Physics spreadsheet - measuring suspension

Discussion in 'Car Modding' started by thoraxe, May 3, 2020.

  1. thoraxe

    thoraxe Registered

    Joined:
    Dec 25, 2016
    Messages:
    149
    Likes Received:
    15
    Was just about to email the shock manufacturer and then realized I didn't understand some of the damper sheet details.

    * Base -- this would be the base force when the damper adjustment is at the softest setting, and then each adjustment level would theoretically increase the force by the "per setting" value?

    So, if the base was 100 N/m/s and the "per setting" was 20, that means that at a damper setting of "0" in the simulator, the force would be 100, and at a setting of 10 the force would be 300? (ballparking)

    * F Threshold - I assume this is the "force threshold" and is the speed at which the damper "switches" from the slow force to the fast force?

    * Compressed length -- the comment says "of damper unit", but does that literally mean from the bottom mounting point to the top mounting point? Or is this really asking about shaft length, or something else? In the case of my shock, there is an eye mount at the bottom that attaches to the control arm, and then there is a plate mount at the top that attaches to the chassis. Would "compressed length" be the measurement from the bottom eye to the chassis mount assuming the bumpstop was removed?

    * Extended length -- kind of the same question: would this be the measurement from the bottom mounting point to the top mounting plate?

    There's a "black line" under the "export HDV" button and then there are a couple of other values (piston speed max, piston speed step size, etc).

    Max seems obvious, as does step size (how many points on the graph). The next one appears to be a multiplier that I don't understand. The offsets seem clear enough.

    Thanks!
     
  2. lordpantsington

    lordpantsington Registered

    Joined:
    Oct 5, 2010
    Messages:
    849
    Likes Received:
    79
    Your understanding of "Base" is correct, but whether it is the softest/hardest depends on your damper manufacturer and the sign on the per setting value. It is possible that as your clicks go up the rate decreases. (In which case you would want Base high and negative stepping per value.)

    F Threshold understanding is also correct. This is often commonly referred to as the kneepoint.

    Compressed and Extended lengths: Yes, the spreadsheet is asking for the 'center of mount to center of mount measure'. It would be easiest to understand this in terms of spherical mounts as the center to center measure is simply a point to a point. An eye (I'm thinking rubber surrounded metal shaft that gets bolted, is basically a limited motion spherical joint. Your plate is likely the same. Just estimate the locations of the center pivots. This is strictly just damper free motion.

    FWIW compression stroke limit has never been defined*. If enough force is being transferred you can compress your springs and damper through what is IRL possible, the limitation on compression comes from the rising rate of the bumpstop. I once assumed the same for extension, but believe one of the patch notes stated this was changed.

    *this is 'old style' info, I've not looked further into the chassis.ini to see if you can limit motion.

    The stuff under the export button is just data generation to be put into the graphs. lbs/in/s looks to just convert units.
     
  3. thoraxe

    thoraxe Registered

    Joined:
    Dec 25, 2016
    Messages:
    149
    Likes Received:
    15
    I'm struggling to understand the shock dyno data that I've received. The behavior of the dampers doesn't appear to be linear at all. The increase in damper force appears to increase as the adjustment is increased. In other words, the "stiffer" you make the damping, the more force increase you get.

    I can share these PDFs, but I can't share the original shock dyno files because they have the valving information which is proprietary to the company (FIGS Engineering) that specified it.

    https://drive.google.com/file/d/1RQnfZS_fmS6I3jf-cj5Rpxc8nfMiuAXG/view?usp=sharing, https://drive.google.com/file/d/1UauyADXgkkP1W46GxZJuYHjuzPus6pgv/view?usp=sharing, https://drive.google.com/file/d/1XsEZZJoA0w1mdI0X58M0ouxrAhXIS1cy/view?usp=sharing, https://drive.google.com/file/d/1hEfIqv9DzYBv1-N94PsYfJFS4ypAZZuq/view?usp=sharing

    These are damper sweeps of the front and rear shocks with compression and rebound, respectively. The other is held constant. For example, if compression is swept, rebound is held constant.

    I can re-export different data from the original test files if anyone has suggestions, but I'm really having a difficult time wrapping my head around how to make these work in-game.
     
  4. lordpantsington

    lordpantsington Registered

    Joined:
    Oct 5, 2010
    Messages:
    849
    Likes Received:
    79
    If you could re-export with Metric units (N and m/s), it is going to help.
    Do you have access to the table of data (force measured) @ specific speeds rather than the graphs generated from the data?

    This may help:
    https://www.meteormotorsport.com/wp-content/uploads/2014/03/Adjustable-Manual.pdf
    Your graphs are Force vs Velocity, and showing curves for acceleration and deceleration. You are going to want to average those (or have the software do it for you?)

    rf wants the rate per setting, [N/m/s] @ Click#.
     
    Last edited: Jun 29, 2020
  5. thoraxe

    thoraxe Registered

    Joined:
    Dec 25, 2016
    Messages:
    149
    Likes Received:
    15
    Here's an Excel spreadsheet (created with Libreoffice) that has one sheet for each sweep. This is the front shock with rebound held fixed and compression damping swept through:
    https://drive.google.com/file/d/1l8_RClMrKMxWWDBLY2gBpkFShkpJtoti/view?usp=sharing

    Yep! see above. I can also re-export/re-swizzle.

    Technically I have 7500 DAs but the manufacturer says the 8700 valving was set close enough that the graphs should be representative. Thanks for digging this up!

    There is a graph of "average force versus velocity". Here is what the shock dyno manufacturer says:

    "Average Force vs. Velocity (F9) - Graphs the average of compression force and rebound forces for a given velocity, with all velocities graphed as positive. Roehrig Engineering does not recommend the use of the graph, as it is a poor representation of the shock data."

    The spreadsheet has numbers and then numbers with A. The A sheets are the output from the "average" graph method. I believe that, for the front shock (where compression is being adjusted) we care about the "CA" column. The shock dyno software help does not indicate what the 2-letter acronyms stand for. I emailed Penske to ask them what Roehrig (the dyno software maker) means with those acronyms. Looking at the graph in the F9 / Average mode, the force value that corresponds to the velocity matches up with the "CA" column. My guess is that's "Compression Average", but who knows.

    If this new spreadsheet gives you any more insight or thoughts, please let me know!
     
  6. lordpantsington

    lordpantsington Registered

    Joined:
    Oct 5, 2010
    Messages:
    849
    Likes Received:
    79
    Ah 7500, that would explain the click and adjustability differences. https://www.penskeshocks.com/wp-content/uploads/2017/03/TECH-MANUAL-PS-7500-DA-08-10.pdf It is great that Penske just puts their manuals out there.

    While what was stated is correct, average graphs do not take hysteresis into consideration, the dampers in rf2 don't take this into consideration. In rf2 you can only specify the slope of the line per setting that is shown on a FvV graph. Average will be somewhere between the open/closed curves, and thus sufficient.

    R=Rebound
    C=Compression
    O=Open
    C=Closed
    A=Average
    *O&C I was previously referring to as accel/decel.

    https://www.penskeshocks.com/wp-content/uploads/2019/01/Dyno-Graph-Exp.pdf

    Where to go from here?
    Find the slope of the line between each data collection point (rise/run). At least you don't have to manually extract data into a spreadsheet or rebuild the graph as a vector graphic. In the past that was the most tedious part of this. 7500DA Dampers are adjusted down from full stiff. There were 17 clicks for Compression, from the previous sheets, 30 for Rebound.

    Not sure if this is needed depending on what options you've set for adjustability, but I would break those rates into High speed and Low speed averages. IIRC some Penske manual stated that <2"/s is low speed. It looks like the graphs 'knee' around 0.050 m/s. This is what you would set BumpStage2 and ReboundStage2 to in the HDV. Note: Rebound should be negative.
    Then figure your increase in rate per click. For the SS provided 17 was at 10963 N/m/s LSC, 7356 HSC; 0 was 10445/6232, So something like this would be your HDV entry:
    SlowBumpRange=(10963, -30.5, 18)
    SlowBumpSetting=0
    FastBumpRange=(7356, -66.1, 18)
    FastBumpSetting=0
    or
    SlowBumpRange=(10444, 30.5, 18)
    SlowBumpSetting=17
    FastBumpRange=(6232, 66.1, 18)
    FastBumpSetting=17

    If you want, I believe you can use special entries to specify exact rates at each click, and what the label is in the garage (so correct -4, -8, nomenclature for Penske shocks).
     
    Last edited: Jul 1, 2020
    Lazza likes this.
  7. thoraxe

    thoraxe Registered

    Joined:
    Dec 25, 2016
    Messages:
    149
    Likes Received:
    15
    Yeah they are no joke!

    Funny you posted a link to this, because the tech at Penske I am emailing never looks at those letters and didn't immediately know what they were. I'll have to send him his own dyno graph explanations! :)

    Yeah, that's part of the difference between the 7500 DAs (my actual units) and the 8700s (the tested units). But since the forces are the same, it's really just doing the click math.

    My particular units have shaft-adjust rebound, not clicks. There are 4 1/2 rotations of adjustment, but they recommend making changes 1/4 rotation at a time. So, one could argue that my rebound has 18 adjustments (4.5 divided by 0.25 = 18). Compression has 40. But that's just adjusting "my" per-click values.

    My 7500 DAs do not have independent high-speed and low-speed adjustment. I believe I'm adjusting the high speed only, but it appears from the graphs that's not _entirely_ true? It's probably close enough, though... And, yes, that 2"/s is the same figure I was quoted. The test graphs I got with my specific shocks show that, but they did not do any sweep tests. Just a fixed performance test.

    I'm lost in your acronyms here and also lost a little bit in the numbers.

    I think you are saying 17 compression (full stiff) was 10963 N/m/s low-speed compression, 7356 N/m/s high-speed compression... but I'm confused why those numbers are inverted? Meaning the low-speed force number is bigger than the high-speed force number? Is that because the difference-per-click on low speed was bigger (higher slope)? I'm also not sure which two values you are comparing to get 10963. And is that because the values are "mm/s" but you are specifying "m/s" (x1000)? You mentioned "slope of the line between each data collection point", but is that the slope of the line of the compression curve for a single adjustment (eg: +17) or is that the slope of the line between +17 @ x mm/s vs +16 @ x/mm/s?

    I'm also not sure why the difference is between your two sets of values. The force numbers are similar but the "per adjustment" numbers are inverted (negative) which I'm assuming is because you are treating one as "start high and decrease as you adjust the setting up from 0-18" and the other is "start low and increase as you adjust the setting up from 0-18"?

    Thanks SO MUCH! Please PM me, I want to send you something for your troubles!
     
  8. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    I'm going to be pedantic and point out you'll want 19 settings here :p
     
  9. lordpantsington

    lordpantsington Registered

    Joined:
    Oct 5, 2010
    Messages:
    849
    Likes Received:
    79
    While your shocks do not have independent adjustment for high and low, as far as I'm aware for rf, the rates always need to be defined for each. If you average the high rate with low stuff it'll throw the numbers off. This is why I split it. You are correct, there seems to be influence on slow even if you are only able to adjust high speed.

    Low Speed Compression
    High Speed Compression.
    Spreadsheet.

    Correct, the numbers are seemingly inverted because you are defining the damper rate/slope of the line segment, at slower speeds the slope is typically higher.

    The rate for each segment from data point to data point was defined as change in Average Force over change in speed. While this isn't exactly the instantaneous rate at each sample point, it is good enough to be in the ball park, and draw a trend from setting to setting.

    10963 N/m/s was the average of those compression rates from cell H57 to H67 (speeds 0 to 50 mm/s)
    6624 N/m/s was the average of those compression rates from cell H68 to H106 (speeds 55 to 250 mm/s)*
    *note that I found a flaw in my high speed averages I was originally only averaging up to 80mm/s. This also changed the High Speed Rate @ setting 0 to 4718 N/m/s. This then changes the stepping to 112.1N/m/s per setting.

    Also correct with x1000. Units on dampers are N/m/s, you can toggle these visible in the Player.json.

    The two sets of numbers are really the same thing. Either you can start with a low rate and add, or start high and subtract. Up to you to decide which approach to take. Those rates have room to wiggle. If you want to work up target rates by theory, I found these to be helpful:
    https://optimumg.com/category/technical-papers/tech-tips/
     

Share This Page