Oil capacity analysis

[blien]

Hello,

I am working on a project to calculate the oil capacity of a vehicle axle housing assembly at various inclinations as would be experienced in the field. The level would remain constant in relation to a standpipe that siphons off excess oil as the machine runs.



I am currently doing it by making a solid model of the exterior of the axle using Exterior Copy Geom features of the external surfaces of the various compenents that I merge together and make protrusion-use quilts of.



I then make assembly cuts along a plane representing the top of the oil supply in the axle through both the axle assembly and the solid model of the axle. I then compare the two numbers and get a resulting empty space volume, which should represent the oil present.



I have a couple questions in relation to this... Is there a better way to do this? I tried merges and cutouts, but the file is so huge that it would take forever to regen, even if there were no accuracy conflicts or geom checks. Would absolute accuracy cure this? How do I turn that on?



Also, using the method I have established, is there a way to automate the comparison of the volume values and generate a report? This would allow a person to set up an analysis and leave it to chug all night while you were at home. Otherwise, the assembly cuts take a horribly long time to regenerate. Why is this? Is there a way to speed it up.



I would appreciate any advice anyone has on this. There has to be a better way to study oil capacity than this... I have pretty much the entire range of PTC modules at my disposal, so that should help.

 

[miked]

Make an empty part and put it in the assembly using the coord sys to place it.

Modify the empty part in the assy by creating a copy of the interior surface of the housing.

Make a surface in the part (not the assy) that represents the upper surface of the oil.

Merge this with the interior surface copy in the part and you will have a quilt that represents the volume.

You can then modify the top surface angle to get different results.

Do you think that would help?

Mike

 

[blien]

I had thought of that, but the small bearings and gears that are inside the axles are hard to deal with. Many of them are difficult or impossible to cut-out, as their accuracy is different or they are packaged by the original designer, rather than fully constrained. If this were a hollow housing, I would have done it exactly as you have explained though.



Another consideration I failed to mention is that the engineer who asked me to do this wants this assembly to be as associative as possible. This means it has to work whether or not parts are replaced inside the axle housings.

 

[dougr]

Think miked is referring to surface merges not merges or cutouts - surface merges don't need part (absolute)accuracies to match.



What he suggests will work well - have done it before.



For additional immersed components just do a surface copy of their solid surfaces and create a use quilt cut to subtract their volume.



All this will be fully associative.

 

[Hacks]

This sounds like a perfect application for the Behavioral Modeling Extension (BMX) in Pro. Using analysis features placed at proper positions in the model tree would allow you to easily compare volumns/capacities by viewing the analysis feature parameter in the model tree before and after geometry features are added.



If you can get a hold of someone with BMX experience and explain to them what you are trying to do, I'm sure your find it easier and more powerful than the current method you are employing.



Brad Hacker

ME CAD Admin

Zebra Technologies Corp.

 

[dougr]

BMX will work well once you have your oil modeled.



Presumably you'll determine your oil volume on the level and use BMX to keep this volume constant throughout your angle range(s)...

 

[blien]

Hi Doug,



Actually, the purpose of the study is to determine how much oil will remain in the axles at various attitudes. Remember this axle is constantly being filled and drained of oil. The drain pipe level is what drives my study information. Kinda the opposite of what you mention in your last post.



Nevertheless, I had intended to use BMX once I found a reliable way to figure the oil volume.



My biggest problem is the complexity of this assembly is terrific, and the regens necessary for BMX would take days, even with a fairly loose criteria window to shoot for. Simply changing the angle of the axle and hitting regen can take 2 to 3 hours to rebuild.



I also have the problem of still not being able to build an oil model, as I need to keep it associative, which is not possible using conventional surface copies, or even merges (were they possible in this case), given that various parts may be replaced later on.



The only option I see is to build an exterior axle model, using Ext Copy Geom surfaces, which I cut at the same angle as I cut the actual axle assembly. I then compare the resulting volumes. This would lend itself well to a BMX comparison, if I could somehow speed the regens.



Any ideas on regen speeds?

 

[dougr]

In your oil model make as many features as possible to be read-only.



Pro/E skips read-only features in regens - may help some..