function thm_sst_dist3d_16x1,ion=ion,elec=elec,time,probe=prb,index=index dat = {thm_sst_dist3d_16x1} spin_period = 3. dim = size(/dimension,dat.data) nenergy = dim[0] nbins = 1 ; dim[1] one = replicate(1.,16) dat.dphi = 360 ;22.5 if keyword_set(ion) then begin dat.theta = 0 ;replicate(1,16) # [52*one,-52*one,25*one,-25*one] dat.dtheta = 90 ; phi16 = (findgen(16)+.5) * 22.5 ; sphi16 = shift(phi16,8) dat.phi = 0 dat.dphi = 360 endif if keyword_set(elec) then begin dat.theta = 0 ;replicate(1,16) # [-52*one,+52*one,-25*one,+25*one] dat.dtheta = 90 ; phi16 = (findgen(16)+.5) * 22.5 ; sphi16 = shift(phi16,8) dat.phi = 0 ;replicate(1,16) # [sphi16,phi16,phi16,sphi16] dat.dphi = 360 ;22.5 endif dat.integ_t = dat.dphi / 360 * spin_period if 0 then begin ; Warning: not final cals! idap_start = [12,19,26,34,44,69,103,150,215,306,506,906,2000,3000,4000,5000] * 1.5 * 1000 idap_width = [ 7, 7, 8,10,25,34, 47, 65, 91,200,400,3060,1000,1000,1000,1000] * 1.5 * 1000 edap_start = [12,19,26,34,44,69,103,150,215,306,506,906,2000,3000,4000,5000] * 1.5 * 1000 edap_width = [ 7, 7, 8,10,25,34, 47, 65, 91,200,400,2000,1000,1000,1000,1000] * 1.5 * 1000 energy = (2*idap_start + idap_width)/2 + 5000. ; midpoint energy dat.energy = energy # replicate(1,nbins) ; total energy width denergy = (idap_width) dat.denergy = denergy # replicate(1,nbins) endif else begin ;not really the final cals either...but closer... thm_sst_energy_cal,energy=energy,denerg=denergy,inst = keyword_set(ion),probe=prb ;single angle means that the anode by anode corrections don't exactly apply. This code uses a kluge, the average energy & denergy across all anodes dat.energy = total(energy,2)/4. dat.denergy = total(denergy,2)/4. endelse ;weights = calc_omega_flt2(dat.theta,dat.phi,dat.dtheta,dat.dphi, 1.) ;dat.domega = weights[0,*,*] dat.nenergy = nenergy dat.nbins = nbins dat.bins = 1 dat.gf = 4. ; correction to notional GF for reduced data. The correction in the spin-plane is account for by increasing the integration ;time of the data, so this correction only includes the theta component. ;The one angle reduced bin contains contribution from all four anodes, so the effective GF along the theta axis is increased by a factor of 4 ;dat.gf=1. dat.units_procedure = 'thm_sst_convert_units' dat.geom_factor = .1 dat.eclipse_dphi = !values.d_nan return,dat end