the primary spring seems ok. But the secondary spring is assisted by the slanted area so when you pull hard enough on it, the pulling will aid it's spring.
Image
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

Thanks Bert.

When you say slipping, do you mean the CVT belt, or the clutch?

I can maintain and control the speed between 0-20 mph quite easily, it just requires more revs than I would think necessary.

Does the secondary side ramps(?) require any lubrication or should they operate by design dry? Is there anything specific I should check on the secondary side.
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

Went out for a test ride just now, Did lots of slow speed riding to feel the changes in the belts etc.

I'm pretty sure its not slipping, both front and rear pulleys are barely warm to the touch (guess 30 degrees c?).

If I give it lots of revs its picks up better, making me think its a ratio thing rather than a slipping thing.

It also pulls strongly from Idle to 10- slows from 10 to 20- picks up from 20 onwards, making me think its just a range thing.
UAofE
I luv the smell of Diesel...
Posts: 225
Joined: Sat Sep 20, 2014 6:39 am
Location: Denver, CO USA

Re: Changing CVT behaviour

Post by UAofE »

Yes, could just be the engine's inefficient range.
2006 Honda Rebel
Kubota OC95, Comet 44C/44D
User avatar
Diesel Dave
Site Admin
Posts: 1016
Joined: Sat Dec 02, 2006 1:21 am
Location: Essex, UK
Contact:

Re: Changing CVT behaviour

Post by Diesel Dave »

The problem with this system is it's trying to do 2 jobs.

1) act as a clutch

2) act as a variator

So your low speed response is with the belt slipping, and after it grabs fully then it will respond as a variator. This is why petrol auto scooters use a centrifugal clutch after the belt.

In it's current form it will be set to pull away from standstill with up to 4 passengers in a lightweight car, with your setup you can afford for the engagement to come in lower. The trick is to get the weights so it engages earlier but doesn't drag at idle. S'gonna need some careful experimentation.

The spring force will control the amount the belt is gripped by the system - again you can get away with a lighter spring because you don't have the full expected load of the car, it will also make the system slightly more efficient in not wasting power.

I've worked on loads of scooters from 50's through to 500 TMAX and 600 Silverwings; they are infinitely tuneable with differing weights, clutch springs etc - after all this is F1 technology from the 1960's

Image
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

Thanks Dave

That makes sense, and also explains why I have to give it more throttle than I would think. In a car, the engine would box down if loaded at a lower RPM.

I am going to a friends house where he has long stretches of empty road tomorrow to have a play with the weights etc.

I addition to the slow speed at the bottom end, at the top end it doesn't seem to shift all the way into top gear, so moving the engagement point lower should sort this as well.
User avatar
coachgeo
I luv the smell of Diesel...
Posts: 2002
Joined: Sun Feb 22, 2009 6:00 am
Location: USA Ohio, Above Cincinnati, Close to Dayton

Re: Changing CVT behaviour

Post by coachgeo »

Two things in play here.

1. The mental sound..... the sound; the one our brain / ear relationship, of acceleration of a vehicle over the years is NOT the sound you want these CVT vehicles to have. CVT works best cause it pushes the engine up to its optimum efficient RPM way more often then we are accustom too. Just sounds wrong to us.... but it is actually right in terms of mechanical efficiency. It is our brain/ear relationship that has been trained wrong thru years of inefficient use of an engine. Auto makers have gone to great lengths in CVT trans. vehicles to get their vehicles to "sound" and "feel" like we are accustomed to, even though its all wrong.

2. If you've not tuned the CVT primary/secondary to your engine/vehicle then yeah you got some weight and spring experimenting to do. USE FACTS, test runs under same conditions measuring speed, acceleration etc. Don't go by your ears or you will set it up inefficiently.
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

Thanks Coach

Its my understanding that during the bulk of the acceleration, and the main shift "point", you want the engine to be at peak torque? is that correct in this instance.

I have a tachometer that I will be using to play with so will find out what its shifting at the moment, and have a play with weights.
User avatar
Diesel Dave
Site Admin
Posts: 1016
Joined: Sat Dec 02, 2006 1:21 am
Location: Essex, UK
Contact:

Re: Changing CVT behaviour

Post by Diesel Dave »

Spot on Alex,

For example many moons ago I tuned a Comet 40/44 setup to run with a Yanmar clone motor:

Using the performance graph the sweet spot was about 2750 (lowest fuel consumption), the torque graph was almost completely flat and therefore the HP was a straight line.

On the old Comet stuff you had variable weights and also a 'garter spring' for the driving pulley and varying springs for the driven pulley, there were 3 types of weights and 2 rear springs, so medium weights and the harder (red) spring set it to 2600 which was close enough.

Looking at the 850 perf graph you have a flat output from 1800 to 2600 so you have a wide choice of setup.
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

Yes indeed, it seems that 1800 is my peak torque, although as you say its pretty flat.

On the flat, do I want it to gear up to the redline at max speed? I think 75+ should be achievable with 22 bhp.

Considering 55 is doable with 10. I belive total gear ratio is for 78 but I haven't done the maths on the pulley sized etc.

I can achieve 60 relatively easily at the moment, and at a guess the engine is at around 3000 rpm based on the front belt travel. It still has a fair bit to go before its parallel with the outer edge of the pulley.
User avatar
Diesel Dave
Site Admin
Posts: 1016
Joined: Sat Dec 02, 2006 1:21 am
Location: Essex, UK
Contact:

Re: Changing CVT behaviour

Post by Diesel Dave »

Ideally max rpm = max speed but more importantly is to set you most used cruising speed at the motors most efficient.

So if your fuel graph goes upwards after 2600 ideally you want 70mph to be below that point if possible but don't overgear it to the point it struggles to hold speed up motorway inclines.

I'd rather had a bike that had good performance between 50-75 and a top speed of 80 than something that will do 90 but is a pain around town.

Unless your fitting a screen, then always remember Edward Turner's words, "The most comfortable speed for an unfaired motorcycle is 55mph, more than this and rider fatigue plays too high a price on the ability to cover distances". He knew a thing or two about motorcycles.
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

Aha I see

my graph is lowest at 2400 and then goes upwards from there up to redline.

The main aim is 70mph (for obvious reasons). At the moment it struggles to do 70 unless on the flat, but i'm sure with some tweaking it can be achieved.

I think based on my fuel consumption graph im going to gear 70 for 2600-2800.

I' sure with a play I can get some good numbers.

You make a good point Dave, I rarely go quickly anyway, I suppose if I where to do lots of motorway journeys then I should fit a screen. I am convinced that this bike will still be better on fuel around town than on the motorway.

I did 150 motorway miles and got 89mpg. Will see what its town tank does!
User avatar
Diesel Dave
Site Admin
Posts: 1016
Joined: Sat Dec 02, 2006 1:21 am
Location: Essex, UK
Contact:

Re: Changing CVT behaviour

Post by Diesel Dave »

You need a tacho...

Image
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

On its way from China. For my testing I will have to make do with a handheld IR tach.
User avatar
Diesel Dave
Site Admin
Posts: 1016
Joined: Sat Dec 02, 2006 1:21 am
Location: Essex, UK
Contact:

Re: Changing CVT behaviour

Post by Diesel Dave »

If you have the Superdream Tach, you can prob. get it to work using a bicycle speed sensor and a magnet attached to the flywheel.

That's how my Tiger one works.... :D
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

The super dream tach is wired up at the moment, but it fluctuates like a pendulum. Not much use. The whole cluster is rather sorry so I have an aftermarket set including lights on its way from China!
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

Added some weights today, engagement /start shif point seems to be 1800 rpm , with a movement through the range at approx 3000.

It has bettered town riding, but motorway riding as now too lowly geared (too many revs), im having to give more fuel for the same speed. Looks like back to the old weights, I only added 5 grams or so, I suspect its previous settings was the best compromise (sorry Tim!)
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

Also my thoughts are confirmed with a tank avg of 81mph instead of 88 as per the last tank. Need to wait for tach before re-tweaking.
gilburton
I luv the smell of Diesel...
Posts: 761
Joined: Sat Apr 21, 2007 11:35 am
Location: UK northants

Re: Changing CVT behaviour

Post by gilburton »

I've not been on the site for some time so I'm just catching up.
The CVT needs a handful of throttle at the start. The engine will be revving away but you won't be going all that fast.
What happens is that the bike speed catches up with the engine revs and you suddenly realise you are rolling along with revs matching the speed.
The engine catches up with the throttle rather than you adjusting the throttle.
You have to forget any relation to a manual gearbox. Try riding an auto scooter and it's the same.
I used to own a DAF/Volvo 66 and basically you'd just floor the throttle until you reached the speed you want.
Fuel consumption will suffer a little as you can't roll along in a high gear at low revs as the cvt will just adjust to the optimum revs/road speed.
Forget weights etc. These diesel engines all give their best power at certain revs. Compare the revs on various diesel engines whether single/twin or triple.
You will find that they are broadly similar on industrial engines. The only thing that changes is the HP
Most industrial engines are not designed to be constantly changing revs like a car/bike engine and give their best at a constant rev setting.
The cvt will allow it to do that!!
If the cvt came off a DIESEL Aixam then it will be set up for those revs.
I would be looking at a sprocket change if you want quicker take off.
A lot depend on your main usage. Possibly you could have a town sprocket and one for when you go on longer runs at weekends or just find a compromise.
An extra rear wheel with a different sprocket would make changes easy lol
Possibly the slight bogging down you are getting is the pulley engaging with the belt coupled with the fact you are not using enough throttle??
Anyway as I said I am late to the discussion so you might have got used to it by now??
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

Many thanks for the response.

I have now done nearly 700 miles, 70% town 30% motorway.

The more I ride it, the more I realise its tuned perfectly.

My tach still hasn't arrived, so I don't know how many revs in doing, but if If I give 100% fuel from standstill it accelerates quite quickly.

Having had the CVT apart in its entirety, Its easy to see that , as there is no clutch, the clutch action is accomplished by the belt slipping on the front pulley.

I have ridden many CVT based engines, and I think its because its a 3 cylinder, you get the impression that the engine is revving much higher than it actually is.
BertTrack
I luv the smell of Diesel...
Posts: 334
Joined: Tue Jun 05, 2012 3:53 am
Location: Netherlands

Re: Changing CVT behaviour

Post by BertTrack »

Correct!

The difference between single and multi cilinder engines can take some getting used to.

Especially when you like to play by ear.
Image
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

Tach installed for a while now, done around 200 miles with it on.

Engagement RPM is around 2500 and then shifts down to around 1750 above 15-20mph.

That side seems pretty good.

at 75mph im at 2500 rpm, and the CVT hasnt shifted all the way up (as I can see from the front pulley).

80 mph is achievable on the downhill. 65-70 on the flat. I am not sure I should change antyhing, as per others, gearing down will likely increase top speed at the expense of fuel economy.
UAofE
I luv the smell of Diesel...
Posts: 225
Joined: Sat Sep 20, 2014 6:39 am
Location: Denver, CO USA

Re: Changing CVT behaviour

Post by UAofE »

alexanderfoti wrote:at 75mph im at 2500 rpm, and the CVT hasnt shifted all the way up (as I can see from the front pulley).
...
gearing down will likely increase top speed at the expense of fuel economy.
It won't because the belt isn't achieving it's overdrive ratio. Higher consumption would come from the greater power needed to maintain a higher average road speed.
If you continue to cruise the same speeds you do now, consumption shouldn't change but you will have the gearing to get higher speed when needed.

Thats why I'm going to gear down from 15 to 14 tooth sprocket tonight. I'm geared for 70mph but its too tall to get there since my clutch won't downshift above 45mph.
2006 Honda Rebel
Kubota OC95, Comet 44C/44D
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

I thought that as well, as its still "shifting" its not hitting the rev limiter.

The only problem problem is that I am belt driven on the rear wheel, to decrease the ratio I assume I can either go smaller on the drive pulley or bigger on the rear pulley.

The front pulley is as small as I want it to be as I do get slippage if I accelerate very hard from a standstill and that would get worse with a smaller pulley, the rear pulley is already pretty big, not sure if I could fit a bigger one on the back, possibly?
UAofE
I luv the smell of Diesel...
Posts: 225
Joined: Sat Sep 20, 2014 6:39 am
Location: Denver, CO USA

Re: Changing CVT behaviour

Post by UAofE »

Certainly possible. I have seen people use absolutely humongous sprockets on the back, some nearly as big as the wheel itself!
2006 Honda Rebel
Kubota OC95, Comet 44C/44D
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

I'll see what ratio I can obtain. The rear pulley is already pretty big. I must count the teeth.
alexanderfoti
Site Admin
Posts: 1290
Joined: Fri Jul 15, 2011 4:22 pm
Location: London

Re: Changing CVT behaviour

Post by alexanderfoti »

OK I am running a 22 tooth front pulley and 112 tooth rear pulley. Annoyingly the next size up is 144 teeth which is too much of a jump, and would reduce my top speed to 68 mph at 3000rpm.

The other option is a custom made pulley but I think that may be a bit too expensive for experimentation.
Post Reply