Help with fuel pressure weirdness and theory

I read that tread and I think there was no follow up. I used compressed air to check it and it sealed. After looking at it I agree with Enki that it would be super unlikely to fail.

So I got it back together and no love. It took much cranking to start and initially idled fine with occasional pops. As it runs and warms up there's a noticable stumble or slight misfire. The pump managed to hit 1936 psi. About 15 minutes in the fuel pressure started to act up. I noticed after revving it up that pressure dropped to 60 psi for a split second. Over the next few minutes it slowly and steadily dropped until it stalled. The reading was 39 psi. KOEO, the pressure went down over a few minutes to 29 psi.
So what? Spill valve solenoid failure? Another PRV replacement?
It's a condition related failure that can be consistently replicated. I'm thinking it's caused temperature or pressure. I really wish I had CEL's to point at something. The complete loss of rail pressure seems to be a spill valve failure. I was surprised how long the engine will run with the relays unplugged. It makes me think that the spill valve is cutting out and the engine runs until the rail is drained. Wouldn't that cause a CEL though?

 

Update time. I'm giving up and I'm going to take this pos to a shop. I've thrown all the parts at it and no joy. I replaced the HPFP, fuel pressure sensor, had the injectors tested and cleaned, replaced the pcv valve, and only kinda cleaned the valves as they were in decent shape. It's still stalling after 20 minutes at idle.
I'm leaning towards it being an electrical problem. I hate to give up but I can't put anymore time in without results and my electrical troubleshooting ability is sad.
I'll post another update when(if) it's fixed.

 

Did you ever post any log files of the issue happening?

Edit: A shop is gonna rebuy all the parts you already bought, charge you double and send you home only for the issue to keep happening.

 

I'll do a data log today. Thanks

 

So it was running, then when it hit 39psi it died? Meaning it was still running while declining like 60->50->40.

The ITFP dead heads into the engine bay, no losses can occur up there unless it's leaking fuel externally.


If the answer to the first question I asked is "Yes, it ran and then died once it dropped to 39psi", then you have an ITFP problem. The pump, the regulator, or it's not seated and leaking back into the tank.

 

This what I was hoping to see represented in the log. If the pressure dips below ~40 ish PSI at any point when running, then it's not sending fuel.

 

It doesn't happen like that. The reading will be in the 600's at idle, then it will dip into the 500's, then stay in the 500's, dip to 400's then stay there and continue that pattern until it drops to around 200 psi and then stall. That decline takes about 3 minutes. After that, KOEO reads 36-44 psi. If I leave KOEO pressure rises very slowly taking about 10 minutes to reach a peak psi around 1300, then slowly drops.

I removed the pump housing and checked everything. It's seated fine. Sock is new and clean. The regulator was crusty so I cleaned it but I didn't really know how to test it.

After putting everything back together after injector cleaning and new HPFP the first pressure reading when cranking was 6 psi. After 10 minutes of priming it rose to 9, 11, 15, 19, then fired up.

 

10 minutes of priming is too much. Should be more like 30 seconds.
Either the reg is bad or the pump is bad.

 

It does, but only after about 40 minutes of driving. The pattern is similar to what happens at idle. The pressure looks good, but it will start to dip lower and lower over several minutes. I'll start seeing 400s, 300s, 200s, down to 40 for the low pressure reading. It runs worse relative to the decline. When it stalls it won't run again for an hour or more.

 

Pump. Nothing else makes sense to me.

 

When I got the car it had a dw65c that I replaced thinking it was failing, but the problem is still there, why I'm thinking it's a pump related electrical fault. I don't know much about the regulator but it's an easy enough job to replace. I'm just sick of dead ends.

 

I don't think I've ever heard of any other part failing, but maybe the resistor thing at the front of the car is torched or something. This has fallen out of my wheelhouse.

 

Same here. I bypassed the resistor to eliminate it as a fault. This car sat for several years undriven. I haven't found any rat nests or corrosion but I gotta be missing something. Thanks for the help!

 

Do these cars have a check valve to prevent the fuel from draining out of the line and back into the tank?

 

Data Logs
I didn't check what was being logged (idiot) so the only relevant thing is Fuel Pressure
Since I had to mess with the AP Manager, I looked at the map notes for the Cobb OTS Tunes and the boost target for the map I'm using is 16 psi. AP showed 18.94 psi. I keep getting a P0103 of course.
Definitely getting odd drops in pressure at times. The car stalled as I turned in to park it. I also did the PRV test after it died and pressure made it to mid 1700's before dropping.

 

Typically doesn't matter. My fuel system is plumbed without a check valve, so it does all drain back. It still hits 65psi with the priming pulse.

 

It should only take a couple key cycles to prime the system. You have something fucked up with the ITFP.

I'd just get a new basket assembly from Mazda and drop it in. No fucking around with shitty aftermarket parts.

 

That's what I'm thinking. And I learned the hard way that oem pumps are the only option.

 

I ask for the opposite reason. I had a buddy with a boat that would do the same thing as op run fine then after a while struggle to maintain anything above throttle (it was a carbonated Chevy 305 so a little easier to keep running with shitty fuel pressure/flow) the problem ended up being that there was a little bit of grit in the line and it would slowly get sucked up to and jam against the check valve restricting flow causing said symptoms. Turn the motor off and wait now since the check valve was jammed open all the fuel would flow backwards and wash the check valve out so that it would run fine twenty minutes later.

$5 check valve later and problem solved

 

Problem solved!
I wish I could say which thing fixed it, but results are good. I replaced the fuel pressure regulator, but while I was taking another look at the in tank pump housing I found that there is another fuel filter inside the pump housing. It seems to be a media filter that's in the cylindrical portion that pump sits in. So it filters the fuel from the pump outlet to the pressure regulator. With the regulator and pump removed I poured Naptha in through the regulator port and shook it. I then blew compressed air into the regulator port to blast the Naptha out of the pump outlet port. A surprising amount of dirt and debris came out so I repeated that process until nothing else came out. I put together and reinstalled it. 3 key turns had my pressure at 68 psi.
I started it up and it ran for 5 minutes then stalled. Pressure was at 19 psi. I pulled the fuel line at the HPFP and there was no fuel in it because it was being sprayed at 70 psi into the backseat cushion. Maybe I didn't seat the line completely or my janky zip tie "no fuel clip no problem" setup didn't hold, either way. I've never been happier to have accidentally blasted gas in my car.
So maybe it was the pressure regulator or too much dirt in the housing filter. Maybe Easter Bunny was right in a roundabout way and dirt was affecting the regulator.
I want to thank everyone who responded and hopefully the bullshit I went through can help someone out down the road.