It's too darned chilly to spend a ton of time in the garage working on the floating saddle project (beginning with extracting the old bushing), so I've decided to spend the cold snap designing a low-profile air filter solution for the Super E carb I've mentioned earlier in this thread (I think; it could've been in another one).

The goal is reducing, if not entirely eliminating the irritating right knee being skewed away from the tank. I know lots of people don't mind this, but I dislike it.

I was thinking of making an adapter that would let me attach the Offenhauser cast aluminum plenum which I found is still manufactured for relatively little money, and isn't gawdawful looking. But I really am not a fan of the forward-facing, conical style filter I'd have to use. Sure, it'd get my knee in closer to the tank and retain great flow, but the aesthetics of a simple, round filter of some sort better compliments the lines of the V-twin engine and of the bike overall, IMHO.

So, I've decided to make up a CNC-machined prototype from 6061-T6 bar, and which will take a K&N universal 'lid' filter--either both pieces or simply the lid (see 1st pic). The pictured filter is a little over 6" in diameter, which will look fine to my eye. I'm leaning toward the latter 'lid only' option, for at least these reasons:

First, having an inverted filter that extends inward over the carb sufficiently to dramatically reduce the existing knee interference would mean having to fashion a new enrichener solution. Not an extreme challenge but it would increase the production cost inordinately. (That said, I'll look at that option when mocking things up.) What I have in mind would have a provision to simply move the existing enrichener lever over from the S&S cover--or people could buy one of those aftermarket jobs that just pulls straight up.

Second, while there would be a reduction in filter media area from the existing ring-style filter, it's looking like the element portion of the lid alone would still be more than double the surface area of the carb's mouth, perhaps as much as three times. I'll do the math when the filter gets here, but that's lots of media area.

Third, it'd look really nice. What I have in mind is a machined plate that puts the lid element as close as possible to the Super E's mouth without inhibiting flow significantly. I wish I still had my flow bench but recall that the studies I did when working on the blow-through supercharger project for the A-Series engine's snorkel plumbing. I was looking at so-called 'cobra head' elbows and the like, but of course those are solid, and so you're looking at friction and turbulence (among other things) resulting from a deck [x] millimeters from the carb or throttle body's mouth.

Here, the air will be coming straight into the carb's mouth through the lid--or very nearly straight through, due to that rubber disc in the center. The effect in terms of flow should be fairly close to no filter at all--or as close as one can get given the need for some sort of filtering media. K&N's lightly oiled cotton doesn't offer much restriction per square inch of area, if memory serves...

The rubber disc will be covered on the outside, and its central hole plugged, by a turned aluminum stud, perhaps with a tapered inward probe similar to the existing Super E's cast cover, but at any rate with a 'baby moon' smooth disc on the outside. Should look great and easily polished if I call out the proper finishing pass.

The filter element will be retained by being sandwiched between the machined plate--with a recess matching the one you see in the ring filter's top edge--and a machined ring that would also cover much of that outer rubber ring on the lid. It'd be easy-peasy for people to polish the perimeter cover if they wanted to do so, to basically match the level of finish I achieved by working my rocker covers (2nd and 3rd pics), though most people's rockers will only need hand-rubbing with Simichrome polish. Mine were in deplorable condition. Well, and people with Pans or Knuckles, etc., wouldn't need to do that work at all...

The filter gets here later in the week, so I'm hoping to meet with my next door neighbor machine shop soon thereafter. I'm excited to get that knee in closer to the tank!

 

So, I got the air filter pictured in post no. 271. I have in mind an ultra low-profile, 'straight shot' unit, where the air flows straight through the element and into the carb's mouth. While I am not racing, I'm still curious about the serviceability of the contemplated design.

To start with, the unit will be servicing a carbureted, naturally aspirated, 74 C.I.D. 4-cycle engine, which according to one calculator I found means an appetite of roughly 70 CFM turning 4,000 rpms (at most; I can't imagine I'll redline the old girl very often). An S&S Super E is estimated to flow as much as 135 CFM with manifold and head. Basically, it's much more carburetor than my engine needs.

The element portion of the filter lid alone is 4.25" in diameter, but less 7/8" for the center rubber disc. That yields a breathable surface area of 14.19 sq. in., less .6 sq. in. for the rubber disc, which leaves 13.59 sq. in.

The carb's mouth is roughly 2", so 3.14 sq. in. Therefore, using only the disc portion of the filter would mean a breathable filter area--directly in front of the carb--of roughly 4-1/3 times the area of the carb's mouth. This would be far less restrictive than the short velocity stacks seen online with a piece of wire mesh at the outer end that is only slightly larger than the carb itself.

But to be more scientific about it, an orifice of 3.5" in diameter (the rough net of what we have with the filter disc) can move 3,217 SCFM with a primary pressure of 10 PSIG. This is very different from a pipe of [x] length of the same diameter, which would flow significantly less--but that's the benefit of the 'straight shot' design: we're basically talking about an orifice immediately in front of the carb's mouth.

So, using just the disc, and since we have no change in direction of air flow to factor in, I do not anticipate any measurable reduction in performance from an 'open face' or 'straight shot' unit... but we'll see!

 

Twin motor gulps about 50% of the time so you more properly want something like twice what they calculator calls for.

Also while the exit on the E is 1 7/8, the venturi is only 1 9/16.

It flows about 200 cfm at 10in H2O.

 

Ah, very good - thanks. Well, even at those figures, I should be fine. The only downside of passing so much air through that disc will be that I'll have to clean it more often than the existing ring element in the stock S&S housing. But that's no biggie, as what I have in mind will be three or four recessed Allen screws to release the ring securing the disc element. Should be a 10-minute job to clean, re-oil, and reinstall.

I'll make a drawing mañana and discuss CNC routines with my neighbor. I think the jaws on the CNC lathe may be large enough to hold 6" round stock, and 5.5" would be just sufficient for a minimalist design. But I am not sure there's active drilling added on that machine (yet, though it has that feature), so we may just make it out of flat bar and use the big mill to get everything done in a single setup to save money.

Well, two setups, actually: (1) make and radius the central bore, create the recess to hold the filter element off the deck, drill the through-holes for the three retaining screws into the carb and the float chamber vent, and the threaded holes to secure the element retaining ring; (2) take the bar out of the clamp, bolt it to a fixture, and do the outer profile and finishing routines. The retaining rings could be cut, radiused, and parted off automatically on the big lathe--as could the little stud to fill that hole in the central rubber disc.

 

So, took my drawing and the air cleaner 'lid' over to my machinist neighbor... and he had me sit on the bike. He pointed out the obvious: the circular intake, while compact and neat-looking, is positioned right next to my calf muscle. We hooked up a shop vac to the filter lid, and anytime my pant leg gets within an inch or so... wham! The engine would be starved of air.

Yes, we could machine fore/aft-oriented ribs into the filter-retaining ring to reduce the occurrence--which would also look cool. But I bet it wouldn't make much difference moving that much air: the pant leg would still get sucked against the 'grille' and starve the air flow.

The other option to get the knee against the tank--machining a heavily recessed billet that essentially wraps around the carb--would be really expensive both from the materials and machining time standpoints, and also would involve a 'dog-leg' enricher setup. Very cool but to my mind prohibitively expensive.

So, we're back to the Offenhauser plenum + K&N cone filter setup I had originally envisioned. We mocked it up (I had rolled the bike into the machine shop), worked on a satisfactory drawing with efficient routines (all machining done from one side to save money), and after a big aerospace job he's got to get out, he'll bang out the prototype from 6061-T6 bar.

The mock-up yielded a dramatic difference in knee clearance--which not only means your knee will be able to rest against the tank like on the L/H side; you'll also be able to reach the brake pedal without the existing, pigeon-toe maneuver. For me, anyway, those two benefits are well worth the effort.

It may be after Thanksgiving but soon enough we'll have what, so far as I know, is the lowest-clearance, high-flow S&S Super E air filter setup in the world. I don't absolutely love cone filters, but the Offy plenum actually looks cool against the v-twin, so I think the end result won't be too unattractive. Also, the aesthetic is very unlike those aftermarket units that have a big elbow jutting out to the side. Instead, everything is tucked in and far less obtrusive.

Finally, I may bend up and polish an aluminum or stainless trim piece that mirrors the shape of the cone filter (kind of like a heat shield in design) if I don't like the way the red K&N mesh looks when the bike is viewed from the R/H side. We'll see!

 

I finally located a 2-5/8” inlet air filter that is very compact. It’s a cone shape but with a smaller O.D. at the base and less dramatic difference between the base and cap diameter than K&N and other popular brands. Evidently used on older BMWs and dual carb, air-cooled VWs and early Porsches. It should alleviate my concern with the base of the cone getting too close to the cylinder cooling fins, due to the ultra low profile of the Offenhauser aluminum intake (whose part no. is 2748, available from a handful of vendors).

It has a washable element like the K&N, and has a chrome cap, so it looks decent. Comes with a hose clamp, but I'll likely ditch that for a better-looking stainless T-Bar clamp.

I’ve ordered one of the filters, so it should be here for a test drive when the prototype adapter is done.

 

So, my machinist and I met again today, tweaked the drawing, and settled on a final design for the adapter. The best aspect of the elegantly simple design is it requires just a single setup (in the big mill; there isn't active tooling set up in the CNC lathe), with all machining done from one side. That way, if anyone likes this solution, I can reproduce the adapters for a very reasonable cost.

The Offenhauser casting will be attached to the adapter with stainless acorn nuts (which I and lots of other folks use at the cylinder hold-down flanges and elsewhere). I am leaning toward sealing done by a large diameter, small cross-section rubber O-ring rather than the standard paper gasket (this no. 2748 casting fit the old Carter AFB and WCFBs which came on old Vettes and other V8-powered vehicles - the Felpro number is 2102). We'll have to see how snugly the casting fits the precisely machined part because the opening in the casting is far from round (pretty common with the sort of piece). If the tolerances are close enough to use an O-ring, there will also be identical cross-section, small O-rings between the standoffs and the casting, so in essence the casting will not only be sealed top and bottom but rubber-mounted, if you will. A plus with my old 'paint shaker' engine, which buzzes pretty good above 65 mph.

The adapter can either be affixed to the Super E (using a standard gasket) using the screws it came with, or these cool McMaster Carr stainless button-head screws my machinist told me about that have nylock inserts integral to the threads. So, you don't need to use loctite because the screws are themselves nylock, like the inverse of the nylock nuts we're all familiar with.

 

Those nylock bolts sound interesting; curious how they hold up to the vibrations.

 

Yeah, very cool. Here's the diagram.

However, I have to confess that I chickened out and ordered regular stainless button heads, for three reasons:

1. the stakes are high. If the top one vibrated loose at speed, there goes my fresh engine. What I'm planning on doing is experimenting with them in other locations--like the timing cover--and if they survive, then perhaps use them on the air filter plate;
2. for the same reason, if I start offering a kit and including or recommending them, I'd arguably be liable for resulting damage (or feel responsible even with a binding disclaimer);
3. they're really expensive. I'm trying to keep the cost of conversion down as much as possible. That's reason for sourcing a casting that is commercially available and reasonably priced, and making just the adapter rather than machining an entire package.

 

As much as I enjoy all the cool fasteners from McMaster
I frequently make my own "patch " bolts with this