CVT test run

[2gnospam]

Sorry I can't help with the clutch tuning, but would want to know where did you get that tach. Been looking for a low cost one compatible with a Vanguard twin.

Thanks

[wfo]

you can get different weights for that primary, i used a 150 gram weight and it seemed to work better

[JohnF]

Totally science. Insanely complex.

get ready >>

Olav Aaen's "Clutch Tuning Handbook" if you want to know the theory of operation.

Sheave angle
ramp angle
spring preload
belt
flyweight
yada yada yada...

I assume that : you've found what you consider the proper belt, that we must tune the CVT you have now, this CVT has fixed sheave angle/secondary helical angle/primary flyweight ramp angle, that the only variable we can adjust are flyweight mass, primary pressure spring, secondary spring preload - like any of the CVTech Powerbloc series CVTs, and that we are pretty close with the CVT we are starting with.

Changing any one of those variables will change all output parameters of the CVT to some extent, as you have mentioned, but each variable has a much bigger effect on one particular output than the others.

. flyweight mass - running RPM
. primary pressure spring - engagement RPM
. secondary spring preload - gearing selection (upshift/downshift)

First tune the primary

The drive pulley (primary clutch) is an RPM sensitive device and we use it to dial in the power output of the engine. I can't tell very much about the tuning of the primary from your video because it doesn't look like you ever ask for max HP.

Keep in mind that your engine performance is subject to the effects of density altitude and that your primary clutch tune is subject to the performance of your engine.

First goal is to get the flyweight mass right. For this we need to see what RPM the CVT runs at as it shifts through gears so you need to create a situation where you can go WOT and have the CVT shift through gearing slow enough to see it. Since we are watching tachometer and not having a computer doing the watching I would try WOT speed run either on a slope or towing a heavy trailer. We want to see: engagement - quickly go to target RPM - then gear up shifting. If RPM is too high you heavier flyweight mass - RPM too low use lighter flyweight mass.

Next we'll tweek the pressure spring. It has two variables that we really look at and that is preload and rate. Preload will change the engagement RPM and rate has a small effect on running RPM. Shimming a spring like Rock Doctor does changes the preload but do so with caution because you can coil bind the pressure spring and limit the movement of the drive pulley sheaves. Engagement at an RPM (and therefore engine HP) too low and you'll just kill the engine, just like when you learned to drive a standard transmission pickup truck. Engagement RPM too high and you'll endure violent launches. Too high/too low depend completely on the what is necessary to get the vehicle mass moving which is completely dependent on the vehicle mass/terrain slope at the time. If you always run light on flat ground you'll prefer a lighter preload and engagement RPM. If you carry the materials for a small house in a single trip... well, you get the idea.

If your CVT is one of the new Powerblocs with the little mass disks then I might cycle at this point a bit to tweek the flyweight mass around the pressure spring weight assuming I am at some middle of the road density altitude. If one of the older Powerblocs, like on the conquest, then I'm done.

And now shifting

The driven pulley (secondary clutch) is a torque device and it is how the CVT chooses a gear. If this were grossly off then everything I told you so far is wrong and you've wasted your time.

Grossly off secondary

The drive and driven parts of the CVT communicate through the squeeze pressure on the belt and the spring preload in the secondary dials in what that pressure needs to be. If the preload is too high (imagine so high it is like the secondary is welded) then there will be gobs of heat generated by the pulleys overcoming the giant squeeze just to turn and the CVT won't shift up. WOT but won't shift out even on a flat (assuming proper HP) would be a clue to this. If the preload is too low then the CVT will be lugging the engine by trying to be in too high a gear for the situation and if grossly off then there will be a gob of heat created in the secondary sheaves and belt squeal because the belt will slip. The "right" rate of upshift/downshift is dependent on use - again, light -vs- loaded.

So, if you are lugging then increase preload; if you are waiting forever to get speed then decrease preload; if you are generating lots of belt/sheave heat/squealing and/or the shifting was grossly off then adjust the preload towards improvement and start the tuning process all over.

Anyway, that's just my opinion. Put some WOT runs on video and post those.

Oh, and totally science, right?

[Bazooo guy]

Great info John.
So many variables ...
I think a big part of tuning is testing until it works best for the driver and conditions.

[thebuggyman1]

Wow, Great info John. I did hit WOT a couple times in that video, but only for 2 or 3 seconds...I was trying to keep the RPMs low, and in the "normal range".
Adding the load, Hmmm, I never thought of towing something heavy. I don't have any large hills, but they really don't bog that motor anyway. I guess I could tow the 1200# Avenger around the yard with the hand brake on. That might bog it down. lol
I think I'm pretty close to where I need to be, maybe a slight decrease in spring rate in the primary.....and maybe the same in the secondary.
Do you have any info on the primary springs? Colors/rates etc.
Do you have any info on how to adjust the Invance secondary? Are the springs replacable/adjustable? Info on rates if they are? The ramps work differently in the Invance secondarys, so I think the only adjustment would be in the spring.

[JohnF]

I bothered a cvtech engineer for about a week digging into this stuff for my clutch.
Go here to the CVTech R&D home page. Under "Technical Information" select "Pulley Adjustment" and there will be a choice that gives "Spring Specs".
I don't know the invance secondary but i imagine there are spring choices.

Shoot a query to info@cvtech-aab.com - I got a technical rep. named David and he will help.

And I think you mean to decrease the primary spring preload but keep the same rate.

Does your drive pulley have the cool mass blocks with the disk masses and screw cap? Probably. Lucky...
And which primary spring do you have?

[EDIT]

Hey, I found a post of yours where you describe the tune being sluggish until 2500-2800 RPM. I think you are describing lugging. Increase the preload in the secondary so it stays in a lower gear longer. Like I said above - it should engage - go to 3600** (oops, that's my target RPM) RPM - and then start shifting up.

** - If you hunt around under "Pulley Adjustment" there is another "Pulley Adjustment" that shows a curve. Ignore the exact numbers. Imagine an ideal curve that has a straight line from engagement RPM to max HP RPM. This line is steep and the vehicle speed is following granny gear up to max HP RPM. Then the ideal curve is perfectly horizontal as the gearing shifts out. Then at max shift out another steep line as you are fully shifted and now the vehicle speed will increase with engine RPM and the CVT can no longer regulate engine RPM.

If you are feathering the throttle because you are driving around sanely you won't see a lot of this. That's why you need some WOT runs.

[thebuggyman1]

Hey, I found a post of yours where you describe the tune being sluggish until 2500-2800 RPM. I think you are describing lugging. Increase the preload in the secondary so it stays in a lower gear longer. Like I said above - it should engage - go to 3600** (oops, that's my target RPM) RPM - and then start shifting up.

** - If you hunt around under "Pulley Adjustment" there is another "Pulley Adjustment" that shows a curve. Ignore the exact numbers. Imagine an ideal curve that has a straight line from engagement RPM to max HP RPM. This line is steep and the vehicle speed is following granny gear up to max HP RPM. Then the ideal curve is perfectly horizontal as the gearing shifts out. Then at max shift out another steep line as you are fully shifted and now the vehicle speed will increase with engine RPM and the CVT can no longer regulate engine RPM.

If you are feathering the throttle because you are driving around sanely you won't see a lot of this. That's why you need some WOT runs.

The "sluggish" upshift I described was with the rubber engine mounts, and probably incorrectly described. The motor has never lugged or bogged. I think what was happening was the center to center of the clutches was decreasing upon acceleration. Slow start, then sudden upshift(leap in speed) when the mounts flexed, then return to normal upshift as the mount stabilized.

I don't know the code(s) of my spring(s).....I'll have to pull the clutches apart to find that out.
I think I need to do some more testing, hill climbing, trailer towing etc.
So, who wants to ride?

BTW There's lots of good info at that link.

[thebuggyman1]

I've put some more time on the machine, tried a few things, and even put it to the torture test IMG_0014.jpgIMG_0015.jpg
There's nothing like taking a 600# machine and towing around a 1200# machine with it.
It was interesting to see the difference in how the clutches shifted.

I also had a nice long chat with Roy @ QDS...most of what he said was way over my head, but the gist is that I needed lighter weights in the primary.
We talked a little about the adjustable type weights, but the range is too small until you get close to the right weight.

My clutch started out with the largest weights they make, 275g... really they weigh 312g, the 275 is just the steel insert not counting the plastic holder.
Roy felt I would need weights about 100g lighter to bring the RPM range to what I wanted. Because the tuning is a trial and error method, I decided to modify the weights I have and hopefully I'll only have to buy the final set at the proper weight.IMG_0017.jpg This afternoon I took the weights to my buddy's machine shop and removed a little steel from them. They now weigh 238g,
that's approx 3/4 of the original weight. WISH ME LUCK

[hydromike]

I'm going to be running the same clutch on the v1505 in the Newt the Bold, but I'm still at the drivetrain layout stage; R&D with clutch engagement is a long way off for me. Nevertheless, it's great to see somebody else toying with the engagment of these clutches to evaluate their sensitivity. I'm probably going to want first engagement around 1100 rpm in mine, especially if I get the range box working the way I want it to.

Definitely keep us updated....
Good luck!

[thebuggyman1]

Gave the lightened clutch weights a test run last night. With the snow in the yard it was hard to tell for sure, but it seemed to change the "fully shifted" RPM by about 200.
I think the full shift was happening at 3k-3100 rpms, and now it seems to happen closer to 3200 or so. My target RPM is closer to 4k-4200, so it's back to the machine shop tomorrow to remove some more weight.
[IMG][/IMG]