Preliminary analysis of aerodynamics of sparrow strainers (or not)


Jay Scheevel
 

Let me preface this with the statement that the current design of sparrow strainers works. You should not deviate from that design. The following is purely theoretical at this point and is not flight proven.

 

Here are my observations so far.

At 2 degress AOA, which is about where the Q2 is at high speed cruise, the standard LS! Airfoil produces a Cl of 0.461 and Cd of 0.0191 as shown below

In order to reverse the effect of the elevator deflecting upward in flight, sparrow strainers are installed. As a result, the span that contains the sparrow strainers ends up with an overall NEGATIVE lift coefficient, Cl of -0.333 and almost double the drag of the clean LS1: a Cd of 0.0317  (shown below). The extreme pressure field of the sparrow strainer, because of its high angle relative to the streamlines, actually kills the lift of the LS1 airfoil over the length of span that is occupied by the sparrow strainer. The result is a negative lift coefficient over that portion of the span. So roughly 1/8 of the total canard span (the portion of the span with the sparrow strainer installed)  is actually providing negative lift at an AOA of 2 degrees, and is creating lots more drag than it would if there was no sparrow strainer present.

 

The catch 22 is that we know the sparrow strainer is necessary for the LS1 profile, as we can demonstrate as follows:

 

If the sparrow strainer is omitted and no stick force is applied (hands off), the elevator portion of the canard will tend to reflex up in flight. But, how much would it reflex up? 

 

Well, it looks like it would reflex up 20 degrees before the pressure field is in equilibrium on top and bottom of the elevator. This would result in the overall “hands off” configuration having a negative Cl = -0.420 over the entire span occupied by the elevator. This would result in a severe “tuck”.

 

If the elevator portion of the LS1 is redesigned so as to not deflect up or down in flight and to be in trail throughout most of the AOA range, then the a result is shown below. This design should acheive a “hands off” Cl of 0.434 (slightly less than the bare LS1 of 0.461) and a Cd is 0.0198, very slightly higher than the bare LS1. No stick force or sparrow strainer would be required for this elevator profile if it performs as modeled, so the draggy and negative lift portion of the span would not be nescessary.

 

But what about a different airfoil, such as the the  Roncz 1145MS airfoil? R1145MS would result in the same Cl as the modified-elevator LS airfoil, but at a lower Cd of 0.01393. As shown below.

However, the Ronz airfoil, if fitted with an articulated elevator in the Q2 configuration, would tend to see the elevator deflect downward before achieving aerodynamic balance with no springs, stick force or sparrow strainer applied. Equilibrium would be achieved with the elevator deflected downward approximately 2 degrees as shown below. This deflection results in a higher Cl and would require that the overall airfoil be installed 1 degree nose down to the LS1 to perform aerodynamically similar to the LS1 equipped Q2. Doing so, would  yield a Cl of 0.458 and Cd of 0.01372 at an AOA of 2 degrees.

 

So we would mount the Roncz airfoil 1 degree lower angle. If we were to do this on a Q2, then the Cl would be similar to the LS1 airfoil, but the drag would be about 30% lower than the LS1: Cd 0.014 at and AOA of 2 degrees.

 

Note that none of these models include the effect of a turbulent stall bubble on the trailing edge of the elevator, so the numbers would be slightly different on a flying aircraft.

 

In summary, we see that the Roncz 1145MS airfoil would probably have been a better choice than the LS1 for a Q2, but it may not have been available at the time the decision of QAC was to go with the LS1. We can also see that the sparrow strainer solves the problem of preventing the LS1 elevator floating up in flight, but at a great drag cost and a reduction in overall canard lift over the span to which the sparrow strainer is applied (roughly 1/8 of the total span of the canard) with the 11.5” sparrow strainer, higher if using the 17.5’ sparrow strainer).

 

As shown above, it is possible to design a more benign elevator shape for the existing LS1 canard that will solve the problem of the elevator floating up in flight without the need for a sparrow strainer, but this solution is not as efficient with respect to drag as would be use of an entirely new airfoil as was done by John Roncz for the Long EZ.

 

This is my analysis so far. I will continue to work on this as time permits.

 

Cheers,

Jay


Mike Dwyer
 

With a lower angle of attack on the canard, wouldn't that reduce the margin of having the canard stall prior to the main wing?
I'd still like to see a better sparrow strainer design.  Move it into the airstream and make it operate in a "not stalled" condition so it could be smaller...
Mike Dwyer Q200
Great work Jay!

Q200 Website: http://goo.gl/V8IrJF


On Sun, Feb 27, 2022 at 6:37 PM Jay Scheevel <jay@...> wrote:

Let me preface this with the statement that the current design of sparrow strainers works. You should not deviate from that design. The following is purely theoretical at this point and is not flight proven.

 

Here are my observations so far.

At 2 degress AOA, which is about where the Q2 is at high speed cruise, the standard LS! Airfoil produces a Cl of 0.461 and Cd of 0.0191 as shown below

In order to reverse the effect of the elevator deflecting upward in flight, sparrow strainers are installed. As a result, the span that contains the sparrow strainers ends up with an overall NEGATIVE lift coefficient, Cl of -0.333 and almost double the drag of the clean LS1: a Cd of 0.0317  (shown below). The extreme pressure field of the sparrow strainer, because of its high angle relative to the streamlines, actually kills the lift of the LS1 airfoil over the length of span that is occupied by the sparrow strainer. The result is a negative lift coefficient over that portion of the span. So roughly 1/8 of the total canard span (the portion of the span with the sparrow strainer installed)  is actually providing negative lift at an AOA of 2 degrees, and is creating lots more drag than it would if there was no sparrow strainer present.

 

The catch 22 is that we know the sparrow strainer is necessary for the LS1 profile, as we can demonstrate as follows:

 

If the sparrow strainer is omitted and no stick force is applied (hands off), the elevator portion of the canard will tend to reflex up in flight. But, how much would it reflex up? 

 

Well, it looks like it would reflex up 20 degrees before the pressure field is in equilibrium on top and bottom of the elevator. This would result in the overall “hands off” configuration having a negative Cl = -0.420 over the entire span occupied by the elevator. This would result in a severe “tuck”.

 

If the elevator portion of the LS1 is redesigned so as to not deflect up or down in flight and to be in trail throughout most of the AOA range, then the a result is shown below. This design should acheive a “hands off” Cl of 0.434 (slightly less than the bare LS1 of 0.461) and a Cd is 0.0198, very slightly higher than the bare LS1. No stick force or sparrow strainer would be required for this elevator profile if it performs as modeled, so the draggy and negative lift portion of the span would not be nescessary.

 

But what about a different airfoil, such as the the  Roncz 1145MS airfoil? R1145MS would result in the same Cl as the modified-elevator LS airfoil, but at a lower Cd of 0.01393. As shown below.

However, the Ronz airfoil, if fitted with an articulated elevator in the Q2 configuration, would tend to see the elevator deflect downward before achieving aerodynamic balance with no springs, stick force or sparrow strainer applied. Equilibrium would be achieved with the elevator deflected downward approximately 2 degrees as shown below. This deflection results in a higher Cl and would require that the overall airfoil be installed 1 degree nose down to the LS1 to perform aerodynamically similar to the LS1 equipped Q2. Doing so, would  yield a Cl of 0.458 and Cd of 0.01372 at an AOA of 2 degrees.

 

So we would mount the Roncz airfoil 1 degree lower angle. If we were to do this on a Q2, then the Cl would be similar to the LS1 airfoil, but the drag would be about 30% lower than the LS1: Cd 0.014 at and AOA of 2 degrees.

 

Note that none of these models include the effect of a turbulent stall bubble on the trailing edge of the elevator, so the numbers would be slightly different on a flying aircraft.

 

In summary, we see that the Roncz 1145MS airfoil would probably have been a better choice than the LS1 for a Q2, but it may not have been available at the time the decision of QAC was to go with the LS1. We can also see that the sparrow strainer solves the problem of preventing the LS1 elevator floating up in flight, but at a great drag cost and a reduction in overall canard lift over the span to which the sparrow strainer is applied (roughly 1/8 of the total span of the canard) with the 11.5” sparrow strainer, higher if using the 17.5’ sparrow strainer).

 

As shown above, it is possible to design a more benign elevator shape for the existing LS1 canard that will solve the problem of the elevator floating up in flight without the need for a sparrow strainer, but this solution is not as efficient with respect to drag as would be use of an entirely new airfoil as was done by John Roncz for the Long EZ.

 

This is my analysis so far. I will continue to work on this as time permits.

 

Cheers,

Jay


Frankenbird Vern
 

 Good point, Mike. And one best avoided!  

Decalage revision, but that is with the R1145MS. Don't know yet with your elevator, Jay. The Frankenbirds canard is bolted in as is the wing, because it is Dragonfly so jus saying there is room for adjustment.. +/- perhaps 2 degrees.  So that means I can "sneek" up on the numbers and not having to do structural workovers. Gap seals and fairing work..minor compared to bonded in surfaces.     

 I suspect your correct about the timing John (Roncz) was working with. Cozy also used the 1145MS..I owned planes set #34
sold to me by Nat and Shirley Puffer. 

 Starships Canard was also the R1145MS airfoil, not a big deal to build it, but the main wing tooling was a major challenge.
 Not like the Long Eze which Beech had one used to train our test pilots.  I seem to remember on Starship there were 9 or 10 transitions 
in profile on the way to the tipsails.  We thought John had been on some serious drugs once the design was released to us. 

Vern     


From: main@Q-List.groups.io <main@Q-List.groups.io> on behalf of Mike Dwyer <q200pilot@...>
Sent: Sunday, February 27, 2022 6:15 PM
To: main@q-list.groups.io <main@q-list.groups.io>
Subject: Re: [Q-List] Preliminary analysis of aerodynamics of sparrow strainers (or not)
 
With a lower angle of attack on the canard, wouldn't that reduce the margin of having the canard stall prior to the main wing?
I'd still like to see a better sparrow strainer design.  Move it into the airstream and make it operate in a "not stalled" condition so it could be smaller...
Mike Dwyer Q200
Great work Jay!

Q200 Website: http://goo.gl/V8IrJF


On Sun, Feb 27, 2022 at 6:37 PM Jay Scheevel <jay@...> wrote:

Let me preface this with the statement that the current design of sparrow strainers works. You should not deviate from that design. The following is purely theoretical at this point and is not flight proven.

 

Here are my observations so far.

At 2 degress AOA, which is about where the Q2 is at high speed cruise, the standard LS! Airfoil produces a Cl of 0.461 and Cd of 0.0191 as shown below

In order to reverse the effect of the elevator deflecting upward in flight, sparrow strainers are installed. As a result, the span that contains the sparrow strainers ends up with an overall NEGATIVE lift coefficient, Cl of -0.333 and almost double the drag of the clean LS1: a Cd of 0.0317  (shown below). The extreme pressure field of the sparrow strainer, because of its high angle relative to the streamlines, actually kills the lift of the LS1 airfoil over the length of span that is occupied by the sparrow strainer. The result is a negative lift coefficient over that portion of the span. So roughly 1/8 of the total canard span (the portion of the span with the sparrow strainer installed)  is actually providing negative lift at an AOA of 2 degrees, and is creating lots more drag than it would if there was no sparrow strainer present.

 

The catch 22 is that we know the sparrow strainer is necessary for the LS1 profile, as we can demonstrate as follows:

 

If the sparrow strainer is omitted and no stick force is applied (hands off), the elevator portion of the canard will tend to reflex up in flight. But, how much would it reflex up? 

 

Well, it looks like it would reflex up 20 degrees before the pressure field is in equilibrium on top and bottom of the elevator. This would result in the overall “hands off” configuration having a negative Cl = -0.420 over the entire span occupied by the elevator. This would result in a severe “tuck”.

 

If the elevator portion of the LS1 is redesigned so as to not deflect up or down in flight and to be in trail throughout most of the AOA range, then the a result is shown below. This design should acheive a “hands off” Cl of 0.434 (slightly less than the bare LS1 of 0.461) and a Cd is 0.0198, very slightly higher than the bare LS1. No stick force or sparrow strainer would be required for this elevator profile if it performs as modeled, so the draggy and negative lift portion of the span would not be nescessary.

 

But what about a different airfoil, such as the the  Roncz 1145MS airfoil? R1145MS would result in the same Cl as the modified-elevator LS airfoil, but at a lower Cd of 0.01393. As shown below.

However, the Ronz airfoil, if fitted with an articulated elevator in the Q2 configuration, would tend to see the elevator deflect downward before achieving aerodynamic balance with no springs, stick force or sparrow strainer applied. Equilibrium would be achieved with the elevator deflected downward approximately 2 degrees as shown below. This deflection results in a higher Cl and would require that the overall airfoil be installed 1 degree nose down to the LS1 to perform aerodynamically similar to the LS1 equipped Q2. Doing so, would  yield a Cl of 0.458 and Cd of 0.01372 at an AOA of 2 degrees.

 

So we would mount the Roncz airfoil 1 degree lower angle. If we were to do this on a Q2, then the Cl would be similar to the LS1 airfoil, but the drag would be about 30% lower than the LS1: Cd 0.014 at and AOA of 2 degrees.

 

Note that none of these models include the effect of a turbulent stall bubble on the trailing edge of the elevator, so the numbers would be slightly different on a flying aircraft.

 

In summary, we see that the Roncz 1145MS airfoil would probably have been a better choice than the LS1 for a Q2, but it may not have been available at the time the decision of QAC was to go with the LS1. We can also see that the sparrow strainer solves the problem of preventing the LS1 elevator floating up in flight, but at a great drag cost and a reduction in overall canard lift over the span to which the sparrow strainer is applied (roughly 1/8 of the total span of the canard) with the 11.5” sparrow strainer, higher if using the 17.5’ sparrow strainer).

 

As shown above, it is possible to design a more benign elevator shape for the existing LS1 canard that will solve the problem of the elevator floating up in flight without the need for a sparrow strainer, but this solution is not as efficient with respect to drag as would be use of an entirely new airfoil as was done by John Roncz for the Long EZ.

 

This is my analysis so far. I will continue to work on this as time permits.

 

Cheers,

Jay


Jay Scheevel
 

Hi Mike,

Here’s the deal on wing mounting angles. Unlike the current dogma about the canard stalling first, the canard in the Q configuration never really achieves a full stall. The full stall angle is something like 13-14 degrees  AOA for the LS1 at the wing root. What happens before it gets to this angle is that the main wing “overpowers” the canard wrt pitch moment and forces the angle of attack lower. This is the pitch bunk oscillation.

This is why the pitch buck is a sort of nodding action instead of something more violent. 

Why is this the case? In intermediate angles of attack the angle versus lift curve is linear, so the ratio of lift contribution between canard and MW is constant for a given elevator deflection over a range of AOA’s. At higher angles of attack and higher elevator deflection, the slope of the lift curve for the canard decreases progressively while the MW lift curve remains linear. The Q is set up so that the MW is still in the linear portion of the lift curve when the canard enters the reduced slope portion (at about 8-9 degrees). At this angle the canard is still not close to stall.  Because the MW is picking up lift vs angle at a higher rate, the MW starts to win the pitch moment battle and it forces a lower angle of attack on the entire plane.  This is the pitch buck. Nothing is fully stalled.  If the canard were fully stalled, it would feel like a true stall and like it wants to depart into a spin.

I mounted my MW at a lower angle relative to the canard so my canard gets closer to stall and is more like 10-11 degrees at pitch buck. My pitch buck is more “exciting”, but this is NOT because the MW is closer to stall, but because the canard is closer to stall.  

So the design of the relative angle of the canard in the tandem wing configuration is intended to match the lift ratios in a way to make sure the canard reaches the non-linear portion of the lift curve first.

Accelerated stalls are another animal and are to be avoided in the Q.

Cheers,
Jay


On Feb 27, 2022, at 5:15 PM, Mike Dwyer <q200pilot@...> wrote:


With a lower angle of attack on the canard, wouldn't that reduce the margin of having the canard stall prior to the main wing?
I'd still like to see a better sparrow strainer design.  Move it into the airstream and make it operate in a "not stalled" condition so it could be smaller...
Mike Dwyer Q200
Great work Jay!

Q200 Website: http://goo.gl/V8IrJF


On Sun, Feb 27, 2022 at 6:37 PM Jay Scheevel <jay@...> wrote:

Let me preface this with the statement that the current design of sparrow strainers works. You should not deviate from that design. The following is purely theoretical at this point and is not flight proven.

 

Here are my observations so far.

At 2 degress AOA, which is about where the Q2 is at high speed cruise, the standard LS! Airfoil produces a Cl of 0.461 and Cd of 0.0191 as shown below

In order to reverse the effect of the elevator deflecting upward in flight, sparrow strainers are installed. As a result, the span that contains the sparrow strainers ends up with an overall NEGATIVE lift coefficient, Cl of -0.333 and almost double the drag of the clean LS1: a Cd of 0.0317  (shown below). The extreme pressure field of the sparrow strainer, because of its high angle relative to the streamlines, actually kills the lift of the LS1 airfoil over the length of span that is occupied by the sparrow strainer. The result is a negative lift coefficient over that portion of the span. So roughly 1/8 of the total canard span (the portion of the span with the sparrow strainer installed)  is actually providing negative lift at an AOA of 2 degrees, and is creating lots more drag than it would if there was no sparrow strainer present.

 

The catch 22 is that we know the sparrow strainer is necessary for the LS1 profile, as we can demonstrate as follows:

 

If the sparrow strainer is omitted and no stick force is applied (hands off), the elevator portion of the canard will tend to reflex up in flight. But, how much would it reflex up? 

 

Well, it looks like it would reflex up 20 degrees before the pressure field is in equilibrium on top and bottom of the elevator. This would result in the overall “hands off” configuration having a negative Cl = -0.420 over the entire span occupied by the elevator. This would result in a severe “tuck”.

 

If the elevator portion of the LS1 is redesigned so as to not deflect up or down in flight and to be in trail throughout most of the AOA range, then the a result is shown below. This design should acheive a “hands off” Cl of 0.434 (slightly less than the bare LS1 of 0.461) and a Cd is 0.0198, very slightly higher than the bare LS1. No stick force or sparrow strainer would be required for this elevator profile if it performs as modeled, so the draggy and negative lift portion of the span would not be nescessary.

 

But what about a different airfoil, such as the the  Roncz 1145MS airfoil? R1145MS would result in the same Cl as the modified-elevator LS airfoil, but at a lower Cd of 0.01393. As shown below.

However, the Ronz airfoil, if fitted with an articulated elevator in the Q2 configuration, would tend to see the elevator deflect downward before achieving aerodynamic balance with no springs, stick force or sparrow strainer applied. Equilibrium would be achieved with the elevator deflected downward approximately 2 degrees as shown below. This deflection results in a higher Cl and would require that the overall airfoil be installed 1 degree nose down to the LS1 to perform aerodynamically similar to the LS1 equipped Q2. Doing so, would  yield a Cl of 0.458 and Cd of 0.01372 at an AOA of 2 degrees.

 

So we would mount the Roncz airfoil 1 degree lower angle. If we were to do this on a Q2, then the Cl would be similar to the LS1 airfoil, but the drag would be about 30% lower than the LS1: Cd 0.014 at and AOA of 2 degrees.

 

Note that none of these models include the effect of a turbulent stall bubble on the trailing edge of the elevator, so the numbers would be slightly different on a flying aircraft.

 

In summary, we see that the Roncz 1145MS airfoil would probably have been a better choice than the LS1 for a Q2, but it may not have been available at the time the decision of QAC was to go with the LS1. We can also see that the sparrow strainer solves the problem of preventing the LS1 elevator floating up in flight, but at a great drag cost and a reduction in overall canard lift over the span to which the sparrow strainer is applied (roughly 1/8 of the total span of the canard) with the 11.5” sparrow strainer, higher if using the 17.5’ sparrow strainer).

 

As shown above, it is possible to design a more benign elevator shape for the existing LS1 canard that will solve the problem of the elevator floating up in flight without the need for a sparrow strainer, but this solution is not as efficient with respect to drag as would be use of an entirely new airfoil as was done by John Roncz for the Long EZ.

 

This is my analysis so far. I will continue to work on this as time permits.

 

Cheers,

Jay


Mike Dwyer
 

Thanks for the explanation Jay.  I'm pulling like 20 lbs of back pressure just prior to stall, but at the stall the force drops to zero.  That made me think that the canard is stalling.  Your explaination makes sense!  
Fly Safe,
Mike Dwyer Q200

On Sun, Feb 27, 2022, 8:26 PM Jay Scheevel <jay@...> wrote:
Hi Mike,

Here’s the deal on wing mounting angles. Unlike the current dogma about the canard stalling first, the canard in the Q configuration never really achieves a full stall. The full stall angle is something like 13-14 degrees  AOA for the LS1 at the wing root. What happens before it gets to this angle is that the main wing “overpowers” the canard wrt pitch moment and forces the angle of attack lower. This is the pitch bunk oscillation.

This is why the pitch buck is a sort of nodding action instead of something more violent. 

Why is this the case? In intermediate angles of attack the angle versus lift curve is linear, so the ratio of lift contribution between canard and MW is constant for a given elevator deflection over a range of AOA’s. At higher angles of attack and higher elevator deflection, the slope of the lift curve for the canard decreases progressively while the MW lift curve remains linear. The Q is set up so that the MW is still in the linear portion of the lift curve when the canard enters the reduced slope portion (at about 8-9 degrees). At this angle the canard is still not close to stall.  Because the MW is picking up lift vs angle at a higher rate, the MW starts to win the pitch moment battle and it forces a lower angle of attack on the entire plane.  This is the pitch buck. Nothing is fully stalled.  If the canard were fully stalled, it would feel like a true stall and like it wants to depart into a spin.

I mounted my MW at a lower angle relative to the canard so my canard gets closer to stall and is more like 10-11 degrees at pitch buck. My pitch buck is more “exciting”, but this is NOT because the MW is closer to stall, but because the canard is closer to stall.  

So the design of the relative angle of the canard in the tandem wing configuration is intended to match the lift ratios in a way to make sure the canard reaches the non-linear portion of the lift curve first.

Accelerated stalls are another animal and are to be avoided in the Q.

Cheers,
Jay


On Feb 27, 2022, at 5:15 PM, Mike Dwyer <q200pilot@...> wrote:


With a lower angle of attack on the canard, wouldn't that reduce the margin of having the canard stall prior to the main wing?
I'd still like to see a better sparrow strainer design.  Move it into the airstream and make it operate in a "not stalled" condition so it could be smaller...
Mike Dwyer Q200
Great work Jay!

Q200 Website: http://goo.gl/V8IrJF


On Sun, Feb 27, 2022 at 6:37 PM Jay Scheevel <jay@...> wrote:

Let me preface this with the statement that the current design of sparrow strainers works. You should not deviate from that design. The following is purely theoretical at this point and is not flight proven.

 

Here are my observations so far.

At 2 degress AOA, which is about where the Q2 is at high speed cruise, the standard LS! Airfoil produces a Cl of 0.461 and Cd of 0.0191 as shown below

In order to reverse the effect of the elevator deflecting upward in flight, sparrow strainers are installed. As a result, the span that contains the sparrow strainers ends up with an overall NEGATIVE lift coefficient, Cl of -0.333 and almost double the drag of the clean LS1: a Cd of 0.0317  (shown below). The extreme pressure field of the sparrow strainer, because of its high angle relative to the streamlines, actually kills the lift of the LS1 airfoil over the length of span that is occupied by the sparrow strainer. The result is a negative lift coefficient over that portion of the span. So roughly 1/8 of the total canard span (the portion of the span with the sparrow strainer installed)  is actually providing negative lift at an AOA of 2 degrees, and is creating lots more drag than it would if there was no sparrow strainer present.

 

The catch 22 is that we know the sparrow strainer is necessary for the LS1 profile, as we can demonstrate as follows:

 

If the sparrow strainer is omitted and no stick force is applied (hands off), the elevator portion of the canard will tend to reflex up in flight. But, how much would it reflex up? 

 

Well, it looks like it would reflex up 20 degrees before the pressure field is in equilibrium on top and bottom of the elevator. This would result in the overall “hands off” configuration having a negative Cl = -0.420 over the entire span occupied by the elevator. This would result in a severe “tuck”.

 

If the elevator portion of the LS1 is redesigned so as to not deflect up or down in flight and to be in trail throughout most of the AOA range, then the a result is shown below. This design should acheive a “hands off” Cl of 0.434 (slightly less than the bare LS1 of 0.461) and a Cd is 0.0198, very slightly higher than the bare LS1. No stick force or sparrow strainer would be required for this elevator profile if it performs as modeled, so the draggy and negative lift portion of the span would not be nescessary.

 

But what about a different airfoil, such as the the  Roncz 1145MS airfoil? R1145MS would result in the same Cl as the modified-elevator LS airfoil, but at a lower Cd of 0.01393. As shown below.

However, the Ronz airfoil, if fitted with an articulated elevator in the Q2 configuration, would tend to see the elevator deflect downward before achieving aerodynamic balance with no springs, stick force or sparrow strainer applied. Equilibrium would be achieved with the elevator deflected downward approximately 2 degrees as shown below. This deflection results in a higher Cl and would require that the overall airfoil be installed 1 degree nose down to the LS1 to perform aerodynamically similar to the LS1 equipped Q2. Doing so, would  yield a Cl of 0.458 and Cd of 0.01372 at an AOA of 2 degrees.

 

So we would mount the Roncz airfoil 1 degree lower angle. If we were to do this on a Q2, then the Cl would be similar to the LS1 airfoil, but the drag would be about 30% lower than the LS1: Cd 0.014 at and AOA of 2 degrees.

 

Note that none of these models include the effect of a turbulent stall bubble on the trailing edge of the elevator, so the numbers would be slightly different on a flying aircraft.

 

In summary, we see that the Roncz 1145MS airfoil would probably have been a better choice than the LS1 for a Q2, but it may not have been available at the time the decision of QAC was to go with the LS1. We can also see that the sparrow strainer solves the problem of preventing the LS1 elevator floating up in flight, but at a great drag cost and a reduction in overall canard lift over the span to which the sparrow strainer is applied (roughly 1/8 of the total span of the canard) with the 11.5” sparrow strainer, higher if using the 17.5’ sparrow strainer).

 

As shown above, it is possible to design a more benign elevator shape for the existing LS1 canard that will solve the problem of the elevator floating up in flight without the need for a sparrow strainer, but this solution is not as efficient with respect to drag as would be use of an entirely new airfoil as was done by John Roncz for the Long EZ.

 

This is my analysis so far. I will continue to work on this as time permits.

 

Cheers,

Jay


David J. Gall
 

Jay,

 

Nice analysis. I’m heartened to see that someone else has independently concluded what I’ve been saying for years: the Roncz R1145MS airfoil is a better choice than the LS-1. The Roncz airfoil also solves other issues such as eliminating sparrow strainers. I think Rutan broke off his involvement with QAC over this very issue.

 

I note that you’ve asserted that the elevator floating angle of the Roncz airfoil would require mounting the canard more nose down by one degree. I believe this is incorrect. The elevator floating angle you found is not intended to be achieved at “high speed cruise” as compared to your first illustration on the LS-1; rather, the elevator floating angle is intended to give a *low* speed cruise so that in the unlikely event of a total mechanical disconnection of the elevator from both the trim system and the control stick the airplane could still be flown to a destination or diversion airport (long-range cruise) yet that trimmed speed would also be slow enough that a semblance of a “high speed landing” without elevator control could be survived on arrival, using only throttle for climb/descent. See Rutan’s CP59; also see the fate of George Mead’s Piper “Pugmobile.”

 

So I would recommend mounting the Roncz canard at the same angle of incidence as the LS-1, not a reduced angle of incidence. In flight, the “down” trim spring should be able to hold the needed nose-down elevator bias to achieve hands-off trim at high speed cruise, and the “up” trim spring should be able to hold the needed nose-up elevator bias for slow flight (but not stall).

 

Besides having significantly lower drag than both the GU and LS-1 airfoils, the Roncz drag is lower still compared to the LS-1 airfoil because of not needing sparrow strainers and also because of not having that crazy negative lift zone in the spanwise lift distribution that is caused by the LS-1’s sparrow strainers.

 

Another benefit of the Roncz airfoil is that it is almost as thick as the GU airfoil (20.5% vs. 21% for the GU). Structurally that makes it almost a direct replacement for the GU canard, not needing the tubular carbon spar but able to be built using glass and a slightly modified GU canard layup schedule (think: Waddelow canard). For a Q1 Quickie or for a Q2 (max. gross weight 1000 lbs.) it could be a direct replacement; for a Q200 the added engine weight (MGW 1100 lbs.) would require additional structure. Of course, with some of these planes currently operating at 1300 lbs., all bets are on a structural redesign using carbon.

 

Finally, Mike’s point about the lower angle of incidence on the canard is actually countered by the elevator deflection stop limits; one would run out of elevator deflection before achieving stall if the canard incidence were set too nose-down. However, it would be prudent to check the maximum CL and the AOA at which that occurs when selecting the canard installation incidence angle. It might be necessary to shorten the canard chord slightly if the new airfoil is a significantly better performer than the one it replaces, in order to prevent a main wing stall. (Trimming the chord is preferable to trimming the span because the resulting surface has a higher aspect ratio that gives it a steeper slope of the lift curve, so it achieves CLmax at a lower AOA than a lower aspect ratio surface would.)

 

Keep up the good work,

 

 

David J. Gall



 

From: main@Q-List.groups.io <main@Q-List.groups.io> On Behalf Of Mike Dwyer
Sent: Sunday, February 27, 2022 4:15 PM
To: main@q-list.groups.io
Subject: Re: [Q-List] Preliminary analysis of aerodynamics of sparrow strainers (or not)

 

With a lower angle of attack on the canard, wouldn't that reduce the margin of having the canard stall prior to the main wing?

I'd still like to see a better sparrow strainer design.  Move it into the airstream and make it operate in a "not stalled" condition so it could be smaller...

Mike Dwyer Q200

Great work Jay!

 

Q200 Website: http://goo.gl/V8IrJF

 

 

On Sun, Feb 27, 2022 at 6:37 PM Jay Scheevel <jay@...> wrote:

Let me preface this with the statement that the current design of sparrow strainers works. You should not deviate from that design. The following is purely theoretical at this point and is not flight proven.

 

Here are my observations so far.

At 2 degress AOA, which is about where the Q2 is at high speed cruise, the standard LS! Airfoil produces a Cl of 0.461 and Cd of 0.0191 as shown below

In order to reverse the effect of the elevator deflecting upward in flight, sparrow strainers are installed. As a result, the span that contains the sparrow strainers ends up with an overall NEGATIVE lift coefficient, Cl of -0.333 and almost double the drag of the clean LS1: a Cd of 0.0317  (shown below). The extreme pressure field of the sparrow strainer, because of its high angle relative to the streamlines, actually kills the lift of the LS1 airfoil over the length of span that is occupied by the sparrow strainer. The result is a negative lift coefficient over that portion of the span. So roughly 1/8 of the total canard span (the portion of the span with the sparrow strainer installed)  is actually providing negative lift at an AOA of 2 degrees, and is creating lots more drag than it would if there was no sparrow strainer present.

 

The catch 22 is that we know the sparrow strainer is necessary for the LS1 profile, as we can demonstrate as follows:

 

If the sparrow strainer is omitted and no stick force is applied (hands off), the elevator portion of the canard will tend to reflex up in flight. But, how much would it reflex up? 

 

Well, it looks like it would reflex up 20 degrees before the pressure field is in equilibrium on top and bottom of the elevator. This would result in the overall “hands off” configuration having a negative Cl = -0.420 over the entire span occupied by the elevator. This would result in a severe “tuck”.

 

If the elevator portion of the LS1 is redesigned so as to not deflect up or down in flight and to be in trail throughout most of the AOA range, then the a result is shown below. This design should acheive a “hands off” Cl of 0.434 (slightly less than the bare LS1 of 0.461) and a Cd is 0.0198, very slightly higher than the bare LS1. No stick force or sparrow strainer would be required for this elevator profile if it performs as modeled, so the draggy and negative lift portion of the span would not be nescessary.

 

But what about a different airfoil, such as the the  Roncz 1145MS airfoil? R1145MS would result in the same Cl as the modified-elevator LS airfoil, but at a lower Cd of 0.01393. As shown below.

However, the Ronz airfoil, if fitted with an articulated elevator in the Q2 configuration, would tend to see the elevator deflect downward before achieving aerodynamic balance with no springs, stick force or sparrow strainer applied. Equilibrium would be achieved with the elevator deflected downward approximately 2 degrees as shown below. This deflection results in a higher Cl and would require that the overall airfoil be installed 1 degree nose down to the LS1 to perform aerodynamically similar to the LS1 equipped Q2. Doing so, would  yield a Cl of 0.458 and Cd of 0.01372 at an AOA of 2 degrees.

 

So we would mount the Roncz airfoil 1 degree lower angle. If we were to do this on a Q2, then the Cl would be similar to the LS1 airfoil, but the drag would be about 30% lower than the LS1: Cd 0.014 at and AOA of 2 degrees.

 

Note that none of these models include the effect of a turbulent stall bubble on the trailing edge of the elevator, so the numbers would be slightly different on a flying aircraft.

 

In summary, we see that the Roncz 1145MS airfoil would probably have been a better choice than the LS1 for a Q2, but it may not have been available at the time the decision of QAC was to go with the LS1. We can also see that the sparrow strainer solves the problem of preventing the LS1 elevator floating up in flight, but at a great drag cost and a reduction in overall canard lift over the span to which the sparrow strainer is applied (roughly 1/8 of the total span of the canard) with the 11.5” sparrow strainer, higher if using the 17.5’ sparrow strainer).

 

As shown above, it is possible to design a more benign elevator shape for the existing LS1 canard that will solve the problem of the elevator floating up in flight without the need for a sparrow strainer, but this solution is not as efficient with respect to drag as would be use of an entirely new airfoil as was done by John Roncz for the Long EZ.

 

This is my analysis so far. I will continue to work on this as time permits.

 

Cheers,

Jay


Jay Scheevel
 

Thanks David for your long note and discussion points. Much appreciated. I will absorb them and continue my analysis.

 

Michael Dunning, feel free to chime in anytime. I watched that video that you linked on the Wasabi flight testing the Raptor. Seems like Elliot has become wiser and more methodical in recent years. Nice to see their approach and discussions. It is a good watch for anyone looking to do a first flight, regardless of the type.

 

On their way out to GA, the Wasabi guys stopped in Borger TX. I landed there about 3 minutes before Charlie and Rob Johnson landed in the Tri-Pacer a few years ago on our way back from FOD in Enid. I have never seen so many bugs on my flight surfaces. Almost could not see the white paint! It was a good test of the LS1 contamination resistance! Charlie’s plane had by then turned into a spray-plane, in an attempt to subdue the bug population using a fine mist of Phillips XC…That made for an eventful ride home for those guys, but eventually worked out fine for Charlie.

 

Cheers,
Jay

 

_._,_._,_


Michael Dunning
 

Jay,

In typical engineer fashion, I don't get the chance to do this at my day job and got my references confused. Appreciate the patience while I double-checked everything....

The NLF development papers are what mention the hinge moment concern. This is the airfoil Somers tweaked for the Lancair 360. You'll notice the Abstract points squarely at correcting the shortcomings in the GAW-turned-LS series of airfoils:
 
 

Totally agree that a new canard using the R1145MS is the "correct" solution. As Mike Dwyer rightly pointed out, that's not really practical for most of us with built hardware. Personally, at that point I'd rather just start a new design with known design allowables rather than guess at the QAC numbers (Waddelow).  However, I will concede that if you have the fuselage shells the wing and canard are largely segregable.

Can you send me the Reynolds numbers, stock LS coordinates, and hinge location you're using as a baseline? XFOIL will calculate the hinge moments for speculative comparison (i.e. relative, not absolute design values).

Footnote: Selig worked with Somers on the NLF development (1995) and later had Ashok as a grad student (1998+), thus I view the AS504X and later airfoils as the current "gold standard" for EAB airfoil comparisons. The AS504X airfoils have the advantage of published (if primitive) wind tunnel test results as compared to the non-existent ones for the NLF(1)-0115.
--
-MD
#2827 (still thinking about planning on visualizing how to finish building)


Jay Scheevel
 

Hi Michael,

 

Thanks for the additional comments, data and history on the evolution of the LS-1 etc.

 

As far as what you are requesting, I am using 28% chord as the hinge point on the LS1 elevator. I am attaching the Javafoil coordinates for the LS1 (BL 15 on the Q2), also the multifoil for the same airfoil with the sparrow strainer as the second foil in the file, and my version of the LS1, with the redesign limited to only the bottom profile of the current LS1 elevator.

 

I am using 1 million as Reynolds number for my models.

 

BTW, I made use of the multifoil capability of Javafoil (XFOIL repackaged in a Java shell, by Martin Hepperle) to do my full aircraft modeling of the Q2xx summarized in the following:

http://n8wq.scheevel.com/documents/All_Text_and_figures_Part1.pdf

http://n8wq.scheevel.com/documents/All_Text_and_figures_Part2.pdf

http://n8wq.scheevel.com/documents/All_Text_and_figures_Part3.pdf

 

Cheers,

Jay

 

 

From: main@Q-List.groups.io <main@Q-List.groups.io> On Behalf Of Michael Dunning
Sent: Friday, March 4, 2022 6:05 PM
To: main@Q-List.groups.io
Subject: Re: [Q-List] Preliminary analysis of aerodynamics of sparrow strainers (or not)

 

Jay,

In typical engineer fashion, I don't get the chance to do this at my day job and got my references confused. Appreciate the patience while I double-checked everything....

The NLF development papers are what mention the hinge moment concern. This is the airfoil Somers tweaked for the Lancair 360. You'll notice the Abstract points squarely at correcting the shortcomings in the GAW-turned-LS series of airfoils:
 
 

Totally agree that a new canard using the R1145MS is the "correct" solution. As Mike Dwyer rightly pointed out, that's not really practical for most of us with built hardware. Personally, at that point I'd rather just start a new design with known design allowables rather than guess at the QAC numbers (Waddelow).  However, I will concede that if you have the fuselage shells the wing and canard are largely segregable.

Can you send me the Reynolds numbers, stock LS coordinates, and hinge location you're using as a baseline? XFOIL will calculate the hinge moments for speculative comparison (i.e. relative, not absolute design values).

Footnote: Selig worked with Somers on the NLF development (1995) and later had Ashok as a grad student (1998+), thus I view the AS504X and later airfoils as the current "gold standard" for EAB airfoil comparisons. The AS504X airfoils have the advantage of published (if primitive) wind tunnel test results as compared to the non-existent ones for the NLF(1)-0115.
--
-MD
#2827 (still thinking about planning on visualizing how to finish building)


David J. Gall
 

Michael,

Other than the general comments about tailoring the elevator hinge moment, I fail to see the relevance of the NLF(1)-0115 or the AS504X to the Quickie line of aircraft. They suffer from the same complaint you have about the Roncz R1145MS except that they are not even of comparable thickness to the GU canard. Therefore, they would require extensive rework to the layup schedules, beyond just the tweak needed to adapt the GU layup schedules for the Roncz airfoil; these new airfoils you suggest would require re-engineered structure before they could be used, even if one were able to incorporate the (unobtainable) tapered tubular carbon spars. What am I missing here?

 

 

~David J. Gall

 

From: main@Q-List.groups.io <main@Q-List.groups.io> On Behalf Of Michael Dunning
Sent: Friday, March 4, 2022 5:05 PM
To: main@Q-List.groups.io
Subject: Re: [Q-List] Preliminary analysis of aerodynamics of sparrow strainers (or not)

 

Jay,

In typical engineer fashion, I don't get the chance to do this at my day job and got my references confused. Appreciate the patience while I double-checked everything....

The NLF development papers are what mention the hinge moment concern. This is the airfoil Somers tweaked for the Lancair 360. You'll notice the Abstract points squarely at correcting the shortcomings in the GAW-turned-LS series of airfoils:
 
 

Totally agree that a new canard using the R1145MS is the "correct" solution. As Mike Dwyer rightly pointed out, that's not really practical for most of us with built hardware. Personally, at that point I'd rather just start a new design with known design allowables rather than guess at the QAC numbers (Waddelow).  However, I will concede that if you have the fuselage shells the wing and canard are largely segregable.

Can you send me the Reynolds numbers, stock LS coordinates, and hinge location you're using as a baseline? XFOIL will calculate the hinge moments for speculative comparison (i.e. relative, not absolute design values).

Footnote: Selig worked with Somers on the NLF development (1995) and later had Ashok as a grad student (1998+), thus I view the AS504X and later airfoils as the current "gold standard" for EAB airfoil comparisons. The AS504X airfoils have the advantage of published (if primitive) wind tunnel test results as compared to the non-existent ones for the NLF(1)-0115.
--
-MD
#2827 (still thinking about planning on visualizing how to finish building)


Michael Dunning
 

David,

The comment was strictly about the tailoring of the hinge moment. The rest was a fun history tangent.

Jay,

I noted a couple concerns in the provided LS(1) airfoil after running through the usual XFOIL file conversion hoops (remove tab delimiters, normalize the X/Y scale). While the max thickness is recovered correctly, the max thickness location, max camber and max camber location are considerably off from the published numbers:


Airfoil Tools (untrusted)


NASA TM-X-72843 (trusted, usually)


Regardless of that, my initial results at 2 degrees angle of attack are off in all respects from the first JavaFoil result you posted when using your (normalized) coordinates:



The two tools are never going to match but I do expect them to generally agree. My XFOIL results for zero lift AoA and where cl = 0.461 occurs are in line with the Airfoil Tools published numbers at RE=1M and the NASA published wind tunnel data, so I feel confident ruling out gross user error with the tool at least. The wind tunnel data in Figure 6 (PDF pg 28) is in general agreement and even the XFOIL over-prediction of max lift occurs at the typical +15%...and max lift AoA prediction is closer than usual at 15 vs 16 degrees actual.

Both my initial checks, the Airfoil Tools plot, and the NASA report show any cl < about 0.5 occurring at negative angles of attack. I'm doubtful but possibly this is due to the plans-template-level-line vs. chord line references? Again, not sure how JavaFoil works.


In the meantime, I need to stop and calibrate XFOIL with the wind tunnel data from the NASA report before I can be of any further use to you:
NASA-TM-X-72843 "Effects of thickness on the aerodynamic characteristics of an initial low-speed family of airfoils for general aviation applications"

Regards,
--
-MD
#2827 (still thinking about planning on visualizing how to finish building)


Jay Scheevel
 

Hi Michael,

Not much time to fully absorb all the items you bring up, but let me hit a couple. Airfoil dimensions were normalized for my multifoil model representation including with the Eppler main wing so absolute chord equal to 1.0 (100%) is reduced for dimensional convenience to accommodate both airfoils and total lift calls. This changes Reynolds numbers slightly as well. So I am just comparing apples to apples within my  personal modeling framework.

 The difference in Angle of LS1 in the QAC  plans relative to the UIUC database was noted and discussed (in addition to actual wind tunnel data) in part 1 of the Q2xx modeling paper that I did. Have a look at that summary and see if it it squares up the discrepancy you observed. 

I am headed out of town for a week or so and I will see if I can wrap my head around the other points you make as time permits. Thanks for your inputs.

Cheers,
Jay 


On Mar 8, 2022, at 9:50 PM, Michael Dunning <dunningme@...> wrote:

David,

The comment was strictly about the tailoring of the hinge moment. The rest was a fun history tangent.

Jay,

I noted a couple concerns in the provided LS(1) airfoil after running through the usual XFOIL file conversion hoops (remove tab delimiters, normalize the X/Y scale). While the max thickness is recovered correctly, the max thickness location, max camber and max camber location are considerably off from the published numbers:


Airfoil Tools (untrusted)


NASA TM-X-72843 (trusted, usually)


Regardless of that, my initial results at 2 degrees angle of attack are off in all respects from the first JavaFoil result you posted when using your (normalized) coordinates:



The two tools are never going to match but I do expect them to generally agree. My XFOIL results for zero lift AoA and where cl = 0.461 occurs are in line with the Airfoil Tools published numbers at RE=1M and the NASA published wind tunnel data, so I feel confident ruling out gross user error with the tool at least. The wind tunnel data in Figure 6 (PDF pg 28) is in general agreement and even the XFOIL over-prediction of max lift occurs at the typical +15%...and max lift AoA prediction is closer than usual at 15 vs 16 degrees actual.

Both my initial checks, the Airfoil Tools plot, and the NASA report show any cl < about 0.5 occurring at negative angles of attack. I'm doubtful but possibly this is due to the plans-template-level-line vs. chord line references? Again, not sure how JavaFoil works.


In the meantime, I need to stop and calibrate XFOIL with the wind tunnel data from the NASA report before I can be of any further use to you:
NASA-TM-X-72843 "Effects of thickness on the aerodynamic characteristics of an initial low-speed family of airfoils for general aviation applications"

Regards,
--
-MD
#2827 (still thinking about planning on visualizing how to finish building)


David J. Gall
 

Michael,

Fun history tangent noted.

Regarding your reported airfoil discrepancies, is it possible that you are inadvertently comparing two different airfoils, the LS(1)-0417 from the NASA TM against the LS(1)-0417MOD that’s used on the Q200? That might explain the differences in location of maximum thickness, etc. 


David J. Gall

On Mar 8, 2022, at 10:54 PM, Jay Scheevel <jay@...> wrote:

Hi Michael,

Not much time to fully absorb all the items you bring up, but let me hit a couple. Airfoil dimensions were normalized for my multifoil model representation including with the Eppler main wing so absolute chord equal to 1.0 (100%) is reduced for dimensional convenience to accommodate both airfoils and total lift calls. This changes Reynolds numbers slightly as well. So I am just comparing apples to apples within my  personal modeling framework.

 The difference in Angle of LS1 in the QAC  plans relative to the UIUC database was noted and discussed (in addition to actual wind tunnel data) in part 1 of the Q2xx modeling paper that I did. Have a look at that summary and see if it it squares up the discrepancy you observed. 

I am headed out of town for a week or so and I will see if I can wrap my head around the other points you make as time permits. Thanks for your inputs.

Cheers,
Jay 


On Mar 8, 2022, at 9:50 PM, Michael Dunning <dunningme@...> wrote:

David,

The comment was strictly about the tailoring of the hinge moment. The rest was a fun history tangent.

Jay,

I noted a couple concerns in the provided LS(1) airfoil after running through the usual XFOIL file conversion hoops (remove tab delimiters, normalize the X/Y scale). While the max thickness is recovered correctly, the max thickness location, max camber and max camber location are considerably off from the published numbers:
<dummyfile.0.part>


Airfoil Tools (untrusted)
<dummyfile.1.part>


NASA TM-X-72843 (trusted, usually)
<dummyfile.2.part>


Regardless of that, my initial results at 2 degrees angle of attack are off in all respects from the first JavaFoil result you posted when using your (normalized) coordinates:

<dummyfile.3.part>


The two tools are never going to match but I do expect them to generally agree. My XFOIL results for zero lift AoA and where cl = 0.461 occurs are in line with the Airfoil Tools published numbers at RE=1M and the NASA published wind tunnel data, so I feel confident ruling out gross user error with the tool at least. The wind tunnel data in Figure 6 (PDF pg 28) is in general agreement and even the XFOIL over-prediction of max lift occurs at the typical +15%...and max lift AoA prediction is closer than usual at 15 vs 16 degrees actual.

Both my initial checks, the Airfoil Tools plot, and the NASA report show any cl < about 0.5 occurring at negative angles of attack. I'm doubtful but possibly this is due to the plans-template-level-line vs. chord line references? Again, not sure how JavaFoil works.
<dummyfile.4.part>


In the meantime, I need to stop and calibrate XFOIL with the wind tunnel data from the NASA report before I can be of any further use to you:
NASA-TM-X-72843 "Effects of thickness on the aerodynamic characteristics of an initial low-speed family of airfoils for general aviation applications"

Regards,
--
-MD
#2827 (still thinking about planning on visualizing how to finish building)


Michael Dunning
 

Funny; I had that same thought after I went to bed...

However, I used the coordinates from Jay's stock LS(1) airfoil file directly with no changes...except one: re-scaling them up to a unit chord on import. I noticed that both the files without the sparrow strainer ended at x = 0.7116. I figured that the one with the sparrow strainer was at a unit chord length (x = 1.0) and the other two just had the strainer deleted. Plotting them all side by side, that doesn't look to be the case :(



I'm also a bit puzzled by Jay's references to the Eppler main wing, so it probably has something to do with tweaks needed for his modeling framework. I'm starting to lean towards this discrepancy stemming from differences in reference chord length. This sort of thing is how I learned about 'Reynold's Number' vs. 'Reynold's Number per foot' the hard way.

We'll just have to wait for Jay to get back.
--
-MD
#2827 (still thinking about planning on visualizing how to finish building)


Jay Scheevel
 

Hi Micheal,

 

Not back yet, but I will give a short response to clarify.. Corrections for the differences in chord were made external to Javafoil. I explain my procedures in the first of the documents that I linked to. I realized that Reynolds number is scale dependent (that is kind of the point, isn’t it?), so I have had to use a Reynolds number in Javafoil, that allows me to reach approximately 1 million effective Reynolds number for the Q2 configuration multifoil. To do this the cord is not unit for either airfoil of the multifoil pairs. So, to achieve the entire airplane in a internally consistent model, I cut the half span down to thirds using three models to approximate taper washout and tip efficiency loss (all of these are consolidated externally in spreadsheets with corrections incorporated there).

 

I modeled multifoils using the LS1 airfoil digitized directly from the hotwire templates in my set of plans. This is overlain in the plot below where the hotwire template is black and the coords used in Javafoil are in red. This figure is from my paper, part 1, that I linked previously. The black square Cp’s are from wind tunnel measurements on the same airfoil at Re=1 e6.  BTW in this comparison, I am using the same relative coords that I sent to you, except below I have scaled it to unit chord in order to compare apples to apples.

 

Cheers,

Jay

 

From: main@Q-List.groups.io <main@Q-List.groups.io> On Behalf Of Michael Dunning
Sent: Thursday, March 10, 2022 7:06 PM
To: main@Q-List.groups.io
Subject: Re: [Q-List] Preliminary analysis of aerodynamics of sparrow strainers (or not)

 

Funny; I had that same thought after I went to bed...

However, I used the coordinates from Jay's stock LS(1) airfoil file directly with no changes...except one: re-scaling them up to a unit chord on import. I noticed that both the files without the sparrow strainer ended at x = 0.7116. I figured that the one with the sparrow strainer was at a unit chord length (x = 1.0) and the other two just had the strainer deleted. Plotting them all side by side, that doesn't look to be the case :(



I'm also a bit puzzled by Jay's references to the Eppler main wing, so it probably has something to do with tweaks needed for his modeling framework. I'm starting to lean towards this discrepancy stemming from differences in reference chord length. This sort of thing is how I learned about 'Reynold's Number' vs. 'Reynold's Number per foot' the hard way.

We'll just have to wait for Jay to get back.
--
-MD
#2827 (still thinking about planning on visualizing how to finish building)