I can't give a complete answer to why you can't get the numbers to line up, but I can give some pointers from my experience with the object. The big thing here is that the WindAndStackOpenArea object, while looking simple, is not actually that easy or intuitive to use. It is essentially a tool to implement the simple ventilation calculations found in ASHRAE fundamentals, or CIBSE Design Guide A (it's designed for ASHRAE, but can be adapted to apply the CIBSE formulae). This means that to understand what it's doing you want to read those documents.
This then raises a number of issues. Jason suggested working with just a single object, and this is always good advice when trying to break down how a model is working but in this case it's also just the way you need to implement it - the effective opening area for both stack effect and cross-ventilation is not simply the sum of the opening areas, so you can't apply an object for each opening. You need to make one object for a zone and work out the effective opening area following ASHRAE/CIBSE. Note that this effective opening area may be different depending on whether or not you are calculating stack or wind-driven ventilation, making the combined object questionable.
Some other points:
- The height difference is not, as one
might naively expect, the difference
between the upper and lower
windows/openings. Rather it is the
difference between the midpoint of
"an" opening and the "neutral
pressure level". The engineering
reference states that "Since the
‘Wind and Stack with Open Area’
object requires the height difference
between the midpoint of the lower
opening and the neutral pressure
level, which is difficult to
estimate, this object should be used
with care (e.g., research only)"
which makes it rather useless.
- The default values used for "opening effectiveness" for
wind-driven ventilation are more than
an order of magnitude higher than the
value used in the CIBSE equation for
single-sided ventilation, and appear
to align roughly with
cross-ventilation scenarios - The default “autocalculate” values for the discharge coefficient for
the stack calculations are, in
fact, specifically for stack
ventilation through a single large
opening rather than separate
openings (ASHRAE, 2009)
My preference is to adjust the implementation to use the CIBSE equations rather than ASHRAE because of both the neutral pressure level problem and the fact that ASHRAE wants you to read the effective opening area adjustment factor off a graph while CIBSE provides an actual equation.
You might want to try testing and computing with a single ZoneVentilation:WindandStackOpenArea object. Having multiple objects just makes it harder to understand individual effects. After that, if you still can't figure it out, maybe set the Height Difference to zero and see if the wind part can match.
Another point that might help is that when you're digging into E+ calculations like this you may want to look at the outputs at the timestep level - the hourly outputs will just be the average of the timesteps and this can lead to some differences in calculations if you try to work off them.