VansAirForceForums  
Home > VansAirForceForums

-POSTING RULES
-Advertise in here!
- Today's Posts | Insert Pics

Keep VAF Going
Donate methods

Point your
camera app here
to donate fast.

  #311  
Old 11-16-2015, 04:59 PM
crabandy crabandy is offline
 
Join Date: Jan 2010
Location: Ottawa, Ks
Posts: 2,309
Default

Pic of my left inlet ramp, I glasses in the inboard portion to give the baffle material something to push against. The left hand forward baffling is removed in this pic, lots of room between the top of the cylinder and the curved top of the cowl.


Right side has the top of the cylinder squeezing together with the bottom of the top ramp, I should have cut some material off the top ramp and moved the curve forward.


I'm planning on cutting both ramps out and using a diffuser boot instead.
__________________
RV 7 500 hours and counting
20 donation done
Reply With Quote
  #312  
Old 11-16-2015, 05:11 PM
crabandy crabandy is offline
 
Join Date: Jan 2010
Location: Ottawa, Ks
Posts: 2,309
Default

I'm not sure of the angle I should make with the bottom of the inlet. The existing inlet points downward, the metal baffling points back to the middle of the cylinder. Should I leave the downward angle initially then angle towards the middle of the cylinder as the top of the new diffuser boot can expand on the top side?



__________________
RV 7 500 hours and counting
20 donation done
Reply With Quote
  #313  
Old 11-24-2015, 09:50 PM
crabandy crabandy is offline
 
Join Date: Jan 2010
Location: Ottawa, Ks
Posts: 2,309
Default

I roughly measured my right inlet with the more restrictive upper cowling ramp in 3 different places, front-middle-aft end of inlet. Due to the shape and access, the cross sections are basically square and I used pieces of paper to gauge the size and measured the papers. Almost the same vertical dimensions with expansion sideways. The top page is the side view and measurement of the inlet in inches. The bottom page shows the inlet from the front-black, middle-red, and rear-purple with rough total areas denoted in their color square.


I've made no progress in trying to pick an inlet, I'm trying to decide between making a smoother larger diffuser shape behind the Van's inlet or make a new round inlet and diffuser. The Van's inlet is harder to seal and make a diffuser for, the round inlet involves reshaping the front of the cowl.

Here's a 5 inch circle mounted in the center of the Van's inlet, I think for asthetics and diffuser shape it should be moved above the cowling split line and outboard toward the outside of the cowling.





__________________
RV 7 500 hours and counting
20 donation done

Last edited by crabandy : 11-24-2015 at 09:56 PM.
Reply With Quote
  #314  
Old 11-26-2015, 09:40 AM
DanH's Avatar
DanH DanH is online now
 
Join Date: Oct 2005
Location: 08A
Posts: 10,353
Default

Good start...measuring, playing with shapes and locations, and thinking.

As velocity ratio (velocity through the inlet divided by freestream velocity, Vi/Vo) goes down, duct shape inside the inlet becomes less critical. As velocity ratio becomes higher, internal diffuser shape becomes more critical.

You've measured the stock inlet and find it to be 21.29 sq in. A 5" diameter round inlet would be 19.625", only slightly smaller. Assuming no other changes (notably exit area) there would be no appreciable difference in Vi/Vo. Any gain in conversion of dynamic pressure to static pressure would have to come from an improvement of the internal shape.

Bill Lane obtained a good text when we were talking about ducts earlier this year. Here's a snip, the first paragraph of the chapter on diffusers, credit Applied Fluid Dynamics Handbook, Robert D. Blevins:



Note the part about boundary layer. Any large, stable separation results in less static pressure, as does a shape or separation that results in jet flow. Again, from Blevins; you want #1 or #2, the latter resulting in the highest Cp. #3 works OK. #4 and #5 are bad news:



So how to apply this very basic knowledge to your application? First realize the stock Van's inlet is sized for moderate Vi/Vo, as would your new round inlet. A lot of the diffusion (conversion of dynamic to static) is happening out in front of the inlet, and actual velocity through the inlet is significantly lower than aircraft velocity. That's why we can get away with the crappy duct shapes...the sharp angles, edges, and steps that would trip a faster flow. I would expect only moderate improvement in pressure recovery with improved duct shape alone.

Still, we'll take all the pressure we can get. Once you settle on which Vi/Vo scheme you wish to pursue (the size question), the inlet shape and location largely depends on what is best for duct shape and sealing. Do all you can with shape, but since here the low velocity ratio makes shape less critical, sealing is where you can probably make a significant improvement. Let's face it, the typical flap sealing around a GA inlet is truly awful.

Bottom line? At this inlet size, simply avoid big pro-separation duct errors, and let seal design be the primary driver of the inlet shape and position.
__________________
Dan Horton
RV-8 SS
Barrett IO-390

Last edited by DanH : 11-27-2015 at 03:09 PM.
Reply With Quote
  #315  
Old 11-26-2015, 11:28 AM
Tom Martin Tom Martin is offline
 
Join Date: Jan 2005
Location: Ontario, Canada
Posts: 1,591
Default

To add to what Dan has stated one thing you can do is move the inlet as close to the prop as you can. This lets you make a slightly longer "diffuser' duct, which will help in keeping with a smooth flow.
If you have a constant speed prop you can turn the blades by hand to get the full coarse condition and then set your inlet 1/4 to 1/2" aft of the blade edge.
Moving the duct closer or farther away from the prop hub might also get you a bit more length due to blade width.
__________________
Tom Martin RV1 pilot 4.6hours!
CPL & IFR rated
EVO F1 Rocket 1000 hours,
2010 SARL Rocket 100 race, average speed of 238.6 knots/274.6mph
RV4, RV7, RV10, two HRIIs and five F1 Rockets
RV14 Tail dragger

Fairlea Field
St.Thomas, Ontario Canada, CYQS
fairleafield@gmail.com
Reply With Quote
  #316  
Old 11-28-2015, 08:13 AM
DanH's Avatar
DanH DanH is online now
 
Join Date: Oct 2005
Location: 08A
Posts: 10,353
Default

Andy,

Previously I wrote "At this inlet size, simply avoid big pro-separation duct errors, and let seal design be the primary driver of the inlet shape and position." You can, however, select other inlet sizes and shapes. Let's explore a bit, from the standpoint of design and fabrication.

As noted previously, as you go smaller, you'll need more and more attention to internal diffusion. If you go larger, you'll eventually arrive at a place where you need nothing more than a hole in the front. The advantage is design simplicity. With mere holes, you no longer need to seal around individual inlets. All that remains is to divide the upper cowl volume from the lower cowl volume, the actual sealing being simple enough that leakage is reduced. That makes a plenum lid unnecessary; it is just a sealing device.

The Mooney Acclaim and Cessna TTX are good examples. Below, I've posted the Acclaim baffling. The Continental has no front alternator, which simplifies the under-the-propshaft tinwork, but it can be adapted to a Lycoming. Look up some photos of Sean Tucker's engine baffling for an example.



Here's an interesting one-shot illustration, Fig 5 from AIAA 80-1242R, which tells a lot about big holes in the front of a cowling. (The paper is available at the NASA documents server: http://ntrs.nasa.gov/archive/nasa/ca...9980214918.pdf)



The researchers set up an entire engine nacelle and wing section in a NASA Ames wind tunnel. They tried three different inlet sizes (41, 61, and 107 sq in total) while measuring upper cowl pressure recovery (Cpu), and the variation in drag (Cd). None of the inlets had any sort of internal diffuser. The climb condition was about 100 knots (and 8 degrees AOA), while the cruise condition was roughly 160, close to typical RV speeds. Wc was an arbitrary 3 lbs per second mass flow, the right ballpark for our Lycomings, set by throttling the flow downstream of the inlet as necessary. The scale across the bottom is inlet ratio, 0.2 being (for example) an inlet area 5x the area of a 3 lbs per second stream tube...a very low Vi/Vo inlet.

There are a number of interesting observations to be made, but here I'll stick to just two key points. First, without good internal diffusers, pressure recovery (Cpu) takes a dive as the inlet ratio is increased. In the 0.6~0.8 range it's is truly awful; the addition of internal diffusers is mandatory at those ratios, and here you see why. However, look how good Cpu is at low inlet ratios, with a very steep rise for the climb condition between 0.4 and 0.2. That performance comes without internal diffusers to complicate design and fabrication, just big holes.

Take a beer:30 survey on any flightline, and you find the usual objection to big holes is drag. However, take a look at the lower plot; the Cd lines are almost flat across a wide range of inlet ratios. For the same cowl and mass flow, inlet size actually has very little relation to drag.
__________________
Dan Horton
RV-8 SS
Barrett IO-390

Last edited by DanH : 11-29-2015 at 10:15 AM.
Reply With Quote
  #317  
Old 11-28-2015, 10:50 AM
crabandy crabandy is offline
 
Join Date: Jan 2010
Location: Ottawa, Ks
Posts: 2,309
Default

Dan,
Thanks again for taking the time to explain this, I understand parts of the conclusions in the studies but still wrapping my head around it. I'll have to grab my notes.......

I get the 3 lbs per second mass air flow, I believe it comes from the lycoming cooling air requirement charts. It's hard to interpet scale but 2 lbs per second mass airflow keeps CHT's in the 400's.

Just putting some thoughts on paper:
-The 3 lbs per second mass airflow will keep things cool wether through large or small inlets.
-The mass airflow of 3 lbs per second is driven by the pressure differential between the upper and lower plenums.
-Large inlets externally slow the air and provide good pressure recovery with lots of sins in inlet/diffuser shape.
-Small inlets have higher velocity air that need very good inlet/diffuser shape to provide good pressure recovery
-Poor pressure recovery may provide less than 3 lbs per second mass airflow and high CHT's

How do you calculate the size of a tube required to flow 3 lbs per second mass airflow at 160 knts? I'm trying to find the formula of how big an inlet is needed to provide .2 Vi/Vo.

So with the inlet size Vi/Vo in the study at cruis of 160 knts the inlet velocity for the different inlets was:
107 sq in = 32 knts
61 sq in = 64 knts
41 sq in = 96 knts
__________________
RV 7 500 hours and counting
20 donation done
Reply With Quote
  #318  
Old 11-28-2015, 12:47 PM
crabandy crabandy is offline
 
Join Date: Jan 2010
Location: Ottawa, Ks
Posts: 2,309
Default

More thinking out loud......

At sea level, standard temp and pressure at 90 knts the theoretical inlet size required to provide 3 lbs/sec of cooling air at a Vi/Vo of .2 =____Area of inlet____.

Vi=18knts (Vi/90knts=.2)
Vo=90knts
90knts=151ft/sec
Volume of inlet=151ft * inlet area

So for the volume of air is it 14.7 psi and interpolating this chart (http://www.engineeringtoolbox.com/ai...ity-d_771.html)
at 60 degrees (close to std) interpolating 10-20 I get .154 lbs/ft3.

I'm having trouble converting/figuring out the volume of 3 lbs of air per second, probably going about it all wrong. Dan, what formula did you use to derive your inlet size of 6 inches?
__________________
RV 7 500 hours and counting
20 donation done

Last edited by crabandy : 11-28-2015 at 01:56 PM.
Reply With Quote
  #319  
Old 11-29-2015, 08:36 AM
DanH's Avatar
DanH DanH is online now
 
Join Date: Oct 2005
Location: 08A
Posts: 10,353
Default

Quote:
Originally Posted by crabandy View Post
Just putting some thoughts on paper:
-The 3 lbs per second mass airflow will keep things cool wether through large or small inlets. Don't really need 3 lbs/sec for a 360
-The mass airflow of 3 lbs per second is driven by the pressure differential between the upper and lower plenums. Right
-Large inlets externally slow the air and provide good pressure recovery with lots of sins in inlet/diffuser shape. Without sin...larger inlets make diffusers less critical. Very low Vi/Vo inlet doesn't need any diffuser at all.
-Small inlets have higher velocity air that need very good inlet/diffuser shape to provide good pressure recovery But they work just fine if you get it right.
-Poor pressure recovery may provide less than 3 lbs per second mass airflow and high CHT's Yep.
Quote:
How do you calculate the size of a tube required to flow 3 lbs per second mass airflow at 160 knts?
Initially design for the most difficult cooling case; full power, slow airspeed. You realized that in your next post, but 90 knots is unrealistic; check dynamic pressure (Q) for 90 knots to see why. At 90, the maximum available Q is 5.28" H2O on a standard day, and per the chart 0-360 chart 3 lbs per sec would require about 12.5" drop across the baffles. Although propwash will push the available Q a little higher (see below), you can't get 3 lbs per second at 90 knots.

120 knot standard day Q is 9.38" at SL, and it doesn't drop below 8" until 5000 feet. With good inlets you might get a Cpu (ratio of upper plenum static pressure to available dynamic pressure) as good as 0.8, so 9.38" available means 7.5" in the upper plenum. The Lycoming cooling air chart says 7.5" will give you about 2.4 lbs per second mass flow. Better start there, as it's what you have.

BTW, you may be thinking "But will that be enough to keep CHT in line?" Well, no guarantees, but there are a few things that can boost the end result.

One, the Lycoming cooling chart data is emperical data, meaning it's what they recorded using whatever they considered to be standard baffles. We don't know exactly what those baffles looked like; that information seems to be in a Lycoming book you can't get. However, judging from the state of standard GA baffle tin design, or even Vans baffle tin, it wasn't anything fancy. We can do things to improve heat transfer at the baffles. For example, I see you've already been busy with glass wraps.

Two, in the climb power regime there is some boost in Q due to propwash. Exactly how much varies with propeller design (notably the shape of the blade roots near the spinner), and the intake location/shape (notably how far outboard and how much radial deltaP across the inlet face). For the BA Hartzell and outboard, low Vi/Vo inlet I use, the difference between 100 knots IAS level flight (low power) and 100 knots IAS climb (full power) at 3000 ft PA was:

low power: upper 6.0 lower 2.5 deltaP 3.5
high power: upper 9.0 lower 3.75 deltaP 5.25

Note that even in the low power case, propwash is boosting Q a bit, as standard Q for this altitude at 100 knots about 6", and no inlet can achieve 1.0 Cpu if the plenum has an outlet. (An inlet with no outlet would be an airspeed pitot.)

In the high power case, propwash is boosting the available Q quite a lot. If I assume a Cpu of 0.8, then available Q was about 11.2", equivalent to about 137 knots.

These numbers should not be treated as accurate in an global sense (because I was using IAS rather than TAS to make the climb airspeed target easier to hit), but the concepts are correct.

Quote:
I'm trying to find the formula of how big an inlet is needed to provide .2 Vi/Vo.
Whoa hoss...first things first. I've seen data taken with a stock RV-8 inlet, and pressure recovery was actually pretty good. Don't you want to determine what you have before launching off into big time glass work? If pressure recovery is good, the design goal becomes improved sealing and heat transfer, not intake design.

I saw piccolo tubes in your photos. Go fly at 120 knots and determine your current Cpu. You need PA, temperature, and upper plenum pressure.
__________________
Dan Horton
RV-8 SS
Barrett IO-390

Last edited by DanH : 11-29-2015 at 10:19 AM.
Reply With Quote
  #320  
Old 11-29-2015, 08:22 PM
crabandy crabandy is offline
 
Join Date: Jan 2010
Location: Ottawa, Ks
Posts: 2,309
Default

Prop is in California, but I did get temps and pressures before I pulled it off. Notes are at the airport, should get there tomorrow.

Here's some historical pressures from a while ago, more detail in this thread:
http://www.vansairforce.com/communit...=112941&page=6

"I made 2 flights 4 hours apart at approximately 2000 Pressure Alt. and 78*F with plenum pressure differential measured between static pressure in in/H2O. Here's the rough averages:
115 knts IAS-upper plenum 7.5 In/H20-lower plenum 1.4 In/H2O
130 knts IAS-upper plenum 9 In/H2O-lower plenum 1.5 In/H2O
160 knts IAS-upper plenum 13.25 In/H2O-lower plenum 2 In/H2O"

"Whoa hoss...first things first. I've seen data taken with a stock RV-8 inlet, and pressure recovery was actually pretty good. Don't you want to determine what you have before launching off into big time glass work? If pressure recovery is good, the design goal becomes improved sealing and heat transfer, not intake design."

Yes, my original plan was to use the Van's inlet with a plenum/diffusers and better sealing to the cowl inlets. Reading the different studies has me slightly rethinking the use of a round inlet. I've been very hesitant about "ruining" any of my current parts leaving me an easy out to put it back the way things were........too late.


Either inlet will still require me trashing the current upper cowling inlet ramps and forward baffling. The thought of major glasswork/re-shaping of the front cowling has had me stalled, I'm currently taking it slow and trying to think through the options. I'm not sure if the simplicity of sealing the round inlet along with reshaping the front cowling outweighs the complexity of sealing the Van's inlet. My equally important goal is easy cowling removal and installation.
__________________
RV 7 500 hours and counting
20 donation done
Reply With Quote
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -6. The time now is 10:03 AM.


The VAFForums come to you courtesy Delta Romeo, LLC. By viewing and participating in them you agree to build your plane using standardized methods and practices and to fly it safely and in accordance with the laws governing the country you are located in.