Let's review a little...It started when I said when making the last connection, you must have an opening to the atmosphere. I didn't say you would have a bad connection, but the implication was that you could.
Then, a demonstation was made proving my point...heating the last fitting on a closed system raised the pressure as the heated air expanded pushing through the fitting to the only path or opening it had. Until the whole system came into equilibrium, the air was expanding and moving through the fitting - the fact that the pressure peaked, broke through and then stayed constant does not mean that air stopped moving through, only that the pressure couldn't rise as it expanded - i.e., it was being relieved. Depending on when that occurs, will depend on whether there's enough pressure to blow through the solder to compromise the joint. Air in a closed body will increase pressure at about 0.1#/degree (this will vary based on what you start with - more at sea level, less at say Denver where you'r nearly a mile up). I don't know how much pressure a liquid solder joint can withstand, but as a joint is being heated for soldering, that heat is being conducted along the piping and heating the air, expanding it. It either increases pressure if it is trapped, or it just flows through the joint. Blowing through the joint at the 'wrong' time will compromise the joint.
Seems to me that steam is a gas, and that it, too, expands as it gets heated. The water also expands considerably when it does a phase change from liquid to gas. I totally agree that water, in whatever form, is the biggest reason for a failed joint. But, let us look at two different situations: liquid water at the joint, and liquid water near the joint, taking liquid at the joint first.
When there's liquid water at a joint you are trying to solder, until all of that water is converted to steam, you cannot get the temperature of the joint hot enough to melt the solder...while boiling, it stays at the boiling point of water which will vary based on the pressure, but let's take 212 as the norm. Solder melts a lot higher than that. In the process of boiling off the water, it creates steam Under adibiatic conditions (constant pressure), water expands by a factor of 1600x it's volume as it changes from liquid to vapor (steam). Guess what, that increased volume in a closed vessel will force its way through the weakest point...the joint you are trying to solder. This can contaminate the joint, wash away the flux, and is likely to create a bad joint, if you ever do get it hot enough to melt the solder.
Then, lets take the situation where there's no water at the joint, but some near the joint...in our similarly closed vessel. You get the joint hot enough to melt the solder, but in the process now have not only the heated air, but the supersaturated air/steam mixture and guess what, it blows through the joint since it isn't open to the atmosphere. So, what caused this? Failing to provide a path for the expansion of the gas, and yes, steam is a gas, which goes back to my (maybe too simplified) statement, that to ensure you have the best chance of a good connection, make sure you have an opening to relieve the buildup of pressure. Mixing water vapor into the equation just makes the whole situation worse, but expansion occurs regardless, and under the right circumstances, can cause a joint to be bad. It comes down to the partial pressure gas laws...you can only mix so much of various gasses together before the pressure rises or the volume increases, and heating it will try to expand it.
Leave the piping open to the atmosphere, and you have a better chance of getting a solid joint, water in the picture or not.
