Cylindrical solar cell panels, also known as evacuated tubular solar collectors, use thin-film solar cells inside glass tubes to efficiently collect sunlight. Tracking is not required. They have high efficiency, are easy to install and clean, and have low roof penetration. Numerical modeling shows good agreement with experimental performance data, validating the design approach. Extensive measurements and comparisons to theory indicate the technology works well.

1. !"#$%&'$()#*)%&*+#),*-.#)'*!.##/(,.'*0/.1/,'$/23*
45/.'"*)%&*678/'$1/%,*
!"#$%#&'(&)*+,#$*+-$./0,1$('$2"#$34(56+-&*7$%8$%*+#5$
*
9:*;.5%*<.25/#=>=?*0'/@*A:*-1/2,)&=B*C)%$/#*-5D##=E*A/,/'*-,/85/%2=E*)%&*4$1*F/)#"E*
*
>*A5.,.%*6%@$%//'$%@=*GG!*
?*!.##/@/*.H*I8J()#*-($/%(/2=*45/*K%$L:*.H*M'$N.%)*
B*-.#*O&/)2*)%&*-.#)'*6%/'@"*P),/'$)#2*)%&*-.#)'*!/##2*
E*6#/(,'$()#*6%@$%//'$%@*C/8)',1/%,=*-)%,)*!#)')*K%$L/'2$,"*
*
I-M*-IGM9*?Q>?*R*6$%&5.L/%=*45/*S/,5/'#)%&2*
*
>*>B*S.L/1T/'*?Q>?*

2. O%,'.&D(J.%*
U !"#$%&'$()#*2.#)'*(/##*,DT/2*D2/*
,5$%VW#1*!O0-*,/(5%.#.@"3*
R 4')(X$%@*%.,*'/YD$'/&*
R 6)2/*.H*1.D%J%@=*(..#$%@=*)%&*
(#/)%$%@*
R !O0-*Z*?Q[*/($/%,*
R ]$%&*#.)&2*'/&D(/&*
R G.^*'..H*8/%/,')J.%*
U -$@%$W()%,*1/&$)*(.L/')@/*.H*,5/*
,/(5%.#.@"3*
R F.^*&./2*$,*^.'X_*
R C./2*$,*^.'X*^/##_*
R ]/*)'/*%.,*$%,/'/2,/&*$%*,5/*
8.#$J(2*`1D(5a*
?*>B*S.L/1T/'*?Q>?*
Photo by Greg P. Smestad

3. P.JL)J.%*
U b/c/'*D%&/'2,)%&*,5/*
("#$%&'$()#*,DT/*,/(5%.#.@"3*
R ]5),*$2*,5/*,5/.'"*T/5$%&*
8/'H.'1)%(/_*
R C./2*$,*^.'X*T/c/'*,5)%*%.%V
,')(X$%@*d),*8)%/#2_*
R !)%*2.e^)'/*1.&/#$%@*T/*
D2/&*,.*)((D'),/*&/,/'1$%/*
8/'H.'1)%(/_*
R !)%*,5/*,/(5%.#.@"*T/*
$18'.L/&_*
U 6&D()J.%)#*,..#*H.'*
D%&/'@')&D),/*2,D&/%,2*),*
-)%,)*!#)')*K%$L/'2$,"*
B*>B*S.L/1T/'*?Q>?*
Photos by Greg P. Smestad

4. q
dDB
D6
!5)')(,/'$2J(2*.H*!"#$%&'$()#*4DT/*A)%/#2*
U 4DT/*&$)1/,/'*f*!g*f*??*11*
U 4DT/*28)($%@*f*"*f*?!g*f*EE*11*
U 4DT/VT)(X8#)%/*2/8)')J.%*f*!#*f*>:h!g*f*BB*11*
U !/##*&$)1/,/'*f*!$*f*>h*11*
U 4DT/*G/%@,5*f*%*f*>QQQ*11*
U M'')"*f*&'()*+*f*EQ*
E*>B*S.L/1T/'*?Q>?*
Direct sunlight
Glass
Tubes
Index-Matching Fluid
CIGS

5. !9:.;<$
b)2/&*.%*,5/*'/2/)'(5*.H*/L)(D),/&*,DT/2*
h*>B*S.L/1T/'*?Q>?*

6. C'$L$%@*9/)2.%*H.'*!"#$%&'$()#*4DT/2*
g*>B*S.L/1T/'*?Q>?*
!
-8 -6 -4 -2 0 2 4 6 8
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
d/D6= 1
d/D6= 2
d/D6= inf
Solar Hour
ED/E0
Equinox
Tilt = Latitude
Replicated:
D. C. Beekley and G. R. Mather, Jr.,
“Analysis and experimental tests of a
high-performance evacuated tubular
collector,” DOE/NASA CR-150874, 1978.
Manuscript Freely obtained at:
http://ntrs.nasa.gov/archive/nasa/
casi.ntrs.nasa.gov/19790008199_1979008199.pdf

7. =>&#,2$(+$?65>+-&>,*5$!@A#1$ 4,*B#&$(+$?65>+-&>,*5$!@A#1$
O'')&$)%(/*9)J.*.%*!"#$%&'$()#*4DT/2*
i*>B*S.L/1T/'*?Q>?*
ED/E0
ER/E0ED = irradiance on tilted cylindrical tubes from direct sunlight
ER = irradiance on tilted cylindrical tubes from backplane scatter
E0 = irradiance on flat panel normal to direct sunlight
E = irradiance on horizontal flat panel from direct sunlight
Solar Hour Solar Hour
Latitude = 40°
Tilt = 28°
d = 2D6
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19790008199_1979008199.pdf
Latitude = 40°
Tilt = 28°
White-Panel Backplane
d = 2D6

8. -8 -6 -4 -2 0 2 4 6 8
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
!.18)'$2.%3*!"#$%&'$()#*M'')"*j*+#),*A)%/#*
k*>B*S.L/1T/'*?Q>?*
0.5(ED+ER)/E0orE/E0
Solar Hour
Flat Panel
Cylindrical
Tubes
The theory
indicates flat
panels should be
better around
noon, while the
tubes are better in
early and late
hours
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19790008199_1979008199.pdf
1.0
-
0

9. :C%:;DE:F!$
P/)2D'/1/%,2*)%&*(.18)'$2.%2*,.*,5/.'"*
l*>B*S.L/1T/'*?Q>?*

10. 45/"*G..X*G$X/*+#),*A)%/#2m*
Greater than the critical angle
Photo by Greg P. Smestad
>B*S.L/1T/'*?Q>?* >Q*

11. m)%&*45/"*G..X*G$X/*b#$%&2*
Less than the critical angle
Photos by Greg P. Smestad
Tubes Shadows
>B*S.L/1T/'*?Q>?* >>*

12. 6&@/V9)"*0/.1/,'"*.H*!"#$%&'$()#*4DT/2*
>?*>B*S.L/1T/'*?Q>?*
Cylindrical Tubes
Performance Region
Critical or
Acceptance
Angle
qa
qa
Flat Panel
Performance
Region
Flat Panel
Performance
Region

13. P/)2D'/&*On*!D'L/2*H.'*-.#)'*A)%/#2*
>B*S.L/1T/'*?Q>?* >B*

14. !.18)'$%@*An*P.&D#/*A.^/'*
First glance: cylindrical tube
panel (Solyndra) performs
better than flat panel (Solar
Frontier)
Second glance: we need to
normalize for differences in
solar cell active area
Third glance: there is the
characteristic kink in the curve
due to critical angle
phenomenon
Electrical
3 September 2012
Latitude = 37.3492°
Longitude = -121.9381°
Altitude = 20 m
White-Panel Backplane
>B*S.L/1T/'*?Q>?* >E*

15. S.,*-.*+)2,3*O,*$2*A.^/'*C/%2$,"*
First glance: flat panel (Solar
Frontier) performs better than
cylindrical tube panel (Solyndra)
Second glance: we are gaining
some power from the open regions
in the panel, but not doubling
Third glance: the kink is still there
due to critical angle phenomenon
Electrical
3 September 2012
Latitude = 37.3492°
Longitude = -121.9381°
Altitude = 20 m
White-Panel Backplane
>B*S.L/1T/'*?Q>?* >h*

16. :G/#&>)#+2*5H$:5#,2&>,*5$
%(I#&$=#+1>26$ !"#(&6H$D&&*->*+,#$;*0($
!.18)'$2.%*.H*45/.'"*)%&*678/'$1/%,*Electrical
>B*S.L/1T/'*?Q>?* >g*

17. !"#$%&'$()#*4DT/*M'')"*!.18)'$2.%*,.*45/.'"*
28 October 2012 (AM1.5 Conditions)
Latitude = 37.3492°
Longitude = -121.9381°
Altitude = 20 m
Black Velvet Backplane
>B*S.L/1T/'*?Q>?* >i*
45/.'"3*&$'/(,*.%*("#$%&'$()#*,DT/2*
ED/E0
Solar Hour
Latitude = 40°
d = 2D6

18. 4.J!KL;:$E.=:MDFN$
n)#$&)J.%*)%&*.8J1$N)J.%*
>k*>B*S.L/1T/'*?Q>?*

19. -.D'(/*-/,D83*S96G*-.#)'*P.&/#*
0
200
400
600
800
1000
1200
9/2/12 7:12 PM 9/3/12 12:00 AM 9/3/12 4:48 AM 9/3/12 9:36 AM 9/3/12 2:24 PM 9/3/12 7:12 PM 9/4/12 12:00 AM
globalsolarradiation(W/m^2)
Bird
Bras
Ryan-Stolzenbach
>l*>B*S.L/1T/'*?Q>?*
Site data and time info
latitude in decimal degrees (positive in northern hemisphere) !"#!$%&
longitude in decimal degrees (negative for western hemisphere) '(&(#%!)(
ground surface elevation (m) 20.0
time zone in hours relative to GMT/UTC (PST= -8, MST= -7, CST= -6, EST= -5) -8
daylight savings time (no= 0, yes= 1) 1
start date to calculate solar position and radiation 3-Sep-12
start time 12:30 AM
time step (hours) 0.5
number of days to calculate solar position and radiation 1
Bird model parameters
barometric pressure (mb, sea level = 1013) 1013
ozone thickness of atmosphere (cm, typical 0.05 to 0.4 cm) 0.225
water vapor thickness of atmosphere (cm, typical 0.01 to 6.5 cm) 2
aerosol optical depth at 500 nm (typical 0.02 to 0.5) 0.26
aerosol optical depth at 380 nm (typical 0.1 to 0.5) 0.3
forward scattering of incoming radiation (typical 0.85) 0.85
surface albedo (typical 0.2 for land, 0.25 for vegetation, 0.9 for snow) 0.2
Bras model parameter
Bras atmospheric turbidity factor (2=clear, 5=smoggy, default = 2) 2
Ryan-Stolzenbach model parameter
Ryan-Stolzenbach atmospheric transmission factor (0.70-0.91, default 0.8) 0.8
Calculation of solar position based on NOAA's functions and clear-sky solar radiation based on
Bird and Hulstrom's model, Bras model, and Ryan and Stolzenbach's model.
On horizontal
surface

20. -.e^)'/*0/.1/,'"*.H*-$%@#/*A)%/#*
?Q*>B*S.L/1T/'*?Q>?*
Solar Spectrum = AM1.5
CIGS QE = NREL literature
3 September 2012
Latitude = 37.3492°
Longitude = -121.9381°
Altitude = 20 m
White-Panel Backplane
Solar noon results only shown
on central tube
Pe,tube (tracing) = 2.53 W
Pe,array = 101.2 W

21. O=$8>1@*5>P*0(+$ Q=$?"*&2$
9)"V4')($%@*9/2D#,2*
?>*>B*S.L/1T/'*?Q>?*
Note: irradiance on back surface of tube
Pe,tube = 2.53 W
comparable to
experiment
1.30
pm

22. P.&/#/&*6#/(,'$()#*A.^/'*0/%/')J.%*
??*>B*S.L/1T/'*?Q>?*
Comparison of Solyndra
cylindrical tube panel:
experiment and ray trace
model – good agreement.
Variations due to solar source
assumptions, such as
atmospheric conditions
Indicates that software ray-
tracing model validates virtual
design method.
Electrical
3 September 2012
Latitude = 37.3492°
Longitude = -121.9381°
Altitude = 20 m
White-Panel Backplane

23. o,/%&D/*
U +'/%(5*].'&3*
R n/'T3*/7,/%&/&*
R S.D%3*'/)(5*
U o,/%&D/*$2*)*@/.1/,'$(*H)(,.'3*
U O,*&/2('$T/2*,5/*dD7*8'.8)@)J.%*(5)')(,/'$2J(2*.H*)%*.8J()#*2"2,/13*
?B*>B*S.L/1T/'*?Q>?*
!! "=
aperture
2
cos ddAn s#E
( ) ,cosˆ,
aperture
!! "=# ddAL s$ar
2
aperture
cos
n
L
ddAL
s
ss
E
=
!=" ## $
dAs
q!
dW!
Arbitrary Source Radiance Lambertian Source Radiance
Aperture

24. A'/#$1$%)'"*I8J1$N)J.%*-,D&$/2*
U I8J1$N/&*(.%W@D')J.%*.H*,5/*
2"2,/13*
R 9/2D#,2*$%&$(),/*"*f*?!,*)%&*!#-f*
>:h!g*)'/*.8J1)#*
R 4$#,*)%@#/*`+a*f*G)J,D&/*`.a=*
.8J1$N/2*8/'H.'1)%(/*T"*
1)X$%@*'/28.%2/*1.'/*D%$H.'1*
.L/'*"/)'*
R C$p/'/%,*1/'$,*HD%(J.%*()%*T/*
D2/&*,.*.8J1$N/*8/'H.'1)%(/*
H.'*)*@$L/%*J1/*.H*,5/*&)"q"/)'3*
U Me/'%..%*,.*5)%&#/*(..#$%@*%//&2*
U P.'/*/YD$,)T#/*.L/'*,5/*"/)'*2D(5*
,5),*^$%,/'*8/'H.'1)%(/*&'$L/2*
&/2$@%**
U A'/#$1$%)'"*'/2D#,2*25.^*,5),*)*25)8/*
,5),*5/#82*,')8*$%($&/%,*')&$)J.%*$2*
T/c/'3*
R !'.22*2/(J.%*.H*r,DT/s*D2$%@*SK9b-:*
I8J1$N/&*^$,5*')"*,')($%@*2.e^)'/*
R 45/*'/(/22/2*,')8*#$@5,*
R 45$2*@$L/2*#$@5,*)*r2/(.%&*(5)%(/*
R -5)8/*&/8/%&2*@'/),#"*.%*1/'$,*
HD%(J.%*
U S.,/*,5),*8/'H.'1)%(/*^$##*)88'.)(5*
)*d),*8)%/#*
?E*>B*S.L/1T/'*?Q>?*

25. ?.F?MR4D.F4$
+D,D'/*2,D&$/2*
?h*>B*S.L/1T/'*?Q>?*

26. ?(+,5@1>(+1$
U !"#$%&'$()#*,DT/2*.%#"*.p/'*1$%.'*
$18'.L/1/%,2*`1.'%$%@*)%&*/L/%$%@*
5.D'2a*
U C/('/)2/&*8/'H.'1)%(/*)'.D%&*2.#)'*
%..%*
U 6p/(JL/*')"*,')(/*1.&/#$%@*()%*T/*
&.%/*
U o,/%&D/*&'$L/2*,5/*&/2$@%3*
4.*$@%.'/*$,*()%*T/*5)N)'&.D2t*
U -"2,/1*()%*T/*$18'.L/&*`2#$@5,#"a*^$,5*
2.e^)'/*.8J1$N)J.%*R*S.,*/%.D@5*
J1/*,.*@.*$%,.*,5/*&/,)$#2*5/'/*
J@2@&#$&#1#*&,"$
U P.'/*.8J1$N)J.%*()2/23*
R P.&$W()J.%*.H*1/'$,*HD%(J.%*
R P.'/*2,D&"*.H*r,DT/s*25)8/*
U O%(#D&/*S)J.%)#*]/),5/'*-/'L$(/*
&),)*,)X$%@*$%,.*)((.D%,*#.()#*2X"*
(.%&$J.%2*`$:/:=*$%*-)%,)*!#)')a*
U -8/(D#)'*/&@/*')"*(.%(/%,'),.'*
T/,^//%*,DT/2=*TD,*,5$2*2,/8*
rL$.#),/2s*,5/*2$18#$($,"*.H*,5/*
,DTD#)'*)'')"*@/.1/,'"*
!.%(#D2$.%*)%&*+D,D'/*9/2/)'(5*
?g*>B*S.L/1T/'*?Q>?*
,cos
aperture
2
!! "= ddAn s#$

27. M*-$#L/'*G$%$%@3*6&D()J.%*
>B*S.L/1T/'*?Q>?* ?i*

28. ]5),*,.*&.*^$,5*>h*P$##$.%*4DT/2_*
www.solideas.com/GlassTubes.html
>B*S.L/1T/'*?Q>?* ?k*