You are unlikely to get this from other sources so hope you find it both useful and also revealing regarding "private" cloud computing and what I actually l think are it limited benefits.

1) New tools and processed needed. The tools and processes used for cloud computing are COMPLETELY different to those that an existing IT team already have. Attempting to apply these existing tools (whilst possible in someways) rails against the liklihood of gaining the maximum benefit of cloud computing.
2) The training needed to learn these new tools and processes.
3) The human resources diverted to developing thew new skills that have to be backfilled. It is likely that they will become dedicated to the new environments.
4) The cost of havingn to run duplicate environments for a period of time or more likely, indefinately. Not all applications can run on the cloud effectively.
5) the cost of modifying your applications to operate on the cloud (not all apllications handle well access to dynamic compute resource allocation like memory, computing power, disk etc). there is a 100% likelihood that you will incur a cost here.
6) The likelihood that the external application license you have will not work in a cloud environment. this will either need to be fixed or the software replaced! If replaced, then the replacement cost!!
7) The additional computing infrastructure and software investment needed
8) the cost and time of the procurement process itself! And,
9) the likelihood that some of your product selections will prove incorrect and have to be again replaced within 12 months! It is a rapidly changing and developing market which only increases the risk of this.
10) New integration API's will be needed
11) New security tools and processes will be needed
12) Application management automation. This one is MASSIVE and is a product of and requirement of cloud computing. You are now scaling your application infrastructure up and down dynamically. To do this you need to build new templates and capabilities to highly "automate" tasks that once were previously able to be handled manually. To not do this at all is to simply not do cloud computing at all. You may as well not bother without this investment. Each app will probably take 2-4 months to set up correctly btw.
12) the cost of internal support. New systems, new tools, likely new applications and certainly new integrations = increased support costs.

I expect that unless your company has in the order of 500 serveres or more, that the benefit of "private" cloud computing will not be realised nor justifyable if ever calculated accurately.

Why not consider public cloud computing (which changes many of the economic impacts in a more positive way)? If your argument is becasue of "security" then I feel you may be allowing yourself to become a victim of the IT vendors "FUD" tactics. Public cloud computing security is not bad, it is just different.

the truth is that done properly, adoption of cloud computing should yeild at least 40% savings. I have seen up to 80% and more of the total TCO, but TCO of itself is not all that you want to measure. Corporate IT agility is a major economic benefit. I do not believe that these benefits can be achieved through private cloud computing unless the company's IT infrastructure is massive (like a global bank for example).

many use private cloud computing as a methods of getting into the cloud - a stepping stone if you will. I think it has great value being used this was as "part of the journey" but not as the destination.

Hope some of this helps. I could easily go on and on!! :-)