-
The Gear Box is the official blog of Jay Tedeschi, the MSD Sr. Solutions Evangelist. Jay will discuss a wide array of topics, including software interoperability, applicability, and hardware/software developments.
-
Spring... or is it Sprung?
January 6, 2008, 02:02 PM Jay TedeschiA quick one this week… but most definitely a good one. Not sure how many of you realize this, but there are all sorts of neat little timesavers built into Dynamic Simulation that are somewhat less than obvious, unless of course you have time to poke around under the hood… which of course I do. [Grin]Consider the following example. Let’s suppose for the sake of this discussion that you are tasked with determining the optimum spring rate for a set of intake and exhaust valves in a high performance inline 4 cylinder engine. As an Inventor Professional (or Simulation Suite) customer you have at your disposal one of the most comprehensive simulation tools available today. You take your nearly finished cylinder head assembly and begin to work on the problem for the exhaust valve.

Let’s assume that you did not work through the tutorials, and so have not used the Spring/Damper/Jack joint previously…, however you do remember seeing a spring looking thingy in the menu of available joints when browsing the list. A brief look confirms this and you quickly select this Force Joint.

A pair of center point selections to align the spring between the two retainers and your spring joint is in place. Yeah… I know, it looks like hell, but please try to remember, this is nothing more than a graphic representation of the spring joint.

If it will make you feel better, you can change the display properties of the joint, so that it looks more in line with what will end up in the final design… you are going to have to go into this dialog box anyway to set the spring’s physical properties, e.g., spring rate, damping, etc. etc. so to change the display characteristics are only a couple of extra selections. Take a look at the figure below… access this by selecting the spring joint from the browser and then selecting “Properties”. As we are looking to minimize any “chatter” when the valve closes, we initially select a weak spring force and then determine the nature of the valve closure… we can work our way to progressively stiffer springs from there.
Regardless, you will note that I have highlighted two types of input for spring stiffness here, Constant Value, and Input Grapher. Constant Value is exactly what you would assume it to be, the 1.0 N/mm here remains consistent through the bulk of the compression range. However, I wanted to point out the fact that you can use the Input Grapher if you choose to use a spring with variable properties, which would be the case if you were to use a progressive rate spring.
So, here we are, ready to begin our research into the optimum valve spring values for our design. It only took us… what… 5 minutes or so to find the joint, insert it, change its properties and get to the point where we can begin to do some real engineering?Well, there is a quicker way… I know this is somewhat akin to impatiently waiting 30 seconds for a hot dog to cook in the microwave, but I am all in favor of keeping things as simple as possible, and there is a way to solve this same problem with one of our existing joints, and forgo the Spring/Damper/Jack joint altogether. The joint that was used to define the translational motion of the valve itself in the context of the valve guide is a prismatic joint, which adds but a single degree of freedom to our model, and therefore contributes greatly to reducing the workload on the solver, and ultimately gives us better solve times. Select the Prismatic Joint from the browser, right click and then select properties.

A note here about working with Joint Properties… Hopefully you have noticed that in addition to the General tab, we typically have a tab for each translational and rotational degree of freedom for that joint. As I pointed out, this is a Prismatic joint, which has a single degree of freedom, hence the single tab “dof 1 (T)”. Select it and let’s take a quick look at just what we can accomplish with this in lieu of the addition of the Spring Joint. First of all, to allow for any modifications to the joint characteristics we have to enable joint force, shown below.

Note that we can now modify several properties of the joint itself… constant or variable force, Damping, Coefficient of Friction and Spring stiffness and Free Position. As we did with the Spring joint, we set an initially weak constant spring force, and have the option, just as before, to simulate a progressive rate spring as well.

One thing before I wrap up. After you accept these changes to the joint and exit the dialog box, you will note that the display of the joint itself in the browser has changed somewhat. Any time you change any of the joint parameters from the default the joint will be displayed with a green pound sign next to it… a nice little feature that allows you to quickly identify the joints that you modified.
I should also point out that if you are using Automatic Constraint Mapping, which most of you probably are, the only way to change many of the joint properties is to modify the assembly constraints which were translated to give us these joints in the first place.I realize that we have eliminated just a single force joint, but try to look at the big picture… in a very complex model, the reduction of joints and degrees of freedom such as we have shown here can have a major positive effect on our overall solve times, and in the end, that is what it is all about… getting an accurate picture of the performance of our assembly in as short a period of time as possible.
Comments
-
January 21, 2008 07:28 PM simon ng
love the dataset!
-
February 5, 2008 12:09 PM Jay Tedeschi
Thanks Dude... I will post it for you as soon as I get a break.
You must be logged in to post a comment.