A quick guide to Torque Spring selection

[EthansDad]

I've gotten quite frustrated by how torque springs are sold. I've personally wasted a lot of time and money trying out springs based on who told me what about their favorite brand spring. 1000rpm, 1500rpm, red, green blue - it never made sense to me. just again, ended up buying and trying.

So, here is a quick guide to spring selection I've come up with. nothing new here, just applied to our brands we normally use. This is just a small table for example purposes. I will gladly test and post more data on any brand spring you have if you care to send it my way.

Background on Torque Spring
The main job of the torque spring is to back shift the CVT when needed. like going up a hill or coming into a corner. that is why when the spring fades, the bike bogs because its stuck in high gear (no backshift). The force the torque spring puts on the rear sheaves is also important because until that is overcome, the CVT will not begin to shift.

Different torque springs has different qualities based on construction, metal used and spring height. basic common sense, if I have the two of the same sized springs, but one uses a thicker metal, it will be stiffer. If I have two springs of the same metal thickness, but one is taller, the taller one will put more initial force or resistance on the rear sheave.

Also interesting fact - when springs ware out, they lose free length or get shorter. try it, go measure the new VS old height of your favorite spring. I've seen some drop as much as 1/4 inch.

so the important qualities of the springs to know are:
-free length - how high is the spring
-spring rate - how much force in lbs/in will the spring produce?


Torque Spring in the rear pulley
What I want to know about the Torque spring when its in my CVT is how much initial tension it puts on the sheaves (resist shift) and how much additional resistance per inch it will put on as the bike starts to shift out.

important data points on the DRR rear pulley that will help -

-When you put a torque spring into the rear pulley, no matter how tall it was to start with, it gets compressed down to exactly 1.75 inches

-when the rear pulley is all the way open, it will compress the rear spring to exactly 1.25 inches

building a chart for spring selection

using the data above, we can start to build out something like this:

https://lh6.googleusercontent.com/_axc6dhltIEE/TVw9z1y2-ZI/AAAAAAAAAGY/g5bKMGGH5X0/s800/Picture%206.jpg

Only three springs here for example, but can tell a lot from this. some observations - the reb hebo is a short spring at only 3.5 inches. when compressed in the rear pulley, it puts an initial force on the sheaves of 77lbs. the measured rate of the spring (change in force over 1 inch) is 44lbs/in

the max rpm 1500 on the other hand is a taller spring, yet softer in terms of rate. the end effect it puts more initial force (resist shift longer), but once the force has been overcome, it opens up (or shifts) quicker.

Also note, the reb hebo and max rpm 1500 get to the same final force at 1.25 inches - they just take a different path there.

Summing it up
I personally don't think there is any one magic spring out there. based on your CVT design goals, you need to pick a spring that matches your desired shift points and it is also matched to overcome the force your roller weights generate at a given RPM on the rear sheaves/pulley halves.

I'd personally like to see this kind of data for all springs BEFORE I purchase, but at last, they do not do that. just red, green this, 1500, 2000 that, etc. Again, if anyone wants to donate some springs my way, I'll continue to fill in this chart as a reference for all. Its helpful if you are looking for that right spring that is just a bit more/less than you are currently using.

-EA

[Logan #34's Dad]

So, Whats the chances that WE could get you to do a complete Torque spring chart for all the springs out there?
I'm not real sure how to use this information as it pertains to clutch tuning but........
It'd be nice to know what is the strongest all the way to the weakest.
Also, So if a "bad" spring gets shrunk - Is it really bad or did these numbers just get changed? I've always figured the springs went bad because the metal got hot and made it more spongey.
If I have an old spring or a broke in spring that is shrunk, could that spring be done with the "break in" process and remain that way forever? As long as a certain "broke in" spring meets your needed numbers it may last forever.
Have you tested an old spring and brand new one of the same spring to see if the pressure numbers a different?

[EthansDad]

be glad to test what you have. I have a new red malossi and ztr 1000 here I can add. if you want to send me one, just pm me. can even give it back to you track side if you want.

As for spring selection -I use this guide in a couple of ways. first, if I have what I think are the right rollers, but the bike is shifting before stall, I need more pre-tension, or a taller spring. If I want to change the rate at which it shifts, I refer to spring rate. I know I need more rate if the bike on take off and over revs (say to 13k), then settles back into the shift speed of 11k. I need more rate or a stiffer spring. I'm under I need less rate if even with the best fit rollers, the bike is slow to get to peak RPMs. I'd also rather use the lowest rate I can that will hold the curve of my CVT design. anything stronger just wastes power in generating heat in the CVT.

Interestingly, spring height, which effects pretension can be adjusted or fine tuned. not a factory setting, but can be done. I've thought that if my spring shrinks, why not shim it up to the original height and use it again? I just have not been that cheap yet to try.

here is a pic of a shrunk spring. As for losing "spinginess" that would be a function of metallurgy and I have not stuck my head in that hole yet. I also have not done a retest on a worn spring, was just looking for a quick way to tell a spring has been run.

With the micrometer, the new spring height is 4.33. the worn one is 4.236 inches.

[edwardsp&b]

Have you tested an old spring and brand new one of the same spring to see if the pressure numbers a different?

Thats what I would like to know! Just how much is being lost.

Eric is going all mythbusters on us. Let us know what happens Eric, I will give you a few springs from testing I have when I see you.
Bryan

[bigdeal00]

Man my head hurts after reading all that!:D What are you gonna do with all that knowledge when you re done with minis ?LOL Seriously though that is a LOT of data ther Mr ahlm. We need to find a way to get them "REBUILT" I got a couple hundred i could save some$ if i could get em rebuilt!!!!

[EthansDad]

ok, so found a worn out spring or two to test for giggles. at least one I thought was worn out. have a max rpm 1500 spring that I know has been run for at least 30 laps on a 1+ mile long national track. when I took it off, the bike was still shifting strong, but was just a hair off on launch. I really just replaced it due to the time on it. geez, 30 laps is like 7.5 motos. refer back up to the chart for this same spring new. here are the 30+ lap numbers on the max rpm 1500:

free length 4.26
rate = 36 lbs/in
compressed to 1.75 in = 93lbs
compressed to 1.5 in = 101 lbs
compressed to 1.25 = 111lbs

I measured the 1.75 in twice, its higher than new, must have been a bit off on the read of the new one.

no matter, what I can see here is that the RATE of the spring is still the same, just shorter and the overall force is down 3 or so lbs. I think I could use this spring some more.

I also tested a worn out reb hebo. this one's seen a couple of races and I think my kid's been playing with it in the garage for about a year.

Free length 3.26
rate = 40lbs/in

compressed to 1.75 = 64 lbs
compressed to 1.5 = 76
compressed to 1.25 = 94

A couple of things I see here. first, the length is shorter, again! (see, I told you!). second, the rate has dropped 4lbs/in. third, the initial tension at 1.75in is down by 9lbs, and at full compression, is down by 20lbs of force. I''d say this spring is a no good any more. what I'd expect to see if this old spring was run was a bog down low (due to failure to downshift) and a higher than design operating rpm at full (since the rate is no longer there to hold the shift back). mainly, it would run at non-peak HP rpms and be slower on the track and/or get stuck trying to climb a hill.

to your point rocky, I don't know if now that its shrunk, if you can redesign the clutch around the shorter, less powerful spring and run forever. the spring is not going to shrink forever. the obvious fix would be to lower your rollers, but could also shim the spring to get back your .3 or so inches. the problem there is you run into coil bind, or you can only compress a spring so much until the coils run into each other. I also don't know if there are some metallurgy properties of the spring that after so long, it goes way off the charts on performance drop. never tested that.

I too have a bunch of dead spring soliders laying around I would not mind bringing back into service.

-EA

[Logan #34's Dad]

Thanks for the in-site Eric. I gotta wonder if we set our clutching to an old "broke in" spring would the spring continue to fade or is the metal set and remain consistent and require a whole lot less spring changes.
I recall that ZTR sells his springs on less fading - due to the metal make up. Then Koso countered the ZTR with the White ones. So, if they use better metal then maybe those springs may last forever.

I see us dad's looking in the trash for old springs! Bumming old springs from people. Lol.

[EthansDad]

so giving that some thought - I'll use the worn out hebo as an example of a resurrection. I've never done this, but here is how it might work out.

I know my rate has dropped 4 lbs/in and is now 40 lbs/in. I also know I'm down 9lbs initially, and 20 lbs at full and that my spring is .24 inches shorter.

Let's say I use something like in the picture below. the yellow one is the malossi torsion controller (which I don't think buys you much as its designed) but will add .175 inches to initial spring compression. Also, the metal thing is designed to help the spring rotate, and again does not work well as designed (at least for me), but adds another .153 inches of spring compression when used.

so, if I put them together, I've got .328 more spring compression. I'm going to use the new spring rate and the shim height to build the table out VS measuring.

compress to 1.75 (actually to 1.42) =73.6lbs
compress to 1.5 (actually to 1.172) = 83.52 lbs
compress to 1.25 (actually to .922) = 93.52

(using FL - compress to * rate or 3.26-1.142 * 40lbs/in)

I got the force back up, but I don't think the spring will compress to less than an inch because of coil bind. if it did work, I'd bet the bike would back shift and launch just fine, but might run over designed rpms due to rate being lower, and I can't shim rate! could lower rollers though and bring rpms back down.

try it with just the yellow malossi spacer

compress to 1.75 (actually 1.575) = 67.6 lbs
compress to 1.5 (actually 1.325) = 77.4
compress to 1.25 (actually 1.075) = 87.4

came up 3 lbs on initial force. maybe enough, don't know.

I've again never tried this and don't know if it would work.

-EA

[rookiewrench]

After reading all the data and the opinions from everyone, here is my thought on the torque spring info............. Thank you Lord for my shifter I will never complain again about jetting.

[nitrofish]

Man... You really need to get racin'. I already have trouble sleeping. A few thoughts of mine to chime in. "stress", and "frequency" which generates heat are the killers of a spring. The closer to it's max travel limit you compress the spring, the more stress (psi) the wire sees. Frequency, how your rider is on and off the throttle will contribute to all. Impurities in the wire itself get hot and break the wire down.

0)A purer metal would last longer say a HG chrome silicon formulation.
1 Spring stiffness? Correct me if I'm wrong, but a stiffer spring will always wear out faster.
2) Spring height? Will a taller spring wear out as fast?
3) Metalurgy and spring design, we can learn alot from "other" engine parts on "other" vehicles. ;)
4) Should we "heat cycle" the springs a few times before use.
5) Could the springs be cryogenically treated.

Ok, I just kinda rambled and did not make much sense. But I would say a spring that has been used and "shrunk" over a period of time should not wear anymore. I think it is "plausable" Mr. Mythbuster that we could in fact use a spring forever or at least create one. As long as you treat the spring in the same fashion each and every heat cycle it could work. Start with a heavier spring than you normally use and create a softer one? I know in a bind we have all lightened our roller set-up to make up for a weak spring. Ok, I'm done for now. Thanks, Mr. Ahlm I need some tylenol. Matt.

[EthansDad]

Nitrofish, some very good questions - I have no clue.

My #1 goal was to get a chart going for spring selection and some guidance on how to choose what's best for an application.

My #2 goal was to be able to visually and quickly identify a spring that was good VS bad, since some times my springs get mixed together and I hate throwing away $20 if I don't have to - there is your shrink, but that might not be 100% until the springs are retested as I saw on that not-so-worn out max rpm I was about to chuck.

on what wears springs, I think the ultimate killer is heat, stress and frequency just make heat faster. I had a metallurgist tell me one day if I took a very hot spring and splashed cold water on it, it would immediately lose its rate. never tried that, but think that shows there is definitely some things going on in the chemistry of the metals. I'll leave that to the spring designers to try out. I know some have done the cryo and it seems to hold the rate a bit longer.

I think with a good chart of spring characteristics, it will become more obvious where the gaps are in spring design if some of the fabrication guys out there care to fill in the blanks.

-EA

[MAXRPM]

Eric,
I will send you all of my springs and you can include all sizes in your chart. I dont know if that will help anyone or not but for guys that like data and information at least they will have it.
The Maximum RPM spring is available in standard style springs of (white) 1000,1500 and 2000 the new over range Maximum RPM spring (purple) is available in 1000,1250,1500 and 2000. We created a new size of 1250 for that in between spring.
You will have on Wednesday.
Please note: Color does not mean what size it is, color is only to determine which spring it is, the number is laser etched on the spring.

[EthansDad]

cool Justin,

I'll PM you my address. I'll take what I have so far, plus all these you send and put it in a spread sheet. been looking at google docs, and looks like I can post that data up public for all to view + post a link for anyone to get to it.

-EA

[EthansDad]

So big thanks to Justin for some more test springs to round out the field. below is the link to google docs, anyone can view, and that is where the good stuff lives. click it and look!

Spring compare on Google Docs (https://spreadsheets.google.com/ccc?key=0Av3_8nTLRJqTdHBfaS05S2FBQmg2dFRlRUhLMm9Yc 3c&hl=en)

A bit about the data:
I hear-by certify my results as "hillbilly accurate". seriously though, I took quite a bit of time to get accurate and repeatable results, bought some expense test gear, and made my own test rig, tested some springs 2-3 times each, but at these force loads, .5mm difference can equal 5lbs or more. for anyone that would like more data on how I produced results, or test rig, let me know and I'll tell you. I can say with confidence when its all said and done, my results are within 3-5lbs of accurate. its more like 1lb off for lighter springs, 3lbs off for mid, and 5lbs off for heavy ones. I did have some springs professionally measured last year, and my numbers are +-1lb, so again,even on the heavier ones, I feel good about them to use as a guide, but would not refer to them as the sole source of accuracy.

how to use the data:
when you look at the doc, a lot of numbers there, can be confusing. I sorted the springs from soft to firm based on rate, going from left to right. remember rate is the change in force over an inch, so think of it as how firm/soft it is. that is the #1 thing to compare in my opinion. second, different height springs have a different initial force, or basically how long it will hold the gear down before shift. therefore, the #2 thing to look at is the initial force, or the force in Lbs at 1.75 inches.

From there, you can see lots of these springs are close and might be an option for you to try based on your design goals.

Follow on
I think there is still quite a bit more to learn on springs, but this is a good start. for example, Max RPMs sells standard and over drive springs (white VS purple). same height, same rate (basically) and same initial force (basically). what's different? can see the purple ones are a thinner metal, but make more turns to get to max height. would be nice to know how that effects the CVT since per the force scale, they really look the same.

I think nothing beats track testing, but hopefully for some tuners looking to try something new, this will help narrow down springs in range to what you are looking for.

More spring testing - sure. if you have some favorite springs not here, just let me know, or better yet, send me one. I can get it added to the list.

[Reimer Racing]

Found this video at High Gain Tunning. Malossi spring rates.

http://www.youtube.com/user/HighGainSR50?feature=mhum#p/u/18/d6FNA3avx1A

[EthansDad]

cool video. got to love the Germans and how they can put techno music to anything.

they don't say in their video what installed and compressed height is for them, so makes me question the numbers for an ATV application. last year, I had a professional shop that uses the ASME y14.5m-1994 standard for testing do a few things for me. First, they used an optical comparator and CMM on the rear pulley. came up with a number of interesting dimensions for me, one of which was the initial or installed spring height of exactly 1.25in and compressed at 1.75 inches.

They also did a malossi red spring for me with a tool that looked much like the one in the video.

per their numbers, they got on a malossi red:
installed (1.75) =68, final (compressed) = 84. I got 66lb and 80lbs for the same spring. which is within the resolution I am claiming.

If you want DBA measurements on these springs, find a testing company that uses something like the ASME standard. for testing / tuning and finding the next stiffer or softer spring you might want to try, this chart is accurate.

-EA

[Reimer Racing]

He is from the USA lives in Col. In the beginning of the video he states installed and compressed when the pulley is full open.
Rich has built Turbo Chargers and Fuel Injection systems for scooter motors. Check out his other videos.

[EthansDad]

not sure where this conversation is going really. I have no doubts this guy knows his stuff, and has a better testing rig than I used for sure. EDIT: no, his test rig is really no better.

my point is he says "installed height" and "compressed height", but does not tell us in the video what distance in inches he uses. If you know him, ask him what he is using for an installed and compressed height (in inches) and you can compare his numbers to mine if that is your goal. or, if he can tell you the difference between his installed and compressed height, you can calculate rate and again compare that to my rate calculations. based on the video, I see two numbers, but without knowing the distance (in inches) the spring traveled between those two numbers, I can't calculate his rate to compare nor can I compare initial to final force.


-EA

[EthansDad]

Also wanted to say, I put these numbers public so all could use, but also so folks could question them just as you are. if they are not agreed on accurate, than they are no good to anyone, so no offense taken from anyone that wants to challenge the numbers.

Second, I took a closer look at his test rig in the video. at first glance, thought it was a one-off high end spring tester (I was distracted by the music and not focusing!). at second glance, its not. its a digital postage scale (pelouze 4010), just like mine. it is more accurate (.1 on his VS .5 on mine), but still just a postage scale. second, his compression rig is a drill press. you can actually still see the drill chuck on there, and they are using it to hold the spring. as for height, I am guessing they used the pre-set drill depths on the press. I'm not sure how accurate that is. again. .5-1mm can, and will throw these numbers off 5lbs or more. third, the springs need to be on a flat surface on both sides. if part of the spring drops into one of the teeth on the drill chuck, you just lost 1-2mm on measuring. I ran into this problem as the spring would slip into the teeth on my vice and/or the bumps on the scale and I would not get the same numbers twice until I put a flat surface on both sides of the rig.

here is a quick video of my red neck rig to compare. can't dance to it like his, but can see the similarity and repeatability of the measurements.

http://www.youtube.com/watch?v=03zwODb1rwY

[familyof5]

The headline would read " mini quad son humiliated by dancing dad on utube video, Poor kid just wanted to race." You should just stick to helping us new quad dads getting it right. Thanks for your post. Between headaches some of it is actually sinking in. I used to talk to the neighbor about sports; now its springs , belts rollers and the next local race.

[EthansDad]

yea, thx familyof5. the smart ***** in me wanted to put some techno music on in the background and dance while I did the video. sometimes I really am thankful my common sense kicks in.

-EA

[nitrofish]

Nice work! Props to ya posting this for all of us newbies. As family of 5 said, we used to talk about more interesting things. Oh well, at least the wives don't hang around long in the garage. I am concerned that you may hurt your little toe, where's your shoes!

[EthansDad]

A quick update on the Max RPM springs. I just spoke to Justin and got some clarification on what's the difference between their white VS purple springs.

Here is the deal - the white springs are for the normal pulley, the purple springs are SPECIFICALLY for the malossi over range pulley kit. turns out the outside diameter (OD) of the malossi pulley is a bit larger than normal. some springs fit OK, but many springs will actually bind on the side wall of the pulley, which adds additional force, and will also will wear the springs out much faster. Justin mentioned they would have to actually grind out the Inner Diameter (ID) of some springs to make them work with the malossi over range kit.

so, the purple or over range springs from MAX RPM have about the same performance as their white counter parts, but the ID is larger and specifically designed to work with the malossi over range kit. if you try to use the purple springs in a normal pulley, you might find they don't fit and are too wide to fit in the bottom cuff.


And thanks for your concern Nitrofish! the joy of work at home days. if you look close, you can see my PJ bottoms too! :)

-EA

[familyof5]

Concerning the PJ comment and not the spring data, TMI( To Much Information as my 10yr. old daughter would say)

[Reimer Racing]

Just wanted to show the video, meant no harm. Great job with collecting the data.

[bignasty]

Originally posted by EthansDad
A quick update on the Max RPM springs. I just spoke to Justin and got some clarification on what's the difference between their white VS purple springs.

Here is the deal - the white springs are for the normal pulley, the purple springs are SPECIFICALLY for the malossi over range pulley kit. turns out the outside diameter (OD) of the malossi pulley is a bit larger than normal. some springs fit OK, but many springs will actually bind on the side wall of the pulley, which adds additional force, and will also will wear the springs out much faster. Justin mentioned they would have to actually grind out the Inner Diameter (ID) of some springs to make them work with the malossi over range kit.

so, the purple or over range springs from MAX RPM have about the same performance as their white counter parts, but the ID is larger and specifically designed to work with the malossi over range kit. if you try to use the purple springs in a normal pulley, you might find they don't fit and are too wide to fit in the bottom cuff.


And thanks for your concern Nitrofish! the joy of work at home days. if you look close, you can see my PJ bottoms too! :)

-EA

We ran into this problem this last weekend..luckily it happened in the pits and not out on the track. I had to grind the spring a little to make sure it wouldn't get jammed up on the clutch....;) ;)

[Bens Dad]

Eric I really like the work you have put into this. knowing what the spring rate is, or knowing what spring is weakest to strongest really takes a lot of guess work out of it. I think the whole 1000 1500 2000 needs to have some sort of measurable tension on the box. Thanks again learned alot.

[Coley'sdad#8]

Bignasty,
Yeah with the overrange the regular springs will actually pack up when fully compressed (wide open run) and not close the rear pulley when you go into a turn causing a heck of a bog making ya think it is a weak spring but I knew it wasn't cause it was brand new. Went with the mach I ,same spring for several races and practices , no problems.

[bignasty]

The red mach 1 spring has to be ground down as well. There is thatmetal ring that the stage 6 clutch comes withthat jams up. not letting the spring go over the hub when the spring is fully compressed. Is there another mach 1 spring out there?

[asadad00]

Really good data on the springs, will be useful in tuning,but i am not as hi-tech as you guys somebody has to just tell me what to run. i run both of the mach-1 springs and don't have to grind them any ,works good on the overange pulley.

[EthansDad]

I also use the Mach 1 spring on the malossi overrange pulley and it fits nicely without grinding. I have not tired this, but was told at the track that the new max rpms over range springs (the purple ones that are a touch wider) WILL work on the stock DRR rear pulley. nice to know since they don't make the new 1250 rpm one yet in white or normal size.

-EA

[rmuscle]

sent u a pm

[o4twinpeaks]

just bumping back near the top

[JustinBoots]

Bringing this back to the top to give praise to EthansDad for opening my eyes on this torque spring topic. Funny how I thought I understood it and then I read your stuff and go "Ohhh!". At any rate thanks for being so helpful on this forum....

Now do have a question as well related to the CVT tuning.....once you find what you believe is the quads "Sweet Spot" would you ever really change it based on track style? For example would you go lighter on rollers to bring RPMs up on a tight hilly track, or would you leave the CVT alone and move up on your rear sprocket tooth count? Basically would you only adjust gear ratios using front/rear sprockets from that point forward?

[fish-d-fish]

If you change sprockets you should change rollers too. Small intervals

[Nichols Atvs]

How do i get a copy of the chart the google docs will not open for me.

[farmboyquad]

Ethansdad,

I just sent you an email about a new spring tester tool. Let me know what you think. Thanks