Loading ...
Sorry, an error occurred while loading the content.
 

Re: Betz limit [was: Re: [AWECS] Re: Makani M1]

Expand Messages
  • dave santos
    To Dave Lang,   It would help to explain why the Betz Limit based on a rotor solid-disc assumption does not apply in a loose general sense to other shapes
    Message 1 of 25 , Apr 8, 2011
      To Dave Lang,
       
      It would help to explain why the Betz Limit based on a rotor solid-disc assumption does not apply in a loose general sense to other shapes like a hollow ring or frontal rectangle. I have seen the Betz Limit principle used in this general sense by leading authors struggling to characterize the limits of non-disc windfarm geometries & it makes considerable sense in those contexts. No one proposed a triple penalty, but you could opine as to whether a small turbine disc on a wing inscribing a larger disc is in fact a Betz "double hit".
       
      Betz efficiency is a rough indication of the far more meaningful AWE limitation of Surface Sprawl. The Makani M1 will be severely limited in economic efficiency by the excessive requirement for ground interconnect infrastructure (roads, fences, buried cables, no-go zones, etc.). Despite a big sky, highest airspace utilization is very desirable,
       
      daveS
       
       
    • Dave Lang
      I was only alluding to two things.... 1. The (conceptually and quantitatively) shaky ground associated with shooting from the hip as regards to efficiency of
      Message 2 of 25 , Apr 8, 2011
        Re: Betz limit [was: Re: [AWECS] Re: Makani M1]
        I was only alluding to two things....

        1. The (conceptually and quantitatively) shaky ground associated with "shooting from the hip" as regards to efficiency of AWE schemes, and the implications thereof.

        2. That ROI and COP trumps ALL the details and implications that might be thrown in the mix, such as efficient use of airspace, etc, etc....all of which will show up in the ROI and COP if done rigorously (could one even design an AWE farm without addressing this issue?). These conjectures will ALL "washout to fact" when you do your solid simulations/testing and begin to approach a final design of something that actually works and has potential attraction to investors!

        Makani's M1 may have an efficiency of 1% for all I know, but who knows (although Makani may well) until one does the simulations and testing and evaluates TOTAL COST (to determine ROI and COP). I would not sell any AWE scheme short based on a "half-baked assessment of how many Betz limits" they are suffering from...what I am driving at is do the math folks (right).....do the sims....do the tests.

        DaveL





        At 11:27 AM -0700 4/8/11, dave santos wrote:
         
        To Dave Lang,
         
        It would help to explain why the Betz Limit based on a rotor solid-disc assumption does not apply in a loose general sense to other shapes like a hollow ring or frontal rectangle. I have seen the Betz Limit principle used in this general sense by leading authors struggling to characterize the limits of non-disc windfarm geometries & it makes considerable sense in those contexts. No one proposed a triple penalty, but you could opine as to whether a small turbine disc on a wing inscribing a larger disc is in fact a Betz "double hit".
         
        Betz efficiency is a rough indication of the far more meaningful AWE limitation of Surface Sprawl. The Makani M1 will be severely limited in economic efficiency by the excessive requirement for ground interconnect infrastructure (roads, fences, buried cables, no-go zones, etc.). Despite a big sky, highest airspace utilization is very desirable,
         
        daveS
         
         

      • Pierre Benhaiem
        Correction: 18133.5 m² ; a little less because of vertical projection of the wing. Note:you put a small turbine with a generator at the tip of each blade of a
        Message 3 of 25 , Apr 8, 2011
          Correction:"18133.5 m²"; a little less because of vertical projection
          of the wing.

          Note:you put a small turbine with a generator at the tip of each blade
          of a conventional wind turbine.Betz limit concerns the whole
          conventional wind turbine,but also concerns each small turbine in
          rapport to apparent wind:Betz limit within Betz limit.

          It is the same thing for AWECS of type flygen (another thing:it is
          interesting that a rotor aloft with a low output produces also a low
          drag (it is a point for examen)).Losses due to the rotor aloft (which
          Betz limit of the rotor aloft) are the price of the conversion,a little
          like losses due to reel-in phase for an AWECS of type reel-in-out.

          Other example:an existing huge wind turbine sweeps about 12000 m²:%
          output within Betz limit gives a main data for the global power. KiteGen
          carousel sweeps about 2 km² and more: % output within Betz limit
          gives a main data for the global power.

          PierreB

          --- In AirborneWindEnergy@yahoogroups.com, "Pierre Benhaiem"
          <pierre.benhaiem@...> wrote:
          >
          >
          > It is told on Makani website that M1 operates between 200 m and 400 m
          > altitude.Span is 35 m;so the swept area could be 18133.5 m².So with
          a
          > wind speed of 9 m/s (to obtain the theoretical 1 MW) the power at 100
          %
          > Betz limit could be 4700203.2 W. If we do not take into account the
          > (big) other losses the obtained power is about 21 % Betz limit.
          >
          > Four remarks:
          >
          > 1) The center of the loop is not swept.
          >
          > 2) Knowledge of the output in rapport to Betz limit is yet useful for
          > the knowledge of swept area then space occupation (surface and
          volume).
          >
          > 3) It is not sure that here the swept area is optimal (perhaps too
          high
          > for M1 potential,that could explain the low value of % Betz limit, or
          > for other considerations like a necessary high value of radius of loop
          > because of high kite speed ).Diehl's formula does not take into
          account
          > Betz limit because it does not take into account the swept area,but it
          > would be possible to make a translation to obtain a rapport between
          > (kite area and kite ratio CL/CD)/(swept area).I think KiteGen already
          > made this calculation.
          >
          > 4) With a less performant flygen (per m² of kite area) like
          > FlygenKite <http://flygenkite.com/> the swept area will be very
          > different than for M1. FlygenKite <http://flygenkite.com/> flying
          > slower the radius can be reduced:it could be a good point to reduce
          > space occupation (this point should be clarified).
          >
          > Doug,DaveS and me are right about the necessary limits of speed
          because
          > of the too high tip speed of the rotor.
          >
          > PierreB,
          >
          > http://flygenkite.com <http://flygenkite.com/>
          >
          > --- In AirborneWindEnergy@yahoogroups.com, Pierre BENHAIEM
          > pierre.benhaiem@ wrote:
          > >
          > > Correction of the precedent message:"To obtain 1000 000 W with 70 %
          of
          > losses".It is 30% of losses,and 70% of the initial value.
          > >
          > > PierreB
          > > http://flygenkite.com
          > >
          > >
          > >
          > >
          > >
          > > > Message du 04/04/11 20:19
          > > > De : "Pierre Benhaiem"
          > > > A : AirborneWindEnergy@yahoogroups.com
          > > > Copie à :
          > > > Objet : [AWECS] Makani M1
          > > >
          > > >
          > > >
          > > > On Makani website datas for the expected prototype M1 are;span =
          35
          > m,and nominal power of 1 MW is reached with wind velocity = 9
          > m/s.Illustrations show a very thin design:the width of the wing should
          > be about 3 m,maybe 5 m,but not more.So with an area of 175 m² the
          > power could be formalized as following:
          > > > 4/81 aD A w3 CL(CL/CD)² with aD = air density;A = kite area;w
          =
          > wind speed;CL = lift coefficient;cD = drag coefficient;4/81 is the
          > transformation of 2/27 (see Diehl's formula and the link below) after
          > conversion from a reel-out system towards a flygen system,including
          Betz
          > limit and drag of an ideal rotor (8/9);
          > > > So 4/81 1.2 175 93 1.2 = 9072; complete formula is: 4/81 1.2 175
          93
          > 1.2 (1.2/CD)²
          > > > To obtain 1000 000 W with 70 % of losses (comprising Betz limit):
          > 1000 000/6350.4 = about 157.47 ; square root of 157.47 is about 12.549
          > > > So (CL/CD) should be the value of 12.549 which is also the ratio
          > kite speed/wind speed.
          > > > So kite speed should be (12.549) 9 2/3 =75.294 m/s. 2/3 is the
          > optimal kite speed after slowing down because of turbines.
          > > > So the tip speed of turbines aloft should be some value like 300
          m/s
          > .
          > > > Thank you for corrections and comments.
          > > > PierreB
          > > > http://flygenkite.com
          > >
          >
          http://www.energykitesystems.net/OrthoKiteBunch/OptimizationOfAManualFly\
          \
          > gen.pdf
          > > >
          > > >
          > >
          >
        • North, David D. (LARC-E402)
          I agree with Dave Lang. The Betz limit sort of loses its relevance with most AWE configurations (flygen or groundgen). Betz developed this law to describe the
          Message 4 of 25 , Apr 8, 2011

            I agree with Dave Lang. The Betz limit sort of loses its relevance with most AWE configurations (flygen or groundgen). Betz developed this law to describe the theoretical maximum amount of energy that could be extracted from a full stream tube of fluid passing within the outer tip diameter of a turbine relative to the total energy passing through it.

             

            I suppose you could look at each turbine on a flygen separately and try to estimate the energy in the complex, helical stream tube that passes through each turbine on the flygen, but it seems like it would be very difficult (although not impossible) to model. Using the Betz limit calc with knockdown factors, or “double Betz” is probably an oversimplification of the problem.

             

            So again, I agree with Dave L. Build the simulations as best you can (hand calcs all the way up to CFD). Build the test articles. Get data. Recalibrate models (or throw them out and use empirical performance curves from the data). Rinse and repeat.

             

            Efficiency does matter (e.g. high L/D, low drag tethers, super efficient turbines/generators, light weight/strong materials, etc.) but In the end ROI (which involves much more than efficiency) will probably be the only figure of merit that matters. The teams that can field marketable systems with decent ROI (better than fossil fuels) will be successful and profit. This is why I was musing before about the really low efficiency Chinese drag system before (don’t slam me Doug S. !). I’m not a big fan of these low efficiency drag systems, but their R&D, DDT&E and materials costs are probably much lower than everybody else’s. And they’ve certainly got low labor costs for sewing enormous square meters of low cost fabric drag buckets.

             

            I think the winning AWE solution may be somewhere in the middle between the high-tech (high cost, high efficiency, high performance, composite construction, complex electronic autonomous controls) and the low-tech (low cost ,fabric construction, low L/D, , large(r) surface area, simple (semi-passive?) controls). Kind of hard to visualize something in between the two that combines the best features of both.

             

            Kind of got off track from the Betz discussion! But anyhow.

             

            Dave North

             

            Disclaimer:
            The views and opinions expressed herein are my own and do not necessarily state or reflect those of NASA or the United States Government, nor do they represent the official position of NASA.

             

          • Dave Lang
            Thanks for the comments DaveN. Even the Betz limit for the highly idealized stream-tube model has been re-visited with more sophisticated analyses and
            Message 5 of 25 , Apr 8, 2011
              Re: Betz limit [was: Re: [AWECS] Re: Makani M1]
              Thanks for the comments DaveN.

              Even the Betz limit for the highly idealized stream-tube model has been re-visited with more sophisticated analyses and realistic CFD models and it has been conjectured that for the assumed Betz-model, a more accurate power conversion limit could be as low as 30% efficiency (rather that the oft-quoted 60%)....this is due to flow escaping/diverting around the (idealized stream-tube) rotor geometry (not included in Betz' very elegant but somewhat over-simplified model) and thus avoiding getting robbed of its kinetic energy (thus the loss of efficiency).

              If one wants a challenge to stretch their AWE analytical wings, a good exercise is to try to figure out what an equivalent Betz limit would be for a close-hauled sail boat (a sailboat on a full downwind leg is pretty simple to do)....many surprises await, even to the point of having to examine what one means by (the blurry definition of) efficiency - it is the prime baseline ingredient if one wants to express power conversion as a ratio of two quantities :-)?

              DaveL



              At 2:54 PM -0500 4/8/11, North, David D. (LARC-E402) wrote:
               
              I agree with Dave Lang. The Betz limit sort of loses its relevance with most AWE configurations (flygen or groundgen). Betz developed this law to describe the theoretical maximum amount of energy that could be extracted from a full stream tube of fluid passing within the outer tip diameter of a turbine relative to the total energy passing through it.
               
              I suppose you could look at each turbine on a flygen separately and try to estimate the energy in the complex, helical stream tube that passes through each turbine on the flygen, but it seems like it would be very difficult (although not impossible) to model. Using the Betz limit calc with knockdown factors, or ³double Betz² is probably an oversimplification of the problem.
               
              So again, I agree with Dave L. Build the simulations as best you can (hand calcs all the way up to CFD). Build the test articles. Get data. Recalibrate models (or throw them out and use empirical performance curves from the data). Rinse and repeat.
               
              Efficiency does matter (e.g. high L/D, low drag tethers, super efficient turbines/generators, light weight/strong materials, etc.) but In the end ROI (which involves much more than efficiency) will probably be the only figure of merit that matters. The teams that can field marketable systems with decent ROI (better than fossil fuels) will be successful and profit. This is why I was musing before about the really low efficiency Chinese drag system before (don¹t slam me Doug S. !). I¹m not a big fan of these low efficiency drag systems, but their R&D, DDT&E and materials costs are probably much lower than everybody else¹s. And they¹ve certainly got low labor costs for sewing enormous square meters of low cost fabric drag buckets.
               
              I think the winning AWE solution may be somewhere in the middle between the high-tech (high cost, high efficiency, high performance, composite construction, complex electronic autonomous controls) and the low-tech (low cost ,fabric construction, low L/D, , large(r) surface area, simple (semi-passive?) controls). Kind of hard to visualize something in between the two that combines the best features of both.
               
              Kind of got off track from the Betz discussion! But anyhow.
               
              Dave North
               
              Disclaimer:
              The views and opinions expressed herein are my own and do not necessarily state or reflect those of NASA or the United States Government, nor do they represent the official position of NASA.
               

            • Pierre BENHAIEM
              Thanks for the comments DaveN and DaveL.They open my eyes about CFD models and the probably great difficulties to obtain realistic simulations of AWE. PierreB
              Message 6 of 25 , Apr 8, 2011
                Thanks for the comments DaveN and DaveL.They open my eyes about CFD models and the probably great difficulties to obtain realistic simulations of AWE.

                PierreB 




                > Message du 08/04/11 23:18
                > De : "Dave Lang"
                > A : AirborneWindEnergy@yahoogroups.com
                > Copie à : "North, David D. (LARC-E402)"
                > Objet : Re: Betz limit [was: Re: [AWECS] Re: Makani M1]
                >
                >  

                >

                Thanks for the comments DaveN.


                >

                Even the Betz limit for the highly idealized stream-tube model has been re-visited with more sophisticated analyses and realistic CFD models and it has been conjectured that for the assumed Betz-model, a more accurate power conversion limit could be as low as 30% efficiency (rather that the oft-quoted 60%)....this is due to flow escaping/diverting around the (idealized stream-tube) rotor geometry (not included in Betz' very elegant but somewhat over-simplified model) and thus avoiding getting robbed of its kinetic energy (thus the loss of efficiency).


                >

                If one wants a challenge to stretch their AWE analytical wings, a good exercise is to try to figure out what an equivalent Betz limit would be for a close-hauled sail boat (a sailboat on a full downwind leg is pretty simple to do)....many surprises await, even to the point of having to examine what one means by (the blurry definition of) efficiency - it is the prime baseline ingredient if one wants to express power conversion as a ratio of two quantities :-)?


                >

                DaveL


                >


                >


                >

                At 2:54 PM -0500 4/8/11, North, David D. (LARC-E402) wrote:


                >
                I agree with Dave Lang. The Betz limit sort of loses its relevance with most AWE configurations (flygen or groundgen). Betz developed this law to describe the theoretical maximum amount of energy that could be extracted from a full stream tube of fluid passing within the outer tip diameter of a turbine relative to the total energy passing through it.
                >

                >
                I suppose you could look at each turbine on a flygen separately and try to estimate the energy in the complex, helical stream tube that passes through each turbine on the flygen, but it seems like it would be very difficult (although not impossible) to model. Using the Betz limit calc with knockdown factors, or ³double Betz² is probably an oversimplification of the problem.
                >

                >
                So again, I agree with Dave L. Build the simulations as best you can (hand calcs all the way up to CFD). Build the test articles. Get data. Recalibrate models (or throw them out and use empirical performance curves from the data). Rinse and repeat.
                >

                >
                Efficiency does matter (e.g. high L/D, low drag tethers, super efficient turbines/generators, light weight/strong materials, etc.) but In the end ROI (which involves much more than efficiency) will probably be the only figure of merit that matters. The teams that can field marketable systems with decent ROI (better than fossil fuels) will be successful and profit. This is why I was musing before about the really low efficiency Chinese drag system before (don¹t slam me Doug S. !). I¹m not a big fan of these low efficiency drag systems, but their R&D, DDT&E and materials costs are probably much lower than everybody else¹s. And they¹ve certainly got low labor costs for sewing enormous square meters of low cost fabric drag buckets.
                >

                >
                I think the winning AWE solution may be somewhere in the middle between the high-tech (high cost, high efficiency, high performance, composite construction, complex electronic autonomous controls) and the low-tech (low cost ,fabric construction, low L/D, , large(r) surface area, simple (semi-passive?) controls). Kind of hard to visualize something in between the two that combines the best features of both.
                >

                >
                Kind of got off track from the Betz discussion! But anyhow.
                >

                >
                Dave North
                >

                >
                Disclaimer:
                > The views and opinions expressed herein are my own and do not necessarily state or reflect those of NASA or the United States Government, nor do they represent the official position of NASA.
                >

                >


                >


              • dave santos
                I also agree with DaveL that ROI rules but there are so many critical dimensions to a real AWE system that elevating any isolated factor (like Betz
                Message 7 of 25 , Apr 8, 2011
                  I also agree with DaveL that ROI rules but there are so many critical dimensions to a real AWE system that elevating any isolated factor (like Betz performance) above the rest  is a distortion. The double-Betz-hit conjecture is still an interesting question, but in the abstract.
                   
                  A personal beef is how often someone gushes about wind power increasing at the cube of velocity, which is true, but then they presume that they will effectively harvest all this increased power, which is not true. For a variety of reasons, including the  runaway capital cost of chasing top performance, its far safer to estimate one can harvest the increase at the square of velocity. This rule-of-thumb, while incorrect text-book physics, is a properly pessimistic design assumption.
                   
                  DaveN is probably right that the winning early AWE formula will be the sum of many middle-of-the-road trade-offs, a balance of performance & cost. Still, the solution may also involve some very clever inventive leaps unforseeable by any conservative design strategy.
                   
                   
                   
                • Bob Stuart
                  I ve read comments elsewhere by guys who feel challenged by the Betz limit and want to beat it. This misses the point on two counts. One, is that there are
                  Message 8 of 25 , Apr 8, 2011
                    I've read comments elsewhere by guys who feel challenged by the Betz limit and want to beat it.  This misses the point on two counts.  One, is that there are seldom any hard physical limits on the section of air available.  The other, more fundamental misunderstanding is that this is not like the problem of increasing efficiency in heat engines, where the limits are not fixed, but subject to improvements in design and materials.  Betz simply warns us about being too greedy, and reducing the flow through our device below the optimum level.

                    Bob Stuart

                    On 8-Apr-11, at 5:45 PM, dave santos wrote:


                    I also agree with DaveL that ROI rules but there are so many critical dimensions to a real AWE system that elevating any isolated factor (like Betz performance) above the rest  is a distortion. The double-Betz-hit conjecture is still an interesting question, but in the abstract.
                     
                    A personal beef is how often someone gushes about wind power increasing at the cube of velocity, which is true, but then they presume that they will effectively harvest all this increased power, which is not true. For a variety of reasons, including the  runaway capital cost of chasing top performance, its far safer to estimate one can harvest the increase at the square of velocity. This rule-of-thumb, while incorrect text-book physics, is a properly pessimistic design assumption.
                     
                    DaveN is probably right that the winning early AWE formula will be the sum of many middle-of-the-road trade-offs, a balance of performance & cost. Still, the solution may also involve some very clever inventive leaps unforseeable by any conservative design strategy.
                     
                     
                     


                  • Pierre BENHAIEM
                    Interesting paper:reading p.6 and 7 about tether-3 wires: Consider a realistic example of what one might try to build, a 100 kW wind generator with a tether
                    Message 9 of 25 , Apr 9, 2011
                    Interesting paper:reading p.6 and 7 about tether-3 wires:"

                    Consider a realistic example of what one might try to build, a 100 kW wind generator with

                    a tether length of 1500 feet (circa 450 metres) running 3-phase output at one of 380, 480, or 600

                    VAC. The corresponding phase currents at full power and full voltage into a resistive load are

                    88.6, 70.2, and 56.1 amperes respectively. In turn, the minimum required cable gauges are 3, 4,

                    and 6 AWG respectively, and greater if the reactance of the load cannot be guaranteed to stay

                    low. Even with a three-wire delta with no additional grounding lead, the weight of the copper

                    alone is 325, 257, or 162 kg for the 1500 ft run. The ohmic losses in the tether at full current are 7

                    kW, 5.5 kW, and 5.6 kW respectively, or 5.5 to 7 percent of total production over 1500 ft."

                    So for 1MW the weight of the copper alone (waiting for nanotube?) is from 1620 to 3250 kg with a tether length of 450 m.In static use M1 (hypothetical 175 m² and wind = 9 m/s) could lift only 850 kg,FlygenKite (700 m²) could lift 3400 kg (it is not yet enough).What are the repercussions about safety and reliability if lifting tether is only possible during dynamic use?

                    After discussions about "to Betz or not to Betz" limit(s),to consider that rough calculations I take is for a very (too) favorable hypothesis,and already requirements seem to be very difficult to satisfy.

                    PierreB
                    http://flygenkite.com  






                    > Message du 04/04/11 20:59
                    > De : "Pierre BENHAIEM"
                    > A : AirborneWindEnergy@yahoogroups.com
                    > Copie à :
                    > Objet : re: [AWECS] Makani M1
                    >
                    >  

                    > Correction of the precedent message:"To obtain 1000 000 W with 70 % of losses".It is 30% of losses,and 70% of the initial value.
                    >
                    > PierreB
                    > http://flygenkite.com  
                    >
                    >
                    >
                    >
                    >

                    > Message du 04/04/11 20:19
                    > > De : "Pierre Benhaiem"
                    > > A : AirborneWindEnergy@yahoogroups.com
                    > > Copie à :
                    > > Objet : [AWECS] Makani M1
                    > >
                    > >  

                    > >

                    > > On Makani website datas for the expected prototype M1 are;span = 35 m,and nominal power of 1 MW is reached with wind velocity = 9 m/s.Illustrations show a very thin design:the width of the wing should be about 3 m,maybe 5 m,but not more.So with an area of 175 m² the power could be formalized as following:

                    > > 4/81 aD A w3 CL(CL/CD)² with aD = air density;A = kite area;w = wind speed;CL = lift coefficient;cD = drag coefficient;4/81 is the transformation of 2/27 (see Diehl's formula and the link below) after conversion from a reel-out system towards a flygen system,including Betz limit and drag of an ideal rotor (8/9);

                    > > So 4/81  1.2  175  93   1.2 = 9072; complete formula is:                                                      4/81  1.2  175  93   1.2 (1.2/CD)²

                    > > To obtain 1000 000 W with 70 % of losses (comprising Betz limit): 1000 000/6350.4 = about 157.47 ; square root of   157.47 is about 12.549

                    > > So (CL/CD) should be the value of 12.549 which is also the ratio kite speed/wind speed.

                    > > So kite speed should be (12.549) 9  2/3 =75.294 m/s.  2/3 is the optimal kite speed after slowing down because of turbines. 

                    > > So the tip speed of turbines aloft should be some value like 300 m/s .

                    > > Thank you for corrections and comments.

                    > > PierreB

                    > > http://flygenkite.com

                    http://www.energykitesystems.net/OrthoKiteBunch/OptimizationOfAManualFlygen.pdf

                    > >         

                    > >  


                    >

                  • Theo Schmidt
                    Pierre BENHAIEM schrieb: ... ... I would think that this much copper or whatever is suboptimal. Using high-voltage DC might be better if the inverter aloft
                    Message 10 of 25 , Apr 10, 2011
                      Pierre BENHAIEM schrieb:
                      ...
                      > Consider a realistic example of what one might try to build, a 100 kW
                      > wind generator with a tether length of 1500 feet (circa 450 metres) running 3-phase output
                      > at one of 380, 480, or 600 VAC.
                      ...
                      > So for 1MW the weight of the copper alone (waiting for nanotube?) is
                      > from 1620 to 3250 kg with a tether length of 450 m.
                      ...

                      I would think that this much copper or whatever is suboptimal. Using
                      high-voltage DC might be better if the inverter aloft doesn't weigh too much.
                      Then one could use two wires and many kilovolts, saving a lot of material and
                      weight in the lines. Of course one also needs a sure method of shorting this out
                      in case of accidents. And remember that insulation weighs, too. An interesting
                      optimisation problem which I'm sure has been done.

                      Theo Schmidt
                    • Bob Stuart
                      ... I wondered about using aluminum conductors, but was recently informed on this list that for a combination of conductivity and tether strength, steel is
                      Message 11 of 25 , Apr 10, 2011

                        On 10-Apr-11, at 1:58 AM, Theo Schmidt wrote:

                        Pierre BENHAIEM schrieb:
                        ...
                        > Consider a realistic example of what one might try to build, a 100 kW 
                        > wind generator with a tether length of 1500 feet (circa 450 metres) running 3-phase output 
                        > at one of 380, 480, or 600 VAC. 
                        ...
                        > So for 1MW the weight of the copper alone (waiting for nanotube?) is 
                        > from 1620 to 3250 kg with a tether length of 450 m.
                        ...

                        I would think that this much copper or whatever is suboptimal. Using 
                        high-voltage DC might be better if the inverter aloft doesn't weigh too much. 
                        Then one could use two wires and many kilovolts, saving a lot of material and 
                        weight in the lines. Of course one also needs a sure method of shorting this out 
                        in case of accidents. And remember that insulation weighs, too. An interesting 
                        optimisation problem which I'm sure has been done.

                        Theo Schmidt

                        I wondered about using aluminum conductors, but was recently informed on this list that for a combination of conductivity and tether strength, steel is better.  If necessary, considerable line losses can be tolerated.  It may be possible to use a capacitor to use high-frequency AC with only a single conductor, saving the weight of insulation.  

                        Bob

                      • Robert Copcutt
                        ... Yes, that was me. The spreadsheet showing the calculations is still there. I promised to update it and post an excel version but have been too busy with
                        Message 12 of 25 , Apr 10, 2011
                          On Sun, 2011-04-10 at 07:26 -0600, Bob Stuart wrote:
                          >
                          >
                          > On 10-Apr-11, at 1:58 AM, Theo Schmidt wrote:
                          >
                          > > Pierre BENHAIEM schrieb:
                          > > ...
                          > > > Consider a realistic example of what one might try to build, a 100
                          > > kW
                          > > > wind generator with a tether length of 1500 feet (circa 450
                          > > metres) running 3-phase output
                          > > > at one of 380, 480, or 600 VAC.
                          > > ...
                          > > > So for 1MW the weight of the copper alone (waiting for nanotube?)
                          > > is
                          > > > from 1620 to 3250 kg with a tether length of 450 m.
                          > > ...
                          > >
                          > > I would think that this much copper or whatever is suboptimal.
                          > > Using
                          > > high-voltage DC might be better if the inverter aloft doesn't weigh
                          > > too much.
                          > > Then one could use two wires and many kilovolts, saving a lot of
                          > > material and
                          > > weight in the lines. Of course one also needs a sure method of
                          > > shorting this out
                          > > in case of accidents. And remember that insulation weighs, too. An
                          > > interesting
                          > > optimisation problem which I'm sure has been done.
                          > >
                          > > Theo Schmidt
                          > >
                          > >
                          > I wondered about using aluminum conductors, but was recently informed
                          > on this list that for a combination of conductivity and tether
                          > strength, steel is better. If necessary, considerable line losses can
                          > be tolerated. It may be possible to use a capacitor to use
                          > high-frequency AC with only a single conductor, saving the weight of
                          > insulation.
                          >
                          >
                          > Bob
                          >
                          Yes, that was me. The spreadsheet showing the calculations is still
                          there. I promised to update it and post an excel version but have been
                          too busy with other things to get it done yet. Hopefully soon.

                          HF AC down a single line would be inefficient. The way to go is 2 lines
                          carrying single phase AC, or preferably DC. Single phase generators tend
                          to vibrate too much so 3 phase is better. It only needs a simple
                          rectifier to convert to DC and a multi-kilovolt version would weigh very
                          little. Inverters are for going the other way.

                          Using 2 wires has an important safety benefit because if one breaks the
                          other can still be used to pull the airborne generator back to a safe
                          place. There is no need to insulate these lines and in fact the
                          insulation would just get in the way.

                          Something that no one else seems to be considering is using a
                          combination of airborne and ground-based generators. The airborne
                          generators/motors only need to be powerful enough to get the wing off
                          the ground and into the wind. Any more powerful than that and their
                          mass, and the mass of the tethers, start to become a problem. The main
                          generator will be on the ground driven by the reeling out of the twin
                          tethers.

                          Another reason for restricting the power of the airborne electrics is
                          the tip speed of the blades - as recently discussed by Pierre et al.
                          Noise from the blade tips rapidly increases as their speed increases
                          (5th to 6th power) so to be a good neighbour the blade rotation rate
                          needs to be kept down.

                          Quadracopters can be controlled very precisely. Putting more than 4
                          generators on a wing would be extravagant. Less than 4 makes control
                          difficult.

                          Robert.
                        • Pierre BENHAIEM
                          Consider a realistic example of what one might try to build, a 100 kW ... ... That is an extract from the attachment (see on the precedent
                          Message 13 of 25 , Apr 10, 2011
                            "Consider a realistic example of what one might try to build, a 100 kW
                            > wind generator with a tether length of 1500 feet (circa 450 metres) running 3-phase output
                            > at one of 380, 480, or 600 VAC.
                            ..."
                            That is an extract from the attachment (see " " on the precedent message).Note:Makani uses a bruschless DC motor.
                             
                            PierreB




                            > Message du 10/04/11 15:06
                            > De : "Theo Schmidt"
                            > A : AirborneWindEnergy@yahoogroups.com
                            > Copie à :
                            > Objet : best voltage [was: Re: [AWECS] Makani M1 [1 Attachment]]
                            >
                            >  

                            > Pierre BENHAIEM schrieb:
                            > ...
                            > > Consider a realistic example of what one might try to build, a 100 kW
                            > > wind generator with a tether length of 1500 feet (circa 450 metres) running 3-phase output
                            > > at one of 380, 480, or 600 VAC.
                            > ...
                            > > So for 1MW the weight of the copper alone (waiting for nanotube?) is
                            > > from 1620 to 3250 kg with a tether length of 450 m.
                            > ...
                            >
                            > I would think that this much copper or whatever is suboptimal. Using
                            > high-voltage DC might be better if the inverter aloft doesn't weigh too much.
                            > Then one could use two wires and many kilovolts, saving a lot of material and
                            > weight in the lines. Of course one also needs a sure method of shorting this out
                            > in case of accidents. And remember that insulation weighs, too. An interesting
                            > optimisation problem which I'm sure has been done.
                            >
                            > Theo Schmidt
                            >


                          • Robert Copcutt
                            We need to be careful with the rather misleading terminology here. Brushless DC refers to the whole package including the electronics. The motor/generator
                            Message 14 of 25 , Apr 10, 2011
                              We need to be careful with the rather misleading terminology here.
                              "Brushless DC" refers to the whole package including the electronics.
                              The motor/generator itself will be 3 phase AC.

                              Robert.


                              On Sun, 2011-04-10 at 20:41 +0200, Pierre BENHAIEM wrote:
                              >
                              > "Consider a realistic example of what one might try to build, a 100
                              > kW
                              > > wind generator with a tether length of 1500 feet (circa 450 metres)
                              > running 3-phase output
                              > > at one of 380, 480, or 600 VAC.
                              > ..."
                              > That is an extract from the attachment (see " " on the precedent
                              > message).Note:Makani uses a bruschless DC motor.
                              >
                              > PierreB
                              >
                              >
                            Your message has been successfully submitted and would be delivered to recipients shortly.