Fwd: boundary flow measurement instrumentation

[j...@oxaero.com]

Begin forwarded message:

From: "j...@oxaero.com" <j...@oxaero.com>
Date: September 3, 2012 1:14:23 PM CDT
To: Jari Hyvärinen <jari.hy...@linflow.se>
Cc: Roy Arnold <2tur...@gmail.com>, Nathan Murray <nmu...@olemiss.edu>
Subject: boundary flow measurement instrumentation

Jari,

Here is my latest probe design.  The tubes will be glued together with epoxy glue.  I will attach it with aluminum duct tape over the nose of the probe and over the feet of the L bracket on each side.

Following is the PCB that will hold the sensors:

It will be taped to the under side TE with the barbs pointing aft, presenting it's narrow dimensions to the stream.  I plan to cover it with a hood for better aerodynamics and to reduce buffetting of the tubing.  The tubing will run aft from the probe(s) then around the TE to the sensors.  The header is oriented to receive a ribbon cable from inside the fuselage.  There is a convenient hole in the fuselage.  A connector will be located just inside the hole, so removal of the wings will not require removing the whole apparatus from the wing.

The sensors (http://sensing.honeywell.com/product%20page?pr_id=142006) have ±0.25 %FSS BFSL (Full Scale Span Best Fit Straight Line) accuracy with 12 bit resolution.

My logger will record these four analog inputs, airspeed, static pressure and temperature at 1 Hz.  I will take humidity readings manually.

I should be able to bench test this next week.

Jim

[j...@oxaero.com]

Begin forwarded message:

From: "j...@oxaero.com" <j...@oxaero.com>
Date: September 4, 2012 11:34:14 AM CDT
To: Jari Hyvärinen <jari.hy...@linflow.se>
Cc: Roy Arnold <2tur...@gmail.com>, Nathan Murray <nmu...@olemiss.edu>
Subject: Re: boundary flow measurement instrumentation

Jari,

Thanks for the insight.  If I come up with nothing, it will be disappointing.  It seems to me, that unsteady nano-scale(?) behavior must transition to a macro-scale time averaged effect in order to affect glider performance significantly.  I look forward to the suggestions on probe design and your help analyzing the data.

Jim

On Sep 4, 2012, at 9:31 AM, Jari Hyvärinen <jari.hy...@linflow.se> wrote:

Jim,

As I see it, the influence on the boundary layer due to the leading edge tape is predominantly of unsteady ("high" frequency) nature. I will be interesing to see if it has any influence on the time averaged boundary layer characteristics. If so, it would surprise me, but you never know.

I look forward in studying your results.

Also, I talked to a friend how has worked many year with boundary layer monitoring. According to him this is difficult and not many people have had success with it. He will send me a suggestion on how one should/could design the prob. I will send you the material as soon as I get it.

Jari

[j...@oxaero.com]

Jari,

Here are my wave length, frequency and harmonic calculations.  Look for peaks in your surface microphone spectra near 1, 2 and 4 kHz.  I'll look at my in-fuselage recordings.  

This may be barking up the wrong tree, but this can give us the mean vortex diameter by adjusting that parameter to make the harmonics match the spectra.  We will have the boundary flow speed as a direct measurement.  I can calibrate my DAQ system airspeed against my airspeed system that Johnson calibrated by mounting my probe next to the glider Pitot in the nose.

In the video taped nose dipping event, the airspeed actually dropped one kt when the nose dipped.  I was trying to hold it constant.  It was essentially a constant velocity nose dip event in smooth air.  I have witnessed this on two other occasions, but only with deturbulator panels installed.  

I have everything I need now except for one connector and some tubing for the probes.

I have not heard any more from you about an adjustable height probe.  Is that because of the changing local aerodynamics from changing the Pitot height?

I am considering stacked, rectangular cross section (2:1 aspect ratio) Pitots with the lowest one on the surface with the bottom wall thickness milled down to a very thin value.  I think that the 2:1 rectangular geometry will reduce effects the probe itself.  I am also thinking of tapering the individual Pitot walls to a thin forward edge on all four sides.  Any ideas?

Jim

On Sep 6, 2012, at 8:46 AM, Jari Hyvärinen <jari.hy...@linflow.se> wrote:

Jim,

Exciting !

More thinking, The LE-tape induced vortices will most likely die out fairly quickly, but the waves induced by them will propagate far downstream (like the waves on a water surface induced by some local disturbance).

The nose dipping would in my mind be due to the acceleration caused by sudden decrease in drag, and then the acceleration generates higher flow speed causing increased lift. So, you do not need the increased camber to increase lift. This is an optional description, the future will
hopefully show us more on what is going on.

Jari

j...@oxaero.com skrev 2012-09-05 17:58:

Jari,

Yes, increased energy is the thing.  I have been imagining this as increased velocity in a prematurely detached flow.  Of course, there must be unsteadiness in the flow, but if performance increases, the unsteadiness cannot be full turbulence in an attached flow.

Actually, I have many oil-flow images of the longer, thinner separation bubble you speak of...some very, very long.  Also, I recall at least one indication of tangential reattachment ahead of a deturbulator panel.  I have often wondered if the dappled (not streamed) oil pattern ahead of the modified bubble is indication of the unsteadiness you mention.  

If I am correct, a bed of rolling vortices must consist of alternately, oppositely rotating vortices.  How would that influence unsteadiness in the flow above?  I can see it imposing a wavelength in the perturbed flow above.

A quick calculation indicates that the vortices would have a flow-wise frequency of 31.5 MHz.  That is too high to show up as a flow disturbance.  But what about harmonics?  13, 14 and 15th order harmonics fall out as 3.85, 1.92 and 0.96 kHz respectively.  Hmm, that's interesting...right around the 2 kHz peak we have recorded.  This may meaningless, since that peak is present with and w/o the tape.  On the other hand, it may be the reason the natural 2 kHz peak grows in amplitude with the tape...it lines up with a harmonic of the hypothised vortical flow.  

My nose dipping episodes argue for a large increase in lift force.  That implies (1) an energized top surface flow (higher velocity, Bernoulli effect) and/or (2) premature detachment of the energized flow that effectively increases the top surface wing camber.  This thinking is why I am searching for evidence of both increased time averaged flow velocity and detachment.  Of course, every step so far has revealed surprises, so I am expecting more of the same.

It gets curiouser and cueiouser.

Jim

On Sep 5, 2012, at 9:18 AM, Jari Hyvärinen <jari.hy...@linflow.se> wrote:

Hi,

An other way of thinking,

The introduced unsteady features of the boundary layer does not need to influence the average thickness but the increased energy in the boundary layer (energy in the waves propagating downstream) will (my theory) create a thinner (and longer) separation bubble using the additional energy. This will decrease drag.

So, even if you do not see significant changes in the BL-measurements things may still be as expected.

Jari

------------------------------------------------------------------------

j...@oxaero.com skrev 2012-09-04 18:34:

Jari,

Thanks for the insight.  If I come up with nothing, it will be disappointing.  It seems to me, that unsteady nano-scale(?) behavior must transition to a macro-scale time averaged effect in order to affect glider performance significantly.  I look forward to the suggestions on probe design and your help analyzing the data.

Jim

On Sep 4, 2012, at 9:31 AM, Jari Hyvärinen <jari.hy...@linflow.se> wrote:

Jim,

As I see it, the influence on the boundary layer due to the leading edge tape is predominantly of unsteady ("high" frequency) nature. I will be interesing to see if it has any influence on the time averaged boundary layer characteristics. If so, it would surprise me, but you never know.

I look forward in studying your results.

Also, I talked to a friend how has worked many year with boundary layer monitoring. According to him this is difficult and not many people have had success with it. He will send me a suggestion on how one should/could design the prob. I will send you the material as soon as I get it.

Jari

-----------------------------------------------------------------

j...@oxaero.com skrev 2012-09-03 20:14:

Jari,

Here is my latest probe design.  The tubes will be glued together with epoxy glue.  I will attach it with aluminum duct tape over the nose of the probe and over the feet of the L bracket on each side.

Following is the PCB that will hold the sensors:

It will be taped to the under side TE with the barbs pointing aft, presenting it's narrow dimensions to the stream.  I plan to cover it with a hood for better aerodynamics and to reduce buffetting of the tubing.  The tubing will run aft from the probe(s) then around the TE to the sensors.  The header is oriented to receive a ribbon cable from inside the fuselage.  There is a convenient hole in the fuselage.  A connector will be located just inside the hole, so removal of the wings will not require removing the whole apparatus from the wing.

The sensors (http://sensing.honeywell.com/product%20page?pr_id=142006) have ±0.25 %FSS BFSL (Full Scale Span Best Fit Straight Line) accuracy with 12 bit resolution.

My logger will record these four analog inputs, airspeed, static pressure and temperature at 1 Hz.  I will take humidity readings manually.

I should be able to bench test this next week.

Jim

-- 
Best Regards / Med vänlig hälsning

Jari Hyvärinen
ANKER-ZEMER Engineering AB
Box 22025
70202 Örebro
Sweden

Phone/Tel: +46-19-291622
Fax: +46-19-291510
Mail: jari.hy...@linflow.se
Web: www.anker-zemer.se
     www.linflow.com

-- 
Best Regards / Med vänlig hälsning

Jari Hyvärinen
ANKER-ZEMER Engineering AB
Box 22025
70202 Örebro
Sweden

Phone/Tel: +46-19-291622
Fax: +46-19-291510
Mail: jari.hy...@linflow.se
Web: www.anker-zemer.se
     www.linflow.com

-- 
Best Regards / Med vänlig hälsning

Jari Hyvärinen
ANKER-ZEMER Engineering AB
Box 22025
70202 Örebro
Sweden

Phone/Tel: +46-19-291622
Fax: +46-19-291510
Mail: jari.hy...@linflow.se
Web: www.anker-zemer.se
     www.linflow.com