In the work presented here, fatigue crack growth (FCG) experiments were performed on three commercial polyethylene (PE-HD) pipe grades with a bimodal molecular weight distribution. In order to investigate the influence of different specimen types on FCG results, tests were conducted using compact type (CT) specimens and cracked round bars (CRB). The effects of R-ratio and frequency on FCG behaviour were also studied. A discontinuous crack growth mechanism was characteristic for all materials investigated. By comparing the materials it turned out that the molecular weight should be the most important parameter for the increase in FCG resistance. On the basis of the results gained so far, testing time was clearly reduced using CRB rather than CT specimens. It was found out that with increasing test frequency there was a noticeable increase in crack growth resistance in CT specimens in the low crack-growth regime, while in CRB specimens no differences could be found because of the large scatter band. For all investigated frequencies, a temperature increase resulting from hysteretic heating was measured at the crack tip in CT specimens. FCG rates showed a great dependence on R-ratio in terms of the stress intensity factor range. There may be a tendency for higher crack growth rates at higher R-ratios as a result of more creep crack extension associated with the higher maximum and mean stress intensity factor levels.