I have built 2x two element beam yagis (dipole with reflector) antennas and added a pawsey stub 1:1 balun onto each. Both have been tuned and the swr is really good. I am just wondering what the correct wavelength spacing would be for maximum gain?

I did manage to find this handy graph online for spacing single dipoles (please see below). Would this be ok? Spacing 0.9 seems good for two dipoles and has a gain of 3.2dbd. Would this spacing apply for two yagis?
What I am hoping to do is build up e.r.p for my relatively low power tx so it can boom

One wavelenght is fine.

You need to get a copy of the ARRL Antenna Handbook. It gives some great designs for 144 MHz and 70 MHz amateur bands which can be easily scaled to Band II. They give the design calculations, so it's easy enough to do.

Thanks guys, I've just been researching this and found an interesting page (although it seems to be related more to dx'ing than tx'ing).

https://www.grantronics.com.au/docs/StkYagis.pdf

Seems to suggest that if you use 1.25 lambda spacing, you will get a cleaner signal (less side lobes) but with slightly reduced gain. I might try this, what you reckon?

Also had a look at the ARRL antenna book, that's got some info and will take a while to read lol

One thing that did come to my attention was in the section about feeding phased arrays. If I'm understanding it right, the author is implying that Wilkinson splitters are not very good for feeding multi antenna systems as each antenna has a slightly different feedpoint impedance so the power being split wont be radiated equally?

I am finding that the two antennas I built are not quite the same. The first one I built has a swr of 1.02 to 1 (according to the vna) and the swr chart showed a very sharp "pointy" type sweep. I was very happy with this and so I built the second one exactly the same. All element and reflector dimensions were copied. When I tested it, I found that it had a completely different sweep, its like I have built the perfect broadband antenna! The sweep doesn't have a sharp point (like the 1st), it has a completely flat response for 10mhz @ 1.2 to 1 swr (at the resonant point).

Weird how you can build two identical antennas and the results are totally different. I would imagine the Wilkinson splitter box I have made will make the two antennas radiate unequally.

I bet you have a short or (more likely) an open in the phasing harness or one of the antennas