Page 1 of 2 12 LastLast
Results 1 to 50 of 90
  1. #1
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174

    Idea! Gathering some shock data

    Starting a thread which hopefully will be useful to some people. My hope is to provide some good theoretical data about shocks, in particular to start off with air spring curves. Then move on to providing some real world numbers from a home build spring measurement system.

    Phase 2 would be a similar idea/concept but for compression damping and rebound. That is to say if Phase 1 seems interesting and successful I'll move on to building a shock dyno. I'd love to have a means to get empirical data on many of the various shocks out there on the market.


    So to start this off.. One night i was on a really boring conference call for work doing some random IT stuff which more or less involved me sitting around waiting for other people to do things for 4 hours. I got to thinking, as I have mused all summer, about how there is very little data released about bike suspension systems. Shock curves, damping data, data sheets in general appear to be totally missing. Perhaps these are available to bike builders/designers (which I am not), but in general it seems like there is an information black hole. So I went out to garage grabbed a pair of calipers and a Fox 7.875x2 RP2 and went to town measuring it and reproducing a digital copy of it.


    It looked more or less like:


    I left out most of the complexities, especially of the top cap, because measuring that would have been super painful. But all told I think I did ok.

    So What do we really need to know about an air spring to be able to calculate it's curve?
    Well I figure its:
    A known starting point with known:
    -positive chamber volume
    -positive chamber pressure
    -negative chamber volume
    -negative chamber pressure

    So fortunately I could bisect the model, get a profile of the empty volume of both the negative and positive chambers at full compression and zero compression states. Those models looked something like:



    The top being a fully compressed shock and the bottom the shock at rest. So I measured out some volumes, and I didn't really keep good notes and it's been a few months since I did all this, but I believe my numbers were something like:
    Shock in uncompressed state - positive volume 4.3 cubic inches.
    -negative volume .07 cubic inches.
    Shock in compressed state - positive volume 1.4 cubic inches.
    -negative volume 1.82 cubic inches.

    So... assuming the the amount of air actually being compressed into the negative chamber when you install the air sleeve we can assume that at full shock compression the volume of the negative chamber is at 0psi for the full 1.82 cubic inches. Which means when that volume is compressed to .13 cubic inches and we apply Boyle's law PV=T we can come up with something like

    1.82*athmosphere pressure (15psi at sea level) (this part I'm a bit flakey on to be honest as pressure at sea level vs absolute pressure screws me up)=.13X
    restated: 1.82*15=.13X
    solve for X gives us X=210.
    Remember to take away atmospheric pressure at sea level as we don't include it in any of our future calcs so X=195.

    Here's another part where my notes get messy (I didn't take any... just have an excel workbook full of ugly calculations...). I apparently threw that number out at some point in the future and started using a value of 240 for the pressure in the negative chamber when the shock is at rest. The difference in volume of the negative chamber is admittedly a large percent, but as a value its very small so in terms of measuring off the model this is a total crap shoot, an initial measure/guess is .07 cubic inches, using X=240 would imply that X=.107 cubic inches. Other numbers I ran went as high as .13 cubic inches. Ultimately it doesn't matter much because we're going to throw the negative chamber talk into the closet for a little while.


    So now I've got pressures and volumes for both chambers in both states and I can measure out the piston area from the model. I got 1.76 square inches of surface area for the positive piston and .87 square inches of surface area for the negative piston. Thus we can put together a little spreadsheet that looks like this:



    I left out some magic that is in another table and needs to be its own discussion about IFP forces. For those who know how to calculate it (not hard) the assumptions I'm making are:
    Shaft diameter .375"
    internal floating piston diameter: .8" (guess work from some cut-away views of rp23's)
    inital pressure: 400psi (more guessing/internet heresay)
    gives a shaft ratio of 4.555 and an ifp stroke of .439 inches.
    Assuming an ifp set height of .65 inches (again from cut-away views and guess work)
    that means likely minimum force from IFP is 44lbs and max would be 136lbs.

    So given all that, we now have a model of a shock that is somewhere in the ballpark or atleast with in a few kilometers of the ballpark.

    Next step I took was to think about "shimming" my shock, aka reducing the positive air volume by inserting a non-compressible solid into the shock body. I was looking at this Changing FLOAT Air Spring Compression which lead me to this table: 2012 FLOAT Air Spring Summary


    So fox is talking about compression ratios, well looking at my shock a 7.785x2 with high volume can. I had done all my modelling to date as if it were a standard volume can and not included the HV sleeve. But I realized that If I knew what the delta between a standard air can and the XV1 can which I have was that I could use the charge Fox has provided to backwards engineer the value for what the positive chamber must be. I mucked around with it for a long time and did a lot of spreadsheet work and the best fit came out to be:




    Now the chart is based on me fiddling with a bunch of stuff and finding the best fit. In reality fox likely rounds up and down those numbers, it's not likely that compression ratios are nice neat numbers to 1 decimal place, I'm sure some rounding or possibly even creative rounding has been taken into consideration in order to get pretty data to release to the public.So take it with a grain of salt just like everything here.

    So a 7.875x2 shock with a standard air can comes out to have a positive air chamber volume of about 4.9-5 cubic inches via approximate measuring or reverse engineering some fox specs. So we're definately in the ball park now. Same goes for the compressed volume of the positive chamber at 1.4 cubic inches.


    Using the numbers above plus some undiscussed IFP calculations we can create a reasonable model of an air shock and spit out a graph and IMO it looks reasonable. We can also use the formula to determine what effect putting a spacer into the positive chamber would have. Though I'm still waiting on my official FOX float tuning kit 803-00-612 I made up a spacer of my own with a volume of approximately .6 cubic inches and threw it into my shock. My butt informed me after the first ride that it did make a very big difference to bike feel, and the data suggests it should. But saying I feel a difference isn't really very scientific.

    So in the next installment of my ramblings I'll document how I built a spring rate tester and I'll start to post some pretty graphs/data of actual shocks with various settings. We can explore things like:
    What impact does the fox float tuning kit 803-00-612 have on a shock?
    How does a DHX air boost valve and bottom out control affect the spring rate?
    How linear is a coil shock?
    What impact does IFP pressure have on a coil shock?
    How many graphs can I make before I lose my sanity?

  2. #2
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    1,412
    I've been through this exercise for the dual air forks. Was wanting to do exactly this for my RP23. I'd been having thoughts of building my own dyno just yesterday.



  3. #3
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    1,412
    Quote Originally Posted by dberndt View Post
    So... assuming the the amount of air actually being compressed into the negative chamber when you install the air sleeve we can assume that at full shock compression the volume of the negative chamber is at 0psi for the full 1.82 cubic inches. Which means when that volume is compressed to .13 cubic inches and we apply Boyle's law PV=T we can come up with something like
    This assumption isn't valid (even though the numbers in your table seem to balance out). There is a bleed port and this balances the chambers and assures relatively low preload on the spring. The bleed port might or might not be up at the end (full extension) part of the stroke, so you either have to assume that the pressure in the two chambers is equal at this point or measure where it really is.

    I find it easier to not get lost with these calcs to explicitly mark values as either psia (absolute) or psig (gauge). If you are converting a pressure to a force, always make sure you are using psia. If you are looking at a balance of forces make sure you include all forces. e.g. atmospheric pressure acts as a constant force on the cross section of the damper body.

    ... but then again I am some crazy Brit engineer who thinks the SI system of units makes sense... even though I grew up using Imperial units and therefore live in a world of mixed units. Most of my spreadsheets have input in Imperial and calculation in SI.

    Your starting point values on the volumes don't agree with Fox's published compression ratios data. I've just done an exercise (because I'm away from home and needed a diversion) to back calculate the positive air chamber volumes based on Fox's compression ratio chart: 2012 FLOAT Air Spring Summary

    Believing the Fox numbers implicitly, I end up calculating a positive chamber cross section area of 1048mm^2 (1.625in^2). Your numbers give a CSA of 1.45in^2. Not sure who is right. My calculation depends heavily on the Fox numbers meaning what I think they mean which is a long stretch for the imagination.

    My approach would be to work out the effective cross-sectional areas and the volumes at rest for all chambers (pos/neg/ifp) and net to atmosphere.

    The cross-sections of interest can all be described relative to the following diameters:

    body inner diameter csa (bi)
    body outer diameter csa (bo)
    can inner diameter csa (ci)
    shaft outer diameter csa (so)

    The relevant pressures are n,p,ifp and atm. In the following I use the above bid/bod/cid/sod symbols to reference the cross-section areas, rather than the diameters themselves.

    The force balance on the whole assembly is:

    F=ifp*so + p*(bi-so) + p*(ci-bo) - atm*bo - n*(ci-bo)

    For the variance of each of the pressures with shaft position, you need the at rest volume and the relevant c/s areas to derive the effective length. The pressure in each chamber can then be calculated using Boyle, just considering the length ratio.

    i.e.
    ifplen=ifpvol/so
    plen=pvol/(ci-so)
    nlen=nvol/(ci-bo)

    ifp=ifp_stat*ifplen/(ifplen-X)
    p=p_stat*plen/(plen-X)
    n=p_stat*nlen/(nlen+X)

    N.B. negative chamber at full extension has same pressure as positive chamber because of the assumption that the bleed between the chambers is at full extent.

    That should be it. If you give me the volumes and cross-section areas/diameters (assumptions where applicable) I can adapt my spreadsheet to do the rest, with graphs all ready to go.

  4. #4
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    I've got to get to work so will digest and respond more to your post later petercam, but for now to keep you occupied i'd ask. If there is a bleed valve between pos and neg how do you explain people getting "stuck down" shocks. A bleed port would not allow that to happen would be my understanding.

    My start points are based on rough guesses and poor spreadsheet math, I probably can't hope to defend them from a full on engineer who seems to understand this a lot more than I do, but I would say that I'm not sure I understand how you went about using the chart on 2012 FLOAT Air Spring Summary to back calculate any sort of chamber cross sectional area? There seems to be no data or linkage between the data their putting up about compression ratios and cross sectional area. Further to that drawing the shock and then determining CSA with a CAD/CAM package seems like a fairly easy way to go about this that should land me within a few % accuracy, its not a difficult part of the assembly to get right... That being said In the bottom of the bottom middle of my first chart I explicitly list the CSA as 1.76 pos and .87 neg.


    Gotta head to work, more later.

  5. #5
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    1,412
    Having read your entire post... looks like we've looking at many of the same sources. Here is where I back calculated from the the (rounded) Fox CRs to suggest a consistent result for the cross-section area:



    What I did was say that the vols for end-space and stroke were x and y respectively.

    The compression ration (no volume reducer) is by definition: CR0 = (x+y)/x

    With a volume reducer in place, this gives: CR1 = (x-vr+y)/(x-vr)

    From known CR0 and CR1 with a known volume reducer, you can solve for x and hence y. Given the stroke you can solve for the cross-section area.


    I'm away from home so not able to measure anything, so using the Fox chart let me do some thought-maths. I'm not trying to negate or disrespect any of your work. I'm amazed at the timing that I literally looked at this yesterday.

    I think we are in great shape to fully describe the air spring characteristics.

    As for stuck down...

    Normal behaviour is for the positive pressure (at any part of the stroke except at full extent to exceed the negative pressure. The seals will wear according to this pattern of pressure loading. Eventually the seal system separating the negative and positive will fail causing a bleed of high pressure over to the negative chamber. The shock now reaches equilibrium at a point that means it doesn't reach the equalisation port (full extent). It is stuck down. Any attempts to force the shock to full extent are fighting against... the negative now having higher pressure than it was ever intended to have and the bleed port sitting right up at the most progressive phase of the negative chamber's compression. Additionally, the seal system is now facing a reversal in pressure differential (the negative is at higher pressure), so the seals will shift in their glands and rest on a non-worn part, so there isn't likely to be any bleed back in the other direction.

  6. #6
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    As I actually have an apparatus for measuring shocks throughout their stroke can we devise a test to establish the existence of this bleed port and it's function?

    If the pressure equalizes at top of travel (full shock compression) then I would assume that if I installed the air sleeve and did two full compression cycles I'd see a delta between the compression stroke on the first pass and that of the second, as the second would have benefited from the equalization of pressure via the bleed port.

    Sound fair? What are your thoughts?

    As to how you came up with CSA that's good work. It certainly makes sense now. I failed to realize that we could go from volume to surface area using stroke length, it should have been obvious. Anyways on CSA I think we more or less agree, my drawn/measured number is within ~10% of your calculated number, I think it'd be fair to assume that this comes from the margin of error in the FOX data sheet and any measurement/drawing errors.

  7. #7
    mtbr member
    Reputation: Vrock's Avatar
    Join Date
    Jan 2004
    Posts
    427
    The bleed port It's really easy to see in Fox shocks, it's just a dimple in the air can at the beginning of the travel.

  8. #8
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    Pic? I dont see it...
    as a reference for example.

  9. #9
    mtbr member
    Reputation: Dougal's Avatar
    Join Date
    Jan 2004
    Posts
    4,405
    Quote Originally Posted by dberndt View Post
    I've got to get to work so will digest and respond more to your post later petercam, but for now to keep you occupied i'd ask. If there is a bleed valve between pos and neg how do you explain people getting "stuck down" shocks. A bleed port would not allow that to happen would be my understanding.
    As mentioned by others, the bleed port is a notch pressed into the air-can from the inside-out. You can see the lump from the outside.
    A stuck down shock is when this bleed port stops working, it can be caused by many things including the seal getting soft enough that it deforms into the port (notch) and stops any air escaping on the way past. Combined of course with positive air pressure leaking past the seal deeper in the stroke.

    My main question is "what model are you using to predict the pressure rise?"

    *edit*
    I just found in your first post that you used boyles law "aka ideal gas law" to predic the pressure rise. Unfortunately that doesn't work unless you compress the shock slowly enough that the temperature doesn't rise.
    You need to use adiabatic compression instead.
    */edit*
    Owner of www.shockcraft.co.nz and NZ Manitou Agent.
    www.dougal.co.nz Suspension setup & tuning.
    SPV Devolve

  10. #10
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    I'm sorry, I have a DHX air and an rp2 here in front of me and I don't see it. Show me the money...errr bleed port. I've looked inside, the air can, outside the air can, there is no obvious depression that I can find.


    Also, for now all I have is an a device to slowly compress a shock against a load cell, giving me an output of force vs displacement, as it takes 30+ seconds to do one compression stroke I'd consider that temperature increase is likely extremely minimal, though i have no data to back that up. So I'm just going to continue to use Bowle's law and do some experiments to compare theoretical data vs real world.

    Eventually I'd like to work at higher speeds and lower forces by isolating the damping circuit (removing the air or coil spring). But the combined/holistic view of the entire shock at once is likely never go be realistic, the forces required to generate the shaft speeds in question with a pressurized shock are going to be a lot more than I'll likely want to deal with or be able to afford to deal with.

  11. #11
    mtbr member
    Reputation: Dougal's Avatar
    Join Date
    Jan 2004
    Posts
    4,405
    Quote Originally Posted by dberndt View Post
    I'm sorry, I have a DHX air and an rp2 here in front of me and I don't see it. Show me the money...errr bleed port. I've looked inside, the air can, outside the air can, there is no obvious depression that I can find.
    I don't have a DHX air to compare, my DHX is coil. I have a few FLOAT's lying around, the bump is just like this one. On the left edge of the can, halfway between the decal lower edge and the last reduction in diameter.

    Same place on this one.



    Quote Originally Posted by dberndt View Post
    Also, for now all I have is an a device to slowly compress a shock against a load cell, giving me an output of force vs displacement, as it takes 30+ seconds to do one compression stroke I'd consider that temperature increase is likely extremely minimal, though i have no data to back that up. So I'm just going to continue to use Bowle's law and do some experiments to compare theoretical data vs real world.
    Your 30 second test is a long way from real world. I've never encountered a 30 second bump.
    Adiabatic compression is what you need. The exponent will likely have to be varied from the 1.4 used for air to get a good fit with accurate experimental results.
    Owner of www.shockcraft.co.nz and NZ Manitou Agent.
    www.dougal.co.nz Suspension setup & tuning.
    SPV Devolve

  12. #12
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    Thanks, I found the bump finally... no external mark on my cans but if you look far enough inside its there. Interesting. Will have to change my calculations.

    I don't disagree that it's not a real world test, but it can be used at low speed to prove the model out and generally goof around. For making real world changes and testing on the bike the numbers can be re-crunched for Adiabatic compression.

  13. #13
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    Quote Originally Posted by petercarm View Post

    The cross-sections of interest can all be described relative to the following diameters:

    body inner diameter csa (bi)
    body outer diameter csa (bo)
    can inner diameter csa (ci)
    shaft outer diameter csa (so)
    Can you define the values you are looking for a bit more? I don't think I follow exactly in particular what bi vs bo is.

    I believe bi would be a reference to the diameter of the IF piston? I haven't disassembled and rp2/float/etc enough yet to know that, but I have a good guess. bo is 27mm, account for some wall thickness, say 1 to 1.5mm and you get a BI D of 24 or 25mm, surface area .76square inches.

    BO diameter 27mm, giving an area of .887 square inches.

    ci or sealhead OD 1.5" dia, or 1.76sq in

    small inner shaft so has Diameter .375" or surface area .110 sq in.

    Hopefully those numbers are what you are looking for, or make sense. Let me know if you're looking for something else. I'd post my excel spreadsheet but to be honest it's a pretty big mess. I can clean it up and post if need be but it sounds like you already have a handle on the maths and could easily share?

  14. #14
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    1,412
    Last remaining measurement of interest is how far into the stroke the pressures are equalised. i.e. where the bump centre is relative to the centre of the seal gland.

  15. #15
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    bump centre is approximately 1.02" from end of can, in that one inch you'd have .57" of bottom of can seals. plus whatever the negative chamber space is, plus i''d assume half the seal head on the which is ~.352/2.

    1.02-.57-(,352/2)=.254. It's a bit of a crap shoot to measure and figure out exactly what that means for the negative chamber, also the start of the negative chamber in the can is not at 90 degrees, there is a chamfer which eats some volume. setting my model at ~.192" nches stroke on the neg chamber accounts for the chamfered area in which I imagine the piston never travels into or bump stops against. Resulting in a negative chamber .20 cubic inches.

  16. #16
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    1,412
    Quote Originally Posted by dberndt View Post
    bump centre is approximately 1.02" from end of can, in that one inch you'd have .57" of bottom of can seals. plus whatever the negative chamber space is, plus i''d assume half the seal head on the which is ~.352/2.

    1.02-.57-(,352/2)=.254. It's a bit of a crap shoot to measure and figure out exactly what that means for the negative chamber, also the start of the negative chamber in the can is not at 90 degrees, there is a chamfer which eats some volume. setting my model at ~.192" nches stroke on the neg chamber accounts for the chamfered area in which I imagine the piston never travels into or bump stops against. Resulting in a negative chamber .20 cubic inches.
    1.02-.57-(,352/2)=.274

    I manage to get this to match your 0.20 cubic inches volume.

    The negative air chamber acts as the top out spring, but it is ramping up pretty fast at that stage. I'm going to assume the centre of the bleed port as the datum for the stroke. It just makes things a bunch easier and I think it makes next to no difference.

    I think I am set with all the numbers I need.

  17. #17
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    As I said the chamfer at the end of the travel means that not all of the air can is available as stroke length.
    That is why 1.02-.57-(,352/2)=.274 doesn't match volume.

    I think you're making reasonable assumptions/choices with starting at 0 stroke being equal pressure.

  18. #18
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    1,412
    I've been looking at the cutaway diagrams myself. Just sense checked a calculated 200psig figure for me on a 2.4 leverage ratio bike with an 8.5 x 2.5 shock.

    seeing as that's the bike I ride, I was happy to see it predict 28% sag.

    I've got a bit of work to do but it is looking hopeful.

    Sent from my GT-I9100 using Tapatalk

  19. #19
    Bicyclochondriac.
    Reputation:
    Join Date
    Jan 2004
    Posts
    13,387
    This a a great thread, I have skimmed it, but it will take me a while to really understand a lot of it.

    I see there is some info Fox floats of various volumes (not sure how to interpret the data). I would like to see some of this spring curve data graphed (pardon me if you've already done this and I don't understand what I am looking at), but more importantly, I would like to know how it compares to other models.

    The practical application of this for me is that I am considering getting a 7.5x2.0 Monarch high volume shock, but I cannot find any info that compared that spring curve of this to the RP2 with an xv1 or xv2 can. I've been told that the Monarch is very "linear" feeling, but I'm not sure if this is due to a more linear spring rate, or better mid stroke damping support. I was very surprised when I asked Push and they could not give me an answer about the actual spring curve.
    15mm is a second-best solution to a problem that was already solved.

  20. #20
    Bicyclochondriac.
    Reputation:
    Join Date
    Jan 2004
    Posts
    13,387
    As a purely practical matter, how do I know what volume reducer (if any) is in my RP2? Is it something I can easily remove or change?

    Thanks
    15mm is a second-best solution to a problem that was already solved.

  21. #21
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    1,412
    Quote Originally Posted by kapusta View Post
    As a purely practical matter, how do I know what volume reducer (if any) is in my RP2? Is it something I can easily remove or change?

    Thanks
    remove from bike, release pressure, place can end hardware in soft jaw vice. Unwind can (as you would for an air sleeve service). Look under the top out washer. If there is a translucent plastic piece there, you have a volume reducer.

    Fox instructions say to pull away the steel washer down the shaft and use a4mm allen key to pry the volume reducer out of position.

    On this thread, I should have a model for the fox spring system by tomorrow (with curves)

    Sent from my GT-I9100 using Tapatalk

  22. #22
    Bicyclochondriac.
    Reputation:
    Join Date
    Jan 2004
    Posts
    13,387
    Quote Originally Posted by petercarm View Post
    remove from bike, release pressure, place can end hardware in soft jaw vice. Unwind can (as you would for an air sleeve service). Look under the top out washer. If there is a translucent plastic piece there, you have a volume reducer.

    Fox instructions say to pull away the steel washer down the shaft and use a4mm allen key to pry the volume reducer out of position.

    On this thread, I should have a model for the fox spring system by tomorrow (with curves)

    Sent from my GT-I9100 using Tapatalk
    Cool, thanks. I've had my old floats and AVA apart, never really though about if there were reducers in there.
    15mm is a second-best solution to a problem that was already solved.

  23. #23
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    I'm still waiting for my set of fox float spacers to come in. I have a few I made myself but I would prefer to only post graphs/charts with data for spacers other people can actually get. Hopefully by mid next week I'll have some real world charts posted up.

    I don't have anything other than a fox rp2, dhx air 5.0 and a van coil rc. Eventually i'd love it if I could get my hands on or temporarily borrow some RS rocks and also some of the newer fox's with boost valve.

  24. #24
    Bicyclochondriac.
    Reputation:
    Join Date
    Jan 2004
    Posts
    13,387
    Quote Originally Posted by dberndt View Post
    I'm still waiting for my set of fox float spacers to come in. I have a few I made myself but I would prefer to only post graphs/charts with data for spacers other people can actually get. Hopefully by mid next week I'll have some real world charts posted up.

    I don't have anything other than a fox rp2, dhx air 5.0 and a van coil rc. Eventually i'd love it if I could get my hands on or temporarily borrow some RS rocks and also some of the newer fox's with boost valve.
    I would like to see some spring curve data with the boost valve. I have read a few people claiming that the boost valve increases the spring rate at the end of stroke, but I would think that it would be so small as to be negligible. It would be good to see actual spring curve data.
    15mm is a second-best solution to a problem that was already solved.

  25. #25
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    1,412
    Boost valve on an rp23 doesn't do anything to spring rate, as far as I can see. It looks like the sense of boost valve between the rp23 and the dhx are the complete opposite.

    Boost on rp23 gives more platform around sag. Boost on dhx gives more bottom out control.

    I'm basing this on some blurry cutaway photos rather than having dismantled the shocks to see the innards.

    I have my graphs ready. Will post as soon as I get off this plane.

    Sent from my GT-I9100 using Tapatalk

  26. #26
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    1,412
    There you go. Not peer reviewed yet, but the curves for an 8.5x2.5 rp23 with and without spacers set to 28% sag with a fat git on a 2.4 leverage ratio frame.

    The force and spring rate are translated back to how they feel at the wheel. Units are Newtons for force and Newtons/mm for spring rate.


    Last edited by petercarm; 12-16-2011 at 05:15 PM.

  27. #27
    PMK
    PMK is offline
    mtbr member
    Reputation: PMK's Avatar
    Join Date
    Oct 2004
    Posts
    2,505
    Guys, the volume vs progression curves are nice. Any chance of having this give real data via an interface with a linkage program.

    This would allow real world net force on the chassis at instantaneous points along the suspension movement. It would also give the true progression of the system design, not solely the damper.

    BTW, don't forget the last portion on many dampers stroke is a spike then mechanical bottom, this is based on the contact with the bottoming cushion, whatever it may be. This being some form of elastomeric will ramp much quicker than the air pressure.

    If you plan to build a dyno, consider that for off-road aplications, you should have 100 IPS for any appreciable data. 150 IPS is obviously much better. 300 IPS will cover the spectrum for almost anything people will ride over.

    PK
    Reps! We don't need no stickin' reps!

  28. #28
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    I could be wrong but testing at those sorts of shaft speeds for the average guy with a garage and the will to build something seem totally unrealistic.

    If we had to apply 5280 newtons (2220N from graph, 2.4x leverage ratio) for a distance of 2.5", 63.5mm.
    Work=F*D, W=5280*.0063M=33.264NM per .025 seconds. or 1330NM/second or 1330 watts. That's almost 2hp without any losses or acceleration taken into account. To accelerate the shaft to 100ips at the start of the travel is going to take a considerably more force. 300IPS is just crazy.

    If you look at something like this: ROEHRIG ENGINEERING INC they are into the right force ballpark but only at 10IPS.

    I'm not really sure what sort of resistance numbers the damper puts out (but I aim to experiment and find out) but in my opinion best case is we measure shock curves and damper/IFP pressures independently.

    Measuring the spring rates in real time isn't going to be realistic either. As has already been pointed out in this thread measuring in real time would be ideal due to the adiabatic process and air having a γ of 1.4 makes the difference between slowly compressing the shock (where slow is anything greater than say .1 second as a guess about how fast the heat might bleed off) but the forces are too large to deal with quickly unless we want to start dropping large weights several feet onto shocks waiting below or instrumenting our bikes with sensors to gather real time data. I've considered jumping from a 5 foot platform onto some sort of a mechanism that transfers the force to the shock, but it just doesnt seem that repeatable or scientific.... We can measure the spring curves over a longer period of time and still get a good sense of how the shock curve looks in comparison to others. Perhaps someone can cook up a somewhat generic correction algorithm to apply to the data collected.

    For now I can slowly measure shock curves. I plan to move on to measuring dampers/IFP without the air or coil spring in place as the next step.

  29. #29
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    Here's an rp2 without air can. Shows IFP force and bump-stop ramp up. Compression done slowly enough that damping should not come into play in any meaningful way (guessing/hoping).


  30. #30
    PMK
    PMK is offline
    mtbr member
    Reputation: PMK's Avatar
    Join Date
    Oct 2004
    Posts
    2,505
    The reason for high IPS is these are shaft velocities seen when riding. As you mentioned, Roehrig has a series of dynos. They also discuss typical applications. Off-road requires higher IPS unlike smooth terrain vehicles that are seldom upset except for broken edges of pavement.

    The last graph is nice, it appears from the 0/0 datum, there is a slight take up of either clearance in the mechanical connection, or this damper is in need of being serviced and has air emulsified into the oil chamber. The IFP ramp up / rod pressure is predictablewith a subtle rising rate from volume change. @ displacement 1.9 is, it appears the elastomeric bottoming cushion (o ring) comes into contact with the sealhead. It offers the spike and slight rising rate until running off almost vertical.

    Again, looks real good and offers the visual representation that people sometimes have a difficult time grasping via the printed media.

    My mention of overlaying your pressure plot into the actual bikes suspension mechanics was not to complicate this. There is a linkage program available, if they have not already added in these type of pressure plots, possibly you could work with them. This would offer the end user the ability to see the true net curve of the entire system.

    It would also allow bikes wth adjustble linkage points to be visually optimized, or any machine to show net effect anywhere in the wheels travel, the changes made by clipping in a volume reduction spacer.

    As mentioned, good job.

    PK
    Reps! We don't need no stickin' reps!

  31. #31
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    Some quick math:

    Chart based on a 2" stroke shock accelerating for 1" of travel to reach peak speed, then decelerating for 1" to stop.

    Soo... as you can see achieving those sorts of accelerations in a lab envrionment are going to be EXTREMELY challenging/unrealistic. Especially since those are just acceleration number's and no damping forces or other losses have been taken into consideration. Just getting a linear actuator to accelerate that quickly is going to be a big challenge.

    If I end up with anything that works at greater than 20IPS I think we'll be doing great. Something that gets past the low speed stack and into the high speed effects so that we can show the characteristics of various shocks and tunes relative to each other. My goal isn't to know exactly what is going to happen with every shock, but instead to create a database of info on various shocks to help people understand the differences in damping tune and spring curve so they can make informed decisions.

    For faster speeds I'd recommend on bike data acquisition, the easiest way to get to 300IPS is probably to just go huck yourself off a 30 foot cliff.

  32. #32
    mtbr member
    Reputation: Vrock's Avatar
    Join Date
    Jan 2004
    Posts
    427
    Are you crazy??????? Dyno tests never go higher than 50 IPS and they cycle only a part of the travel not all of it.

  33. #33
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    I'm assuming that was directed at PMK as he's the one advocating for 300IPS. I'm also trying to make the point that he is crazy.

  34. #34
    PMK
    PMK is offline
    mtbr member
    Reputation: PMK's Avatar
    Join Date
    Oct 2004
    Posts
    2,505
    Crazy or not, those are the numbers that count when looking at HSC values.

    20 IPS is still well on the LSC of the adjuster and the large diameter shims / crossover if installed.

    20 IPS factoring in suspension linkage is equivalent to around 60 IPS wheel movement.

    Let's see, hypothetically, a bike with 150mm at each end has to get across a 3" diameter log, rock, or root with virtually no effort to unload the wheels. It was a bad line poorly executed. It's a slow trail but slightly downhill so minimal pedaling is involved, let's say, 20 MPH.

    Obviously some of the impact will be felt in the bars, so we are not expecting a true plush experience.

    You guys are smart, where does this fall in regards to inches per second of wheel movement to absorb the hit?

    I'm lazy so I grabbed this from the internet

    12 inches in a foot
    5280 feet in a mile

    12*5280 inches in a mile

    3600 second in an hour

    1mph = 12*5280/3600 = 17.6 inches per second

    1 inch per second = 1/17.6 = 0.056818 mph (approx)


    I could be and probably am wrong, but if the bike is moving at 352 inches per second and has a wheel displaced 3 inches, the wheel must move upward pretty fast.

    So to base the impact on say 10" before the log and 10" after the log.

    For easy numbers, lets just use the bikes speed as 352 IPS, the span of the impact is 20", so in 1/17.6 of a second, 3" of impact is absorbed. 3 x 17.6= 52.8 IPS for a minor trail feature at medium speed.

    This is felt in the front and rear suspension.

    If the forward speed increases, the feature becomes larger, or downward force is added in either from weight shift, landing or repeated whoops / brake hack, the numbers get absurd high.

    When I shopped for a dyno years ago, it became obvious that lower IPS are more for smooth ground. Roehrig mentions in some of it's literature recommended uses. 20 IPS is still often good for asphalt stock cars and other smooth surface tracks. Their neatest dyno, which I posted in action a few months ago, has stupid high IPS and cost, but is what is needed to see how things are working.

    Yes, there are some shops that will make multipliers for these low force bicycle dampers, but still, big numbers are better but often cost prohibitive. As Roehrig says, some data is better than no data.

    All the best with this, it is good stuff.

    PK
    Reps! We don't need no stickin' reps!

  35. #35
    PMK
    PMK is offline
    mtbr member
    Reputation: PMK's Avatar
    Join Date
    Oct 2004
    Posts
    2,505
    FWIW, a good watch, and understanding

    Mentioned in the video, 8 meters / second (8 MPS = 314.96 IPS)

    Dyno Run Part 2 | Facebook

    PK
    Reps! We don't need no stickin' reps!

  36. #36
    Save Jesus
    Reputation: beanbag's Avatar
    Join Date
    Nov 2005
    Posts
    2,731
    tl;dr

    Someday when I am not so busy toiling I can come back and give comments.
    In short, to get spring curves with a load cell/dynometer device, you should slowly move the shock back and forth in tiny increments while gradually moving thru the entire range. For example, x(0-1)=.001 sin(t)+.01t. This lets u separate out the spring rate part from seal drag components. I was gonna make a post about this a long time ago, but figured that nobody on MTBR cares about the spring curves for an older fork anyway. Plus I don't have my matlab data on this computer anyway.

    The spring curve will be a combination of a whole bunch of different air spaces here and there, so for all intents and purposes you may as well do a polynomial fit, and subtract this fit from your dynometer data to get the true damping curves.

  37. #37
    Save Jesus
    Reputation: beanbag's Avatar
    Join Date
    Nov 2005
    Posts
    2,731
    Also, as to dyno IPS, I am of the opinion that you only need about one order of magnitude above the characteristic knee speed, i.e. in the transition between high speed and lo speed damping. Often, this is set at the frequency of the sprung rider weight, so figure about 1 Hz x suspension travel, so a few ips at most. Most valving is pretty predictable - low speed progressive bleed, high speed linear shim stack, very high speed progressive piston port flow limit. You can then extrapolate above your dyno's max speed. So worst case the shock gets progressive again at very high speeds, but what were you gonna do about it anyway, make a new piston? LOLOLOLOL

  38. #38
    PMK
    PMK is offline
    mtbr member
    Reputation: PMK's Avatar
    Join Date
    Oct 2004
    Posts
    2,505
    Quote Originally Posted by beanbag View Post
    Also, as to dyno IPS, I am of the opinion that you only need about one order of magnitude above the characteristic knee speed, i.e. in the transition between high speed and lo speed damping. Often, this is set at the frequency of the sprung rider weight, so figure about 1 Hz x suspension travel, so a few ips at most. Most valving is pretty predictable - low speed progressive bleed, high speed linear shim stack, very high speed progressive piston port flow limit. You can then extrapolate above your dyno's max speed. So worst case the shock gets progressive again at very high speeds, but what were you gonna do about it anyway, make a new piston? LOLOLOLOL
    A new piston, oh my...

    Since he is into calculating rear shock / damper info, and is working with his RP2, the digressive piston settings will probably show with fairly low IPS. The RP2 is pretty simple, and no frills design. If this same air spring data is carried over to say a DHX, or possibly a newer RP23 with boost style valves, the damping curve will be changing relative to damper shaft position.

    Regardless this is still a good topic, fun and entertaining. Somehow though it still always come down to sit on it and ride it to get it the best it can be.

    PK
    Reps! We don't need no stickin' reps!

  39. #39
    mtbr member
    Reputation: Vrock's Avatar
    Join Date
    Jan 2004
    Posts
    427
    Quote Originally Posted by dberndt View Post
    I'm assuming that was directed at PMK as he's the one advocating for 300IPS. I'm also trying to make the point that he is crazy.
    Yes, my comment was directed to PMK

    Quote Originally Posted by PMK View Post
    FWIW, a good watch, and understanding

    Mentioned in the video, 8 meters / second (8 MPS = 314.96 IPS)

    Dyno Run Part 2 | Facebook

    PK
    That's a fork, not a shok, and even there the is no need to that hight.

    Quote Originally Posted by beanbag View Post
    Also, as to dyno IPS, I am of the opinion that you only need about one order of magnitude above the characteristic knee speed, i.e. in the transition between high speed and lo speed damping. Often, this is set at the frequency of the sprung rider weight, so figure about 1 Hz x suspension travel, so a few ips at most. Most valving is pretty predictable - low speed progressive bleed, high speed linear shim stack, very high speed progressive piston port flow limit. You can then extrapolate above your dyno's max speed. So worst case the shock gets progressive again at very high speeds, but what were you gonna do about it anyway, make a new piston? LOLOLOLOL
    Totally agree with you, The LS-HS transition is happening at 1-2 IPS, so a 10 IPS Test is fine, 20 IPS is nice and 50 IPS is Perfect.

  40. #40
    PMK
    PMK is offline
    mtbr member
    Reputation: PMK's Avatar
    Join Date
    Oct 2004
    Posts
    2,505
    Quote Originally Posted by Vrock View Post
    Yes, my comment was directed to PMK



    That's a fork, not a shok, and even there the is no need to that hight.



    Totally agree with you, The LS-HS transition is happening at 1-2 IPS, so a 10 IPS Test is fine, 20 IPS is nice and 50 IPS is Perfect.
    Yes that is a fork, wheel to damper rate is 1:1. Linkage systems will lower the IPS if there is sufficient HP to drive the shaft,

    Rest assured crazy is not the worst I have been called.

    VRock, when this topic first started, your linkage program,

    Linkage Design

    and the ability to interface with it immediately came to mind.

    For a rear damper to be analyzed as this, with no corrections for linkage, regardless of who makes the dyno, or the speed it travels, they have still pretty much done no different than sticking a fork in a dyno.

    The information is good, but it is still read as a 1:1 rate. As far as I know, there are no linkage or rear suspension designs that can follow a true straight line graph on account of everything moving in an arc.

    Again, this is all good info. I am not knocking, condemning, or dissing anyone for gathering the data. It is all good.

    PK
    Reps! We don't need no stickin' reps!

  41. #41
    mtbr member
    Reputation: Vrock's Avatar
    Join Date
    Jan 2004
    Posts
    427
    The program is not mine... Anyway, the variation of the Leverage Ratio is the easier part of the equation, there is no need to worry about it now. Gathering information about spring curves and damper forces is always very interesting.

  42. #42
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    1,412
    I've updated my model having taken my rp23 apart and made a few measurements. The big difference is that the bleed port is 7.5 mm into the stroke. This makes a big difference and is what I would have expected to keep the
    preload sane with the mismatch in cross section areas for the positive and negative chambers.

    The graphs earlier in the thread have been updated. I'm showing the difference between a Boyle's law model and adiabatic.

    Happy to supply data to any linkage program. Happy to make my spreadsheet model available generally.

    Sent from my GT-I9100 using Tapatalk

  43. #43
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    Quote Originally Posted by PMK View Post
    A new piston, oh my...
    If this same air spring data is carried over to say a DHX, or possibly a newer RP23 with boost style valves, the damping curve will be changing relative to damper shaft position.


    PK
    I'm not sure why we would carry over air spring data from an rp2 to dhx etc. We would re-measure or re-calculate these. To the second point the sentence seems to be making, I'm not sure why you would expect the damping curve to be changing on the RP23 or DHX air relative to shaft position. Sure there are some gas forces but I believe what you're talking about is the boost valve which is a function of the air spring and has nothing to do with damping as far as I know.

  44. #44
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    I don't have a copy of linkage professional which seems to be the only way to mess with shock curves. Can anyone post a sample of the sort of data linkage wants for input? I can't imagine it'd be hard to create...

  45. #45
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    1,412
    Quote Originally Posted by dberndt View Post
    I don't have a copy of linkage professional which seems to be the only way to mess with shock curves. Can anyone post a sample of the sort of data linkage wants for input? I can't imagine it'd be hard to create...
    For the spring data, you can model just by dividing your forces by the leverage ratio and the spring rate by the square of the leverage ratio. By doing this you end up with the wheel rate (which gets it down to what you feel as a rider anyway.

    As a start I intend to model these curves. The dw-link and vpp are evidently generalised and I ride a Yeti SB-66. The linearity in the SB-66 is something of a problem for shock selection:


  46. #46
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    Hmmm, that doesnt seem to answer the question of how do we produce data for those who wish to utilize it in linkage (Is it a csv file? what does it look like? maximum entries? etc.) Are we as participants in this thread going to aim to make the data generated and collected useful to the average Joe?

    Put it all in a format accessible via linkage or other apps, allow people to view data data/formulas and generate graphs themselves? Keeping it all in our heads, all relative to a particular frame and looking at wheel rates seems counter-productive on the whole and isn't there I want to go.

    I'd prefer to advance the state of general internet shock ramblings and by gathering a sufficiently large sample size to share the data about spring rates for various manufactures and shock sizes as well as damper forces. Put it all in a thread, put it all in a database, whatever works. There are only X shocks with Y tunes from Z manufactures and if we could get some reasonable samples and see how they differ from each other (or don't) that'd probably be useful information to everyone.

  47. #47
    mtbr member
    Reputation: Vrock's Avatar
    Join Date
    Jan 2004
    Posts
    427
    CSV files are very easy, they look something like this: Travel, Force.

    1 40
    2 80
    3 120
    .
    .
    .
    .

  48. #48
    mtbr member
    Reputation:
    Join Date
    Nov 2007
    Posts
    1,412
    I've taken those curves and plotted sample points and then fitted a polynomial trend line to them to get a continuous curve. It is not the same as linkage but it helps move me ahead.

    I don't know anything about linkage but if there is an interface spec or file format I can probably work towards that. As you say, this shouldn't be difficult and there is lots of potential.

  49. #49
    mtbr member
    Reputation: thuren's Avatar
    Join Date
    Jul 2009
    Posts
    789
    I'm just going to throw a couple quick notes about the shock dyno.

    At 25mph, if the front forks did a perfect job of absorbing a 4" tall rock, they would travel at about 140IPS. A good average rear linkage ratio is about 3:1, so the same rock would give about 46IPS to the rear shock.
    Bend, Oregon

  50. #50
    mtbr member
    Reputation:
    Join Date
    Apr 2010
    Posts
    174
    Put the DHX coil on the spring meter today.



    Going to have to get a bigger load cell and recalibrate I guess. 1598lbs is max force. Load cell is 1000lb capable with a maximum overload of 1500lbs, electronics on the input max at 5V in which happens at 1598. So time ot get a new cell and recalibrate I guess if i want to keep going. Though not sure if there is that much value.

    Personally I ride a Giant reign currently with a 700lb spring for most XC days, it would be nice to get the full plot on that shock without going off the scale, so probably a 2500lb or 5000lb load cell.

    Edit: you'll notice in most of the graphs I post there is a bleed down effect every .125" or so, this is the bleed down that happens with every stroke of the hydraulic ram I'm currently using. I believe some of this is seal drag, some of it is low speed damping forces (just a few lbs or less) and some are possibly the ram losing pressure.
    As the issue is a lot less prevalent when using my RP2 I have to assume the downward spikes are pressure bleeding off from the damping circuit or something settling inside, the shock isn't un-compressing, it stays at the same travel (within .001" anyways) but the forces drop around 10lbs.

    Some of the stair step also have to do with data sample rate as well but I won't get too deep into the electronics/software problems I havent fully sorted out yet, but the affect should be minimal.

Page 1 of 2 12 LastLast

Similar Threads

  1. Help! What GPS data do I need?
    By driftwood in forum GPS, HRM and Bike Computer
    Replies: 5
    Last Post: 02-10-2011, 06:21 PM
  2. Cadence Data?
    By QuenteK25 in forum GPS, HRM and Bike Computer
    Replies: 18
    Last Post: 09-05-2008, 11:36 AM
  3. We need some GPX data.
    By cr_mmurillo in forum Lights DIY - Do It Yourself
    Replies: 18
    Last Post: 11-27-2007, 10:15 PM
  4. GPS Data?
    By SoloRider in forum GPS, HRM and Bike Computer
    Replies: 5
    Last Post: 05-24-2006, 04:09 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •