Bear in mind there wasn't enough information provided to complete help resolve your issue. However, based on what you've provided, there could be several reasons why your WCF service wound up in a faulted state when scaling up to multiple instances. Remember, when scaling out your WCF service has to be configured to run simultaneously.
One possible cause is your service isn't initially designed to run to on multiple instances. This exposes your service to conflicts between instances, race conditions, and other synchronization issues. To resolve, look at implementing synchronization mechanisms such as a distributed cache and/or ensuring it's completely stateless.
Another possible reason could be a configuration issue. This could be using a binding that is not compatible or better suited with multiple instances. Bindings that are better suited are NetTcpBinding, NetHttpBinding, WSHttpBinding, and BasicHttpBinding.
There may also be an inner exception that was logged that led up to the faulted state; feel free to share any log snippets should the above suggestions not be helpful.