Anybody use SolidWorks?

[84_Carrera]

We're using SurfCam & I've heard of no issues with using SolidWorks files. And trust me, the CNC department knows how to complain when they want to.

[viejopatron]

I love SW. Use ProE and Catia on the job. Catia feels like SW on steroids and airframers like Boeing and Lockheed use it.

I was told years ago that a SW model has "gaps" that made it difficult to run FEA or toolpath generators. I not sure if that is still the case.

[YTNUKLR]

Are you guys engineers by trade?

Finished my project:

[MattKellett]

YTNUKLR

No - Technically I'm a modelmaker/machinist.

I work for a prototype company in the bay area, working on fit check models/parts, mechanical engineering approval parts, drop test, visual models and low run production parts.

I trained as an injection mold maker after leaving high school and then moved into this field, where nearly all the work is now CNC parts. Once in a while I'll use a lathe to turn some parts, but mostly I'm programing and using a CNC milling center.

We also have SLA and SLS rapid prototyping machines and a tooling , casting and painting deptartment.

Matt

[YTNUKLR]

Cool. I am majoring Neuroscience but right now doing a minor (about 1.5 years) in Mech. Engineering. I would love to go into a field doing this kind of stuff.

[YTNUKLR]

Finished transparent engine ...block is just a fixture, everything else is made with 911 specifications...and just a few liberties here and there

[dw1]

Great work.

I'm a research engineer/scientist in medical R&D, I've been using cad since the mid-80's (Autocad, Designcad, Cadkey, Pro/E, SDRC Ideas, ME10/30, Mecanical Desktop, etc, etc.). I've been successful in getting my current company to standardize on SW and for the last several years I've used SW quite a bit, along w. Mastercam, for modeling of biological systems and for prototyping medical devices. We're currently on SW '08 sp4. BTW, I'm also the CAD/CAM/CAE sys. admin. for my organization.

How big are the "pack & go" zip files? If I pm you an e-mail address (or an ftp, if the files are large) can you send them? I've actually been working on the fan, fan housing & intake plumbing in my spare time (and for practice using some of the more advanced features).

[joecrum]

I'm a Pro/E...

Here's today's project (an in-line particle detector for oil lines). It's amazing what you can do with 3-D modeling. Many times, I've received prototype parts without ever creating a 2D drawing. I just email the file to the manufacturer and they convert to their native format and go.

One downfall is that people (sales, marketing, managers) see the images and assume the product is already finished. They often want fancy images for presentations. It's nice to take off the Mechanical Engineer hat once in a while, but when the wrong people see the images, it's hard to convince them that there's been no engineering involved in creating the image.

It's also too easy to create impossible to manufacture shapes.

I love my job, though!!

[RazorRacer]

I use SolidWorks and CosmosWorks. If you need it, I can do mechanical FEA and thermal FEA stress analysis on your SolidWorks parts.

[Foxford]

I have a love/hate relationship with SW. It is a very powerful tool, but if I have to spec. things for overseas on 2-D drawings...well it could be better. For the most part we build all of our tooling (plastic injection molds) from SW generated parasolds files.

FEA for plastics is very hairy, but we do some comparitive analysis with Cosmos.

Very nice modelling work on the flat six! Waitin' now for the animated assembly file showing the thing pumping away.

[MattKellett]

Scott

Looks like it's coming along nicely. We make quite a few medical device prototypes from intersting brain implants for seizure control, to instruments and other devices.

If you'd like I could throw your model onto an SLA machine (when I have some down time) and give you a scale model of your engine design, if you would allow us to keep a sample to show our clients. Anyway, how much further are you going to take this design study? Are you thinking of modelling the block too?

Matt

[YTNUKLR]

Thanks for all the kind words.

I will put together an animation later and post it on youtube.

dw1 & Matt-Sounds like you are where I might be headed. I've been working as an intern doing MRI scans at the Neuroscience institute on campus...medical devices may be in my crystal ball. BTW, I made a zip folder; the entire assembly and all parts is about 3.5MB as a SW ASM file.

Matt-That sounds like a really cool idea! I would be much obliged. I am basically done with this project. I was only required to make seven separate components, and I was above that before making the block. I made the block in order to limit the degrees of freedom of the crankshaft to rotation--the block is strictly a fixture to illustrate how the engine rotating parts work.

I do want to do a cylinder head, however, it has some interesting sweeps and things that would be good practice. That's for a later date-been really busy with math classes lately.

[RazorRacer]

Scott,

It would be cool (no pun intended) to see how heat dissipates through the air cooled fins. It could be as easy as applying a temperature to the inside surface of the cylinder and a heat transfer coefficient to the fins. You could also find out how much power is lost in heat. Also, it could be modeled with and without flowing air to see how hot the inside of the cylinder gets without air flow. Furthermore, you could do a transient model to see how long the engine can run without air flow before the inside of the cylinder gets too hot.

The hard part would be getting the cylinder wall material and aluminum cylinder properties just right. Also, the heat transfer contact gap between the cylinder wall material and aluminum cylinder would be almost impossible to estimate. However, if we know the actual inside wall temperature and average fin temperature with air flowing, the model could be confirmed.

You have a talent with SW. Thanks for sharing.

CT

[hcoles]

Originally Posted by RazorRacer

Scott,

It would be cool (no pun intended) to see how heat dissipates through the air cooled fins. It could be as easy as applying a temperature to the inside surface of the cylinder and a heat transfer coefficient to the fins. You could also find out how much power is lost in heat. Also, it could be modeled with and without flowing air to see how hot the inside of the cylinder gets without air flow. Furthermore, you could do a transient model to see how long the engine can run without air flow before the inside of the cylinder gets too hot.

The hard part would be getting the cylinder wall material and aluminum cylinder properties just right. Also, the heat transfer contact gap between the cylinder wall material and aluminum cylinder would be almost impossible to estimate. However, if we know the actual inside wall temperature and average fin temperature with air flowing, the model could be confirmed.

You have a talent with SW. Thanks for sharing.

CT

Not sure exactly for air cooled engines but an old rule of thumb was ~30% of the fuel heat goes into the water.

I don't think on most of our air cooled P cars there is an interface between the cylinder and liner - it is all one piece.

On this CAD package - wouldn't you want the source surface to be power and not temperature?

I would suggest that an cfd model would be the place to start - the airflow might be interesting - e.g. some cylinders get less air than others.

[RazorRacer]

hcoles,

Good points. I was thinking about the Alusil or Nikasil lining on the cylinder and modeling the contact gap, but it is probably insignificant because as you said, "it is all one piece".

If Scott would send me the cylinder only, we could start with a simple thermal FEA and based on your 30% power we could put 5% power on the cylinder wall and see what happens and how close it is to real world. If that works, we could add the heads and finally the crank assembly. We know how hot the oil is around those parts. Then we could do the transient studies and finally, if all of that works well, we could start with CFD modeling. I think Scott's presentation would have some additional appeal if there was a complete thermal FEA result to add to the presentation.

CT

[mjw]

Scott, those are some nice looking models. I like the subject matter!

I'm a mechanical engineer in the aerospace industry and have been a Catia V5 designer/modeler for the last four years. Before that it was V4, V3, Computervision CADS3 ... all the way back to Gerber IDS in the early 1980's.

Solidworks has a lot of functionality for the money and has many similarities to Catia V5. STEP files are useful for sharing between different systems; translating via STEP files, I'm able to read CAD files from suppliers that create in Unigraphics, Pro-E, Solidworks, etc. It's not perfect but get's you most of the way there.

My V5 models are typically sent off for rapid prototyping using stereolithography and CNC part production. It's pretty slick, especially as the capital costs of the equipment comes down. I'll still always admire the skills of an experienced machinist and what can be done with a traditional lathe and mill.

[YTNUKLR]

CT, hcoles,
Interesting points. I would love to see what we could do in spare time with these models. The project is already turned in and it's all good. We haven't gotten to much FEA and CFD in SolidWorks but I would absolutely love to learn what you can do. Let me know your email and I will send you that Cylinder part file to look at.

Thanks.
Scott

[RazorRacer]

The e-mail seems to be down on PelicanParts. I'll post a list of aluminum properties from which I have to choose.

[jim72911t]

YTNUKLR,
Looks great. While I don't use SW, I do that sort of stuff everyday with Keycreator (formerly Cadkey). When it was time to upgrade our CAD software a few years ago, I decided that I didn't need the parametric modeling that SW offered, and went with Keycreator instead. (I may be regretting that decision now)

FWIW, I'm a manufacturing engineer, and get the opportunity to take sketches made by the owner/designer (and yes, they are sketches), convert them into solid models, run them through Mastercam, design the tooling, and put them on the VMC's for prototyping and eventual production. My job is probably 75% desk work and 25% shop work, and I'll tell you that I like the shop work just as much if not more. Sitting in front of a computer doing solid modeling, quotes via Excel, production schedules, etc., doesn't compare to actually making a part and being able to hold it in your hands.

Good luck,
Jim

[RazorRacer]

Scott,

The model made it so easy to simulate heat transfer. The part already had alloy 6061 so that remained the same.

hcoles,

30% power in heat loss worked fairly close. I ended up using 5/6ths of that amount to get a ballpark power.

All,

The first slide shows steady state air cooled with the assumption of internal wall temperature between 1000-1200'F. Can someone verify that temperature on the internal wall? The heat transfer is 80 W/m2/'K on the fins and walls. 5000 watts is input on the cylinder wall, but only where there are fins on the outside of the wall. At this point the assumption is the other 1000 watts * 6 of heat drains from the engine and heads.

The second slide shows 60 seconds of no air flow or what would happen after 60 seconds if the fan belt broke. The internal wall temperature rises from 1000-1200'F to 1200-1400'F in 60 seconds. This is based on a non-air flowing heat transfer of 20 W/m2/'K.

These are ballpark guesses that could be improved upon with empirical data. Also, if you cannot read the temperature gauge, the bottom number is 750'F (blue) and the top number is 1500'F (red)..