Large Negative V_oc Values

[Diego Camarillo]

Hi all,

I'm working through trying to recreate some results from PVSyst via PVLib. I've been able to recreate the irradiance fairly well, and am now working on recreating the power. However, I'm currently running into an issue during the singlediode computation.

The _lamberw_v_from_i is returning large negative values of v_oc (on the order of -400, -500) correlating to what appears to be irradiance values below 100 W/m2. As soon as the irradiance values increase above that threshold, v_oc returns to expected levels. 

These negative values then cause an error in _golden_sect_DataFrame, as the upper bounds are smaller than the lower bounds (line 344 in pvlib.tools.py).

This is a little outside of my wheelhouse, so I'm looking for a little help in determining what could be driving the v_oc's to be negative here, and how I can fix it.

Thanks in advance,

Diego

[cwh...@sandia.gov]

This is likely to be an issue with the `module_parameters` can you post the values you are using?

Cliff

[Diego Camarillo]

Parameters for calcparams_pvsyst:
alpha_sc
7
gamma_ref
1.004
mu_gamma
-0.0004
I_L_ref
13.99
I_o_ref
3.03E-11
R_sh_ref
503.6781
R_sh_0
1400
R_s
0.196
cells_in_series
72
R_sh_exp
5.5
EgRef

1.121

PAN file parameters:

Technol=mtSiMono
  NCelS=72
  NCelP=2
  NDiode=3
  SubModuleLayout=slTwinHalfCells
  FrontSurface=fsARCoating
  GRef=1000
  TRef=25.0
  PNom=550.0
  PNomTolLow=0.00
  PNomTolUp=3.00
  BifacialityFactor=0.700
  Isc=13.990
  Voc=49.80
  Imp=13.120
  Vmp=41.95
  muISC=7.00
  muVocSpec=-132.0
  muPmpReq=-0.340
  RShunt=500
  Rp_0=1400
  Rp_Exp=5.50
  RSerie=0.196
  Gamma=1.004
  muGamma=-0.0004
  VMaxIEC=1500
  VMaxUL=1500
  Absorb=0.90
  ARev=3.200
  BRev=3.200
  RDiode=0.010
  VRevDiode=-0.70
  AirMassRef=1.500
  SandiaAMCorr=50.000
  RelEffic200=-3.70

[cwh...@sandia.gov]

One problem is the alpha_sc = 7 value. pvlib expects alpha_sc to be in units of A/C. PVsyst doesn't specify precisely the unit for that parameters; the help files suggest it is in %/C (% of Isc at STC), but a value of 0.07 x 13.99 = 0.9793 A/C seems far too high.

I can't help translate the muISC = 7.00 value in the PAN file to A/C. Maybe a PVsyst user here can help?

alpha_sc = 7 will result in enormous photocurrents. I don't think that also results in negative Voc. Can you give provide one effective irradiance and cell temperature value where you see negative Voc?

Cliff

[Diego Camarillo]

Reducing alpha_sc to a smaller value (I used .03 after a quick google search, open to better estimates) eliminates the problem, and puts the computed power output very close to the expected values.

Is there a way to compute alpha_sc analytically?

[Mark Campanelli]

Do you have the module spec sheet? Should be able to find it there, but you may need to convert the units.