Hi Keith,
A while back, Ray asked for a photo of the valve guide oiling Y piece for a '35 motor, but his post was in a long thread about other stuff, and I had trouble finding it, so I started a new thread.
The fitting you have in your crank case is for this purpose. It allows oil mist to make it's way to way (well in theory) to the inlet valve guide of exposed valve heads, via a drilling in the side of the head.
If I have it right, this was only for the inlet valve, except for '35, when Ariel tried feeding the exhaust valve as well.
Ray explained that the temperature of the exhaust valve caused the oil to burn and gum up the guide, causing the exhaust valve to stick. This is sort of verified in the photo above, where the exhaust valve oil tube has been removed and soldered over.
You should be able to pick the heads of various years in the '30's by the castings / drillings for this oiler.
The photo's are of the 1937 version. As you can probably see, it's an unfinished project, so I don't really have any experience with these motors, other than building this one. Anything below is thoughts rather than practice.
The real question is what to do to stop (reduce) the tendency of the exhaust valve to wear out the guide. Because of the heat and maybe the pressure of hot exhaust gases on the guide, it doesn't last long.
Ray's idea of grease nipples and high temperature grease is above.
Other things that come to mind would be to not use a piston with a scraper ring - I don't think they were in use then.
I don't want to be involved in an oil debate, and I know that PFTE / Teflon oils have questions about their efficiency in engines, but for this purpose only, I wonder whether their high melting point, low friction properties would be worth considering? I'm not sure about whether oil or grease would be better, but some kind of drip feed from an oil can to the guide would be relatively easy to set up and would be controllable.
The other thing that will wear out the guides more quickly is the angle of the rocker arm on the valve stem. There is a section in "tuning for speed" (page 21) on this and a post by Vic Horley in the knowledgebase
http://www.barkshire.co.uk/AOMCC/Knowle ... spx?id=749
Maybe someone has solved the problem already?
Brenton