I'm not sure I completely understand your question but if you use an individual structural image for your 3D source reconstruction SPM will fit the elctrode cap to it and then you can get the electrode locations from D.inv{1}.datareg.sensors
I have a question about EEG images: is It possible to coregistrate EEG and fMRI images so we can identify the position of the electrodes on the fMRI images?