CVT Dax 10T Doesn't Fit, But...

Discussion in 'Transmission / Drivetrain' started by Nuttsy, Apr 6, 2009.

  1. Nuttsy

    Nuttsy Member

    I got my Dax 10 tooth drive sprocket on Fri. as well as my 56 tooth driven sprocket (not from Dax; Ebay item) so a weekend of tinkering ensued.
    1st, the 10T didn't fit the shaft of the pocket bike CVT. It was slightly too small. That's a good thing! Actually, it is exactly the same size as the shaft which is why it wouldn't go on. A few minutes with a file and BINGO...perfect fit. If you do this, take it easy with the filing as the metal is really 2nd rate (not hardened) and files away quickly. Be careful!
    I used a large nut drilled out to the shaft size (9/16") as a spacer/shim, to align the chain on the CVT shaft, which needed one link removed for the smaller dia. sprocket. Also, this sprocket is straight cut so I touched up the tips of the teeth with the file as I was getting some rub inside the chain.
    I started off with the original 44T driven and was impressed with the crispness of the take-off. I could FEEL the MOMENT the clutch grabbed the bell and the bike actually JUMPED forward. NO, no wheelies, lol, but a definite PUSH off the line. I could feel it in the seat as the clutch engaged. Top speed came down only slightly, from high speed of 28MPH on flat to 26MPH. That was kind of expected. I did NOT notice ANY difference in RPM (Subaru/Robin 33.5) though from the 12/44 to the 10/44.
    Next up was the 56T driven swap for the 44T. 1st problem there was the mounting (rag) holes were off and I had to re-drill 9 new ones to match my existing mount. I don't know if that is normal with these different size sprockets or not. Simple enough fix though. The rest of the mount was typical. I cut a new, longer hunk of #41 chain and was ready for a test run.
    Immediately, I was getting a rythemic popping sound from the chain as the wheel went around. The chain was wanting to rise off the ring and then settle again! Strange. Turns out the teeth on the sprocket looked more like a saw blade with some teeth SET. A few were bent inward; others were bent outward. Sub-par quality here. As a minor note, the chrome was OK on one side and lacking and scratched to heck on the other. Definitely not a quality piece of work. Straightened the bent teeth with pliers which was almost too easy. All back together and time for the test ride; again.
    Off the line was only marginally better than with the 44T. Still no wheelies. High speed fell to an unacceptable 20MPH. Still no RPM change but hill climbing was slightly (only very slightly) better than with the 44T although the 44 did very well in my opinion. On my local hills the bike climbs good and accelerates as it goes. Good in my book!
    I was in hopes of getting the engine RPM to go up and therefore maybe a little bit better performance and less stress on the CVT, but it wasn't to be. I guess between the small 4 stroke, the tank it's mounted to, and my 200lb. self, ya can only do so much.
    Overall, I was happiest with the 10/44 combo which kept my top speed and added low end AND climb so I put the 44 back on. I was really hoping to get a solid 30MPH going for those times when you need a little extra, but I never had it before so I won't miss it. I'm satisfied with what I've got and although I was not really too fond of the chain drive, it feels more solid and reliable now. I guess only time will tell. Still no wear on the CVT belt with about 200 miles so I guess the RPMs are agreeing with it.
    I will probably order another 10T from Dax just to carry with me as a spare along with some extra shaft clips.
    Hope this helps others looking into these pocket bike CVTs.

  2. heathyoung

    heathyoung Member

    This is some great info for all of us messing with these CVT's (or planning to) - good to get a baseline figure on what works well and what doesn't.

    I'm looking to run one of these CVT's to the gear side (using a stripped down 7 speed cluster - now a 6 speed) on an 8 speed hub - gives a LOT of clearance for the sprocket, and also allows the chainline to be very short - avoiding problems with tensioning.
  3. loquin

    loquin Active Member

    Nuttsy: It is good to get some feedback. I have a few questions to clarify your setup though.
    1. is this the CVT install on the Western Flyer cruiser?
    2. If not (all the following questions)...Which version of CVT is this?
    3. If it's one of the pocket bike CVTs, where did you find it?
    4. If it's a pocket bike CVT, what is the gear ratio of the gearbox on the output?
    5. Are you using a jackshaft, or is it direct from the output of the CVT to the rear sprocket?
    6. No freewheel, correct?
    Last edited: Apr 8, 2009
  4. Nuttsy

    Nuttsy Member


    Yes, it is on my Cruiser which is a TRUCK! (as in heavy)
    But I'll address the other Q's anyway as others may want to know.

    Q# 2-3 -- It is the CVT for the X2 Ninjas Keep an eye on Fancy as sometimes they have sales on Ebay and you can save $10-20!

    Q #4 -- I have HEARD/READ that the ratio changes from 6.5 down to 2.5. Don't know for certain but hand turning this thing it does 7:1 for low. And using the ratio calculator from JSitton coupled with my actual speeds, I'd say it goes to ABOUT 3:1...maybe less.

    Q #5 -- No jackshaft; direct to the 44T rear sprocket. Probably enough drag on this box without adding another layer of gears. If one day this CVT goes fubar, I may disect it and remove the gears completely and try and fashion a direct output shaft off the final pulley.

    Q #6 -- No Freewheel YET...That's my next expenditure. Coming in the next couple of weeks if I can find the hub dimensions for the kit. They don't seem to know how to measure either so I'll have to get it, measure, and then look for spokes to lace a wheel with. Would have liked to have done this part in the original build but I didn't know if I would like the end result and I was trying to keep the expense down at the time.

    This weekend I'm going to be adding a Tach. from Northern Tool so I'll be able to report back on ACTUAL RPMs. However, I have no doubt that this CVT will do much better with a decent 2 stroke or even the Honda or HS 4 stroke 2HP motor. Hmmmm, future upgrade?
    I like the Robin engine for reliability and simplicity of maintenance, but its power IS limited.
  5. Nuttsy

    Nuttsy Member

    Epic Fail!

    Well a couple of weeks and about 200 or so miles and the DAX 10T gear is toast. It was actually STILL pushing the bike along in that condition. I didn't notice any alignment problems but note the center where it mounts to the shaft. A whole bunch of stress happened here that the cheap metal wasn't made for. Makes me wonder what it does on the TITAN gearbox!?
    It pays to do 'pre-flight' inspections. That can NEVER be stressed enough. Found this before I left on a small trip.
    Am running the 12T again until I can find a QUALITY hardend 10T that fits the CVT shaft.
    Anyone that's got a line on a 10T HARDENED sprocket that fits the pocket bike CVT or Titan gearbox...OR a 10T with 1/2" (12mm) center bore that can be bored out, give me a shout.

    Attached Files:

  6. heathyoung

    heathyoung Member

    Ouch. Didn't the ease of filing it initially give you a clue?

    Buy another one, file it to fit, heat it cherry red with a blowtorch and quench in used engine oil. Easy. Thats if its steel...
  7. Nuttsy

    Nuttsy Member

    Yeah, I knew it was really soft metal (of some sort, lol) but figured if it was meant for chain drive...
    I never expected total meltdown!. I was gonna buy another just to have. But since I'm running the 12T again, I'm gonna 1st try the 56T wheel sprocket. It's close to the same ratio and I have it here, now!
    However, it's SOOOOO much easier to change drive sprockets on this tank than the driven sprocket.
    And, BTW, Ace Hardware (if you have such near you) sells the 1/2" external circlips for the output shaft for 30 cents US each. Not $2-$5 like some sites!
  8. Shrapnel

    Shrapnel New Member

    It wouldn't be too hard to step up to the 8mm chain on your bike. If you changed to a flip-flop hub (are they reverse threaded on the left side?) you could use a reverse freewheel sprocket mount like what they have here :

    (the two different freewheels are at the bottom of the page)

    It accepts the standard 4-hole pocket bike sprockets like what you can find here: Parts

    I've seen as low as an 11 tooth pinion sprocket (which is what I run) and as large as an 84 tooth sprocket for the freewheel using the 8mm chain. That gives you some serious gear reduction options. The 8mm chain is really strong, and the sprockets should last a long time too.

    The best advantage of a hub-mounted sprocket is that it is always centered and doesn't wobble. I've got a pretty long chain on my bike and I haven't had any problems with it.
  9. heathyoung

    heathyoung Member

    Unfortunatly flip-flop hubs are not reverse threaded - BMX ones are 30X1 and ISO (35mm or so) but both RH threads (clockwise to tighten).

    A true track flipflop hub has two RH threads, but a stepped lockring that is LH threaded.

    That parts site is a goldmine - wish we had stuff like this in Aus...
    Last edited: Apr 22, 2009
  10. Shrapnel

    Shrapnel New Member

    Staton-inc sells a hub with the left side of the hub having left-hand threads. It doesn't say if the left side threads are the standard 1-3/8" that normal freewheels use though. Here are the details:

    If it is then all of your problems are solved for $119.:eek:

    Of course, then you'd have to buy the freewheel, sprockets and chain to make it work.:whistling:
  11. loquin

    loquin Active Member

    True. But, after the heat treatment, it will be brittle. To relieve the stress inside the sprocket, and make it less brittle, after the heat treatment, pop it in the oven and bake it at 450 deg F for about two hours, then take it out, and let it cool to room temperature, gradually. Afterwards, it'll still be hard, but, it'll be less likely to break under stress.

    Shrapnel: The left-hand thread freewheel that staton (and others) sell IS the standard 1-3/8 - 24 TPI. Just in left hand threads.
    Last edited: Apr 23, 2009
  12. Nuttsy

    Nuttsy Member

    Tach Update

    I finally got around to installing the Northern Tool Tach on my heap. It seems to work fine albeit the really small readout for these aging eyes. It's quite tiny. It is similar in function as the speedometers where the wheel goes around a few turns before the speed change registers. There is a small lag between the throttle movement and the readout. This is normal though in digital meters and quite liveable.
    It shows the Robin engine popping up to the mid 5000RPM range on throttle up. It shows me cruising in the 5500 to 5700 RPM range. Yet my speed is around 25MPH. Using the handy dandy gear calculator with my CVT, 12T to 44T and 26" wheel, the CVT is only using about a 4.5 primary ratio. Nowhere near the 2 or 3 range it's supposedly capable of. That is somewhat disappointing. Unless the ratio calculator is off. OR the tach. is off. OR I and my tank just flat out weigh too much.
    I'll be trying the 56T final sprocket when time permits to see if there is an improvement. Maybe its time to play with different variator weights.
    I need a chase person to follow me and view the CVT in action with the cover off. Alas, I'm alone here.
  13. DetonatorTuning

    DetonatorTuning Active Member

    hmmm, wonder why it's not pulling all the way up to 'high"

    if you had a 30T on there, even with it only pulling 4.5 from the cvt you would be doing 37mph.

    i have heard of the secondary reduction being such that it wouldn't allow the cvt to get enough rpms to pull on up, maybe it was in Large Phillipino's Buggy Bike gets a cvt thread.

    i'll try to find it.


    ps, yep that's the thread, from post #10 on, very good stuff !!
    Last edited: May 3, 2009
  14. If you're not shifting right, that's all about CVT tuning.
    Welcome to tuning experimenting he11.
    This is what I was speaking about.

    It helps to make a log and write down what changes with each thing you do.

    I would try much lower gearing first.
    (at least 56tooth, but more like in the 70s....)
    The gearing has to be right on to have these CVTs shift right.

    Then try heavier rollers.
    If it's not shifting all the way to the top of the ratio,
    that's because either the gearing is too tall and the motor can't pull it,
    or the rollers are too light.
    Heavier rollers help it shift into the higher gears earlier and easier.
  15. loquin

    loquin Active Member

    44t driven sprocket
    12t drive sprocket
    apx. 3:1 in the gearbox

    total ratio is 11:1. The overall gear ratio is too low - this is why the CVT won't ever get out of the mid range. (it's a 50 MPH top end)

    If you have a 56t rear sprocket, the overall ratio jumps to 14:1. This is probably still too low (38 MPH top end.)

    at 18:1 (30 MPH top end,) it should work. This means that you need about 6:1 reduction on the sprockets, so, with a 12t drive sprocket, you would need a 72t driven sprocket.
    Last edited: May 4, 2009
  16. DetonatorTuning

    DetonatorTuning Active Member

    i'm thinking more like 80T and 20:1.

    the main problem here is limited rpm range. if you want to cruise in the 5500-5700rpm range by the time you get the gearing to support the cvt you aren't going to go over 25mph anyhow.

    these cvt's were designed to give small high rpm engines a resonable take off after they were geared for the best topend they could give.

  17. loquin

    loquin Active Member

    Yeah. My approach has been to
    1. Calculate the high end speed/gearing first, knowing that approximately 10-15% of the available power may be lost in the CVT. (In other words, knowing that you won't be able to achieve the same max top end that you could without the CVT)
    2. Then, back off to the 5500 RPM range, to determine the cruising speed, when the CVT pulleys get to a 1:1 state.
    3. Finally, calculate the max torque situation (at 5500 PM,) as if you're climbing a steep hill. (You divide the 'cruise' speed, from step 2, by the pulley ratio of 2.2)

    With this approach, if you wanted to hit 30 as a top end, then you wound need 18:1 total reduction, cruise would be at 23.6+ MPH, and max torque at 10.7 mph.

    Now, if there are lots of hills, and your main objective would be to have hill climbing ability, then you could work 'backwards.' Suppose that you needed to have max torque at 9 MPH. because of steep hills. Multiply by 2.2 to get the minimum 'cruise' speed of 19.8 at 5500 RPM. The gear ratio could be calculated to be 21.5, meaning that you would need an additional 6.7 gear reduction after the gearbox on the CVT. and, with a 12T sprocket, you would need a 81T hub sprocket, or, a jack-shaft. The top end would then be about 25 MPH.

    I think the key thing to remember is that the CVT isn't the cure-all. Yes, it can increase acceleration (and hill climbing torque,) but it can't do this AND maintain the same top end that you could without a CVT. There are some losses in the CVT (on average, over the life of the CVT belt, about 10%) that will reduce the top end that can be achieved.

    And, I think that you need to get the system gearing right, before you start playing with the variator weights.

    Possibly, the chart below will help in understanding what the CVT does for you, when the bike is geared for climbing, per above.

    Assuming that you peddle up to 5 MPH, then gun the engine. Assuming the clutch pulls in at 2000 RPM, engine RPMs will quickly rise to the 5500 RPM level, and will then level off until the bike's speed reaches the minimum 'cruise' speed breakpoint, and then, since the CVT has no further pulley ratio changes to make, the engine RPM then starts rising as the speed increases.

    Adjusting the pulley weights has the effect of lowering or raising the 'flat' portion of this rpm-speed curve. Adjusting the system gear ratio has the effect of shifting the curve to the right or left (shifting the the first two 'breakpoints and the top speed to the right or left.)

    Attached Files:

    Last edited: May 4, 2009
  18. loquin

    loquin Active Member

    The other thing to keep in mind is the legalities of the situation. In AZ, with the 20MPH MB limit, it may be beneficial to have the cruise 'break point' at about 20 MPH. Then, when your motor starts to increase RPMs, you know you'll need to back off the gas a bit to stay ... legal.
  19. Nuttsy

    Nuttsy Member

    Legalities be ****ED!

    At least for me.
    AND...its a Subaru EHO35...ALWAYS keep this in mind when I post MY findings.
    I was somewhat surprised and at the same time humbled that my posts are not falling on deaf ears. I'm glad there is still some interest in these CVTs.

    So, regarding the Subaru EHO35... the MAX HP of a whopping 1.6hp is achieved at 7000 RPM. Max Torque is achieved at about 46-4800 RPM.
    There is NO WAY IN HE11 that poor little B****RD is gonna hit 7k with (200lb)me and a 100lb bike. The chart is showing No-Load figures.
    Likewise, if I'm jumping from 2k to 5k on throttle up, I'm IN the power band for this engine.
    In reference to my gearing, Yeah, I think it should get lower. Although, if I've got to go to a 70 or so tooth rear driven...well, I think you could see that it would be a HUGE sprocket. That having been said, I've been looking at ways to use a belt system with this critter. Whereby it would have a small 1-2" driver coming off the final to a Whizzer or GEBE type driven pully. Which would be about as big as a 70-80 or larger chain sprocket! AND, I HATE CHAIN DRIVE!
    I really do.
    Then, even IF a 10T or 12T to 70-80T got the engine to rev higher, in order to reach the magic 7000 RPM, this CVT wouldn't get me any more speed than I could pedal on a really BAD day! Nor, do I ask it to.
    What I keep in the back of my mind is that it gives me gearing options I never had before with the friction drive AND the engine is in its BEST operation zone. Meaning it (the engine) should last longer than one that is driven at WOT just to satisfy some performance criteria.
    I'm not blasting anyone here for their input and concern. I'm just trying to remind some that the EHO35 is NEVER going to perform like the Tanaka or similar 2 strokes that can hit 10-12k or more RPM, CVT or not.

    NOW! All of that having been said, I have tried preliminary 'tuning' of the CVT by removing 3 of the weights. FAIL! Don't bother if you are using a 4 stroke of limited HP. Results are: it pulls fine off the line but that's all folks! No top end at all. My next tuning trial would be to go with HEAVIER weights as CBR suggested. If I can get the RPMs up, even slightly say to 6k, it seems the heavier weights would throw out faster to pull the variator in and get closer to the Holy Grail figure these CVTs can produce. All the while keeping in mind this is a EHO35. JEEZE! I wish someone would try this with a high HP 2 stroke. AND on a bike, not a GoPed. I'm limited in my research materiel without a grant of some sort. (hint) LOL
    Lighten UP!!! This is supposed to be FUN!
    Anyway, I'm STILL enjoying my ride. Just playing and reporting. AND, on the side, going CRAZY with all my lighting stuff that I'll soon be posting in the appropriate forum.
    All BS aside...I'm glad I put the RPoMeter on my bike (as well as the CVT) as it gives me some really valuable input that I was only guessing at before.
    Thanks for watching...
  20. Nuttsy

    Nuttsy Member

    Yeah, I got a log...'s called a credit card statement!
    Thanks for the input CBR. Stop being such a stranger. AND... What are you working on of late???