The principle of Coulomb charging effects in single electron tunneling (SET
) devices is well understood in terms of the 'orthodox theory'. However, kn
own deviations arise due to dissipative heating effects, higher-order tunne
ling or coupling to the electromagnetic environment. Those can be accounted
for by more sophisticated theories, which are unfortunately rather unhandy
and too general for the analysis of a realistic experiment. Based on exten
sive DC measurements on metallic SET transistors, we are able to identify a
ll significant deviations (in the weak tunneling limit) and to discuss the
relevant theoretical models (i.e. 'horizon picture', self-heating) needed f
or an appropriate description and an intuitive understanding of the experim
ents. (C) 2000 Elsevier Science B.V. All rights reserved.