Rocker Design

[Dustin T]

I am curious, how does everyone design their rockers? How do you set up the cad drawings and measure shock displacement and motion ratios. Do you measure one wheel bump, or parallel wheel travel, and how do you measure shock displacement? Do you use a 3d parametric model or use just a simplified 2d sketch?

[Dustin T]

I am curious, how does everyone design their rockers? How do you set up the cad drawings and measure shock displacement and motion ratios. Do you measure one wheel bump, or parallel wheel travel, and how do you measure shock displacement? Do you use a 3d parametric model or use just a simplified 2d sketch?

[Denny Trimble]

We use our SolidWorks model, and move one wheel (at each end) through its full range of droop and bump travel. We look for the installation ratio average and rising/falling/constant nature of the curve, as well as ARB link installation ratios and linearities.

Here's a video of our '03 design:
AVI (http://students.washington.edu/dennyt/fsae/bellcrank17.avi)

If you use SolidWorks, here's a start to automating it:
http://www.extremedp.com/casestudies.htm

And, originally posted as a "monthly macro" from that site:
download zip (http://students.washington.edu/dennyt//fsae/Excel-VBA-SW.zip)

[Angry Joe]

We used a CAD solid modelling program combined with Mitchell suspension geometry software. The suspension software was simple, but made designing these sort of things easier.

[Jon Huddleston]

Thanks for the automation info Denny! Nice video as well. A few weeks ago I was working on an ackermann model in SW and was wondering if I could import motion data into excel...

[jack]

guess and check!!! http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

[Denny Trimble]

That's why CATIA is so much better than SolidWorks, right WWU guys? http://fsae.com/groupee_common/emoticons/icon_rolleyes.gif

[James Waltman]

I'm not sure but I think that Catia can do the stuff that you have VB and Excel doing. I really don't know anything about it but one of the guys on our Baja team has the Catia kinematics package pretty well figured out now. He can get all kinds of data and graphs from his kinematics assembly.

An animation of the WWU Mini-Baja car. (http://dot.etec.wwu.edu/fsae/HostedPics/Animations/bajaWWU2005.avi)

[MikeWaggoner at UW]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Denny Trimble:
That's why CATIA is so much better than SolidWorks, right WWU guys? http://fsae.com/groupee_common/emoticons/icon_rolleyes.gif <HR></BLOCKQUOTE>

You used to be one of those guys that would flame people on messageboards about whether macs or PC's were better, weren't you?

BTW, Catia will do it. I think Catia has every feature solidworks does, the only reason one might prefer solidworks is because it's so much cheaper, or because you prefer the UI. http://fsae.com/groupee_common/emoticons/icon_razz.gif

[Denny Trimble]

No, it's just more of a backlash against all the Catia hyperbole I hear, compared to the work actually being done with the software by students.

Feel free to prove me wrong, show me installation ratio data for springs and swaybars being automatically entered into an Excel file. I know Catia should be able to do it, but I haven't seen it yet. And, if you're designing a formula car, you should already be doing it, right?

SolidWorks can also do this through the Cosmos/Motion addin, but it's more work than it's worth in my opinion. We've customized our spreadsheet macro a lot, and it's very easy to set up and iterate now. Just add a controlling dimension for the hub height, and "watch" dimensions for shock and ARB displacements, specify min/max travel and number of steps, and off you go.

I guess I'm a little grumpy on the first day of the quarter... http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Buckingham]

In ADAMS/Car it is as simple as running a vertical wheel travel simulation and plotting damper displacement vs wheel displacement. The trace tool will tell you the slope (motion ratio) at as many points as your simulation ran.

......Did I just use the words ADAMS and simple in the same sentance.....hmmmm.....I guess my head must still be cloudy from New Year's....

[Mark Bacchetti]

dhaidinger-

I'm running that same simulation in ADAMS/Car... however my total damper displacement is 1.1 inch smaller than my physical measurments on the car (1.3" on A/Car versus 2.4" physical/solidworks measurements).

Is there a trick to getting the damper displacement to come out correctly?

[Buckingham]

Mark B-

Is there a trick to getting the damper displacement to come out correctly?

In that same simulation is the wheel traveling the correct amount?

You should be able to better diagnose the problem, or at least come up with more symptoms, by watching the animation of the event. For example, does your bellcrank ever swing wildly out of plane? Does it seem like the control arms are moving off of their bushings?

[Marc Jaxa-Rozen]

The default part attachments in the ADAMS template are apparently set up for a pullrod (but it models a pushrod, go figure...), which makes the rod follow the upper A-arm. Depending on your suspension geometry it might not be obvious- but it will screw up your data if you have a pushrod, so take a close look at the animation.

I think it only affects the front suspension template though.

Marc Jaxa-Rozen
École Polytechnique de Montréal

[Buckingham]

If that is the case, the simple solution is to change the pushrod attachment in your model to the lower control arm http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Mark Bacchetti]

Marc-

You are right. The prod link was attached to the UCA. I switched the part and everything is good now.

THANKS!!!!!!

-Mark
Cal Poly Pomona

[fsae racer]

Denny, nice constraint video, but bending loads on a rod end, as an LBJ, naughty naughty. jk. I gotta hold it down for Inventor, as it will make all the cool constraint videos too, (got some steering ones that will blow ur lil bump model away, lol http://fsae.com/groupee_common/emoticons/icon_wink.gif) eventhough i was never able to actually figure out how to get any kinematic analysis out of it.

[Mike T.]

That same design had been used on our previous 3 cars as well, and was a disaster waiting to happen. We put many more hours on the 2003 car than the previous 3, and ended up with this as a result:

Busted Suspension (http://students.washington.edu/dennyt/fsae/a-arm_failure/)

However, those pictures can be a bit misleading, as it was actually a weld that failed, due to bad design, rather than the rod end. It was repaired, and type of design will never see a new car again at the UW. There was a thread about this when it happened:

Fatigue Testing (http://fsae.com/eve/forums?a=tpc&s=763607348&f=125607348&m=7226097154&r=7226097154#7226097154)

I find that the Excel/Macro route is very convient, and for such a simple type of analysis, I agree with Denny that it isn't worth the time to deal with CosmosMotion to get a simple set of data. Takes a few hours at most to get a shock linkage model and macro set up, and from there takes no time at all to make a quick change and rerun the analysis. We set up a run log as well, so that all critical dimensions along with the travel data for each iteration made could be tracked automatically within the excel file. I'd recommend integrating this type of feature if you are going to go this route, since it's relatively easy to do and you can't make a mistake or forget to save some data. Also, it allows you to go back to a previous setup if you end up getting further from your goal after a few iterations. So basically, for each iteration it is only necessary to change a few dimensions in SolidWorks, hit a button in excel, and examine the graphs and data returned to excel. In determining our IRs for the shocks and ARBs this year we were able to get through a ridiculous amount of iterations and compare a few different layouts, to acheive our targets almost exatcly, in a very short period of time.

Mike T.

[ben]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Marc Jaxa-Rozen:
The default part attachments in the ADAMS template are apparently set up for a pullrod (but it models a pushrod, go figure...), which makes the rod follow the upper A-arm. Depending on your suspension geometry it might not be obvious- but it will screw up your data if you have a pushrod, so take a close look at the animation.

I think it only affects the front suspension template though.

Marc Jaxa-Rozen
École Polytechnique de Montréal <HR></BLOCKQUOTE>

I fell into this trap as well. What they've actually done is used something called a switch part that in theory allows you to toggle between push or pull rod without changing the template.

I could never get it to work that well and just wrote my own templates. It would be fairly easy to just remove the switch part and modify the standard template if you're not confident enough with ADAMS to start from scratch.

Ben