Don,
Thanks for the reply. Great points. Actually response time would play a major part of the "price" since having in-house service engineers would essentially make downtime of instruments minimal. That is currently our practice for HPLCs which are in-house at this time.
The hardest part I am finding to price out is estimated parts, since as long as everything is running smooth, the only parts you have to have on hand is PM parts, and consumables.
The real issue is trying to estimate how often large hardware falures occur such as turbos, control boards, acquisition boards, rough pumps etc.
With only a few MS systems it is probably cost effective to get the service contract but once you have a large number of these systems it becomes a lot more financially viable to consider an alternative. The problem is, what is that magical breakpoint # of instruments when in-house service becomes cost effective.