What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

TOOBUILDER question on plenums ?

L

Larry DeCamp

Well Known Member
There are many comments regarding all the screws required to secure a plenum. Toolbuilder has referenced the use of piano hinge. This is very appealing.
The metal baffle wall component will probably need to curve a little. Is there a better way to curve a hinge than drilling and notching ?
Is there a better way to secure plenum than screws,1/4 turns or piano hinge ?
 
Larry DeCamp said:
There are many comments regarding all the screws required to secure a plenum. Toolbuilder has referenced the use of piano hinge. This is very appealing.
The metal baffle wall component will probably need to curve a little. Is there a better way to curve a hinge than drilling and notching ?
Is there a better way to secure plenum than screws,1/4 turns or piano hinge ?
Click to expand...

I use #6 screws and nut plates spaced a little less than 3". I took a piece of fiberglass and screwed it to a wood block with varied spacing then pressurized the back side to determine the spacing. With a proper screw gun it does not take very long to remove it. It is 10 minutes, not 10 seconds. YMMV
 
Concerning the curved hinges - note that only my rear wall is curved (and only the center 1/3). The front walls are straight (just worked out that way on my airplane)

2i6jf5i.jpg


...but no matter, the hinges lie flat anyway so they will follow a curve without modification. The "L" angles (.016 titanium in this case) on the front and back walls stiffen things up a bunch and provide a rudementary seal surface. The stiffness helps when installing the lid as even the relatively short side walls need a bit of "steering" to keep them in line when the hinge pin is sliding home.

5kle0k.jpg


Here's another view of the straight plenum end wall. And yes, the inlet is rigidly mounted on the wall. The inlet bell sits in front of the cowl and moves with the engine.

ta2am1.jpg
 
Last edited:
Always thought that inlet structure could be the optimum for high Vi/Vo, as the diffuser shape be be fixed/perfect, not disturbed by internal couplers and torque misalignment.
 
Definitely an advantage for internal aerodynamics, ease of fabrication and assembly.

A little bit of motor-mount sag and the inlet rings will conspicuously not match the cowl very well.
 
Steve, if that happens, given the shape of the cowl in the vicinity and aft, would that have an adverse effect?

I see it as having no effect on engine or oil temperatures, assuming the oil cooler gets its air from the plenum, and having almost a zero change in airplane drag. But you'd have a better idea about that than I would.

Dave
 
Engine mount sag can be shimmed as required to maintain alignment (just like most of us do to keep the spinner aligned). The degree to which any misalignment gets offensive depends on the treatment directly behind the rings. If you expect it to fair absolutely perfectly and look like one piece, it's going to be a mess. Treat them like two separate pieces and it gets easier. My cowl was built with much bigger rings in mind so I have a bunch of sanding and rework at the transition, but even in rough form it looks pretty good.

2v2vzlu.jpg


One other benefit to rigid inlets is that you can extend them far forward. Mine are breathtakingly close to the TE of the prop, but wont hit no matter how much the engine shakes around.
 
Mike, we were speaking of misalignment of internal ductwork. It's often seen with plenum systems which try to link a machined aluminum inlet ring with a fiberglass inlet duct using a soft tube of some kind. A James plenum setup would be a common example.

Since the engine twists on the mount in reaction to the prop, even perfectly concentric alignment in the shop won't be concentric in the air.

If low Vi/Vo, the low velocity through the misaligned section should not be a big deal. On the other end of the spectrum, a really good high Vi/Vo inlet is small diameter, with velocity potentially higher than freestream. That duct must lead directly into a diverging diffuser. It's my understanding that bumps, ridges, edges, and too-rapid changes of section will trip the flow in turbulence. Pressure recovery will be poor.

Mike, your Rutan inlet (I dunno, did he invent it?) looks great to me. The nice fat lips should make it insensitive to external separation, the spillage problem, with the caveat that I have absolutely no idea regarding the effect of the surface interruption between ring and cowl.
 
I understood Steve's comment in post #8 was relating to my current "rigid" inlet drooping in relation to the OML of the cowl.

In any case, my current configuration is a bit of a mashup. I started with the idea of the low VI/VO and a cowl flap but changed to smaller inlets and exhaust augmented fixed outlet. I can't expect it will be perfect, but I have room to modify as I go.
 
DanH said:
On the other end of the spectrum, a really good high Vi/Vo inlet is small diameter, with velocity potentially higher than freestream. That duct must lead directly into a diverging diffuser. It's my understanding that bumps, ridges, edges, and too-rapid changes of section will trip the flow in turbulence. Pressure recovery will be poor.
Click to expand...

A lot of time spent on that Mike. Nice work.

I've done a lot of flow studies on ducts using big blowers and alignment grids, guide vanes, wool tufts and boom manometers looking at localized pressure. From these hours spent observing, I'd opine that you'll have massive separation inside a typical RV cowling, especially in the middle where the case is. How much this matters in cooling efficiency and drag is the big question that's hard to answer without careful flight testing.

The large change in section/volume just aft of the inlets just can't be good. Seems to me that the case area should be negated from the plenum and the efforts directed at getting cleaner flow and better pressure recovery over the cylinder and head fins only.

Or perhaps a fairing inside the plenum aft of the spinner to minimize the abrupt change in section and volume. Would be interesting to build a full scale, instrumented test rig complete with scrap engine in place and a couple of 5000 CFM blowers. You could prove or disprove a lot of ideas quickly on the ground.

Would such an effort in time and money find 5 knots or 10, plus better cooling? Maybe or maybe very little gain.

I very interested to see what Mike finds out here.
 
Last edited:
Toobuilder said:
I understood Steve's comment in post #8 was relating to my current "rigid" inlet drooping in relation to the OML of the cowl.
Click to expand...

I stand corrected.

rv6ejguy said:
I've done a lot of flow studies on ducts using big blowers and alignment grids, guide vanes, wool tufts and boom manometers looking at localized pressure. From these hours spent observing, I'd opine that you'll have massive separation inside a typical RV cowling, especially in the middle where the case is.
Click to expand...

Ross, are you referring to a particular inlet configuration?
 
Last edited:
DanH said:
I stand corrected.



Ross, are you referring to a particular inlet configuration?
Click to expand...

I'd surmise that round or rectangular inlets doesn't matter much with regards to downstream turbulence in the plenum. Once the air passes the aft end of the inlet into the massively greater volume, I'd expect huge turbulence despite the flow blockage the fins impose downstream.

Would be interesting to mount a lipstick cam on the outer edge of the inlet and observe wool tufts inside the plenum, in flight. I'd do it on mine except I don't have a Lycoming...
 
Last edited:
I guess the real question remains just how long of a duct do you need to get the velocity low enough to dump it into the plenum without flow separation? And the follow up:. Is a shorter than optimal duct better than nothing?

My sense is that the velocity slows pretty rapidly. In that case you want as much volume as possible so that you don't get localized flows, but a nice even average. Doing a smallish plenum over the heads alone seems counterintuitive to my feeble mind. But I'm willing to stick a camera in there once I'm flying to document it.
 
Last edited:
rv6ejguy said:
I'd surmise that round or rectangular inlets doesn't matter much with regards to downstream turbulence in the plenum. Once the air passes the aft end of the inlet into the massively greater volume, I'd expect huge turbulence despite the flow blockage the fins impose downstream.
Click to expand...

Round or rectangular doesn't matter. Either can be used with external or internal diffusion.

Downstream of the inlet ring (low Vi/Vo) or inlet ring and diffuser (high Vi/Vo), after velocity has been traded from pressure, in-plenum turbulence is good from a heat transfer standpoint. There's a whole lot of exposed fins up there.
 
I can only comment from general trends observed in my testing and the shapes I was using related to ducts with radiators at the other end. The Lycoming setups are quite different shapes but much further from "optimum" in that the inlet so volume vs. length ratios are much more favorable to trigger separation.

Both Russell Sherwood and myself found the transitions had to pretty gradual to avoid separation and efficient turning of the flow required guide vanes. Think how loop type wind tunnels often use guide vanes in each corner to turn the flow with less turbulence and losses. In our cases, we are only turning the flow less than 30 degrees to meet the top edges of the rads. In the Lyc, the flow is turning 90 degrees over a shorter distance while also increasing substantially in cross section/ volume on entry to the plenum.

There is nothing like real world testing however and it will be very interesting to see what you find.
 
Last edited:
Flow between plenum sides ?

Ross,
You bring attention to the ideal of keeping the incoming flow directed over the cylinders. Clint B is developing such plenums like some racer pics I have seen here on VAF.
This raises a question for me with your coil and Don?s flow distributor mounted on top of the case split. How much cooling of these components is required, if any ? Turbulence could be a good thing or unnecessary. I can make either scenario happen when I mold the plenum top.
 
Larry DeCamp said:
Ross,
You bring attention to the ideal of keeping the incoming flow directed over the cylinders. Clint B is developing such plenums like some racer pics I have seen here on VAF.
This raises a question for me with your coil and Don?s flow distributor mounted on top of the case split. How much cooling of these components is required, if any ? Turbulence could be a good thing or unnecessary. I can make either scenario happen when I mold the plenum top.
Click to expand...

It's good to have some airflow past fuel blocks and coil packs.

The theoretical and practical requirements of optimized cooling are sometimes a bit different and always involve other compromises.
 
I question the effectiveness of inlet velocity "blasting" or "aimed at" specific areas of the fins. I think the Lycoming is designed for "top down" flow. In my mind the ideal tractor cooling configuration is divergent ducts that decelerate the flow to minimum velocity at maximum density right as they dump into a huge plenum sitting above the entire engine. This large volume of dense air then flows from the upper deck to the lower deck at the appropriate Delta P. The large volume of stable upper deck pressure means that all properly sized passages (fins and baffles) get ample air. Localized flow management like the "blocker" walls we so often see become less important.

With a huge resevior of fat upper deck air, cooling becomes a simple matter of regulating the overall delta P to the lower deck.
 
Larry DeCamp said:
Ross,
You bring attention to the ideal of keeping the incoming flow directed over the cylinders. Clint B is developing such plenums like some racer pics I have seen here on VAF.
This raises a question for me with your coil and Don?s flow distributor mounted on top of the case split. How much cooling of these components is required, if any ? Turbulence could be a good thing or unnecessary. I can make either scenario happen when I mold the plenum top.
Click to expand...

I?ve also been following Clint?s development of his plenums. The question must be asked as to which type of plenum is better/more efficient... the individual plenum(s) like Clint is designing that incapsulates each cylinder bank (1,3)(2,4) or one that?s more traditional that basically puts a lid on top of the entire engine?
 
Mark33 said:
The question must be asked as to which type of plenum is better/more efficient... the individual plenum(s) like Clint is designing that encapsulates each cylinder bank (1,3)(2,4) or one that?s more traditional that basically puts a lid on top of the entire engine?
Click to expand...

I'll play devil's advocate...

There is heat rejection from engine case to cool upper plenum air, because there is significant deltaT. Given individual plenums, that same case area would be bathed in much warmer air, with less turbulence, so less rejection.

A plenum lid is just a sealing device. As a practical matter, individual plenums tend to be leaky in the pushrod tube and cylinder area.

Pressure recovery is a function of the inlet (and the diffuser when one is necessary). The designer of an individual plenum faces the same choices as the designer of an overall plenum.

There are no attach points on the inboard side, so how do you fasten it to the engine? There is considerable pressure attempting to blow it off the cylinders, about 0.8 x available dynamic pressure if done well. The design point would be about 150 lbs per sq foot if a little safety margin is desired.
 
I can't speak to airflow part of the discussion, but I did fabricate a 3 piece cover, full doghouse plenum for my O-360 in my -6A.

Mechanically, the covers are held on with hinge / hinge pins. The center section on top of the engine is fixed, and the left and right top covers are easily removabl. The issue I had was with the top covers in place, I had to cut holes above the #1 and #2 cylinder top spark plugs and plug wires to get the covers to fit. To fill these holes I had to fabricate fiberglass blisters and attach them to the top covers. There is not a lot of room between the top of the blisters and the bottom of the cowling. It took several iterations of fiberglassing to get to a form where the blisters did not contact the top cowling with the engine's "wet dog shake" on shut down.

From one of the pictures in an earlier part of this thread, it looks like auto spark plugs are being used. That would have eliminated the need for the blisters as the auto plugs are shorter and the plug wire caps provide a right angle bend.
 
All the photos I have are old school film...no digital images. I will look and see what I have in the more recent digital realm, but I am not hopeful.
 
You must log in or register to reply here.
Back
Top