Aerodynamics - Blade for Reverse Velocity Utilization
Thread Starter
Join Date: Apr 2003
Location: Vancouver, BC, Canada
Posts: 1,635
Likes: 0
Received 0 Likes
on
0 Posts
Aerodynamics - Blade for Reverse Velocity Utilization
The purpose of this thread is to place into the public domain the boring concept of a Reverse Velocity Rotor Blade. This blade is designed to operate on a rotor system that incorporates Reverse Velocity Utilization.
Join Date: Nov 2004
Location: Cambridgeshire, UK
Posts: 1,334
Likes: 0
Received 0 Likes
on
0 Posts
Dave,
The concepts you propose are not boring, it's just that folks are starting to wonder if these concepts are ever going to be anything other than just that. Makes it difficult to feedback about the design "practicalities". I thought you now had some tooling to layup test blades, and were considering a rig?
Mart
The concepts you propose are not boring, it's just that folks are starting to wonder if these concepts are ever going to be anything other than just that. Makes it difficult to feedback about the design "practicalities". I thought you now had some tooling to layup test blades, and were considering a rig?
Mart
Passion Flying Hobby Science Sponsor Work
Join Date: Apr 2004
Location: Belgium
Age: 68
Posts: 461
Likes: 0
Received 0 Likes
on
0 Posts
Reverse velocity blade
Dave,
Afraid that this beats me.
My simulator take reverse flow only 'marginally' into consideration, in that sense that it does 'statically (so to speak)' incorporate stall and reverse flow characteristics. This is already a small improvement over just ignoring it, as some of the math models that are published do (for instance first order harmonic models). Differences between ignoring it and incorporating it as in my model are small (I compaired both), because not only the region is small, but its lift contribution is even smaller.
Once this reverse flow region gets massive, my gut feeling is that aeroelastic models should be used, in order to get any reliable aerodynamic model. As I can't see how the h??l I could have that model work in realtime, I pass.
d3
Afraid that this beats me.
My simulator take reverse flow only 'marginally' into consideration, in that sense that it does 'statically (so to speak)' incorporate stall and reverse flow characteristics. This is already a small improvement over just ignoring it, as some of the math models that are published do (for instance first order harmonic models). Differences between ignoring it and incorporating it as in my model are small (I compaired both), because not only the region is small, but its lift contribution is even smaller.
Once this reverse flow region gets massive, my gut feeling is that aeroelastic models should be used, in order to get any reliable aerodynamic model. As I can't see how the h??l I could have that model work in realtime, I pass.
d3
Thread Starter
Join Date: Apr 2003
Location: Vancouver, BC, Canada
Posts: 1,635
Likes: 0
Received 0 Likes
on
0 Posts
Graviman,
The project is moving ahead, but it is no small undertaking, particularly for one person. It calls upon so many disciplines, and the only interesting one is rotor aerodynamics. Below are a couple of pictures. One shows some of the components for the CNC workstation and the other shows the development of the workstation's controller.
It is coming together, slowly. If the prototype and concept eventually shows promise then outside parties might be interested in accelerating the development.
Delta3,
Very little development work has been done on Reverse Velocity. Perhaps, if over the past sixty years the majority of helicopters had been built with two main rotors, there would have been twice as much incentive to pursue this field of research.
The only work, which I know of, that might be close to your project is a thesis done in 2003 and entitled 'Investigation of Increased Forward Flight Velocities of Helicopters Using Higher Harmonic Stall Control and Reverse Velocity Rotor Concept'. He has modified the basic JANRAD computer program to included the above. This booklet has a 78-page appendix that lists his coding to modify the basic program.
Dave
The project is moving ahead, but it is no small undertaking, particularly for one person. It calls upon so many disciplines, and the only interesting one is rotor aerodynamics. Below are a couple of pictures. One shows some of the components for the CNC workstation and the other shows the development of the workstation's controller.
It is coming together, slowly. If the prototype and concept eventually shows promise then outside parties might be interested in accelerating the development.
Delta3,
Very little development work has been done on Reverse Velocity. Perhaps, if over the past sixty years the majority of helicopters had been built with two main rotors, there would have been twice as much incentive to pursue this field of research.
The only work, which I know of, that might be close to your project is a thesis done in 2003 and entitled 'Investigation of Increased Forward Flight Velocities of Helicopters Using Higher Harmonic Stall Control and Reverse Velocity Rotor Concept'. He has modified the basic JANRAD computer program to included the above. This booklet has a 78-page appendix that lists his coding to modify the basic program.
Dave
Join Date: Nov 2004
Location: Cambridgeshire, UK
Posts: 1,334
Likes: 0
Received 0 Likes
on
0 Posts
"The project is moving ahead, but it is no small undertaking, particularly for one person."
Apologies offered, Dave. I wasn't sure how the blade layup CNC was going. Agree with you on rotor dynamics being the interesting bit.
For a lot of these ideas you could depend pretty well on wooden scale models, high speed film or strobe, and a smoke producer. Run the film/strobe slighlty out of sync to see the airflow during the course of the blade rotation. The idea of having the aerofoil near the root having twin trailing edges makes a lot of sense with RVU. The airflow in the zero velocity circle may present the hardest to get right.
----
D3, for the level of sophistication using aeroflexing you need what are known as "time explicit" techniques (ie solutions that iterate in intervals of time). Rather than trying to (say) combine static aeroflow predictions and a rotor blade modal analysis. The simplest breaks the blade into discrete aerodynamic elements, and follows the blade around the azimuth. The time explicit part comes into the fact that the stiffness between the elements and the mass of the element, in addition to the serodynamic loads, affect the element movement at each time cycle.
The level of sophistication grows very directly with computing power. Whole helicopter flows may now be taken into account:
http://www.cfdrc.com/bizareas/aerosp...edictions.html
The license cost of these is usually very prohibitive to a lone engineer, so i recommend scale models!
Mart
Apologies offered, Dave. I wasn't sure how the blade layup CNC was going. Agree with you on rotor dynamics being the interesting bit.
For a lot of these ideas you could depend pretty well on wooden scale models, high speed film or strobe, and a smoke producer. Run the film/strobe slighlty out of sync to see the airflow during the course of the blade rotation. The idea of having the aerofoil near the root having twin trailing edges makes a lot of sense with RVU. The airflow in the zero velocity circle may present the hardest to get right.
----
D3, for the level of sophistication using aeroflexing you need what are known as "time explicit" techniques (ie solutions that iterate in intervals of time). Rather than trying to (say) combine static aeroflow predictions and a rotor blade modal analysis. The simplest breaks the blade into discrete aerodynamic elements, and follows the blade around the azimuth. The time explicit part comes into the fact that the stiffness between the elements and the mass of the element, in addition to the serodynamic loads, affect the element movement at each time cycle.
The level of sophistication grows very directly with computing power. Whole helicopter flows may now be taken into account:
http://www.cfdrc.com/bizareas/aerosp...edictions.html
The license cost of these is usually very prohibitive to a lone engineer, so i recommend scale models!
Mart
Last edited by Graviman; 4th Dec 2005 at 11:51.
Passion Flying Hobby Science Sponsor Work
Join Date: Apr 2004
Location: Belgium
Age: 68
Posts: 461
Likes: 0
Received 0 Likes
on
0 Posts
Solution techniques
Graviman
If I understand you correctly you are referring to blade element techniques.
This is what I used (on the average 25cm, 3degrees). For the blade element aerodynamics I used a modified Prandle-like approach, that was calibrated on pointwise Xfoil data, and also checked against stall and reverse flow data, mainly obtained by the power windmill rotor designers. So I have a full 360° lift and drag model, in the full mach range.
But that model is still static. At high rates of pitch changes, not only angles off attack, but rates in change of these are important, especially when stalling occurs, this is what I understand aero-elasticity is about.
I looked at some (empirical) dynamic models, that are implementable, but this would augment the 'total numerical dimension', by at least one order, which in my case would explode the dimensionality. Right now I just make the simulator work realtime. It took quite some numerical optimisations to achieve this on a PC: dynamics take 60% cpu, rendering 20%, remains only 20% and that the PC needs for itself for task scheduling. Remember that also body aerodynamics and dynamics need to be calculated and parts of these operate many times in stall mode.
Coming back to Dave's proto:
One could even doubt if aero-elasticity will suffice, because one might expect that vortices will be more present, and simulating that is to my knowledge hardly implementable in realtime.
Does that makes sense ?
As a pilot -understanding my Heli and not just the Rotor behaviour was the main goal of my project- I am satisfied/saturated with the results, explaining (quantitatively precise) many things including the boundaries of the enveloppe.
As stated before I think Dave's concept opens new aerodynamic challenges, which are not small imho. This is not implying that it may not been done, but it looks like a leap.
d3
If I understand you correctly you are referring to blade element techniques.
This is what I used (on the average 25cm, 3degrees). For the blade element aerodynamics I used a modified Prandle-like approach, that was calibrated on pointwise Xfoil data, and also checked against stall and reverse flow data, mainly obtained by the power windmill rotor designers. So I have a full 360° lift and drag model, in the full mach range.
But that model is still static. At high rates of pitch changes, not only angles off attack, but rates in change of these are important, especially when stalling occurs, this is what I understand aero-elasticity is about.
I looked at some (empirical) dynamic models, that are implementable, but this would augment the 'total numerical dimension', by at least one order, which in my case would explode the dimensionality. Right now I just make the simulator work realtime. It took quite some numerical optimisations to achieve this on a PC: dynamics take 60% cpu, rendering 20%, remains only 20% and that the PC needs for itself for task scheduling. Remember that also body aerodynamics and dynamics need to be calculated and parts of these operate many times in stall mode.
Coming back to Dave's proto:
One could even doubt if aero-elasticity will suffice, because one might expect that vortices will be more present, and simulating that is to my knowledge hardly implementable in realtime.
Does that makes sense ?
As a pilot -understanding my Heli and not just the Rotor behaviour was the main goal of my project- I am satisfied/saturated with the results, explaining (quantitatively precise) many things including the boundaries of the enveloppe.
As stated before I think Dave's concept opens new aerodynamic challenges, which are not small imho. This is not implying that it may not been done, but it looks like a leap.
d3
Join Date: Nov 2004
Location: Cambridgeshire, UK
Posts: 1,334
Likes: 0
Received 0 Likes
on
0 Posts
"But that model is still static. At high rates of pitch changes, not only angles off attack, but rates in change of these are important, especially when stalling occurs, this is what I understand aero-elasticity is about."
Ah OK, my understanding is that it also takes into account the mechanical properties of the lifting surface for a full simulation of the forces and movements. Certainly time explicit techniques are not for real time evaluation! You can however, easilly include pitch rate dependancy, and any other non-linear aerofoil characteristics.
I still think a scale model is best to evaluate the principles Dave is considering - i recall Dave mentioned a 1:1 model
. Not least since computational techniques may just not capture all of the dynamics going on in such a complicated system - especially at tips, roots and zero vel circle...
Mart
Ah OK, my understanding is that it also takes into account the mechanical properties of the lifting surface for a full simulation of the forces and movements. Certainly time explicit techniques are not for real time evaluation! You can however, easilly include pitch rate dependancy, and any other non-linear aerofoil characteristics.
I still think a scale model is best to evaluate the principles Dave is considering - i recall Dave mentioned a 1:1 model
. Not least since computational techniques may just not capture all of the dynamics going on in such a complicated system - especially at tips, roots and zero vel circle...Mart
Passion Flying Hobby Science Sponsor Work
Join Date: Apr 2004
Location: Belgium
Age: 68
Posts: 461
Likes: 0
Received 0 Likes
on
0 Posts
Real simulation
Graviman
I agree, personally I would also build a scaled rotor and put it in the windtunnel.
Too much going on that we can't/may not model correctly.
d3
I agree, personally I would also build a scaled rotor and put it in the windtunnel.
Too much going on that we can't/may not model correctly.
d3
Thread Starter
Join Date: Apr 2003
Location: Vancouver, BC, Canada
Posts: 1,635
Likes: 0
Received 0 Likes
on
0 Posts
Delta3
You mentioned " one might expect that vortices will be more present,"
I am 'hoping' that for a given forward velocity the rotor vortices will be less than that of today's rotors. This is because the blade loading is distributed over 6 or 8 blades and that much of the forward component of thrust will be offloaded onto the propeller.
Graviman,
You mention " I still think a scale model is best to evaluate the principles Dave is considering."
I think that Sikorsky actually tested a scale reverse velocity rotor in a wind tunnel, many years ago. Then in 2002 RVR was proposed in Sikorsky's Reverse Velocity Rotorcraft Proposal. In addition, here is a 1976 Boeing Report on Reverse Velocity.
I strongly believe that latterly located twin main rotors represent the future for rotorcraft. In addition, the emerging Very Light Rotorcraft categories may be a significant part of this future. The ability to produce small, but full-size, blades should benefit one or both of the above. IMHO, the CNC Workstation must have the ability to produce strong, lightweight conventional blades in addition to producing prototypes of advanced concepts.
Dave
You mentioned " one might expect that vortices will be more present,"
I am 'hoping' that for a given forward velocity the rotor vortices will be less than that of today's rotors. This is because the blade loading is distributed over 6 or 8 blades and that much of the forward component of thrust will be offloaded onto the propeller.
Graviman,
You mention " I still think a scale model is best to evaluate the principles Dave is considering."
I think that Sikorsky actually tested a scale reverse velocity rotor in a wind tunnel, many years ago. Then in 2002 RVR was proposed in Sikorsky's Reverse Velocity Rotorcraft Proposal. In addition, here is a 1976 Boeing Report on Reverse Velocity.
I strongly believe that latterly located twin main rotors represent the future for rotorcraft. In addition, the emerging Very Light Rotorcraft categories may be a significant part of this future. The ability to produce small, but full-size, blades should benefit one or both of the above. IMHO, the CNC Workstation must have the ability to produce strong, lightweight conventional blades in addition to producing prototypes of advanced concepts.
Dave
Join Date: Nov 2004
Location: Cambridgeshire, UK
Posts: 1,334
Likes: 0
Received 0 Likes
on
0 Posts
"I strongly believe that latterly located twin main rotors represent the future for rotorcraft."
Whatever the config, counter-rotating is definately the shape of things to come.
"In addition, the emerging Very Light Rotorcraft categories may be a significant part of this future."
Although not as exciting an application, UAVs seem to be an increasing market - particularly as a multirole platform.
"IMHO, the CNC Workstation must have the ability to produce strong, lightweight conventional blades in addition to producing prototypes of advanced concepts."
The biggest source of resistance to commercial use of RVU is going to be the control system. I would be tempted to detail up a system, to understand the package implications of twin swash plates. This gets the possibility of a test bed, and an understanding of various options for different configs.
Mart
Whatever the config, counter-rotating is definately the shape of things to come.
"In addition, the emerging Very Light Rotorcraft categories may be a significant part of this future."
Although not as exciting an application, UAVs seem to be an increasing market - particularly as a multirole platform.
"IMHO, the CNC Workstation must have the ability to produce strong, lightweight conventional blades in addition to producing prototypes of advanced concepts."
The biggest source of resistance to commercial use of RVU is going to be the control system. I would be tempted to detail up a system, to understand the package implications of twin swash plates. This gets the possibility of a test bed, and an understanding of various options for different configs.
Mart
Thread Starter
Join Date: Apr 2003
Location: Vancouver, BC, Canada
Posts: 1,635
Likes: 0
Received 0 Likes
on
0 Posts
If you're remotely interested,
delta3
In an earlier post in this thread mentions a thesis done in 2003 and entitled 'Investigation of Increased Forward Flight Velocities of Helicopters Using Higher Harmonic Stall Control and Reverse Velocity Rotor Concept'. It has 3D lift distribution graphs like the ones you have produced, plus it has the computer code.
This thesis cost me $46.00 US from a company called Storming Media.
Well I'll be damned.
It's available on the Internet for free by doing the following; 
Go to http://stinet.dtic.mil/
Enter "reverse velocity" in the 'Search for' box.
Select the 'View Full Text pdf - 2 MB" of the first item.
The file will download but it will be totally blank.
Save the file to disk as an .htm file.
Change the suffix from ".htm" to ".pdf".
Open this saved file. "
In an earlier post in this thread mentions a thesis done in 2003 and entitled 'Investigation of Increased Forward Flight Velocities of Helicopters Using Higher Harmonic Stall Control and Reverse Velocity Rotor Concept'. It has 3D lift distribution graphs like the ones you have produced, plus it has the computer code.
This thesis cost me $46.00 US from a company called Storming Media.
Well I'll be damned.
It's available on the Internet for free by doing the following; Go to http://stinet.dtic.mil/
Enter "reverse velocity" in the 'Search for' box.
Select the 'View Full Text pdf - 2 MB" of the first item.
The file will download but it will be totally blank.
Save the file to disk as an .htm file.
Change the suffix from ".htm" to ".pdf".
Open this saved file. "
Last edited by Dave_Jackson; 13th Dec 2005 at 07:25.
