Engine driven fuel pump anomaly

My engine driven fuel pump pressure exhibits an anomaly during climb. Initial climb conditions are 135 kt IAS, WOT, mixture rich, Electric fuel boost pump on. Fuel pressure is 22 to 25 psi. At 1500 feet AGL fuel boost pump off, climb power set, 1000 feet a minute climb indicated. The fuel pressure will drop to 12 to 15 psi setting off the PFD alert. Engine runs smooth no RPM drop etc. boost pump back on fuel pressure returns to normal. Upon leveling off at altitude boost pump comes off, fuel pressure remains normal on engine driven pump alone.

Engine is a Lycoming I/O 390 with about 270 hours installed in an RV 8.

Couple of things...

The engine driven pump gets hot during a climb. You are asking it to work the hardest with the least amount of cooling. I always leave the boost pump on until top of climb, no matter how high.

Second, go back and tighten all fittings between the inlet of the boost pump and the fuel tank. A leaking fitting will show those same symptoms. It is much easier for a suction pump to draw air rather than fuel, so the smallest leak at any of the AN fittings will allow air into the line. Having the boost pump on eliminates the suction requirements, which is why the pressure goes back up.

Vic

I have the same components (390, AFP, RV-8). Testing finds it might drop as low as 18 psi in a hot takeoff at WOT/2700 without the electric pump, deck angle for 100 knots, and even then, not every time. Reducing fuel flow, or reducing deck angle, or turning on the electric pump will put pressure back at the usual 23.

SOP here is boost pump for all departures, and turn it off at the first RPM reduction.

Hmmm...similar, but not identical, situation here. IO-360-M1B, Andair boost pump (on during take-off until 1000' and first power reduction).

I'd say maybe 4 or 5 times over the last several *years*, for no apparent reason, the fuel pressure dropped over a period of a second or two (not instantaneously, so it's not likely instrumentation error) to somewhere like 12-16 psi, low enough to trigger the alarm on the EFIS, then without delay, it just climbs right back up to normal. I can correlate this with no other parameter, really. It happens in level cruise, with nothing changing, and it's SO infrequent it's hard to say how to proceed.

All fuel line fittings nice and torqued appropriately throughout the whole system, and have been since first flight with the exception of the ones that have to get removed for annual inspection (which are then torqued back down and the system leak tested...i.e., I don't think this is a line/fitting problem).

Ideas? It sure gets your attention when it alarms!