Small Holes

I remember, not too long ago, there was a post from someone that designed parts with lots of small holes. If I remember correctly, it was something like a plate with 8000, Ø.008 holes. I think the topic was related to SW performance with such features, but I'm not sure.

Does anyone remember who this was, or know of someone that designs parts with small holes regularly? I have got a few questions that I would like to ask off-group. I tried to do a Google search but came up short....

Thanks,

Reply to
Seth Renigar
Loading thread data ...

it was titled 'thousands of holes' from a few months ago.

a poster named iQ had a way to model the holes.

sjm

Reply to
sjm1.pitt

iQ here. yes this was my group.

we do alot of parts files with round features. we have tried doing this in a single body part and kills the computer. we tried in a multi-body part and got the same results, extreamly slow to work upon. here is my latest stab at making this function.

let me describe one situation @ a part level. it is a =D88.0 disk that has about 10000 holes in it. there are 12 groups of patterns arrayed around the part. for this item they were symetrical and identical. now what we did was make one piece of the pie, 1/12 of the disk. we populated the the single location with the pattern. we then arrayed the the body to complete the part and combined them. it did make it workable, not great yet. on a part that had multiple pattern we figured that one could assemble different bodies and then again array the body. kind of like making a pizza with different slices put together. remember to array the body not the features.

the new situation is a Semiconductor Probe Card. it is a =D812.0 disk that has 20000 small diameter pads. in a single part we could not array the patterns around the part, it was impossible. so we drew a single part that was one pad. we put that in an assembly and made the primary pattern, again as an assembly. then we put that assembly into another assembly and arrayed it to complete the total pattern on the card. we then put this assembly into another assembly that had the disk of the PC Board. and we were able to work and detail this item. we tried this in a single part/multi body and the same concept failed so it has to be done assembly to assembly.

the real problem is with how SWx deals with round items. if you look down into a hole it looks round but it is made up of triangles, you can see this in some exports. one triangle and another inverted triangle, also noticable by the faceting of the edges as set in the image quality in tools/options. a simple hole can have 10-20 facets in it, same with round pads or any round features. so when you model 1000 holes you can actually have 20000 facets. i do not have a number where this becomes unmanageable but you will see that, depending on the computer, you will hit a point where you cripple your computer and make the model un-workable.

so another work-around is to not make round objects. if you make them square sided, triangles squares hex, you can make 100k or more holes or round features. now on our probe cards we model the pads as squares to validate design intent and detail the probe card with these small squares. we do one detail that actuall shows the real pad detail. now we can actually do work instead of sitting around wondering if your selection actually does something after a minute or two after we do the selection. =20

hope this can help. iQ

Reply to
iQ

I'll email you Monday.

Reply to
Seth Renigar

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.