N62588 Phase 1 Flight Testing

[Bcone1381]

Edit: 03Apr2025. Lets organize my flight test thread into useful data. Issues I have are common to experimental flight testing. By exposing my strengths and weaknesses others can learn and maybe avoid common errors.

So issues will be started in a post. We will keep the individual issue discussion on my aircraft within each post using the comment feature.

When I fly I find issues. I have many small issues to fix. Every issue gets a file folder. I print information to help me fix an issue and place them in the file plus a piece of paper to take notes on, so as I work on resolutions I can record changes an nail down scientific data that produced resolutions or failures. I keep the folders together on a clip board. I take them to the airport and back home so I can research solutions.

Index:
Post #9 - Engine Breakin

Text below the line is original and dated the end of Feb2025.
____________________

Phase one flight testing. I am following the EAA Flight Test Task Based. But that program by the letter does not address Engine Break-in. So when Engine break-in is complete, then the program gets implemented in earnest. The first flight was about 45 minutes with the goal of running the engine hard and returning to check things over for pending infant mortality issues.

Issues we found in flight
-Prop RPM needs adjusting
-Rigging - needs slight left rudder in cruise, and has slight L heavy wing.
-ADS-b failed to record anything.
-IAS seems to read slow ~ 10 kts.

The next two flights were dedicated to fixing ADS-b issue as the aircraft is based in ADS-b airspace. This was fixed thru adjustment of settings of the Trig Transponder. In those two flights a ground test found that Prop low pitch stop was set perfectly by Hartzell and so a govenor adjustment was made. The rudder rigging is being addressed. Once that is good, we will address the roll rigging. A trim tab is commonly installed to fix this.

With my focus on Engine Break-in I am not spending much time playing with stuff at various airspeed, then stopping, adjusting, then seeing what effect it has. I flew a 2+ hour flight yesterday and do one thing at a time. I am also going to test out a simple rudder trim device that uses an adjustable bungee to apply some rudder pressure. This is what my colleague calls a MPV design.... "Minimally Viable Product".

Airspeed. The pitot/static tube I installed is from Steen Aero Lab. It sticks out about 10" from the leading edge. We are getting low airspeed readouts....Data says about 14 kts slow. The Husky has a very very similar pitot static tube and they have had the same issue and have a fix for it. I am on there group and getting some good data/ideas that we will implement to improve the error. I might end up getting it fixed with modifications to the pitot tube or move the static port to the fuselage. Right now I see 97 KTAS on the Garmin G5. When I fly a GPS box pattern and average the speed of the four legs I get 111 kts ground speed. I think the compression of the air as it runs into the leading edge of the wing is increasing the pressure at the static holes in the probe, so the airspeed reads low. I will play with various treatments ahead of and behind those holes just aft of the pitot tube "nozzle" and see what happens.

 

[rodsmith]

Good job working through the issues Brooks!

 

[svyolo]

There are terms used in rigging that I was/am not familiar with. The biggest being "heavy wing". I had a slight left roll tendancy, that I tried to fix with a "heavy wing" fix. That caused a heavy wing.

So what I think is a "heavy wing", is the aircraft will roll some set amount, in my case 10-15 degrees. Hands off the ailerons, and the ailerons are displaced into that roll. The aileron displacement causes adverse yaw in the opposite direction. The plane reaches equalibrium, and just sits there, one wing down, yawing in the opposite directions.

Is that an accurate description?

I fixed the roll with the flaps. I undid most of what caused the "heavy wing", and most of it went away. Right now I have a small trim tab, but will still try to get rid the small amount of heavy wing without the tab. Flies hands off.

 

[jaredyates]

Hey John it is great to hear that you are out putting hours on your plane! There can be some complexity to the roll trim, because there are lots of things that can impact it. Since there is coupling between the yaw and roll controls and their axes, that doesn't help either. Sometimes what seems like a heavy wing (out of trim in the roll axis) is actually a rudder thing rather than a wing thing. Asymmetry in the flaps at their full-up position can cause a roll, but the ailerons can too, for a few different reasons. A very slight height difference in the ailerons is a common trim point. This is accomplished by adding washers under the aileron mounting points at the rear spar, either at the bottom, or the top, to raise or lower the aileron accordingly.

​​​The key to happiness is figuring out what the cause of the roll is, and fixing that. In my case it was aileron related. I adjusted a flap, and that would fix it at one speed, but if I was faster or slower than that speed, the uncommanded roll input was still present.

 

[svyolo]

The roll tendency was definitely fixed with flaps (ball centered manually). It does appear to be a slight height difference in the flaps.

The ailerons were more "fun". A slight height difference caused the neutral position(not neutral) of the ailerons to cause a slight left roll, and adverse yaw the other way. When they reached equilibrium, the airplane would just sit 10-15 degrees left wing down, and yawed to the right. I am thinking that is what is called "heavy wing". If I am wrong with that I hope someone sets me straight.

I created the heavy wing by misdiagnosing a rolling moment as a heavy wing, and trying the washer under the aileron hinge. I have almost completely eliminated that.

I have a fixed rudder trim tab, with little deflection. And a temporary small aileron trim tab, but I think I will be able to zero out the heavy wing without the tab.

Right now it flys hands off at normal cruise, just needs rudder if going slower or faster. Small amount of fine tuning required.

 

[Bcone1381]

Seems like the Weather has been IFR or windy so no more flights since #4. I enjoy the discussion on rigging.

John; I don't know what i'm looking for to help trouble shoot my rigging. I jump in and fix it before it gets that far. Feet on the floor with hands off the stick I think it yaws right. Not sure about roll. If I center the ball with left rudder I need right stick to keep the wings level.

What does it mean in my wings are level, the ball is centered, and I study my hood ornament (my kids may not know what a hood ornament is). my heading does not stay on target....it drifts to the right. I suspect that might mean my ball indicator on my G5 needs adjustment, but I'm not sure. I think if I release a bit of R rudder so the ball shows 1/3 ball left then the aircraft holds a heading and the ornament stays on target.

Another data point... I talked with Mark Goldberg a few days ago. He said another check that could effect roll was the H-stab rigging. I checked mine and the R & L H-stabs are identical....with one exception. My R. outboard incidence was about 1/16 bubble off using a old fashion bubble level. The H-stab spar checks good. The H-stab has a stream lined strut that rigs the outboard leading edge of the H-stab up/down. I adjusted mine, raised the Leading edge a bit by lengthening the strut end one turn and have it nailed now. I also rigged up an internal rudder trim because I woke up in the night with an idea, and I did not like the shape of my trim tab.

My plan is to rig this like we use to trim the 727. Start with the tail (rudder) and move forward to the wing. But I'm open to advice from the experts. That means I'd trim out the rudder so it stays on target, note the G5 ball indication so I can change the setting when I land. Then I would work on trimming out any roll that I still have.

 

[jaredyates]

Just a refresher on some of the content that we have in the ecosystem on these topics:
https://bearhawk.tips/3537
https://bearhawk.tips/3722
If you don't have access to these just let me know and I'll send a pdf.

 

[svyolo]

I have never rigged an airplane before, but my thinking is similar. Use your feet to center the ball, and then see what the airplane wants to do in roll. Included in the "what does it want to do in roll", is release the stick and see where the ailerons sit with no pressure on the stick. If they are not neutral, the will induce a roll AND adverse yaw. If they stay neutral, and you still have a roll, it is "something" else. On mine, I fixed that with the flaps. After I sorted the roll out, I set the rudder trim tab to be neutral at normal cruise, which for me I think will be 115-120 knots IAS, at least for now on small tires.

Interesting about your yaw on the G5. I had a bad remote magnetometer. It did not effect yaw, but it showed up as the heading changing, but visually, my heading was not changing. That was an easy diagnosis and fix. I still have one minor issue that I believe is the Left to Right accelerometer built into the internal IRS in my GRT Horizon 10.1. Airspeed, groundspeed, TAS all seem good, referenced against an external GPS. Crosswind component seems off, and a bit erratic. Sort of like checking your IRS's after a flight and seeing 15 knots. It is displayed to me as erratic crosswind indication, and flight path marker displacement. Vertical is good. I can fly and navigate no problem. I will work with GRT to figure this out next month.

 

[huntaero]

Hi Brooks Make sure you ball is exactly level by leveling the airplane and adjust the gage to read level first, then go out and work from the rudder forward you could be chasing an instrument error.

 

[Nev]

What does it mean in my wings are level, the ball is centered, and I study my hood ornament (my kids may not know what a hood ornament is). my heading does not stay on target....it drifts to the right. I suspect that might mean my ball indicator on my G5 needs adjustment, but I'm not sure. I think if I release a bit of R rudder so the ball shows 1/3 ball left then the aircraft holds a heading and the ornament stays on target.​

My understanding is that this is known as a steady heading sideslip. If the heading is kept steady then the ball will indicate out to one side. If the ball is kept centered then a gradual heading change will occur. On my Bearhawk it was indicative of a "heavy wing", and corrected by installing washers under the affected hinge to raise or lower the hinge line as required.

 

[Bcone1381]

I have been thinking that same thing huntaero. Today I leveled it in the hangar (a level says my floor is level) and re-did the attitude set-up but did not check the slip-skid ball indicator. I'll do that tomorrow.

 

[Bcone1381]

I've got 10 hours on the aircraft now and so I am scaling back engine break-in focus and moving towards Rigging. Kevin D's Beartrack Articles from 2018 were brought to my attention yesterday. I read thru them and fabricated some tools and a set of instructions.

One H-stab's streamline strut was adjusted one turn on the fitting to get it lined up. I also checked the H-stab's angle of incidence against the wing's. It is within .100" of being exact. In cruise flight with me and 40 pounds in the rear seat the elevator is displaced out of neutral 1 inch nose up. So when it gets loaded up it will be less. It will be interesting to see how it changes.

I made headway on my airspeed indication. The Husky group had me install an oring on the pitot tube. The IAS : TAS difference with the oring improved from 14 kts to 4 kts. WIth this modification a dynamic static test was performed, where I took off, stayed just above the runway and watched the altimeter. The altitude went down 40' at 100 kts.

I want to get the IAS nailed so I decided to enlarged it a bit more. I tried to taper the approach to the oring and make it look pretty with shrink tubing.



Kevin D's article says my first task is to take measurements, then start with some pitch tests to expose the aircrafts pitch stability and this reveal if the h-stab's incidence is righteous or not. Before reading Kevin's article I was nosing around and found that I should retract my R flap 1 degree to match the left flap. That may be the cause of the rudder and aileron issue. But its windy and rainy. My local body shop will be painting my engine cowl. I drop it off tomorrow so I have a list of things I'll do with the cowl off.

 

[svyolo]

Here is a <2 minute youtube on pitot/static testing on the ground. There are longer ones but the technique is the same. Worked for me.

https://www.youtube.com/watch?v=DN-74HluxC0&t=514s

 

[Bcone1381]

Post #9 is dedicated to Engine Breakin. Breakin Theory, what I did, experienced, why, and learn about truth and best practices of Engine Breakin.

My primary focus up to now has been Engine Breakin. I have 10 flights and 15 hours on N62588. I wanted to see the CHT's drop and Oil consumption stabilize. Those are the two standard items to watch for. CHT change is easy to document. Oil Consuption is not.

There was no change in CHT's after 7 flights and 10 hours at 75% power. I consulted a few others and decided to try 100% power. I now have 10 flights/15 hrs completed and my data shows CHT's dropped 20+ degrees in all four cylinders. So I am moving on from break-in to final rigging, static port solution, and acoustic noise attenuation.

I will add Breakin data and thoughts later using the comment tool. Lets keep breakin discussion here to organize the data.

 

[Bcone1381]

A scientific data driven break-in procedure should exist that is widely accepted for a normally aspirated engine with data monitoring that produced no glazing and oil consumption of 1 qt per 25 hours. It does not.

Lycoming Break-in documents seem vague and have changed over time. The current Lycoming document never says to run the engine at 100% power except for take off. Lycoming SI 1427C dated in 2010 gives conditions of when its proper to do so. Mike Bush likes 100%. A plethora of places I research said that 100% power gives industry best results as long as CHT's and oil temps do not get too hot. No one said 100% power with normal temperatures was harmful. "Too Hot" is not well defined and opinions exist and some engines are not equieped to measure for it. I don't like that information for common installations is weak. I suspect that Lycoming's current procedure accommodates all aircraft engines and all airframe instrumentation. All engine installation are not equal. A normally aspirated engine with a engine monitor is different than a TSIO-540 without and an industry best practice for 360 CI lycoming with modern temp and pressure indications should exist.

 

[Bcone1381]

oil Comsumption...I don't really know what it is yet. I measure oil on my dip stick using a ruler in mm and the 6 qt. mark is my standard datum. Oil level varies with temp and time after shutdown. I think oil level on the dipstick can drift with temp induced oil expansion and case expansion. I am trying to document the oil level in MM and the oil temp after sitting 24 hours to find out my consumption without waiting for 10 hours of tach time.

 

[Bcone1381]

Post #10 is dedicated to Acoustic Noise Attenuation:

At full power my cockpit has high noise levels. The data...
-DC one-x headset: It is noisy with this headset. The intercom squelch and volume was drowned out by the cockpit noise at 100% power. Earplugs made little to no change. Earplugs and the headset made the comm radio volume unsat to talk to ATC. At 75% power it was okay.
-Bose A20: same for the Bose. Earplugs seemed to make no difference or even hurt the noise attinuation. At lower power the Bose seemed better with no earplugs than the DC One-X. More data is needed but I have enough to say very good heasets are overwhelmed by the noise at 100% power.

I am researching cockpit noise attenuation. My .090 acylic side windows drum. I felt them flex with my fingers maybe 1/8". Hard to measure. I am sure my .060 lexan skylight drums. The 180 hp engine, big prop, and vetterman exhaust (w/mufflers) combine to put my head at the center inside a big drum. I felt the side windows drumming and flexing. I will stiffen the acylic side windows, lexan skylight, and place acoustic soundproofing on the firewall, boot cowl, wingroots and floor.

This is my information source for soundproofing. Anyone have other ideas?

howtosoundproof.pdf

 

[gregc]

I tried a variety of approaches on a C172 and found that it is very hard to reduce sound levels without a lot of weight gain. My advice would be: don't believe sound proofing material vendor marketing claims, measure your sound levels and frequency distribution rather than rely on subjective impressions, target specific areas, and experiment with temporary, easy to remove solutions first. To be honest, I would consider a comfortable level at !00% power in my Companion (angle valve O-360) as unrealistic. YMMV

 

[Steve W]

Did you shorten the exhaust stacks? Longer the better to get the blast away from the belly. Have you dynamically balanced your prop?

 

[svyolo]

Post #10 is dedicated to Acoustic Noise Attenuation:

At full power my cockpit has high noise levels. The data...
-DC one-x headset: It is noisy with this headset. The intercom squelch and volume was drowned out by the cockpit noise at 100% power. Earplugs made little to no change. Earplugs and the headset made the comm radio volume unsat to talk to ATC. At 75% power it was okay.
-Bose A20: same for the Bose. Earplugs seemed to make no difference or even hurt the noise attinuation. At lower power the Bose seemed better with no earplugs than the DC One-X. More data is needed but I have enough to say very good heasets are overwhelmed by the noise at 100% power.

I am researching cockpit noise attenuation. My .090 acylic side windows drum. I felt them flex with my fingers maybe 1/8". Hard to measure. I am sure my .060 lexan skylight drums. The 180 hp engine, big prop, and vetterman exhaust (w/mufflers) combine to put my head at the center inside a big drum. I felt the side windows drumming and flexing. I will stiffen the acylic side windows, lexan skylight, and place acoustic soundproofing on the firewall, boot cowl, wingroots and floor.

This is my information source for soundproofing. Anyone have other ideas?

howtosoundproof.pdf

I had at least the same problem, if not worse. At full power (probably only 90%) there was so much noise I couldn't communicate. It was OK at lower power. I could hear the noise thru the headphones, which I misdiagnosed at first as ignition noise. So my first attempt at a "fix" was a noise suppressor on the power and ground of the radio. No help.

What it turned out to be was just the microphone (lightspeed and Bose did the same) picking up noise. The fix, which worked, was turning the mic gain, on the mic, all the way down, and the setting on my MGL V6 radio, also labeled "mic gain", almost all the way down. It is useable now.

For now I have AL floorboards. Between that and the boot cowl I am sure that isn't helping with the noise. One trick is to attach a small amount (1/2" strips) of sounding insulation to the AL to deaden some of the resonance. I have not done this yet, but very soon. You can buy it commercially as sound insulation (car stereo retail) or just use roofing tar paper, which I have used in the past. I will cut strips, and glue a few to the bottom of the floorboards and bootcowl. It should help at least a little.

I also put thin strips of Gorilla tape on top of the 4130 tubing where the floorboards sit.

There are also a few tailpipe "tricks" I may try.

 

[Bcone1381]

My advice would be: xxxxxxxx measure your sound levels and frequency distribution rather than rely on subjective impressions, target specific areas, and experiment with temporary, easy to remove solutions first. To be honest, I would consider a comfortable level at !00% power in my Companion (angle valve O-360) as unrealistic. YMMV

I like your "experiment temporarily with easy solutions" text. Is there a tool to measure sound intensity and freq? My phone can give me a Db reading. Not a Hz reading. Can calculate the frequency? (each rotation of the prop produces two prop pulses) ( 2700 rpm x 2 prop blades = 5400 pulses/min = 90 pulses per second, or 90 Hz?) How does freq data point help me? I like the data collection idea. I will do that. Then we can measure results of the experiments.

What it turned out to be was just the microphone (lightspeed and Bose did the same) picking up noise. The fix, which worked, was turning the mic gain, on the mic, all the way down, and the setting on my MGL V6 radio, also labeled "mic gain", almost all the way down. It is useable now.

For now I have AL floorboards. Between that and the boot cowl I am sure that isn't helping with the noise. One trick is to attach a small amount (1/2" strips) of sounding insulation to the AL to deaden some of the resonance. I have not done this yet, but very soon. You can buy it commercially as sound insulation (car stereo retail) or just use roofing tar paper, which I have used in the past. I will cut strips, and glue a few to the bottom of the floorboards and bootcowl. It should help at least a little.

I also put thin strips of Gorilla tape on top of the 4130 tubing where the floorboards sit.

There are also a few tailpipe "tricks" I may try.

​This post helped me to find the Trig Comm Post Installation Set-up for Mic Gain. I'll work on that.

Sound is waves of compression/rarefaction of elastic air. My acylic windows are thin and flexible. The prop induced pulse hits the acryic and it flexes inward like a stick hitting a drum. Sky light gives in too.. I felt it flexing maybe 1/8" with my fingers on the last flight. Its coming from three sides from just the acrulic/lexan and my ears are in the center of it. Stiffening the acylic/lexan will reduce or stop that part of the noise. Science says it's got to. I'll stiffen them up, test it, record the data, stiffen some more, and I'm sure end up with thicker acylic & lexan after I see how stiff they need to be.

I've got a good noise absorbing foam left over from me seat seat fabrication. That will get installed various places too.
I corked my fuselage tubing that my floor boards sit many moons ago. Kevlar I think might be better than Alum for sound wave absorbtion...a guess. Thats down the road. The exuast is the floor noise generator I suppose.

I think I'll loose hearing if I dont fix it and it would cause my my wife to avoid flying with me. Its fixable. "Build it light and get a good head set has limitations."

 

[svyolo]

My headset experience was buy one and use it. I never knew I would have to adjust it. I had to do BOTH the radio and Mic on the headset. Improvements were always on the table. I wanted to get it flying first.

 

[Nev]

My firewall was drumming early on. I added several beads of RTV to adhere it to the airframe. This made an audible difference. Also the lightweight carpets reduced the drumming on the aluminum floor panels.

 

[Astro John]

I used this product on the face of wall studs when I was building a theater room 15 or 20 years ago. I wonder about marking the tubing locations on the floorboards, and then sticking the tape on the floorboards: https://www.greengluecompany.com/noiseproofing-products/joist-tape

I also plan on doing carpet.

 

[rodsmith]

I placed 1/16th rubberized cork strips over the tubing before installing the aluminum floor. Can't say how much it helped. I used 5/8th open cell foam on the backside of the firewall and on top of the exhaust tunnel, and the grey insulated blanket material on the inside of the boot cowl. That is the extent of my sound proofing and even at full power I can talk on the radio and hear responses ok. My entire front doors are covered with .100 lexan and skylight is .125 lexan.

 

[svyolo]

The tunnel and Firewall are definitely going to get something.

 

[Bissetg]

Hi Brooks, I too had horrendous ambient noise levels. I was running straight exhaust pipes, no muffling. After trying all sorts of things and worrying about not hearing traffic calls I bit the bullet and got MotorSport Solutions to custom build me a 4 into 1 tuned exhaust.

Thats what made the real difference, I could dial back the volume on the radio 25% and still hear it better than before.

Soundproofing and going to Bose ANC headsets all helped but the big win for me was the exhaust.

Good luck, noise is annoying, fatiguing and a safety issue.

 

[Bcone1381]

Acoustic Noise:

The source of noise on my ears ended up being mitigated by Bose A20 headset without earplugs. I did not immediately identify that the noise overpowered the intercom using the David Clark One-X. The easy way to test this is to unplug the mic and experience quiet. I suppose one might blame my intercom manufacturer for not having a mic gain adjustment.
​
The Bose A20 with no ear plugs and max engine power is satisfactory. It seemed negatively effected with headset+ear plugs which I don't understand.

Additionally, I tested the noise source by controlling the drumming with a felt covered board then acoustic foam. I pushed it against each window, the floor, the sidewalls, and firewall to stop the drumming while I monitored the noise on an iphone acoustic sound app that measured Hz, and dB. There was no change heard or recorded.

Hot Fuel:

My engine has an issue running smooth when returning home from a flight and it's not cold outside. I experienced it in the traffic pattern and idling on the ground at 60F (15C). It's been diagnosed as hot fuel in the fuel lines (Bendix Fuel Injection system.) The fix is to get better airflow to the fuel pump cooling shroud and change the fuel injection plumbing so fuel lines are shorter. My tight baffling resulting in little extra airflow in the lower cowl may be a contributor also. We need more data after we fix those two items. Vetterman Crossover exhaust also has a lot of hot tubing plumbed in the area. We'll see. This is experimental aviation!

 

[Nev]

Brooks, regarding the hot fuel, when you notice it in the traffic pattern, is it with the engine at idle or under power ? Mine runs rough at idle on the ground when hot, but runs smooth when I increase to take-off power. In other words I've never noticed an issue inflight. Just a data point for you.

 

[Bcone1381]

I notice it shortly after I bring the power back to 1800 RPM when the OAT is not cold. I reduce power abeam the runway threshold on downwind. It seems the fuel flow decreases, so the speed the fuel travels through the system slows down and gives it time to cook. My firm data points are a heat soaked fuel pump that was hot to touch, and the Fuel Flow transducer mounted to the firewall that was hot as well. Those are Heak Sinks. One fix moves the transducer into cool air.

I would guess your issue of rough idle on the ground when hot is the same as mine, but not to the degree.

 

[Bcone1381]

Hot Fuel:

I moved the Red Cube Fuel Flow transducer off the firewall and re-designed the fuel line route to reduced the length of line going thru the cowl area. The Red Cube now resides in the cool high pressure side of the cowl between cylinders 1&3 surrounded in cool air when we are flying. I have fixed the hot fuel issue in flight at 60F OAT. I will stay vigilant as summer heats up.

You can see that my Patrol IO-360 installation has vertical induction and Vetterman crossover exhaust. This setup is not ideal. The best practice with Bendix is to keep fuel cool. So keeping the fuel injection lines short prevents heating up the fuel as it travels thru the engine cowl.

To have minimum length in my installation requires the lines are threaded them between the exhaust and oil sump. If you look for my red arrows you will see the issue. I avoided this installation in version 1 which had the fuel line exit out the back of the servo to the firewall mounted red cube. Then the fuel traveled up the back of the cowl area thru the aft baffle at cylinder 3 and to the flow divider. The hot firewall communicated heat to the red cube, and long lines soaked up more heat.

What you see here is typical industy best practice for the fuel line route and I am managing the risk of threading the fuel line thru the exhaust - oil sump pinch point. The exhaust shield is covered with fiberfrax and reflective aluminum tape to limit heat from radiating aft towards the sump. The fuel line is wrapped in fiberfrax to insualte and protect it. The line is extreamly secure and has clearance there. it cant touch anything (but its tight). I then installed a cooling tube to keep air flowing into the pinch point during flight.

How hot does it get there? We need that data to feel safe. It stabilized at 153F running 75% power and 1500F EGT. We gathered data using thermocouple placed in this choke point. One arrow points to one of the five thermocouples I installed on the fuel line fitting at the servo input.

Red Arrows below point out the modified heat shield, an example of a thermocouple (white/blue wire) , Loop Clamp for securing the line, and the cool air supply line.​

 

[Bcone1381]

I like data. Gathering heat data is easy and inexpensive. I never touched a thermocouple before last week. My thermocouples look like a 3 meter long single strand of 20awg wire. One end plugs into a meter and the other wire end is put in a specific place in the engine cowl to display data on the temperature meter. I took heat data from five points in the engine cowl during a 40 minute flight and learned a lot. I learned from Arborite (who is active here) to go gather data, and the common error is to try see the data while I collect it and or try to do something about the data while I fly, or change things while I fly to make data confusing. So we made up a flight plan and I stuck to it. Results were good data points now exist.

I learned that my concerns about cowl heat were from heat data I felt on my fingers after the engine cowl soaked up heat after a flight. I learned that my cowl temps during flight (OAT was 60f/15c) were about 125F. I saw fuel temp data as it entered the cowl area, passed thru the fuel pump, and to the servo. I saw the temperature at the area where the fuel line is pinched between the exhaust and oil sump. We figure that the fuel temp issue in flight is possibly solved, and the trend on the engine oil temperature will be unsat as summer time temperatures arrive.

The last attachment has a graph of the temp data that Arborite put together.

I built a Patrol. But what do I know? It seems like decisions I made when I built were backed up with data I gathered and were not just made up. I thought building was complete when I performed my condition inspection in January. ha! Nope, I'm still building. We build, we fly, we learn and make adjustments to perfect our craft. That is experimental aviation. Maybe we don't stop crafting and adjusting our ship, trying to make it a bit better....something you cant do with a type certificated production design.

 

[rkennell]

Brooks: For some reason the photos did not download correctly.

 

[Bcone1381]

A recipe for Heat

I had a 7 row oil cooler. I removed it and replaced it with a 13 row oil cooler due to 200 degree oil temperatures at 60f (15C) OAT. That larger cooler increased the cowl air temperature near the oil cooler from 125f to 150f. Now I can fly at OAT of 75 - 80F at reasonable oil temperatures.

The Vetterman crossover exhaust has attributes. It's quality and my Patrol really goes! But I've got about 8' of exhaust in my cowl. If my EGT's are 1450F those pipes are hot (guess is +900F?), and radiating heat to the rest of the cowl.

My fuel system is Bendix mechanical fuel injection. It is simple and has lots of attributes over a carburator, but it has no fuel return system, and at low power fuel travels slow through about 4 ft of lines. Fuel that heats up to 140f starts (the expert says) causing issues.

The Patrol has a smaller firewall than a side-by-side aircraft so the cowl area has less volume that any other Bearhawk.

I built my baffling very tight because air that does not cool the engine is wasted drag. This decreases air going thru my cowl than if it had a less efficient installation.

When I fly at higher power settings everything is fine. When power is then reduced fuel speed thru fuel lines slows down but engine heat starts to decrease and the fuel has time to heat up.

Larger oil cooler + Vetterman crossover exhaust + Bendix + a smaller cowl volume + tight baffling that maximizes efficient cylinder cooling = warm fuel when fuel flows are reduced.
_______________
I fabricated fittings on a lathe to let me attach a 1/4" hose on both sides to measure cowl pressures. If I know what cowl pressures are above and below the engine, then I can learn how a cowl modification effects the pressure differential betweenthe upper and lower cowl areas. For example, I'll add a lip at the cowl exit when I get this set up and see it's effect on the pressure differntial.

What I dont know....
-does the fuel in the line under the floor that goes to the firewall get heated by the tunnel?
-How much, and If so can I insulate it?
-If ceramic coated exhaust decreases temps by 25% will it fix my warm fuel issue?

Yesterday I removed the exhaust and sent it out to have it ceramic coated.

It seems to me like a bunch of what I believe are industry best practice decision making results in cowl heat. I've not seen folks talk about this. less airflow, more heat, with a injection system sensitive to it.

On the bright side...my CHT are good. Yesterday I saw 340-360F with 8F spread between them.

My next steps...brainstorming now...
-Ceramic coat the Vetterman exhaust.
-Add a lip on the cowl exit.
-change the cowl exit design...double the exit area.
-replumb my fuel supply system so the firewall exit is closer to the fuel pump to decrease the distance that the line travels in a hot cowl area by 10".
-I have two fuel lines....firewall to pump, and pump to the servo....they add 20F to the fuel temperature at the servo. If I put a shroud around one or both lines and blow ram air thru it maybe I can get some joy. But that seems like a band-aid to fix a fundamental issue.
-Take excess upper plenum air and shoot it all around the lower plenum.

I build slow. I want to fly to Oshkosh. Its okay if I don't. Time pressure makes me grumpy, removes the fun, and makes working on the Patrol onerous. So I'm not counting on it anymore.

​

 

[Nev]

I'll add a lip at the cowl exit when I get this set up and see it's effect on the pressure differential.​

Brooks, do you have a cowl lip in place at present (of any size) ?

Early on, I was experimenting with different sized cowl lips while flying with a manometer. I decided to remove the lip completely for one flight and establish a base line. The result was that my CHT's temperatures increased faster than normal and to higher than I would have expected early into the flight, so I only completed a large circuit before landing and refitting the cowl lip. So if you don't have one at all, I think you will be pleasantly surprised at the difference.

 

[Alaskanf1]

It seems like your main concern at this point is fuel temperature? You could run the fuel lines through some SCAT hose (1"?) and have ram air through that hose. I recall seeing a certified plane that did this (C-206?) for the line from the servo to the spider. The SCAT hose went from the baffling to the servo (which also blew some cool air on the servo).

 

[svyolo]

I wonder how much heat the mech fuel pump itself is inputting to the fuel. I am making a couple of smaller changes (eliminating two aluminum bulkhead fittings thru the baffling) to mitigate the same thing. I have EFI feed and return lines going thru the upper baffling. I will use hogged out plastic spark plug passthru in their place.

I have always wondered why planes don't use hot air coming off the oil cooler for cabin heat. It seams much safer than heat muffs, and simpler and lighter. I removed one heat muff (for engine cooling purposes) and will give it a shot. The SCAT hoses were routed in such a way they probably detrimental to airflow out the exit.

I also moved my oil cooler hose flange 2" up on the baffling. I have a hot #2 and 6, cold #1. On average they are OK.

 

[Bcone1381]

I have no lip on my cowl exit yet. My plan is to set up my manometer and get a baseline first. Then I'll fit up a lip and measure the change with the lip and what size angle and shape maximizes pressure differential.

 

[Bcone1381]

You read my mind! You are correct that at this point fuel temperature is the concern. I now understand why Lycoming made an oil sump with a rear entry inlet for aircraft like the Piper Arrow. That makes fuel lines nice and short.

 

[Bcone1381]

I wonder how much heat the mech fuel pump itself is inputting to the fuel. I am making a couple of smaller changes (eliminating two aluminum bulkhead fittings thru the baffling) to mitigate the same thing. I have EFI feed and return lines going thru the upper baffling. I will use hogged out plastic spark plug passthru in their place.

I have always wondered why planes don't use hot air coming off the oil cooler for cabin heat. It seams much safer than heat muffs, and simpler and lighter. I removed one heat muff (for engine cooling purposes) and will give it a shot. The SCAT hoses were routed in such a way they probably detrimental to airflow out the exit.

I also moved my oil cooler hose flange 2" up on the baffling. I have a hot #2 and 6, cold #1. On average they are OK.

Fuel Pump......The fuel pump indeed heats up fuel. My pump has a cooling shroud fed air to it by 1" SCEET fed from the upper plenum. On the ground at idle after I land the fuel is moving slow due to idle fuel flows so I doubt much air flow. Normal cruise temp rise thru the pump is about 9F. Fuel temp rises from 107F at the in fitting to 134F at the out fitting. At 20"/2250 and 5 gph the temp rise thru the cooler was 17F. At 23"/2300rpm and 9 gph temp rise thru the cooler was 5.5F. Moving Fuel doesn't have time to heat up. I visualize a hotter engine that is trying to cool and lower fuel flow going slower in the hot pump soaking up heat.

Heat Muffs....The shrouds (not the muffler) is getting Ceramic coated along with all the pipes. Then I'll take my temp data sample, then remove one shroud and take another data sample.

Fittings......You are right to remove mass in the fuel lines by removing the bulkhead fittings. Your Aluminum bulkhead fittings transmit heat efficiently, and they soak up the heat from the baffles if they are hot. The Red Cube is another heat sink. I had mine on the firewall. That firewall gets hot. The cube soaks up the heat.

I am asking myself "If the Red Cube has a positive head pressure of fuel and if it fuel flows to meet the 125% then why cant I move it into the cabin? Today I did a fuel flow test and it flowed 25 gph with the cube. I only need 20gph. So unless EI can tell me why its a bad idea (I asked them this morning) it is getting moved.

Oil Cooler Heat....I know of a gifted fabricator/builder who built a Harmon Rocket. He did used the cooler as his heater and it worked out well for him. I happen to know my engine need no oil cooler at 30F. Burrrrrrr. Gotta keep my wife warm on a cold day.

 

[Nev]

I wouldn't be surprised if an exit air cooling lip resolves several issues all in one go. Might be just the biz.

 

[Nev]

Brooks I would install the cooling air lip first before moving the Red Cube. Reason is, they get quite inaccurate when installed aft of a fuel pump (or between fuel pumps) - I found that out the hard way. There are a number of references on the forum that show the transducers installed aft of the firewall. I did the same in good faith, not realising that several of those weren't working reliably. A phone call to the manufacturer resolved the issue and mine is now mounted on the engine mount, and the issues immediately resolved.

 

[svyolo]

I am also going to Fab a SS mount (vice AL) to reduce heat transfer to an EFI version of the Bendix flow divider. I don't know what is used to mount the Red Cube. This is unique to SDS and doesn't exist in the EFII type.

What I will say now might make me a heretic on a GA forum. Flame suit on. The best way to improve a Bendix style MFI system is delete the engine driven fuel pump, and just install 2 electric boost pumps, in series.

Hot start problems reduced or eliminated. Fuel heating problems reduced. Fuel pump replacement cost is 120$ instead of 1500$.

But, I know, electrically dependent.

I think overall it is a reliability improvement, and cheaper. And Bendix works really well other than that.

 

[Bcone1381]

Thanks for your feedback Nev. I'll follow your advice and hold off moving the Red Cube.

 

[Bcone1381]

It seems like your main concern at this point is fuel temperature? You could run the fuel lines through some SCAT hose (1"?) and have ram air through that hose. I recall seeing a certified plane that did this (C-206?) for the line from the servo to the spider. The SCAT hose went from the baffling to the servo (which also blew some cool air on the servo).

Welcome to the Forum, Alaska!! I had not visualized the installation you saw on the C-206. I had in mind putting the SCEET around the hose from the pump to the servo. I fit up the C-206 today with some 1" SCEET and boy is the fit nice and simple. Very Nice!! Thanks so much for your collaboration.

 

[svyolo]

Brooks, I just checked my fuel temps. I have EFI with a 1/2 gallon header under the right front seat. I probably have 3-5 times as much fuel line North of the firewall than Bendix FI. I have both electric fuel pumps, fuel pressure regulator, feed and return lines. All firesleeved Aeroquip hoses/AL hose ends. My pumps are flowing 43 gph continuously, so the fuel is making 3-4 passes thru the engine compartment, and back to the little header tank. I can feel the temp of the tank right next to me. In flight, I don't think I am getting more than a 10 deg F increase in fuel temp at the header tank. And that is with 3-4 round trips thru the engine compartment, that has probably 15 feet of fuel line.

Extended ground running, I am probably seeing 30 deg increase. But slow or fast cruise, 10 over ambient.

My attempts at improving cooling and MAP were in vain. Looks like cowl flaps are in my near future.

 

[rkennell]

Svyolo do you have a cowl lip installed?

 

[svyolo]

Yeah I have a 1.5" lip set at 30 degrees.

 

[Steve W]

Loop You might need 2.5-3” lip and more angle.
And check your timing. If it’s at 25*, maybe back it off to 20*. I know that was a fix for some of the angle valve engines in some airframes

 

[svyolo]

Angle valves are normally set at 20, but it might also work on mine. I have EFI/EI and can adjust the timing per rpm and MP. I may try dropping the high MP down to 20. I heard it makes little difference in power at very high power.

 

[Bcone1381]

The Patrol has been down a month due to service and work. The two issues wer of course hot cowl, but also starter kick back.

I have a lot to write, but just thought I'd share real quick some data. I flew yesterday and learned that the air temp under the floor is warm. The tunnel heats it up. I measure the bulkhead temp as my air temp indicator. It reached 107F. The fuel line temp entering the "basement" was ~ 83f and it exited ~96f rise thru this area was 13f @ about 9 gph and rises to 15F @5 gph.

I'll be insulating those lines and the Gascolator today.

I also got Cowl Pressure data with my unmodified cowl. Now I can make changes and see the change.

 

[rodsmith]

The Patrol has been down a month due to service and work. The two issues wer of course hot cowl, but also starter kick back.

I have a lot to write, but just thought I'd share real quick some data. I flew yesterday and learned that the air temp under the floor is warm. The tunnel heats it up. I measure the bulkhead temp as my air temp indicator. It reached 107F. The fuel line temp entering the "basement" was ~ 83f and it exited ~96f rise thru this area was 13f @ about 9 gph and rises to 15F @5 gph.

I'll be insulating those lines and the Gascolator today.

I also got Cowl Pressure data with my unmodified cowl. Now I can make changes and see the change.

I insulated the top and sides of the tunnel area with the same 5/8" closed cell foam I used on the inside of the firewall. I didn't want the heat radiating up through the floor, didn't consider the fuel line traversing that area.

 

[Bcone1381]

Fuel Temp As my fuel wandered thru the fuel system I saw the temp was increasing under the floor of the cabin 13f due to a hot tunnel I suppose. So I insulated 80% of the fuel lines and the gascolator and now have fuel increasing 3F under the floor. I place the temp sensors on fuel tee above the floor before the fuel enters the "basement", a sensor on a line as it exits the basement, one on the bulkhead forward of the gascolator as an air temp indicator, and a sensor on the gascolator itself.

Cowl Lip and Piccolo Tubes. I installed a 1.75" wide lip angled at 45 degrees at my cowl exit. My exit area is 90 sq inches. (Sven's DA-40 w/ a 180 hp IO-360 in a nearby hanger seems to have an exit area of about 80 sq in.) I installed piccolo tubes in my cowl. Two tubes were installed above the cylinders, and two below. the procedure has me connecting hose from them to an airspeed indicator to measure the pressure change, then I use a table to convert the indication to inches of water pressure.

Cowl Pressure: At 100 KIAS my cowl pressure went from 5.67 in h20 to 7.62 in h20. That is an increase of 34% pressure differential across the cylinders and oil cooler.
​
Oil Temp: My oil temp with a big oil cooler went from 200-205f (with an OAT of 65 to 80f) down to about 182f with an OAT of 86f on the ground.

When I taxi to the hangar after I land my engine would stumble and miss and burp due to hot fuel in the system. Last month I was hearing it do that in the traffic pattern when I reduced power abeam of the numbers at 70 F OAT. Yesterday it was about 80 F and it ran smooth during the taxi in. I never thought I'd get the fuel that cool.

 

[jaredyates]

Great news Brooks, how did the differential cowl pressure reflect in the CHTs?

 

[Nev]

Good work Brooks - a differential of over 7" at 100 kts is excellent. I'd be interested to see a photo of your lower cowling exit lip.

 

[Bcone1381]

Great news Brooks, how did the differential cowl pressure reflect in the CHTs?

Good Question. I dont have apples to apples data but we can get close. I'll keep it quick, but the school house might offer some argument. Call it 30 F decrease. At 24"/2400 rpm I saw 380 F before and 360 F but since I can lean it now without getting into vaporization of fuel I can run it leaner which mask the results. SO its fair in my laboratory to call it 30 F not 20 F.

Photo shows the Minimally Viable cowl exit lip version 1. 1.750" lip at 45 degrees. Yes, that is Gorilla Tape.


I am still experimenting. I want more. I have an idea that if the lip can introduce vorticies at flying speed then when it mixes with the exit air it will draw more out. Since the deflector is proven I copies it for Version 1.

Version 2 has a saw tooth edge. The B787 and B747-8 inspired this idea. We'll see. It might fail to produce. It might work. I'll keep you updated. I'm also curious about a saw tooth trailing edge of the cowl without the 45 degree angle.

 

[Bcone1381]

Can you see all three photos?

 

[Viking]

Hey Brooks, I’m getting ready to start my baffling installation on my patrol, and I’m curious about your larger oil cooler installation. Is your 13 row cooler mounted behind the #4 cylinder on the baffling? If so, how is the figment compared to the 7 row? Do you have any pictures of your cooler installation you could share?

 

[Bissetg]

Brooks, this is an interesting concept and I’m following closely! Please keep the info coming on this one.

 

[Bcone1381]

This is my response to Viking's comment in post 35.2

Viking;

My 13 row cooler install is a minimally viable installation. Its packaging is an after thought and not an industry best practice and I don't recommend you follow my installation. There were numerous workaround space issues that I dealt with. The Packaging of the install at your stage was hard for me to get my head. Then making changes we learn that we wished we packaged it differently.

You must attach that environment to this 13row cooler installation. Here is a Question I’m now asking... with my improved pressure differential do I need a 13 row cooler now? Would the origonal 7 row cooler work, or a 9 row? The 13 row choice was made with this attitiude "Go big to know the issue is resolved." The O-540 often uses the 13 row cooler.
​
I like this placement of an oil cooler.

https://bearhawkforums.com/forum/bearhawk-lsa-quickbuild/95635-north-carolina-lsa-kit-build-underway

(in that install, one might Consider a device to allow the air flow to gently transition from the SCAT to the cooler, but I like this placement/packaging)​



Air enters thru the baffle behind #4 cylinder where I had my 7 row cooler attached. Baffle seal material is attached to what you see so that it seals the area when my L cowl door is closed. I was able to use the same oil lines that served my 7 row cooler. Its a pain to remove and install but I'm flying.

 

[gregc]

When trying to decide whether or not to have a cowl lip on my airplane and if so, how big and at what angle, I wondered why there is so little consensus on this issue. This paper by John Thorp helped me understand the basic issues:

https://www.scribd.com/document/536076050/Cowling-and-Cooling-of-Light-Aircraft-Engines-J-Thorp

Thorp was not fan of a large stalled flap on the outlet and pointed out that separated airflow is largely unpredictable (or at least was for folks in 1963 without digital modeling systems and still is for most of us) and that it may well create an area of high pressure just where the pressure needs to be low. I suspect this is why something that works on one cowl design does not work on another even though it appears similar.

The other key point in his presentation is that cooling drag is a big deal and that even well designed systems require 8-12% of available engine power for cooling so there is plenty of opportunity to make things worse!

In the end I concluded that there is no "right" answer. Perhaps over time we can accumulate a set of configurations that have worked experimentally but I would not be surprised to find that they differ for each Bearhawk model.​

 

[Bcone1381]

Here is the feedback on the Cowl Lip....

The standard cowl lip (Version 1 viewd in post #37) had better performance than the Saw Tooth lip. At 100 kts, the Pressure differential dropped from 7.38" to 6.56" with the Saw Tooth.

 

[Viking]

Thanks for the update and the pictures!! That’s a tight area to fit the big cooler in, so I was curious how you made it work.