What about doing a degree or MA through the OU? They are not as cheap as they used to be (sadly), but cheaper than the 9k tuition fees. You can sometimes go to MA level with them without the BA/BSc degree.
I think if you are serious about doing a PhD, you really do need to get some academic qualifications in the field. A Masters in particular teaches essential research skills that you can't really do a PhD without nowadays - searching for academic papers is only the tip of the iceberg, because without a grounding in the field, how will you evaluate them or be able to place them in a disciplinary context? You mention a survey - but gathering data is only the first step: you need to be able to make it fit within a whole academic field (survey results alone - ie. collecting new data - won't get you a PhD; the originality of a doctoral project lies in making a theoretical and methodological case for your work and the significance of your data within the wider discipline). It's pretty hard all at the same time to learn all the research skills needed, plus learn about all of the discipline you're in, plus design and collect data, test out your ideas at conferences and in draft, and then write it up (as you go along, these days - no frantic pulling it together in the last six months any more).
Almost everyone starts a PhD with an ambitious idea they think is very original, and has to change it; a (very) few people do have a genuinely original project, but the vast majority of PhDs are what historians of science call "normal science", eg. they examine a problem that fits squarely within particular disciplinary parameters, and situated itself as very much part of the wider field - there really aren't many brilliantly original interdisciplinary doctoral theses, because by and large the point of a doctorate is to "earn your spurs" in a particular discipline. Plus brilliantly original interdisciplinary projects tend not to be particularly practical and completable within the 3-4 years required for submitting a thesis at most UK universities (those people I know who did have brilliantly original doctoral ideas actually spent a lot of time rewriting and/or refining their work after the PhD - as part of a postdoc research fellowship, for example - the brilliant published book or whatever tends to take more like 6-7 years to appear: the thesis is only the first stage).
Everyone starts out with an idea they end up having to change a lot and refine - because when you start out with your initial research statement, you'll be making a claim that "no-one has hitherto done X" - but in reality, you have no idea at that stage whether no-one has done X because it's a stunningly original idea, or because it doesn't really work, or because it's just not an interesting question, or sometimes because someone else has done it, either a long time ago, or (horrors!) because someone in the US is just about to publish a monograph right on your topic. And you need to have the disciplinary knowledge and flexibility to resituate yourself and work out why some hunches lead to genuine knowledge, and why some just don't. Having the academic background is not just about the research skills, but about an essential core knowledge which informs what you do, so it's worth thinking about putting that foundation in place first.
Have you thought about applying for a Masters that includes a research dissertation, and testing out your idea there? Many doctoral theses start life as MA/MSc projects, and you can in some cases do a Masters without a degree in the relevant field. You would then have a much better idea whether it would work as a PhD thesis, and would also be more likely to be taken on by a department. I would have to say, honestly, that most departments wouldn't take on a doctoral student without the relevant prior qualifications - it would be too much of a risk, and very unfair on the student concerned, as PhD students are generally left to get on with it, often with minimal supervision, and most departments are just not set up to provide the kind of support someone would need coming to the PhD without, say, a research-led Master's degree.