heinzboehmer's 2002 Topaz 6MT Coupe

[heinzboehmer]

I was referring to varying the thickness of the 3D printed part. I believe the lower half doesn't have any role in supporting the cabin air housing.

Ah gotcha. Redesigning the 3D printed part is a lot of work for not much benefit. Since you can print (mostly) hollow volumes, a thinner part is not necessarily lighter. Also, thinner would lead to higher thermal conductivity.

[George Hill]

Round 2

[heinzboehmer]

Round 2

Awesome, you test fit the first print just to validate that I didn't get something (e.g. scale) wildly wrong?

[George Hill]

Hard to say for sure overall, but the bolt hole and dowel line up.

[heinzboehmer]

Also, have a plan for the filter.

Looks like I can reposition it so that it will clear the braces and still get somewhat acceptable airflow across it (hopefully).

Here's a view from the back:



Here's how it fits into just the housing:



As you can see, the filter is completely above the fasteners, so that's nice. Means I can keep the locations of those unmodified and then extend a wall up for it to seal against, like so:



Above view also makes it clear that it will interfere with the top locating features on the firewall plug. Those serve to locate both the housing and the filter. I plan to trim them back, so that they don't interfere with the filter, but still serve to locate the housing. I'll add some locating features for the filter to the housing.

You can also see that the filter does not poke through the housing into the engine bay, so no risk of interference with engine bay components. Just need to thin the housing slightly up front.

Since the filter is moving down, the front section will also no longer seal against the cover. My plan is to keep the mounting bosses for that cover where they currently are and just have a wall on the housing that the filter can seal against, like so:



Note that this wall will not extend all the way down to the upper surface of the housing, as that would block all airflow to the filter. I'll add some ribs to make sure the wall doesn't flop around, should be fine. Might also need to add some protrusions to the underside of the cover, so that the filter stays in place.

Now onto airflow.

Unfortunately, I think this change + the clearance needed for the braces will negatively impact airflow across the filter quite a bit, but I don't think there's a good way around that without completely redesigning the housing and potentially using a different filter(s). Regardless, I'll start with this approach and if it really sucks, I'll figure something else out.

Here's a sectioned view to show how much airflow will be affected. Rough eyeball math says that a third or so of the intake area is going to be significantly blocked:



Keep in mind that the space between the braces is essentially useless now. The housing will have to come up to clear the braces (roughly where the pink line is), but I can't just add two notches for each brace, as water would pool in the middle and have nowhere to drain out to. Only saving grace is that the cross section of the big notch will get smaller the closer you move to the firewall, so the blocked filter area is not as much as the above image makes it seem.

Maybe adding a third drain is something I can explore if a bigger redesign is needed?

And lastly, some fit checks against the cowl + a comparison against the stock filter location (in yellow):




These views also show why the filter has to move down. If it doesn't, it'll interfere with the cowl and block the filtered air coming from the front.

So yeah, unfortunately a fairly big compromise in airflow with this bracing. I'll have to see if this design is adequate enough or if I need to figure out a better solution.

[heinzboehmer]

Hard to say for sure overall, but the bolt hole and dowel line up.

That looks pretty close to me! Awesome.

Also looks like the print came out pretty good. Was a bit worried about the smaller features, but doesn't seem like they were an issue.

[George Hill]

Ya, so far so good. I didn't mess with much just cut it into (3) pieces in a way that would fit and provide the most support when printing and then loaded my generic Sunlu PETG and hit print

[George Hill]

Second one finished well, haven't fitted it yet, just jumped into printing the 3rd (middle) section.

[George Hill]

After a couple of failed prints (filament tangled on the spool) we now have (3) pieces... tomorrow we glue!

[heinzboehmer]

Awesome! Excited to see how it fits.

[George Hill]

glued and drying

[George Hill]

It fits!

I don't have a spare seal so I've just got it laid up on the firewall opening as is. This afternoon I'll fully assemble it into Eileen.

[George Hill]

Its in and it fits 99%. The only point of "issue" is the center tab for the microfilter is to high. But because its been glued together and printed in multiple pieces I don't know if this is the design or just the way I assembled it creating this extra clearance. Either way though its in!

[heinzboehmer]

Its in and it fits 99%. The only point of "issue" is the center tab for the microfilter is to high. But because its been glued together and printed in multiple pieces I don't know if this is the design or just the way I assembled it creating this extra clearance. Either way though its in!

Hell yeah, that's awesome! Thanks a bunch for testing.

Does the center tab look perpendicular to the mating surface behind it? It's possible that the print warped a bit.

If it is perpendicular (and you have the time), could you measure the parallel distance between the two highlighted faces? Would be nice if you could grab that measurement from both the printed and stock parts.



CAD says the center tab should be 2 mm lower than the side tabs. That looks reasonable to me based on pictures and the scan, but it's possible that I got that measurement wrong.

[George Hill]

It did look 90* but I didn't spend a bunch of time staring at.

When it cools down outside I'll pull it apart and compare to a stock one.