Jump to content

Cad/ Designing Competition


NVWOCI WVS

Recommended Posts

What FEA package are you using? I had a student this year analyse the stress in a bicycle wheel using Abaqus - he took it several notches beyond the 2D simulation in 'The Bicycle Wheel' by Jobst Brandt :)... Some boundary condition issues in setting prestress in the spokes though. I built a proMechanica model of the same thing years ago and it worked very well, but won't deal with plastic deformation...

Next year I'll see about modeling the stress relieving process (Plastic deformation in the spoke/hub interface) for another project :).

In this instance only cosmosexpress, we have very little need for a dedicated fea package at work (its all heavy engineering and typically based on emperical designs + theyre too tight to pay for it :D) but its usefull for checking simple brackets and fabrications :)

The joys of having an engineering suite available to mess about for hobbies :D

Link to comment
Share on other sites

Did you really design that?! I thought sewage plants would be pretty standardised things that are all pretty similar. Good job though (Y)

You couldnt be further from the truth :D

Typically on refurb jobs youre designing equipment to fit a 30-70 year old concrete tank, alot are in imperial measurements and nearly every one is different. The concept and design is usually the same but each one is different.

That tank from seafield is one of the largest rotating half bridge scrapers in europe, 55m diameter tank and almost 4 meters deep, tis a lot of shit :D I was working on it for 18 months and Im pretty proud to say that it worked very well and thames water are very happy with it. It sucks sludge from the bottom of the tank and sends it off to the sludge processing plant elsewhere on the site - capable of 72litres per second when cranked up :)

Link to comment
Share on other sites

Here's the model I did ages ago - I only have a printed report from the more recent solid model based work... The circles/squares around the beams show their assumed cross section. The model was only fed an approximate cross section for the rim.

Wheel_Analysis_model.gif

Te stress plot is for the case where the wheel is loaded by the bike + rider weight only. Note the spokes pointing downwards lose stress while the rest stay quite even. This is consistent with other work done in the field. Most people assume that the hub hangs from the upper spokes, when in fact due to the prestress in the spokes before loading, the only significant stress change is for the downward spokes to lose tension (All other spokes gain a small amount of tension - not visible in the result plot as it's much less than the change seen on the downward spoke.

Wheel_Analysis.gif

Link to comment
Share on other sites

The fall off in stress in the spoke pointed downwards is due to the rim being bent flat at the point where it contacts the ground, bringing that point in the rim closer to the hub centre. This local bending in the rim is the reason the downward pointed spoke sees such a huge tension change. The hanging effect such that it is is much less because all the spokes pointing to the upper half of the rim will have to see a stress change in order for the hub to move downward relative to the rim (Without a point force to bend the upper half of the rim locally the change will be massively less than seen by the single lower spoke pointing downwards).

Radially laced wheels will show the effect of stress reduction in the bottom spoke more strongly as only one spoke will point straight at the ground for any given wheel position rather than for tangentially laced wheels where there will be several spokes pointing in about the right direction that share the load. Radial wheels show better lateral stability (Smaller angle between the hub axle and the spoke means the cosine component of the spoke stress is higher). Radially laced wheels are statically indeterminate under disk brake/pedalling loads (Spoke tension will skyrocket as the wheel tries to resist twist between the hub and rim) and will loosen more quickly if the spoke tension is too low as the side forces the spokes see where they cross each other act to resist unscrewing at the spoke nipple (Not sure if this is a very significant effect though).

I haven't time to take spoke by spoke values off this, but if this thread is still running in a couple of days I'll have time to work on it then.

Link to comment
Share on other sites

When I get home from work I'll post some bits up from what I'm working on at the moment. Not bike related, but those into engineering design and analysis might find them interesting :)

EDIT:

Here we go

post-34-1209573634_thumb.jpg

The mesh for a cyclic thermal analysis I'm currently building. The part in question is a piece of the exhaust manifold from a 50litre v16 diesel engine which puts out roughly 2000bhp. This kind of thing is a day-in, day-out occurence for me now.

post-34-1209573667_thumb.jpg

Some old work I was involved with. Had to remodel the block for our 78litre engine after changes were made to the casting molds by our supplier. I can't take credit for the whole model, but sometimes it's easier to start from scratch than work with (in this case) a poor initial model.

Link to comment
Share on other sites

Sweet CAD models - Some of those engines are almost big enough for the average American pickup truck :P...

I presume they're marine or stationary engines? What company are you working for - Perkins are the only company I can think of in the UK that do engines that size?

ProE is my CAD package of choice too. You must be on a pretty serious computer to run models like that (and not spend most of your day waiting for the PC to catch up) though :).

Edited by psycholist
Link to comment
Share on other sites

Here's an easy one to start with (Though maybe not as easy as it looks :) ):

ball.jpg

A C60 Atom - built using the assembly functions, so the program calculates all the angles to make it work:

perspective.jpg

A light bracket for high power LED's for night XCing (Built in my garage):

Front.jpg

A thermocouple mounting into a flow test rig - extra marks for working out how I drew the twisted wires :P...

Thermocouple%20mounting1.jpg

A test rig we built ages ago in the lab - replaced by a simpler one now...

PIV_rig.jpg

And in conclusion I think I must like bright colours :D...

Link to comment
Share on other sites

Sweet CAD models - Some of those engines are almost big enough for the average American pickup truck :P...

I presume they're marine or stationary engines? What company are you working for - Perkins are the only company I can think of in the UK that do engines that size?

ProE is my CAD package of choice too. You must be on a pretty serious computer to run models like that (and not spend most of your day waiting for the PC to catch up) though :).

I work for Cummins. Actually most of the engine go into mining applications, though we do do versions meant for power generation and marine apps. I'm more on the analysis side than the CAD design side (I'm the one that validates the new designs) so ProE is only really used to simplify models and build the necessary assemblies for whatever analysis I'm running. My true weapons of choice are Ansys Workbench, Ansys Classic, and FE Safe.

PC isn't too bad, but it does lack punch when dealing with complete engine schemes (dual 3ghz opterons, 8gb ram and 512mb Nvidia workstation graphics card) and for larger analyses we have a dedicated FEA server (16 processors, 128gb ram) in the building.

Link to comment
Share on other sites

Lofting is way too much work and you wouldn't end up with round wires unless you were using elliptical cross sections or possibly worse... The twisted wires are driven off their centreline as follows:

Look in the help files for the trajpar variable. Basically it varies between 0 and 1 as you move along a trajectory. If you use trajpar with a relation to drive an angle change (Just type something along the lines of ANGLE = TRAJPAR*360*NUMBER_OF_TWISTS into the text box), you can sweep a surface along the previously defined curve (In 3 dimensions - I think I made a spline through a set of points dropped at random for that one) and make it spin around the trajectory as it goes. Then use the edges of the surface as sweep trajectories for round cross section sweeps normal to trajectory and you're there...

Edited by psycholist
Link to comment
Share on other sites

thats lush i would love a frame like that no joke if you could find somewere to make them with the options of your own choice of geo and thickness of the tubing and bracings you would be earning a lot of money

I'm building them, and am currently waiting on delivery of the last few tubes before they hit the shop floor, then onto testing, then into sales. The FEA is being run now.

If I can get the pricing right Adam will sell stock them I hope.

I'm also very into building custom geo frames for people who know the difference. They would obviously be built to order.

toprear75.jpg

rear75.jpg

untitled3.jpg

untitled.jpg

untitled2.jpg

and she's light 1.6-1.7kg...

That's the plan anyway.

There can be rim brake mounts aswell.

post-14732-1209591333_thumb.jpg

Edited by Dang!
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...