Understanding condition of road surface is very important especially for road maintenance and asset management. There are many approaches to obtain road surface condition data, however almost all of them are either low speed with intensive human intervention techniques (visual inspection) or techniques that require advanced measurement equipment (sophisticated profilers), which usually comes with high costs and requiring skillful operators. Using smartphone to collect data is a promising alternative because of its low cost and easy to use features in addition to its potentially wide population coverage as probe devices. This paper explores features and relationship of acceleration vibration that may be useful to express or estimate road roughness condition, which is the main focusing road surface condition for this paper. Results from our experiment and analysis show that acceleration data collected by smartphone sensors at different driving speeds has different significant linear relationships with road roughness condition.