Re: Flow splitting

You basically got it.  I've seen that coupling have some rather, uh,...
"interesting" effects on steam headers in power plants.  Depending on how
well you control the motor, you may have one degree of freedom too few.
There are a couple solutions:

1. Consider using a centrifugal pump which is a bit large for your
application, connected to a large header.  The centrifugal pulp will give a
relatively constant pressure.  It will have to be sized to hold that
pressure at full flow on both pipes, possibly plus a tad.  I think this will
work the best.

2. Put one flow control after the pump, then a flow control on one of the
two legs.  Set up a cascade with the downstream loop about 3 to 6 times
faster than the upstream loop.

3. Limit the ramp rate  on each valve, preventing rapid changes.  This will
minimize hunting as they both respond slowly together.  Possibly get bonus
points for making the loops ramp at different rates.

I think this is basically the same issue one gets when they use too small a
header on the household water, and connect the toilet tank and the shower to
the same line.

Michael


"bendel boy" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> If I have a centrifugal pump, and want to split the flow down two
> pipes, what controllability issues might I have if I use two
> flow-controlled valves, one on each leg?
>
> My first degree-of-freedom analysis indicated that I have two degrees
> of freedom - so I can control the total flow and the flow down one
> leg. But I feel uneasy about this, and have a mental picture of the
> two valves endlessly hunting as they interfere with the upstream
> pressures caused by their operation.
>
> Can anyone offer their thoughts?
>
> Thanks.