odd dips in fuel pressure, pump problem?

Well, for anyone waiting to hear the resolution of my odd fuel pressure issue, I'm afraid you'll have to keep waiting :-) 

For those who haven't followed along, after 77 hours of perfect fuel pressure behavior, I started seeing it drop to around 3 psi, then go back to normal, then drop, and repeat.  This is usually after 30 min or so of cruise.  I saw it today after only 15 min or so though.  Turning on the electric pump 100% stops this behavior.  The fuel line doesn't run near anything hot, and it has plenty of cool air, so I've resisted the usual vapor lock theories. 

Since I last typed, I replaced the mechanical fuel pump with no change.

Fuel filters have been replaced more times than I can count.  I've always had a gascolator with a 115 micron screen, followed by the electric fuel pump.  For the 77 hrs I operated with no issues, I had a small 10 micron stainless screen filter after the mechanical pump.   That was later removed, and replaced with a 40 micron stainless screen filter before the mechanical pump.  I know Rotax says 70-100, but that goes against conventional recommendations for 40 microns in carb applications.  I do have a nice 85 micron filter I could install in place of the 40 micron, and it should at least catch small pebbles :-) 

Finally, I replaced the electric fuel pump with the more often recommended Facet 40105.  I previously had the 60300 which has a check ball, and I wondered if that could be sticking.  No difference at all when I changed it. 

I do NOT have a one way valve around the electric fuel pump.  I bought one, but can't justify the 10 or so extra connections I'd need to install something I know I shouldn't need.  If testing this becomes necessary, I'll just bypass the electric pump for a local flight near the airport to see if that makes any difference.  99% sure it won't.

I just ordered one of these to appease the vapor lock guys.  I could pretty easily install it temporarily, and re-purpose my coolant temp sensor input to measure fuel temp before the mechanical pump.  https://www.amazon.com/AC-PERFORMANCE-Aluminum-Anodized-Adaptors/dp/B0BLNJBDS8?th=1

I've resisted testing 100LL, because I just don't believe this is vapor lock.  I can be stubborn that way.

Cheers,

Rusty

Have you replaced the fuel pressure sender? Have you put a seperate mechanical gauge in line to make sure the true fuel pressure is being reported on your instrument  panel?

I haven't replaced the sender recently. It's a mechanical VDO sender which is known to be crap, but it's been well behaved since I moved it from the engine to the firewall.  I certainly consider that it could be causing false readings, but normal fuel pressure is pretty close to the same if the electric pump is on or off.  The dips in pressure 100% will not happen when the electric pump is on.  As another test, I shut off the fuel valve and watched the pressure drop.  The engine started to stumble around 2 psi as expected, so the reading certainly seems accurate enough. 

Rusty

Another random thought;

Could you have a bit of "something" (scientific term for unknown loose stuff) in your tank/fuel supply, that periodical partially blocks the system.

Personally I don't like the traditional sieve/strainer type tank exit fittings. To me they have been superseded by the availability of in line, fully sealed, transparent, light weight,  gauze fuel filters like Hengst H102 WK, H103WK, Baldwin BF 7725 / 7850 / 7863  /9906.

The in tank strainers are impossible to monitor for blockage and when they do, they are a pain the the proverbial to service/clean. Far better to have one of the above filters on the delivery line from your tank(s) that can be visually checked & if needs be replaced in seconds.

On the Onex, I can see the pickup with an inspection mirror.  It's a course screen, and raised a bit above the floor of the sump area, so I know that there's nothing that could be blocking that.  Also, if it were blocked, and I turned on the electric pump, I wouldn't expect that to un-block it. 

One bad thing about covering hoses with fire sleeve is that you can't see the actual hose. I've wondered if the hose is starting to kink, or maybe I attached a tie wrap a bit too tight and I'm constricting it.  I can easily see that being something that got changed when I did some other work around the engine.  If there was a restriction, the mechanical pump would cause more of a vacuum on the suction side, and the lower pressure would lower the boiling point.  I'm thinking of just pulling the hose from the electric pump to the filter at the mechanical pump to see if it's degrading inside, or softening/kinking.  Cheap and fairly easy to replace. 

One final thought is whether this could be some boiling in the return line.  From the mechanical pump, the hose goes to a manifold with 4 ports.  Two ports go to the carbs, and the 4th is for the pressure sensor and return.  The hose goes from that manifold at the front of the engine to the remote sensor manifold on the firewall.  The sensor is connected there, but there's also a port with a restrictor fitting that returns to the gascolator.  While the main fuel lines have lots of flow in normal cruise, the return line would have very little flow due to the restrictor.  If fuel had enough time in that hose to boil, it might make the sensor reading vary depending on whether liquid fuel or vapor was moving through the restrictor.  As with most theories, I could point out that this hasn't been changed, so why is it a problem now vs before. 

Rusty