Never thought I'd quote Rumsfeld in a favourable context, but it's not the work you know about that pushes up the timescales - it's the unknown unknowns.
All this doubling and adding 50% shouldn't really be necessary if you can be certain [;-)] that your requirements describe exactly the work to be done, and you have experience in all those subject areas.
However, any different way of doing things does add a potential complication, and you often end up forgetting about something that needs to be done.
I find it helps to retain the credibility of your boss (if it's an internal project) to indicate those items where there are "unknown unknowns". Then if she's experienced , she can mentally add in the slack there.
As soon as you run up against another task, you need to add it in - you should quantify, or at least itemise, every single job that stands between you and project completion, even if you can't put a figure on every one.
Those you have experience of will be much more solid than those where you are a novice; you are also much more likely to know the "hidden aspects" of known areas.
You should by now have realised that project estimation is quite a job in itself, even if you don't get the business, you have to put the work in, maybe several days worth, if you're being serious, and write it off (but reuse the experience gained - file it away to save you work the next time) if you don't get the order.
But your rates need to include costs for the work involved in quoting, with a percentage to also cover work you spend time estimating but don't get, and time between projects, if any.
