What's new
Van's Air Force

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

Lower Cowl Exit - Firewall Bump

bjdecker

Well Known Member
Ambassador
First off, thanks to CRABANDY for his original work 5+ years ago.

This past week, I decided to add the firewall bump to my RV-7 (as depicted in builders manual, section 12, page 12-2, figures 12-1 and 12-2.)

I've been gathering test data for over 2 years now, and have a pretty solid data set for a variety of operating conditions, power settings, etc.

With the "bump" installed, there appears to be an increase in TAS of 3 kts. This was measured at 6500MSL, 2500RPM, 24.1" MAP, 9.7GPH - 50°lean of peak, 5°C OAT, 30.20"inHg. TAS was 168kts.

CHT's and OT was a few degrees warmer, but all within limits and nothing really concerning:
No Bump - With Bump
285 287
314 314
290 293
276 279
OT:
185 187

I attribute the increase in airspeed to a reduction in cooling drag due to "shrinking the exit".

Next steps -
Change to stainless steel and extend from 15" to ~18" to cover more of the area above the cowl exit.

Cheers!
 

Attachments

  • IMG_1858.jpg
    IMG_1858.jpg
    209.6 KB · Views: 658
  • IMG_1859.jpg
    IMG_1859.jpg
    173.1 KB · Views: 575
  • IMG_1861.jpg
    IMG_1861.jpg
    357.1 KB · Views: 1,030
  • IMG_1862.jpg
    IMG_1862.jpg
    332.8 KB · Views: 1,079
  • IMG_1863.jpg
    IMG_1863.jpg
    289.4 KB · Views: 1,088
Last edited:
Andy,

I think it's a bit "smoother" -- Unfortunately, I don't have a frequency plot before/after to know for certain.

B
 
I attribute the increase in airspeed to a reduction in cooling drag due to "shrinking the exit".

Due to increased exit velocity.

Now think about getting all those tubes and hoses out of the exit flow.
 

Attachments

  • PB090005.JPG
    PB090005.JPG
    414.3 KB · Views: 1,303
Last edited:
Nice Dan -- where did you end up putting the oil cooler exit diffuser?

In that early construction photo, it is not cut in yet.

First duct was 4" sceet with glass transitions. Later I built a glass duct with a little larger exit. Been flying it a long time now.
 

Attachments

  • IMG_20190915_183800680.jpg
    IMG_20190915_183800680.jpg
    140.6 KB · Views: 977
  • OC Foam 1.JPG
    OC Foam 1.JPG
    262.4 KB · Views: 954
  • 10611 and Ductwork.jpg
    10611 and Ductwork.jpg
    204.5 KB · Views: 858
The BUMP...

Since I had the cowling off to do surgery on the alternator, I took the opportunity to upgrade the cowl exit ramp/bump from the aluminum version to stainless steel.

I made it a bit wider, taller and included some adel clamps (WDG21019-DG12's) to attach to the engine mount.

Thanks again crabandy, DanH for your work here...
 

Attachments

  • IMG_2740.jpg
    IMG_2740.jpg
    294.9 KB · Views: 804
  • IMG_2741.jpg
    IMG_2741.jpg
    282.5 KB · Views: 675
  • IMG_2742.jpg
    IMG_2742.jpg
    307 KB · Views: 790
Making it wider is good. You would like to span the full width of the cowl exit opening.

Making it 'longer' is not really necessary. You can accelerate flow happily around a fairly tight radius. Extra length adds more surface skin friction drag without really doing anything beneficial.

The reverse bend that you now have to facilitate mounting with the Adel clamps is not so good. The goal is to create a smooth ramp for acceleration, like a trumpet bell. If anything, you would like to roll a small lip on the edge that goes back the other way. Can you cut the portion out in between the mounting points, or notch it just inboard of the mounting points and roll that middle portion back? In doing so, you may lose some strength, so you might want to put some doublers locally at the mounting points.

Your speed increase is gratifying. You have increased the effective exit area by getting good high-velocity flow across the whole cross section of the exit (assuming the bump spans across the whole exit). You have eliminated the vena contracta in the exit flow that is caused by flow around the sharp corner at the bottom of the firewall. I am surprised you saw slight increases in CHT, since if anything, you have more cooling flow. Perhaps just variation from slightly different run time or other test condition.
 
There is a picture partway down the first page of Vetterman's website showing an alternative method of securing the exhaust pipes from side to side movement. The alternate method gets the horizontal rubber hose out of the airflow path near the cowl exit. Here's a link to the page:

http://www.vettermanexhaust.com/

It might be possible to combine the alternate exhaust mount and a firewall bump with a little metal creativity. Might be worth another knot or two and a little better cooling.
 
The reverse bend that you now have to facilitate mounting with the Adel clamps is not so good.

Can do! I already had to "notch" it to clear the throttle cable -- removing the material between the adel clamps shouldn't be an issue, aside from the hazards associated with working with stainless steel :/

The 1 or 2 degree delta in the CHTs was probably caused by an increase in the OAT during the test (Morning vs Afternoon) down here in He-double-L's front porch.
 
Last edited:
Just wondering - would this also provide a heat or maybe even flame shield to the bottom of the firewall and floor?
 
Making it 'longer' is not really necessary. You can accelerate flow happily around a fairly tight radius. Extra length adds more surface skin friction drag without really doing anything beneficial.

Steve point out this detail in a PM long ago, and I've tried to pass it along. Please note I kept the big internal trumpet seen in posts 4 and 7 because it created a low pressure termination (less than lower cowl pressure) for the ducted oil cooler, which increases cooler deltaP.

The reverse bend that you now have to facilitate mounting with the Adel clamps is not so good. The goal is to create a smooth ramp for acceleration, like a trumpet bell. If anything, you would like to roll a small lip on the edge that goes back the other way.

I've added a sketch below for those who may not entirely follow what Steve is describing. Reverse bend on left, not so good. Two examples of rolled lip on right. Large radius lip notched for mount tubes is probably best. If interested in fire protection or floor heat reduction, a stainless steel panel could be extended over fiberfrax felt.
.
 

Attachments

  • Cowl Outlet Fairing.jpg
    Cowl Outlet Fairing.jpg
    25.4 KB · Views: 707
Steve, What would you consider to be a optimum radius in this firewall application?
Thanks

It would be difficult to identify any kind of optimum. Too small a radius and the flow may still separate. That depends on a lot of details, such as the distribution of incoming flow. Too big a radius and you have a lot of wetted area that is just friction drag. My instinct is that the radius on the standard RV-8 installation is close to the minimum that should be used. The shape in Dan's sketches looks really good to my eye.

Note also that Dan's installation has side panels too.
 
Radiused...

Bent the "leading" edge over --- now, I think I'll remake the whole darn thing and roll the edge end-to-end.
 

Attachments

  • IMG_2747.jpg
    IMG_2747.jpg
    252.6 KB · Views: 674
Getting rid of the reverse bend for a radius makes so much sense, if I ever take mine off I’ll make a new one with the rolled upper edge.
 
Bent the "leading" edge over --- now, I think I'll remake the whole darn thing and roll the edge end-to-end.

While you're in there, get rid of the transverse rubber hose, hose clamps, etc. Maybe just relocate it well forward, out of the exit flow.
 
While you're in there, get rid of the transverse rubber hose, hose clamps, etc. Maybe just relocate it well forward, out of the exit flow.

Gotcha -- I would need to engineer a new exhaust hangar "system"; round cross sections perpendicular to flow are-the-worst drag-wise. Maybe I can find a tear drop shape and use it for the transverse mount/isolation. NACA 0024 would suffice...
 
Last edited:
Gotcha -- I would need to engineer a new exhaust hangar "system"; round cross sections perpendicular to flow are-the-worst drag-wise...

Check the link I posted earlier in the thread. There's a picture on that page which shows how to eliminate that hose.
 
Check the link I posted earlier in the thread. There's a picture on that page which shows how to eliminate that hose.

Ahhh yes -- thanks Kyle.

Unfortunately that would interfere with the exit radius material. Also, it would seem that would impart more movement (read: stress) to the exhaust stacks as it is fixed to the firewall instead of hanging off the engine...
 
Ahhh yes -- thanks Kyle.

Unfortunately that would interfere with the exit radius material. Also, it would seem that would impart more movement (read: stress) to the exhaust stacks as it is fixed to the firewall instead of hanging off the engine...

Did you read the description on the web page? You still suspend the pipes from the engine or engine mount with flexible mounts. But you firmly mount the exhaust exits to hold the lateral spacing. The slip joints handle the relative movement.

Vetterman is pretty smart.

To make your exhaust fairing work, you'd just need to cut a couple of slots in it for the clamps.
 
Performance update...

Ok, I've had some time to fly with the modified cowl exit ramp and I am happy to report a pretty substantial decrease in CHT's, and modest increase in TAS/IAS.

Aircraft:
RV-7
Lycoming IO-360-A1B6 stock except for Dual Lightspeed PIIIs @ 20°
Hartzell 7497-2 (72")

Flight Test Parameters:
6500Ft MSL, 30.05InHg
~17°C OAT
2500RPM
~24"MAP (WOT)
9.8 GPH
60°F LOP

--

Before:
CHT °F 1: 312, 2: 340, 3: 312, 4: 294
Oil Temp: 190°F
TAS 161Kts

After:
CHT °F 1: 297, 2: 322, 3: 311, 4: 278
Oil Temp: 186°F
TAS 165Kts

--

Your mileage may vary, but I'm liking the results and temperature margin that this widget provides.

Next changes(s):
Remove and replace horizontal exhaust stack brace with NACA 0024 shape -or- similar.
It's been suggested that using Vettermen's alternative firewall mount for exhaust stacks is the way to go, but I respectfully disagree; there's just too much lyco-shake going on and I fear the ball joints don't have enough range or ease of motion to absorb it.
 

Attachments

  • IMG_2747 2.jpg
    IMG_2747 2.jpg
    252.6 KB · Views: 730
Nice!

How did you get the 'before' data? It does not appear to be the same as the 'before' data you reported on with the original narrow aluminum prototype. Did you remove the first one, fly, then install the second one, and fly again?
 
Nice!

How did you get the 'before' data? It does not appear to be the same as the 'before' data you reported on with the original narrow aluminum prototype. Did you remove the first one, fly, then install the second one, and fly again?

About 10 times in fact :)

I never trusted the data from the first aluminum prototype, and since I was working on my IAS/Static issue that came up post-paint, I figured this was an excellent opportunity to re-baseline the performance numbers, build versions 2 & 3 of the shape and try again.
 
This sounds like something to try for a little less drag!
Does it need to be stainless, after all the stainless firewall is still in place?
Pete
 
My first attempt at a cowl bump fabricated from Aluminum. I’m sure improvements can be made and will consider it a work in progress.
 

Attachments

  • IMG_3541.jpg
    IMG_3541.jpg
    289 KB · Views: 621
  • IMG_3542.jpg
    IMG_3542.jpg
    300.2 KB · Views: 580
This sounds like something to try for a little less drag!
Does it need to be stainless, after all the stainless firewall is still in place?
Pete

Stainless is so much fun to work with [sarcasm] tho...

I originally switched to stainless because the exhaust stacks were touching the aluminum version and causing a bit of wear...

I adjusted both the shape and the exhaust to there's no impingement now, so I suppose I could switch back to aluminum.
 
Trivia

The bump goes back at least to Tom Cassutt (Cassutt Racer) in the mid 50's. Possibly he got the ides from someone else. He talked about it in his forums around 69/70.
 
I've been flying with the version seen below on my 180HP RV-7 for over 8 years. Stainless steel, fastened with nutplates on the firewall at the top and the firewall-to-fuselage lip at the bottom. Spans nearly the full width of the exit.

At the time I fabbed and installed it I only had about 100 hours on the airplane. I did not keep good before/after records but I do recall the CHTs dropping at least 10 degrees in cruise at my usual power settings. Seems to me I picked up a couple of knots but I don't recall for sure, and at this point my back tells me not to consider groveling so I can repeat a before/after measurement test. I have installed something similar on several other side by side RVs since then and everyone reports noticeably dropped CHTs, but with one exception, no quantified results. That one exception noted 10-12 degrees.

I mentioned to a Van's engineer I know that they should consider cranking out a part like that and selling it as an option, both for nosewheel and tailwheel versions, since the arguably simple part makes an obvious difference.

i-m9PWZr9-XL.jpg


i-XnfJDQm-XL.jpg


i-wzcvVPn-XL.jpg
 
RV-9 Builder Interested In Learning More

It was mentioned earlier that this was depicted in builders manual, section 12, page 12-2, figures 12-1 and 12-2. Does anyone have that in a format they can share? I would like to read up more about it.
 
interesting concept, which I still might retrofit. On the other side rather sceptical about any speed increase... at least without reducing the size of the inlets and a reshape of the upper cowling.
Presently no issue with my temps, even when flying in hot weather, but would commit for any kind of speed (read efficiency) gain :D
 
Adjustable bump

At Reno this year, I saw a guy had an adjustable bump.. a cable to raise and lower it I assume to control the outflow?
 

Attachments

  • A973DE4E-D44E-49DE-83DF-357FA032C0A5.jpg
    A973DE4E-D44E-49DE-83DF-357FA032C0A5.jpg
    267.9 KB · Views: 496
  • 06183741-2DE4-4F5F-AF1C-A496C1C44DA6.jpg
    06183741-2DE4-4F5F-AF1C-A496C1C44DA6.jpg
    309.8 KB · Views: 468
At Reno this year, I saw a guy had an adjustable bump.. a cable to raise and lower it I assume to control the outflow?

I seem to recall early RV-8 kits having a piano hinge at the trailing edge of the ramp.

Thanks Dan, reduce cooling drag and frontal area (also wetted, ever so slightly)?

We've addressed cooling drag quite a lot in the past. The reasons are complex, but for now accept that a mere reduction of intake area does not in itself decrease cooling drag.

In the context of this thread, note drag reduction requires a decrease in mass or higher exit velocity. The exit lip fairing seen here is claimed to both reduce CHT and increase speed, which suggests both a mass flow increase and an exit velocity increase. I think that's reasonable. See Steve's notes, last paragraph, post #9, and then go back and look at various video of tufted exits. The stock exit is low velocity with a messy wake. This one simply flows better. Higher velocity means a bit more mass flow per unit time.

Changing intake area does not change frontal area. A few quick sketches below. The triple examples, with three different intake areas, all have the same frontal area. To reduce frontal area, (1) shrink the canopy, (2) eliminate the RV's classic coal shovel hanging below the belly line, and (3) drop the wheel pants so the total area of the pant and tire is reduced.

BTW, those three different intake areas all result in roughly the same mass flow if they all feed the same exit area.
-
 

Attachments

  • Intake Areas vs Frontal.jpg
    Intake Areas vs Frontal.jpg
    61.6 KB · Views: 155
  • Frontal Area 4.jpg
    Frontal Area 4.jpg
    29.9 KB · Views: 180
Last edited:
Exit ramp

Perhaps the exit ramp helps to clean up the aerodynamic mess that is the compartment? Considering the speed of the cooling airflow thru the engine compartment, I think a little baffling to keep the flow away from the firewall “stuff” would really help the efficiency. But that is just me.
 
Perhaps the exit ramp helps to clean up the aerodynamic mess that is the compartment? Considering the speed of the cooling airflow thru the engine compartment, I think a little baffling to keep the flow away from the firewall “stuff” would really help the efficiency. But that is just me.

Bob Axsom (RIP) tried a bunch of permutations of baffling and exit mods years ago and reported on them here. Photos are likely long gone unfortunately, but he also provided verbose, detailed descriptions.
 
Bob Axsom (RIP) tried a bunch of permutations of baffling and exit mods years ago and reported on them here. Photos are likely long gone unfortunately, but he also provided verbose, detailed descriptions.

Yep at 2 AM. I have his pictures and Alan Judy's too. Bob concluded that they did not help his 6A. I think he would admit that they may provide benefit for others. I think Dave Anders has some of this too, but not sure. Dave cleaned up the exit area under his heads/barrels.

I am thinking about tufting the inside of my cowl, adding a GoPro and blowing it with a backpack blower through the cooling rings as I have no clue what is happening in there. Fresh out of guesses.

Sorry, is that too detailed?
 
Yep at 2 AM. I have his pictures and Alan Judy's too. Bob concluded that they did not help his 6A. I think he would admit that they may provide benefit for others. I think Dave Anders has some of this too, but not sure. Dave cleaned up the exit area under his heads/barrels.

I am thinking about tufting the inside of my cowl, adding a GoPro and blowing it with a backpack blower through the cooling rings as I have no clue what is happening in there. Fresh out of guesses.

Sorry, is that too detailed?

That sounds like a really neat experiment. I was about to write "cool experiment" but, well, punny.

Bring on the detail, this engineer would love to learn more.

I thoroughly enjoyed reading Bob's stories. In fact one of his is probably why I installed a half-round outlet ramp on the cowl exit of my RV-7. With similar results to what others have described.
 
I tried posting something about my own experience yesterday but it disappeared.

I fabbed and installed a half-round exit on my RV-7 almost 9 years ago. I didn't make any careful before/after speed measurements, but at my normal cruise settings it appeared to drop the CHTs around 10-12 degrees. It's such an easy mod that I'd recommend it to anyone with a side by side RV.

i-m9PWZr9-L.jpg


i-XnfJDQm-L.jpg
 
Thanks for the feedback...

I replaced the large, round cross brace with a more aerodynamic shape. No flight test data yet, but I don't expect much difference.

Edit: The large holes are to fit a home brew rubber bushing.
 

Attachments

  • IMG_2870.jpg
    IMG_2870.jpg
    402.2 KB · Views: 265
  • IMG_2871.jpg
    IMG_2871.jpg
    385 KB · Views: 242
  • IMG_2866.jpg
    IMG_2866.jpg
    204.6 KB · Views: 296
Last edited:
Just me

I replaced the large, round cross brace with a more aerodynamic shape. No flight test data yet, but I don't expect much difference.

I am not sure about taking the flexible hose type mount and putting in a rigid strap. I would contemplate using a flat strap of reinforced rubber instead?

What do I know I am staying at a Holiday Inn.
 
I am not sure about taking the flexible hose type mount and putting in a rigid strap. I would contemplate using a flat strap of reinforced rubber instead?

What do I know I am staying at a Holiday Inn.

There are elastomeric bushings in the ends...
 
Training wheel version?

Is there any estimate if the would have similar effects for -A’s? The big tube down the center, purely round, I expect minimal, but it never hurts to ask.

And is there an imaginative way to get around that the bottom cowl pins running through there?
 
Last edited:
Is there any estimate if the would have similar effects for -A’s? The big tube down the center, purely round, I expect minimal, but it never hurts to ask.

As builder/pilots we would NEVER accept attaching an un-aerodynamic shape to the exterior of our planes (exception: GoPro Camera), yet we seem to forget about the airflow in & around the engine (except for DanH, ScSmith, et al)

So, yeah I think there are some opportunities in the -A models in this area, maybe tapered/triangular "after body" shape glued to the tubes, other strategic places...Hard to quantify without test data tho. :/
 
Last edited:
Another cowl outlet baffle

Did some cowl mods recently, copying ideas already thrown out on this post. My cooling was mostly acceptable, even in Moab this summer (4000'+ and 90+ oat) occasionally running warm, but there is always room for improvement. No great data collected demonstrating additional cooling or speed, but my impression is that my CHT's are running lower and dropping faster. Weather is pretty chilly up here right now, so we'll need to wait for the next warm up (June sometime!) to measure the results.

Thanks to the all forum posters for their insight and examples...

Cheers!
 

Attachments

  • 2021-10-29 16.27.57.jpg
    2021-10-29 16.27.57.jpg
    322 KB · Views: 467
  • 2021-10-29 16.28.09.jpg
    2021-10-29 16.28.09.jpg
    257.5 KB · Views: 378
  • 2021-10-29 16.28.23.jpg
    2021-10-29 16.28.23.jpg
    262 KB · Views: 428
Do any of you know what provides better results, the rounded 180 ramp or just adding a curved plate?

In theory a sort of "S" shape giving a nice flow from the vertical firewall around the corner and out should be the best shape. A "D" shape would be next best, followed by a curved plate. The curved plate still has the issue of the edge tripping the flow and creating turbulence and lower flow rates.

All of those should be an improvement over the abrupt transition on the bottom of the firewall.
 
Back
Top