There's more info on this than any person could hope to digest (yeah, that's not all good) at the infopop.biodiesel site, i.e. on two-tank WVO systems - maybe you've seen it.
I've been mulling/planning/putting off some of the same engineering concerns you raise here. Last few years I think people are avoiding mass-heating of WVO tank due to polymerization problems - WVO that has contact with air, when heated, tends to form a plasticky gunk that can take things down pretty fast, or at least that's my capsulated take on things. Better to heat the WVO as much as possible in a one-pass mode on its way to the IP - methods vary according to your climate. In my case, I hope to get away with (at minimum) a flat-plate heat-exchanger, maybe a coiled-copper WVO-filter heater, and maybe electric injector-line heaters (those are a concern to me, since there seems to be a great potential for heat loss while the WVO is working its way through the injector lines.) Much of this is to be determined empirically with IR laser thermometer and note-pad in hand, etc. as installation progresses.
I wouldn't propose sharing same filters btwn WVO and petro if that's what you meant - much better to have two standalone systems for tweakability as you get to where you need to be in terms of heating, as well as for the capacity to simply switch back to tried-and-true petro or biodiesel tank when stuck at roadside, zombies approaching. PUSH your fuel through your filters, don't try to pull it with whatever pumping scheme you plan.
My own slowly-emerging system uses basics consisting of 2ea three-port solenoid valves. One toggles the inlet to the IP, such that fuel will come from either the stock petro tank/filters, or from the WVO filter (new) and FPHE.
The valve on the return side of the IP is less straightforward, to me anyway: that 3-port toggles all IP return (makes sense when you realize this fuel is gonna be petro, WVO, or a mix, depending on when you look at it) back to either the WVO or the petro tank. Driver should purge to the WVO tank for at least as long as it takes to get all WVO out of the IP before shutdown, in order to err on the side of caution. Switching both of those valves at the same time could/will eventually lead to buildup of WVO in the petro tank. Small amounts wouldn't be a problem, but the stock in-tank electric pumps on my M51 pump far more than the engine actually uses, so there's a LOT of return flow. (You have to also watch that you don't overflow one tank by not keeping track of this transfer...)
You'll want to monitor temp of the WVO where it hits the IP, if not the injectors themselves, so that you'll know when you've reached a high enough WVO temp to switch over.
An augmentation I plan for this that I think is not widely used is to insert a sight-glass/fuel chamber in the WVO return-to-tank line, with its exit port on the high side. I will install this at a high point in the return WVO path, so that any air coming through will tend to rise to that chamber, and then be forced back to the tank where it will bubble out. (Likely place would be up high on the firewall, with tubing routed upwards from the IP directly to it, so there's no air-traps in the way.) The top of this chamber will also be fitted with a needle-valve, which I will use to (greatly) choke off the return flow of WVO. The idea here is to maximize the heat-transfer to the WVO upstream, allowing only a minimal return flow in order to purge air. Sight-glass feature isn't really necessary but I think might be useful in verifying purging progress - seeing the fuel change from WVO to petro, that is, to get a sense of how fast the return flow is without having to pull off a line at the tank or something like that.
Anyway, the WVO portion of that website is pretty active in these considerations. I'd be curious to know what you end up doing/learning in this process...
--Dave
