How do junctions (and other components) affect fuel flow?

Hi, I would like to raise this issue regarding fuel flow through junctions
once again. I am experiencing a slightly different issue where the fuel flow
through multiple junctions is dropping with no opening or closing of junction
output lines or valves. In this system, there are lines from two pumps feeding
a junction with multiple output lines. Only output line has throughflow and it
leads to another junction. There is also only one output line in operation in
this junction, which leads through a valve to a fuel tank. At the first
junction (where the flow from the two separate lines with fuel pumps (each
putting out the same pressure) is combined, the flow rate is, as expected,
twice the flow rate of each input line. However, the flow rate from this
junction through the only output line in use (going to the valve before the
destination fuel tank), is half the flow rate of the first junction, although
it will jump between this value and the full flow rate of the first junction.
And then the flow through the valve to the destination fuel tank jumps around
from values between 1/4 of the input flow for the first junction up to the
full value of the input flow at the first junction. (This is all with the
default fuel line parameter values of fuel flow at 1PSI and volume.) Would it
be possible to include more information on how fuel flow is computed from the
fuel line volume, pressure, and fuel flow at 1PSI parameters, including any
resistance or other effects from any of the fuel system components? Is the
fuel flow in all parts of a given path restricted to the lowest fuel flow of
any line in that path (or to that required by what it is feeding in the case
of an engine or APU)? Relative to that, does the line fuel flow simvar show
what the line is capable of at any given moment, or does it show the actual
line fuel flow at that moment? (I ask this because for the case of the path
feeding the engines, the simvar line fuel flows from the tanks are much higher
than the simvar line fuel flows feeding the engines.) Another thing I noticed
is that for the options setting, all input lines are always open with no need
to specify them in the specific options. I assume this was intended, but it is
not mentioned in the SDK. Something to add perhaps?

Hello @donstim I followed your instructions and I think I know what is
misleading here. I don’t think there’s any problem in the end result, if I
empty both Feed1 and Feed2, the total fuel quantity doesn’t change so there’s
no loss and both tanks are being filled at the same speed.
But the fuel flow simvars seem
to be more and more irregular as you go down the lines. You have constant 7409
g/h after the pump and the value right before the feed tank can oscillate a
lot (bottom at 1850 and I’ve seen it raise to 62000 for an instant). You can
see that even more if you increase the frequency slider on SimVarWatcher. It
looks like there’s discontinuity between fuel transfer from one element to the
next and the element at the end of the line has all the accumulated
discontinuities. The devs in charge of the fuel system explained to me that in
some situations, they have to make guesses on what the FF will be for a given
component and correct the following frame. I think what we see here is the
accumulation of those guesses at the end of a series of components. This will
be investigated further so we can see how this can be improved. Regards,
Sylvain