Most liked posts in thread: SuperTurbo Mini Project

I've done one "tuning day" so far & plan on doing more soon. But it is hard to set up the environmental corrections properly with only a single day of testing.

Also this lets me check how everything's running on a particular day at the track & make quick incremental adjustments if necessary - without screwing something else up in the process.


Someone once said: "There are no stupid questions - only stupid silence."
(can't remember who it was)



.

 

New muffler & next project

Wohoo! My muffler arrived yesterday!

For something that's going under the car that noone will ever see this thing is a piece of *art*. Because it is repackable, you need to be able to take the muffler on & off. So I had them weld two tabs onto a metal band clamp. I can then bolt this onto braces attached to the car:



I went with the perforated steel mesh option. It's the most resistant to heat:



A good excuse to use the new lift....


And just FYI here's a sneak peak at my next project:






.

 

It is a Coast muffler & it is re-packable (see the small screws at one end). But it isn't their new "two stage" model. I'm not really aiming for huge muffling out of this...

 

Re-uploading to the Motec just takes a couple of seconds. Having a real USB link to the computer makes a *huge* difference. More importantly I can get live data from it continuously.

That reminds me I need to take a picture of my new in-car laptop mount...

 

Actually it looks like this (forgive the grainly low-light image):



It's positioned such that I can get to the keyboard even while strapped in, yet it doesn't block any of the center controls.

The mount is light, but sturdy. It bolts to an attachment I made on the right downtube (with a steel backing). And it's designed so that I can take it out in about 10 seconds if I want to run with a passenger.

 

Valve Cover Mod

So moving the coils definitely fixed the issue of melting of wires. And it helped reduce the melting on the valve cover considerably. But there's still a small amount of melting regardless. I'm sure I could run a couple of weekends before it gets too bad, but the risk vs. reward is considerable.

So I've decided to solve the problem the right way, i.e. install an Aluminum valve cover. Since there are no Aluminum valve covers for Minis (that I can find), I'm having to design one myself. Because the shape varies in all dimensions, it's best to have it machined out of a solid block vs. welded. Below are some pictures of my in-progress design.

The trickiest part of the design is the baffles for the PCV. I could leave the space & try to fabricate my own baffle system, but that would be complicated & potentially dangerous (if something came loose & fell down). Instead, I've lowered the top of the cover to leave room for an external PCV system attached through a hole. This considerably simplifies the overall design & gives me even more airflow back to the turbo.

One slight problem: the software I've been using is somewhat limited. Finishing the remaining portions of the design (e.g. rounding the remaining edges, cutting the groove for the seal, cleaning up some of the corners & faces) would be extremely tedious & error-prone. So I'm going to try switching to a different tool to do the rest.





The x-ray render mode is very useful for seeing flaws in the walls and intersections:



That's my new desktop background image.


As always, comments & inputs are welcome!


.

 

An experiment

Here's a link to an interactive 3D version of the model. You'll have to install the plugin - it's from Dassaut Systems (makers of some of the top MCAD software).

http://www.3dvia.com/models/03E62A390B1D2F01

 

I considered doing a casting from a pattern, but decided against it for various reasons:

1. I knew the design would have to be modified no matter what, so I couldn't use the original as an exact model.

2. Given the shape, producing a good physical pattern would take a very large amount of effort.

3. Also given the shape I was slightly worried (perhaps incorrectly) about the potential of leaks from imperfections in the moulded part & mating surfaces.

One alternative I thought of for producing the pattern was to use software to stich photos of the cover into a 3d model & then print the the prototype from that. But that ended up being even more work due to cleanup needed in the model & moulds.

In the end this was really the quickest & easiest approach. I'll probably spend a total of 10 hours on the model - most of that because I'm learning the software at the same time. I can then give it to a shop with a CNC machine & they can quickly produce a clean, precise part. Moreover it's easily repeatable if I need another one in the future.

 

Err, I don't quite have enough projects like this to justify the expense *yet*....