Roclink 800 pid loop

I am using a PID loop instruction in a Floboss 107 RTU for an oil and gas application. There are two loops, a primary loop and a override loop - primary is for flow control, and override loop for pressure control. The PV's of the primary and override loops are from a USM meter and pressure transmitter respectively. The pressure transmitter is downstream of control valve, and USM is upstream of control valve. The output of the loop is a 4-20ma signal to a GE oil and Gas DNGP valve. I'm using a low select between each loop, so which ever output is lowest should be in control. If flow or pressure is below set point, we want valve to open, valve will close when either PV is above set point.

I'm having an issue with the loop select toggling between the two loops. The PID controller was in pressure control for the past several weeks with sp's around 684 PSIG and 165000 MSCFD. The flow was around 40000 MSCFD, pressure stayed at 684 PSIG.

When operators raised pressure SP to 720 PSIG (flow sp remained the same), the pressure got to maybe around 702 PSIG and flow was around 100000 MSCFD when the loop select started toggling between primary and override. What appeared to be happening is primary flow control loop was trying to close valve and pressure loop was doing just the opposite. Valve position remained at around 59 % open. The PID controller is prematurely selecting the primary loop. What I would like to see is the override pressure loop remain in control until the flow is around 10000 MSCFD or so from SP.

Any suggestions as to how I should tune the PID loop so that the flow loop is less responsive?

Here are other configured variables in the PID.

Primary Variable EU Span 0 - 200000 MSCF/DAY

Secondary Variable EU Span 0 - 1500 PSIG

Output Span 0 - 100 % (0% = close)

Loop Period - 0.5 Seconds

Primary Deadband - 300.0 MSCF/DAY
Override Deadband - 0.00 PSIG

Override Gain 27.0
Override Reset .1
Override Derivative 0.000

Primary Gain 2.3
Primary Reset 1.4
Primary Derivative 0.00

Is this a fail open valve or fail closed?looking at the scale factors in negative. makes me think it is a fail open valve which means you should actually have a high select on it which means which ever one is higher closes the valve more. Thator you can swap the scaling on the actual analog output to be reversed and then the PID loops can be mormal again.
In effect possitive is direct acting. More output equals more pressure and more flow. negative is reverse acting. More output for less pressure and less flow. Also the gain looks high on the override.

The valve is fail close - 4 ma = close, 20 ma = open.