heinzboehmer's 2002 Topaz 6MT Coupe

[heinzboehmer]

Okay, some more progress on the cabin air filter housing.

I was dead set on not adding any material to the firewall plug (only making clerance for the braces), since that part is already kinda tricky to install. Figured any extra features would only make it worse. However, this put me at a point where I was fighting for fractions of a mm on the housing to find a compromise between space for the filters, clearance for the braces, clearance for the wiper linkage and clearance for install/removal.

Eventually, I realized this made up constraint was negatively impacting the usability of these pieces, so I decided to transfer two small features onto the firewall plug. Yes, this means I need to reprint what I already have, but oh well.

Here's my test fit prototypes, attached with the most professional of methods:



In place:



Moving those two bumps to the firewall plug means I have miiiles of clearance for installing the filter housing. The middle tab on the windshield cowl does slightly interfere if you try to just lower the housing on, but tilting it up lines the tab up perfectly with the bore for the middle fastener, so it all works out!




And here's a quick screenshot of the assembly in CAD:



You can see I still need to model in all the sealing surfaces on the filter housing, but the overall geometry of the two pieces is feeling pretty final at this point.

Honestly not thrilled about the amount of sealing surfaces I've had to add, but I guess that's what you get when your design interferes with EVERYTHING.

[0-60motorsports]

Amazing work yet again brother.

[heinzboehmer]

After many iterations, I think I'm finally there.

First off, don't think I ever updated on this, but managed to get the filters I wanted from FCP:



For anyone following along at home, you want the Corteco CC1449, not the Corteco CP1449.

Anyway, printed a last prototype out of some scrap spools I had laying around. Managed to squeeze all parts into one build plate, which was fun to see:



Printed with just 1 wall and basically no infill, so the thing is dumb flimsy, but it works fine!





Couple small tweaks needed, but it's looking great.

Have started printing the PET-CF parts. Will update here when they're all ready.

[Bry5on]

Looking great man! Are those two filters used as just a pair in their OEM application, or are there more of them? Thinking of flow area/pressure drop. I remember how tough packaging was for hepa filters (extra pressure drop for hepa) when we were designing some of the Nio cars.

[heinzboehmer]

Looking great man! Are those two filters used as just a pair in their OEM application, or are there more of them? Thinking of flow area/pressure drop. I remember how tough packaging was for hepa filters (extra pressure drop for hepa) when we were designing some of the Nio cars.

There's actually two sets of filters in the F25 X3. One set is at the fresh air intake in the engine bay and the other set (this one) is behind the glovebox. I'm not positive if these ones are used in series when pulling air from the exterior or if they're only used for filtering recirc air. From videos, it looks like the two glovebox filters live next to each other, not stacked (i.e. they're in parallel), but again, not 100% on this.

I can't feel any difference in flow or hear any difference in blower rpm with and without the filters, but that observation is terribly unscientific, so not sure it's worth anything.

Can you think of a good setup to measure pressure drop in the garage? Would be good to quantity these against the E46 filter, but I was unable to find those specs anywhere.

[Bry5on]

There's actually two sets of filters in the F25 X3. One set is at the fresh air intake in the engine bay and the other set (this one) is behind the glovebox. I'm not positive if these ones are used in series when pulling air from the exterior or if they're only used for filtering recirc air. From videos, it looks like the two glovebox filters live next to each other, not stacked (i.e. they're in parallel), but again, not 100% on this.

I can't feel any difference in flow or hear any difference in blower rpm with and without the filters, but that observation is terribly unscientific, so not sure it's worth anything.

Can you think of a good setup to measure pressure drop in the garage? Would be good to quantity these against the E46 filter, but I was unable to find those specs anywhere.

A manometer through a hole in the filter shroud is probably how I’d do it. Easy enough and cheap.

[heinzboehmer]

A manometer through a hole in the filter shroud is probably how I’d do it. Easy enough and cheap.

Oh, duh. For some reason my mind went directly to building some sort of test setup that could seal to both sides of the filter, pull air through it and get a bunch of telemetry out. Been building a bunch of testbeds at work recently, so might be a bit stuck in that mode still

Appreciate the guidance, as always. Will have to find someone to steal the stock filter housing from for testing, since I cut mine up.

Edit: You know, now that I think of it, I'm gonna have to swap the firewall plug too, so it won't be an apples to apples comparison.

Eh, I guess it's fine. Definitely not building the custom bench testing setup.

[George Hill]

I bet it’ll be fine, it’s not going to be perfectly sealed to truly matter. The inlet into the blower case isn’t much bigger than those filters, and honestly I bet it’s smaller.

and those are recirc air only filters in the F25, the others are fresh air only.

[heinzboehmer]

I bet it’ll be fine, it’s not going to be perfectly sealed to truly matter. The inlet into the blower case isn’t much bigger than those filters, and honestly I bet it’s smaller.

Yeah fair point. It definitely doesn't feel like these are a restriction at all, but I do love me some data.

WestBankM4 said I can borrow his stock housing this weekend, so will get data soon.

and those are recirc air only filters in the F25, the others are fresh air only.

Good to know!

I'm glad Corteco makes this activated carbon version of the filters, even though BMW specs them as standard paper ones. It's like they knew I was gonna need them

[heinzboehmer]

Alright, experimental setup ready.

Got myself a cheapo digital manometer, since I thought it would be a bit more versatile than the fluid ones. I really have no way of checking precision or accuracy, but the readings are very repeatable, so I'm gonna say it's good enough for this test.

Started by sealing the prototype with some random hardware store window seal tape. This is as good as it's gonna get, since the printer has been busy with the PET-CF parts and I have yet to get to the gaskets. Don't think this would hold up to engine bay temps, but that doesn't really matter for a static test with the engine off.



I then drilled a hole in the filter cover and sealed one of the manometer hoses against it.



(Sorry, what was that? No, I didn't drill the first hole in the wrong spot and then had to cover it up. No idea what you're on about...)

Plan is to take all measurements with the hood closed, to replicate real operating conditions. Easy enough to route the hose out through the hood grill and then connect it to the manometer. Reference hose got taped to the windshield to make sure that's the same throughout the tests:



(Ignore the reading for now, I'll get some real data tomorrow.)

I also spent some time messing with the climate control settings to find what resulted in the biggest reading on the meter with all the plastics installed, but no filters. The readings were fairly small even if I covered a large portion of the intake area, so I figured that getting the unrestricted pressure difference as high as possible would give me more reliable data, as I wouldn't be running into the resolution limits of the measurement device as much. This is what I landed on:



Lastly, just want to write something down that was on my mind throughout this setup process:

I don't think the way I've set it up will give accurate data on the pressure drop across each filter. I played around with putting the reference tube under one of the filters, but it was hard to put it in a spot where the readings were repeatable. Also, swapping out the housings means moving the reference, so I just decided to instead take relative measurements with ambient pressure as the reference. This means that my results will really only tell me how the stock and X5 filters behave in relation to one another. However, comparing to stock (i.e. what the system was designed for) is kinda the only thing I care about, so this testing should still result in some valuable data