An extraordinarily bizarre airspeed mystery that no A&P can solve

[kicktireslightfires]

I have a minor mystery of the universe on my hands. My airplane always flew at 120+ KTAS (120-126). Then one day, a few weeks ago, it began flying about 10 knots slower INDICATED and no one can figure out why. It's as if I'm now towing an open parachute while I fly. I have spoken to multiple A&Ps and everyone is stumped so I'm hoping maybe someone on here has a bright idea we haven't thought of.

As long as it's not against the rules here, I will send one $100 Starbucks or Amazon gift card (your choice) to whoever is the one who figures this out or prompts the question that figures this out! (Moderator: if this is against the rules, please just delete this section of my post rather than deleting the whole thing.)

It's a Rotax 912iS engine with about 500 hours SNEW, ground-adjustable prop, with Dynon SkyView D1000 Touch PFD and MFD. E-LSA (was S-LSA) low wing.

Same power setting, same RPM, same fuel, same altitudes, same weights.

At the bottom of this post is a table showing the before that I flew with for months after getting a new prop installed, and the after showing what I'm suddenly getting now and have been every day since it all started when we replaced the Dynon GPS antenna -- it had poor signal strength so we swapped it out for a new one. But we cannot identify any connection between replacing the GPS antenna and having the IAS being different.

The groundspeed is accurate -- I cross-compared it to the groundspeed my iPad and iPhone are showing so that's double verified.

The indicated airspeed showing on the PFD perfectly matches the standby airspeed indicator.

I just had a pitot test conducted and the $60,000 machine was pumping 120 knots into the pitot tube and my Dynon PFD was showing precisely 120 knots IAS. The pitot-static testing guy was also completely stumped by this mystery of how I have lost 10+ knots of airspeed. He did say while the pitot was perfect, I was registering a 200fpm leak on the static line, but for an unpressurized VFR-only airplane, it's meaningless.

Since I had a new prop installed a few months back, I also verified that the prop pitch didn't change (suspected not anyway because the RPMs were still all the same). Prop pitch is still exactly what it was when it was installed down to the tenth of a degree.

I thought maybe perhaps the engine is producing less horsepower despite the same RPM, but my A&P who is intimately familiar with Rotax says no.

EGTs are the same before and after.

I thought perhaps maybe I was always flying at these slower speeds and I was just seeing artificial speeds. But that can't be because my TAS was always accurate when I compared to the groundspeed and winds aloft, and my ETE times always bang on.

Next you might be thinking my winds aloft may have been inaccurate due to a calibration problem with my digital compass. But my ADAHRS primary and secondary were properly calibrated before this airspeed loss happened and while I was still flying at 120+ KTAS. So it couldn't be a compass calibration problem, either.

The only other thing changed was the rudder trim tab was given about 15 degrees more of angle to max it out as I was about half a ball out of coordination in cruise. That extra 15 degrees of deflection on the rudder trim tab got it a little better, but still about a quarter of a ball out when in cruise. YES, the extra rudder trim will add drag, but everyone I've talked to agrees there's no possible way it could cause 10 knots worth of drag and maybe just a knot or two. Is there anyone out there who thinks 15 or even 20 degrees of additional rudder trim angle could really slow the plane down by 10 knots? If so, I'd be very curious to hear your reasoning. The only reason we haven't bent it back the other way to test is because it's aluminum and if we bend it back and find it didn't bring the airspeed back, then we bend it BACK to how we had it, that will probably seriously weaken it and risk it snapping off. The rudder trim tab on my plane is already pretty aggressive. But the fact that I was flying at 120+ KTAS with it just ~15 degrees shy of it being maxed out makes me believe it's nearly impossible for just going ~15 degrees more to max it out equate to a 10 knot airspeed reduction.

Truly cannot figure this one out. Have spent many hours and hours and multiple sleepless nights racking my brain. I'd greatly appreciate any help as it just doesn't make any sense to anyone how an airplane can instantly lose so much airspeed.

Please hit me with every question you can think of so we can brainstorm this.

 

[Daleandee]

Last time I had a similar occurance it was because I forgot to pickup the 10º of flaps I used for take-off.

I do notice that the cooler weather gives me better numbers than the high DA & temps of the summer months.

Appears you have covered the possibilites pretty well. I'm interested to see what the answer is ...

 

[kicktireslightfires]

Last time I had a similar occurance it was because I forgot to pickup the 10º of flaps I used for take-off.

I do notice that the cooler weather gives me better numbers than the high DA & temps of the summer months.

Appears you have covered the possibilites pretty well. I'm interested to see what the answer is ...

Haha, yes I've forgotten to raise flaps once or twice, but that's not the problem here. 0 degree flap reading on PFD and verified with looking over both wings that flaps are always at 0. And yes, the cooler weather should have been giving me even better numbers. I was flying at 120+ KTAS in the dead of Florida summer. Now that it's cooler out, I should be seeing even slightly better, not 10+ knots worse.

 

[midwestpa24]

Well we did just go through Thanksgiving and Christmas...did your weight and balance change?

 

[kicktireslightfires]

Well we did just go through Thanksgiving and Christmas...did your weight and balance change?

Merry Christmas!

 

[charheep]

Rigging all good?

 

[Half Fast]

I was registering a 200fpm leak on the static line,

There may be a clue here. Post a diagram of your pitot-static system.

Question for you - could your altitude be incorrect instead of IAS? Do you have a before & after comparison of altimeter reading vs GPS altitude?

 

[wanttaja]

Remember that the airspeed does not just measure the pressure at the pitot. It measures the pressure differential between the pitot and the static line. If you've got a leaky static, it'll affect both the altimeter and the airspeed.

Which gets worse as the airplane speeds up, and as it climbs.

Ron Wanttaja

 

[Jeff767]

The first thing I would do is double check your true airspeed. The link below gives an accurate method to determine what the aircraft is doing.

True Airspeed Calculator VIA GPS

Winds posted by non inertial systems like the Dynon are notoriously inaccurate. I would view those winds as a mere suggestion if that. Once you determine if the aircraft is performing at book values or is actually slower you can seek out a fix. I suspect you are making book values and simply getting bad data from the D1000. Proper IAS display requires both accurate pitot data but also accurate static data. Your static leak might be a hint as to what is happening. I would also take a good look at the entire static system. Make sure there is no water in the system and there are no static port obstructions. Even a bit of wax could cause what you are seeing. Carefully insure nothing has changed forward of the static ports that might have created an airflow change. This can also have a big effect on IAS.

Again step one is to use the calculator posted and fly the aircraft to get the exact data. Your current methods are not going to give you the accuracy you need. If in fact the aircraft is slower you are either down on power or picked up airframe drag. Given you also report rudder trim issue if this is a nosewheel aircraft I would take a close look at it. It’s possible you have an issue that is allowing the nosewheel to rotate right or left as airspeed increases. That will create yaw that has to be countered by the rudder and create drag. Make sure everything on the nosewheel is up to spec including preload if it is that type.

 

[nrpetersen]

Ding or bird poop on the leading edge or top of the wing? Ding on leading edge of the propeller? It does appear to me that something aerodynamic is at fault. The new need for rudder trim suggests it may be one wing, and closer to the tip rather than the root.

 

[iamtheari]

Did the pitot mast get bent or twisted?

The obvious possibility is that the static leak is related, either as a cause of the airspeed error or as a related symptom of the underlying cause. Understanding the static system might let you formulate some experiments using alternate static air or other tools to diagnose the problem.

 

[kicktireslightfires]

There may be a clue here. Post a diagram of your pitot-static system.

Question for you - could your altitude be incorrect instead of IAS? Do you have a before & after comparison of altimeter reading vs GPS altitude?

Attached is a diagram of my pitot-static system. My altimeter reading and GPS altitude has always been within 75 feet of each other before and after the airspeed change. Some times it's within 15 feet, other times more, but I've never seen it more than 75 feet apart.

 

[kicktireslightfires]

Remember that the airspeed does not just measure the pressure at the pitot. It measures the pressure differential between the pitot and the static line. If you've got a leaky static, it'll affect both the altimeter and the airspeed.

Which gets worse as the airplane speeds up, and as it climbs.

Ron Wanttaja

Ron, the pitot-static technician told me many VFR-only unpressurized airplanes just have a 6 inch tube coming off the back of their altimeter and effectively taking in cockpit static pressure. So could a leak in the static really be causing this?

 

[iamtheari]

Attached is a diagram of my pitot-static system. My altimeter reading and GPS altitude has always been within 75 feet of each other before and after the airspeed change. Some times it's within 15 feet, other times more, but I've never seen it more than 75 feet apart.

It looks like you have both steam and glass altitude and airspeed indications, and the glass is dual redundant. Are all 3 of the altitude indications and all 3 of the airspeed indications the same (to within the degree of calibration you had before the anomaly began)?

 

[Zeldman]

Anything crawl up the pitot tube and die.??

 

[Salty]

Ron, the pitot-static technician told me many VFR-only unpressurized airplanes just have a 6 inch tube coming off the back of their altimeter and effectively taking in cockpit static pressure. So could a leak in the static really be causing this?

You're asking about a change. If a leak was introduced, this would certainly change things. There can be a big difference between ambient under the console, and ambient where the port outside the plane is. Heck, there's a huge change in pressure in the cabin if you have a small vent open or not, which is why it's very rare to not have a static port outside the plane.

 

[kicktireslightfires]

The first thing I would do is double check your true airspeed. The link below gives an accurate method to determine what the aircraft is doing.

True Airspeed Calculator VIA GPS

Winds posted by non inertial systems like the Dynon are notoriously inaccurate. I would view those winds as a mere suggestion if that. Once you determine if the aircraft is performing at book values or is actually slower you can seek out a fix. I suspect you are making book values and simply getting bad data from the D1000. Proper IAS display requires both accurate pitot data but also accurate static data. Your static leak might be a hint as to what is happening. I would also take a good look at the entire static system. Make sure there is no water in the system and there are no static port obstructions. Even a bit of wax could cause what you are seeing. Carefully insure nothing has changed forward of the static ports that might have created an airflow change. This can also have a big effect on IAS.

Again step one is to use the calculator posted and fly the aircraft to get the exact data. Your current methods are not going to give you the accuracy you need. If in fact the aircraft is slower you are either down on power or picked up airframe drag. Given you also report rudder trim issue if this is a nosewheel aircraft I would take a close look at it. It’s possible you have an issue that is allowing the nosewheel to rotate right or left as airspeed increases. That will create yaw that has to be countered by the rudder and create drag. Make sure everything on the nosewheel is up to spec including preload if it is that type.

Interesting calculator, Jeff. Thank you. I will try that on my next flight.

Also, very interesting suggestion regarding the castering nosewheel and that perhaps it is going askew while in cruise. Nothing was changed on the nosewheel though so I can't fathom that it would suddenly be going askew. I will get a GoPro with a suction mount so I can stick it on the belly to verify what the nosewheel is doing in cruise.

 

[kicktireslightfires]

Anything crawl up the pitot tube and die.??

It's a brand new pitot tube as of 2-3 months ago (before airspeed loss) and always covered, so no. I also took a very close look with a flashlight and no obstructions found.

 

[kicktireslightfires]

Ding or bird poop on the leading edge or top of the wing? Ding on leading edge of the propeller? It does appear to me that something aerodynamic is at fault. The new need for rudder trim suggests it may be one wing, and closer to the tip rather than the root.

No dings or bird poop. It's an extremely clean airplane and ceramic coated. The prop is brand new and carbon fiber. The only place on the airplane I noticed a ding was towards the root of the horizontal stabilizer, but I don't see how that could have any effect.

 

[kicktireslightfires]

It looks like you have both steam and glass altitude and airspeed indications, and the glass is dual redundant. Are all 3 of the altitude indications and all 3 of the airspeed indications the same (to within the degree of calibration you had before the anomaly began)?

They are all the same, indeed. I have cross-compared both ADAHRS and they agree on airspeed exactly, altitude within 60 feet, and heading within 2-3 degrees. Standby airspeed is exact to both ADAHRS. Standby altimeter is within 50 feet.

 

[Half Fast]

The airspeed indication is derived from both the pitot and the static lines, and it was calibrated without a static leak. Try fixing the leak and see if it makes a difference.

 

[kicktireslightfires]

You're asking about a change. If a leak was introduced, this would certainly change things. There can be a big difference between ambient under the console, and ambient where the port outside the plane is. Heck, there's a huge change in pressure in the cabin if you have a small vent open or not, which is why it's very rare to not have a static port outside the plane.

I will get the static leak fixed. But two reasons why I don't think that's the problem:

1. I seem to be actually flying 10+ knots slower, not just an instrument error. Evidenced by my groundspeed. But I'll do the calculator test Jeff suggested to verify.

2. If I had an instrument error and was not actually flying slower, I suspect it would effect the quality of my landings. Because if my instruments are showing 55 KIAS and I'm actually going 65 KIAS, I would have significant float on landing and I'm not. My landings seem to be the same as they've always been at the IAS I always fly them at.

 

[Bell206]

what I'm suddenly getting now and have been every day since it all started when we replaced the Dynon GPS antenna

The only other thing changed was the rudder trim tab was given about 15 degrees more of angle to max it out as I was about half a ball out of coordination in cruise.

Do I understand correctly these are the only 2 changes you performed immediately prior to the loss of indicated airspeed?
Did either change take you in the same area of the static or pitot systems?

I just had a pitot test conducted and the $60,000 machine was pumping 120 knots into the pitot tube and my Dynon PFD was showing precisely 120 knots IAS.

Was the machine connected to the pitot tube AND the static system during this test?

He did say while the pitot was perfect, I was registering a 200fpm leak on the static line, but for an unpressurized VFR-only airplane, it's meaningless.

While a 200 foot leak may be moot to a VFR aircraft, it can be the difference of 10 mph to an airspeed indicator especially given the acceptable error rates for the indicator can range from 3 to 5 mph.
Is the airspeed indicator a TSO item or non-TSO?
Is the airspeed ind same age as aircraft or was it installed used?

My initial SWAG is this might be an internal leak in the airspeed indicator but might be able to narrow it down with answers to the above.

 

[midwestpa24]

It's a brand new pitot tube as of 2-3 months ago (before airspeed loss) and always covered, so no. I also took a very close look with a flashlight and no obstructions found.

That would raise my suspicions. Work was recently done to the pitot static system. A couple of months (and unknown flight hours) later there seems to be a pitot static issue. I would definitely start looking there. Something could have come loose, something could have found a way into partially blocking the system, something could have kinked after flying a bit.

 

[kicktireslightfires]

Was the machine connected to the pitot tube AND the static system during this test?

It was indeed connected to both the pitot tube and static tube (which is right underneath the pitot tube). There's no fuselage mounted static port on my airplane. It's just a static tube directly underneath the pitot tube.

The standby ASI is non-TSO, but both Dynon ADAHRS are TSO. And all 3 indicate the same. Nothing was installed used.

 

[kicktireslightfires]

That would raise my suspicions. Work was recently done to the pitot static system. A couple of months (and unknown flight hours) later there seems to be a pitot static issue. I would definitely start looking there. Something could have come loose, something could have found a way into partially blocking the system, something could have kinked after flying a bit.

I agree and we've checked the connections at both ends. Something actually was kinked after the pitot tube was replaced. I wasn't getting airspeed on the takeoff roll so I aborted and we found that the static line had kinked just a couple inches above the static tube on the wing where it makes the 90 degree turn inboard. We replaced that section of vinyl tubing with a stiffer walled polyurethane tubing and it solved the problem; no more kink there and airspeed returned.

A leak is suspect at this point, but how could a blockage be suspect? If during the pitot-static system test I was getting a 200 fpm leak rate, wouldn't that mean there's a tear in the line somewhere, and not a blockage? Because if it was a blockage it would clog up and be more pressure not less?

 

[kicktireslightfires]

Also, a leak in the static system wouldn't make me *ACTUALLY* fly slower. Merely make the instruments say I'm flying slower. So if the groundspeed checks and I'm *actually* flying slower, then it can't be a leak in the static system causing this problem. Right?

I will conduct Jeff's True Airspeed Calculator VIA GPS test on my next flight to double verify I'm actually flying slower.

 

[iamtheari]

I will get the static leak fixed. But two reasons why I don't think that's the problem:

1. I seem to be actually flying 10+ knots slower, not just an instrument error. Evidenced by my groundspeed. But I'll do the calculator test Jeff suggested to verify.

2. If I had an instrument error and was not actually flying slower, I suspect it would effect the quality of my landings. Because if my instruments are showing 55 KIAS and I'm actually going 65 KIAS, I would have significant float on landing and I'm not. My landings seem to be the same as they've always been at the IAS I always fly them at.

#2 doesn't rule much out because the error could be less at lower altitude and is almost certainly less at lower airspeed. So if it's reading 10 knots slow in cruise, it might not read even 2 knots slow at Vs.

That being said, #1 was not clear to me from your first post. Sorry about that. You should do a triangular groundspeed-based flight test to get a better idea of what's going on. But if the plane really is going 10 knots lower true airspeed, then you have a problem. Rigging, powerplant, or both.

 

[kicktireslightfires]

You should do a triangular groundspeed-based flight test to get a better idea of what's going on. But if the plane really is going 10 knots lower true airspeed, then you have a problem. Rigging, powerplant, or both.

By "rigging" do you mean rudder cables out of plum, or something else?
Hard to believe there would be any problem with the powerplant because we have been backwards, forwards, and sideways over the engine this year.

 

[Bell206]

a leak in the static system wouldn't make me *ACTUALLY* fly slower.

Well that was your first discrepancy noted above as shown below. And we have a know leak on an in service indicator. Plus with the pitot/static tester connected on both systems you'll never actually "see" the internal airspeed indicator leak as both ends are basically capped. Just the pitot tube to indicator needs to be checked. And yes an indicator can all of sudden develop a leak changing its readings.

However, if your aircraft is actually flying 10 mph slower then ensuring all your indicating systems are properly calibrated and functioning will be the 1st step in troubleshooting your performance issue.

it began flying about 10 knots slower INDICATED

 

[Ed Haywood]

Not sure you can conclude anything from ground speeds, unless you are consistently correcting for winds aloft, density altitude, etc.

If you are putting in the same RPM with the same prop, but getting a different airspeed, IMO that suggests either:
A) Indication error, or
B) drag has changed.

You identified two changes that happened at that time. GPS antenna might cause indication error in PFD if GPS data somehow used to correct indications (I dunno how those things work). Rudder trim might cause drag. So those two would seem like the logical paths to examine. Per Bell's comment, I'd start with making sure your indication is correct first.

 

[455 Bravo Uniform]

Maybe you were getting wrong indications before, and only questioned things when you appeared to slow down? What plane and what are the “book numbers”?

Can you get it to fly faster “uncoordinated”? Not sure how a simple ball can be wrong. But you made a change to rudder trim and here you are. It doesn’t take a whole lot of flying sideways to knock off a bunch of speed.

 

[kicktireslightfires]

Not sure you can conclude anything from ground speeds, unless you are consistently correcting for winds aloft, density altitude, etc.

If you are putting in the same RPM with the same prop, but getting a different airspeed, IMO that suggests either:
A) Indication error, or
B) drag has changed.

You identified two changes that happened at that time. GPS antenna might cause indication error in PFD if GPS data somehow used to correct indications (I dunno how those things work). Rudder trim might cause drag. So those two would seem like the logical paths to examine. Per Bell's comment, I'd start with making sure your indication is correct first.

A&B is exactly my conclusion, too, and that is what I have been looking into, without success. And I have been correcting for winds aloft and also density altitude -- even checked against an E6B for density.

But Jeff mentioned "winds posted by non inertial systems like the Dynon are notoriously inaccurate" so perhaps even with my calibrated digital compass my winds aloft speeds aren't accurate. HOWEVER, I will say that I have every indication that my winds aloft speeds being displayed are correct. Because multiple times during each flight I check METAR winds aloft speeds near the airport I'm closest to at the altitude I'm at, and they are always very, very close to what my real-time Dynon winds aloft are displaying. And also because whenever my Dynon displays a significant crosswind, the airplane is indeed significantly crabbed in order to track straight. So I have every indication that my Dynon real-time winds aloft display is accurate.

 

[kicktireslightfires]

Maybe you were getting wrong indications before, and only questioned things when you appeared to slow down? What plane and what are the “book numbers”?

The airplane was flying numbers very close to book. Now the airplane flies under book speeds.

 

[MauleSkinner]

Not sure how a simple ball can be wrong.

If it’s not mounted level in the panel, it will indicate wrong.

 

[kicktireslightfires]

Can you get it to fly faster “uncoordinated”? Not sure how a simple ball can be wrong. But you made a change to rudder trim and here you are. It doesn’t take a whole lot of flying sideways to knock off a bunch of speed.

Do you mean hold the rudder pedals one way or the other for 30 seconds while in cruise and see if that results in positive airspeed gain?

 

[Ed Haywood]

Not sure how a simple ball can be wrong.

HMMM. What ball is he using ... analog, or the AHRS ball on the PFD? I believe some of the glass panel systems use GPS data to error correct their AHRS systems.

The indicated airspeed showing on the PFD perfectly matches the standby airspeed indicator.

Do you know for a fact they perfectly matched BEFORE you changed the GPS antenna?

 

[kicktireslightfires]

If it’s not mounted level in the panel, it will indicate wrong.

It's a Dynon SkyView so it's a digital indication.

 

[kicktireslightfires]

Do you know for a fact they perfectly matched BEFORE you changed the GPS antenna?

The IAS on the Dynon PFD and on my standby ASI? I sure do. I take pictures of the panel every flight and when my TAS was showing 120 and IAS 102, my standby ASI was showing 102.

 

[Daleandee]

Could the addition of more rudder trim reduce drag? What would case more drag ... a bit more rudder deflection or the entire plane flying a bit sideways?