"If DNP 3.0 had been implemented on the 8052 platform I could make the conversion for much less (I think)."

OK, so I'll admit that I have no idea what DNP is (let alone v3.0 of it).
However, I assume we're talking about a comms protocol here - and I do know about them.
If you were to charge a true commercial rate for your time - not forgetting all the testing, support, documentation, etc - to develop a protocol into a marketable product, I doubt very much that you'd improve much on $6000.

[rant]
A significant amount of that $6000 would go towards coping with the bodges necessary to work around all the problems caused by people not doing the job properly; implementations which look quite close to the spec at first examination, but actually have serious problems when examined closely - especially under "exceptional" conditions! :-(
[/rant]

I doubt I'd take on a fixed-price protocol development project at GBP6000 - let alone US$6000!!