This paper presents the pulse-type hardware neural networks (P-HNN) which could output the pulse width modulation (PWM) waveform of PWM servo motor for robots. Basic components of P-HNN were pulse type hardware neuron model (P-HNM). P-HNM consisted by the cell body model, the axon model and the synaptic model. P-HNM had the same basic features of biological neurons such as threshold, refractory period, spatio-temporal summation characteristics and enabled the generation of continuous action potentials. We used the synaptic model which could vary the synaptic weights by controlling voltage. As a result, we showed that P-HNN could control the PWM servo motor from 0 to 180 degrees.