The advice given is sound, suggest warming the engine for a couple of minutes to minimise the condensate being produced, then stop the engine, empty the displacement lubricator of water and refill with good quality compounded steam oil 220 grade or 440 grade for our small engines and the re start the steam and let the engine run - there should now be minimal water condensate - just enough to force the oil into the steam line rather than a deluge that washes all the oil away. The approach seems to work well and makes the oil last a lot longer.
There is something wrong with the "cold system forming excessive condensate and washing out the oil" theory: It is physically not possible for a cold start to empty a full lubricator.
For the lubricator to work, there needs to be condensation IN the lubricator. Since there is only ONE opening connecting the lubricator to the steam line, and the lubricator being full, any liquid water CAN not enter the lubricator to displace oil, since pressure in the lubricator is identical to the pressure in the steam line.
That whole principle of oil being displaced by condensate, is ONLY possible if steam enters the lubricator and condenses there, contracting and drawing in more steam.
Any water present in the steam line, if that would enter the lubricator, it won't condense because it already IS water and cannot contract any further, so it won't draw in more steam. In that sense, a displacement lubricator is pretty much selfregulating.
With a filled lubricator, the first lubricating is caused by the oil warming up and expanding. By the time the "condensing" part of the working principle comes into effect, the steam lines are allready dry.
Typically, the steam line is allready warm and water-free within a few seconds of running. The engine takes much longer to warm up (in my case roughly 30 seconds) but the engine is downstream, and a cold engine does not affect a hot steamline or oiler.
My observation WRT displacement oilers, is as follows:
At the start of a run, my lubricator is filled to where oil level is at the needle valve. No water present.
At the end of a typical run, about 1/3rd to 1/2 of the oil content is displaced by water, BUT... when everyting is cooled down and I look inside, the oil level is at least 5 mm BELOW the needle valve. In other words: the oil expands and contracts considerably when heating up or cooling down.
Wind or ambient temperature increases oil dosing, during a colder or windier day, about half the oiler empties, while during warmer or windstill days, about 1/3rd is consumed.
I have spent HOURS sitting next to the bathtub observing, and my oiler has a glass body.
There is NO inrush of water in the lubricator at cold start-up. There IS a significant "first shot" of oil upon starting a cold engine, which I can only explain by the oiler being cold, and the oil expanding when it heats up rapidly, but that "first shot" is at best something like 1/10th of a ml.
I have experimented a bit with both "filling to the brim" and "filling to the level I found after running and cooling down" (appr 5 mm below the needle valve).
In first case (filling to the brim) the first shot becomes larger, but after that I had the distinct impression that the displacement did not really come into effect, which is logical if it is considered that there is no open space above the oil where steam can enter and condense.
When filling the lubricator to 5 mm below the needle valve there is no, or barely any first shot, because the oil has to first warm up before its level reaches the needle valve aperture. But it is very hard to tell when exactly lubrication becomes effective.
If an oiler depletes too rapid and there is no needle valve, one could consider insulating the body of the oiler, since the oil feedrate is dictated in part by rate of heatloss at the oiler (if less heat is lost, less steam condenses, and less oil will be displaced).