Negative drag on airfoil model 2

[inoycavalera]

This is my first thread, i want to say thanks to this great forum.

I have a problem with negative drag on an airfoil model. I have a strain gauge balance which can measure 3 forces and 3 moments. I attach an airfoil model to the balance and did some angle sweeping on a wind tunnel to capture the data, i expect i will get a V curve of drag force but it is not. the curve is straight from negative to positive, i mean when the angle of attack is positive the drag is positive and when the angle of attack is negative the drag become negative too. The real drag is must be always positive no matter its in positive or negative angle of attack (please correct me if wrong).

I still can't figure out why this happen, is it possible the problem is come from the design of the strain gauge balance?

Any body has an experience about this? any help would be appreciated.

Thank you

 

[btrueblood]

Sounds like you have a problem in the data reduction routines you are using to interpret the strain gage data.

How did you calibrate the balance? Did you use combined forces and moments, and of different signs, in order to find a full correction matrix?

Get a copy of

http://www.amazon.com/Low-Speed-Tunnel-Testing-Jewel-Barlow/dp/0471557749

 

[MiketheEngineer]

You have re-invented the "Lost Wing of the Incas"

 

[moon161]

>>You have re-invented the "Lost Wing of the Incas".<<

Indeed, I heard how the last remaining sample was lost- it was untethered by an intern when the slightest eddy came through the open door of the storage hangar at the smithsonian. The negative drag fed back into ever increasing thrust. It was out of sight in no time at all.

 

[rb1957]

are you sure it's "drag" or is it the resultant force in the freestream direction ?

is it the streamwise force from the airfoil (drag) or is it the streamwise component for the resultant force (ie the streamwise component of lift ??

 

[inoycavalera]

Hi btrueblood,

some peoples said that the problem is come from the software. I have 3 software, calibration software, check load, and data acquisition. The forces are in kg and the moments are in kgm, when calibrating the moments i enter the force too e.g when i calibrate the pitching moment i enter the lift load too because the calibration software was design to be like that.

Is it possible the problem is come from the drag calibration process? when calibrating drag force, the load was in negative direction...

 

[rb1957]

to calibrate, don't you apply a known load, possibly in both directions ... that's tell you something of the linearity of your balance ... and of course if you're seeing non-linearity you'll need to calibrate over the range of force you're expecting.

 

[inoycavalera]

Hi rb1957,

Yes of course when calibrating i apply a known load. the drag was calibrated in negative direction and of course the apply load was in negative too. is it a problem if i apply the load in negative? but the checkload give me a good result. it seems like an aerodynamic problem somewhere in the software maybe?

 

[btrueblood]

You should be able to calibrate with a combined load, e.g. a lift force (in Newtons, please, kg is not a unit of force), a drag force, and if possible a lift moment (in N-m) of the proper sign. It's quite possible that the lift moment is confusing the data reduction, i.e. it's showing up as a net thrust due to an error in the calibration.

For a wind tunnel balance, it is not a bad idea to calibrate the applied loads "thru zero", i.e. with a negative lift, positive drag (aka thrust), and a negative lift moment, and to touch the opposite corners of the matrix as well, i.e. all the possible combinations.

You did run a bare (tare) case, without a foil, just the stinger? Are you subtracting the tare drag of the stinger, and is that subtraction changing the sign of the drag force?

" because the calibration software was design to be like that. "

You should have a copy of Rae's book. You should also be able to write the calibration and data reduction matrix equations yourself...else how do you know the software isn't completely whacked? I know of more than a few "commercial" wind tunnels whose "came with the machine" software is ignored in favor of internally generated data reduction routines...though this is as much due to users upgrading instrumentation.

 

[inoycavalera]

Hi btrueblood,

thank you very much for your explanation. actually i'm not an expert in aerodynamics area. all the software was developed by a programmer in my office. he said he sure that all his aerodynamics formula in the software is correct. this problem is really confusing for me.

by the way thank you so much btrueblood, i'll try to get the book

 

[btrueblood]

Inoy,

If I was your programmer, I'd say the same thing. But I've written wind tunnel data reduction software before, and screwed it up in a lot of different ways. Sometimes it's just a sign convention that changed from the person who applied the loads to the person coding the software. Sometimes the same person did both jobs and still made the mistake...personal experience.

 

[btrueblood]

Also, it's good that the programmer is in your office (two heads are always better than one). Get the book, and then sit down with the programmer, both of you sitting in front of the tunnel, and walk through the calibration process. It's painful, but once you've debugged the darn thing, you will have both learned a great deal.

 

[rb1957]

can you verify two things, pls ...

1) during calibration you applied a -ve drag load and got a -ve result in your instrumentation ?

2) are you looking at drag as an aerodynamicist (ie along the freestream direction) and lift normal to the freestream,
or as a stress analyst (ie lift normal to chord, drag parallel) ?

 

[inoycavalera]

hi rb1957...

its been so long but still can not find "why"...

1. yes i got negative result
2. i'm not an aerodynamicist so i don't know about your 2nd question. what i can tell you is the force balance is used in a wind tunnel.

is it possible the problem is come from the center of gravity of the balance? i mean center of gravity has different value at different angle.