Like Refik is saying, you need to add pacemaker currents. Another good one to try would be Ih. You could base your model on HH equations and try adding the h-current and t-current components.
From a dynamical systems point of view, a pacemaking neuron needs to have a stable limit cycle. You could build such a neuron using the dynamical systems approach, for example by tweaking the computationally efficient Izhikevich model (http://www.izhikevich.org/publications/spikes.htm). Also see Eugene Izhikevich's excellent textbook on the subject: "Dynamical Systems in Neuroscience. The geometry of Excitability and Bursting".
As an additional resource, have a look at the BioModels Database: http://www.ebi.ac.uk/biomodels-main. You will find lots of interesting models to play with, such as this one: http://www.ebi.ac.uk/biomodels-main/static-pages.do?page=ModelMonth%2F2012-03
Cardiac myocytes are a good example of pacemaker cells and arguably the best characterized. These endogenous pacemaker currents will activate and inactivate time- and voltage-dependentl. As long as the cell's membrane potential is at resting level, the cell will fire rhythmic action potentials.