Loss fuel pressure prior to start

[yellow3800]

This is for someone familiar with the inside of the fuel tank.

I recently replaced the fuel pump motor... and must have [may have] bumped something.

now, with good fuel pressure when running, it must be started first. The problem occurs when the prime fuel pump run stops when key fwd, prior to start. the fuel pressure goes from 30 to 0psi in a second. in other words, fuel pressure only exists when the pump is on. I have to crank the car 3x in short bursts for a start.

of course all else is fine, no codes, no probs. no external leaks, just this.

30psi is at the fuel rail when pump is on and it will maintain is during operation. Fuel tank will pressurize as evident when fueling up at the pump... wooosh as the cap is unscrewed.

What did I mess up in the fuel tank? I assume its in the tank anyway... the fuel pres. reg. holds vacuum.

Help!

[3X00-Modified]

It could be in the tank, and if it is it's the line going from the pump to the top of the pickup, OR the pump itself has a bad check valve.

The regulator holds pressure in the feed line and if the line between the pickup and pump is bad/loose or something like that it will bleed off the pressure there, and second if the check valve in the pump is bad it will just bleed back through the pump and drop pressure.

Typically I would go after the regulator in this instance, BUT since you were JUST in the tank changing the motor then I would have to believe you disturbed something there.

[1988GTU]

Did u crack the plastic feed snout fitting when clamping on the new hose? How about the feed tube at the sender, did u maybe bend it and if rusty, manage to break the fatigued tube?

[Rettax3]

I am 100% with Jon on this one. Regulator is easier, but... Yeah. I don't know what engine you have, but unless you have the TBI 2.0, 30 PSI at the test-port sucks (and I don't think the TBI 2.0 even HAS a test-port IIRC...). You should be looking for ~47 PSI at idle. Inside the tank of higher pressure (Multi-Port, Tuned-Port, Sequential Port, etc) injection systems, there is usually a small rectangular box, usually black plastic with a metal cap, this is what the fuel-pump outlet plugs into, it is a pressure accumulator. If that is broken or leaking, that could be your culprit, and you may not even know just by looking at it. Otherwise, my first guess would be a bad pump -it wouldn't be the first time a new pump fresh out-of-the-box was simply junk. What brand did you buy?

[3X00-Modified]

He's got a 3800 in that car... And pressure should be max of 43.5 Where are you getting 47 at idle? That would only be true if his max pressure is 55 and I'm not sure if the 3800's were using that pressure or at least not the generation motor he has.

[yellow3800]

hmmm... we can rule out bad lines and hoses. I finally had to cave in and replace the fatigued lines in back on the tank unit. ...they were trimmed half-way, so the metal lines are still penetrating the cap as they were originally. new rubber and new metal.

yes, this is the lemon, 3800 using a stock pump motor for the 2.8/3.1 beretta, as is for the last 30kmiles. tank inards are all beretta.

the pump motor was one I had on the shelf... from a fuel tank unit pulled from a junkyard. I figured it was okay when I got my fuel pressure back during a post-install test.

I suspected there is a valve to prevent backflow via pump line side, I just couldn't obviously see it on the motor. I'll open the tank here when I get some time and find this pressure accumulator - black plastic box thingy with a metal cap wait, is this also known as a pulsator? its suspended by mere connection of tubes and is just in between two lines that plug into it?

oh, I probably put this on backwards or it failed a thousand miles later. Why is it called a pulsator? It sounds like one of those examples of things called an oxygen sensor when it should be called and Exhaust Gas Temperature probe.

I say 30psi because if memory serves, is the 3800 pressure, which it gets from a beretta pump. If it climbs higher as rpm climbs, I'll have to get back with you. my idle check gave me 30psi and greater.

[Rettax3]

He's got a 3800 in that car... And pressure should be max of 43.5 Where are you getting 47 at idle? That would only be true if his max pressure is 55 and I'm not sure if the 3800's were using that pressure or at least not the generation motor he has.

Chilton Beretta Service Manual #28480, page 5-8. "On 1988-94 2.8L and 3.1L engines... ...The 2.3L engines also use Multi-port Fuel Injection... ...On 1995-96 3.1L engines, a Sequential port Fuel Injection (SFI) system is used... ...A pressure regulator maintains 40.5-47 psi (271-315 kPa) in the fuel line to the injectors and the excess fuel is fed back to the tank."

Haynes Service Manuals give 41-47 psi for the MPFI and SFI systems.

Chilton's Service Manual for W-Body cars 1988-'96, book #28682, states MFI engines should receive 34-47 psi of fuel pressure, this applies also to 3.1 Turbo engines, 3.4 DOHC 'Twin Dual Cam' engines, and 3800s, through Series Is and early Series IIs. I think 34 is a bit too low, and indicative of an imminent system failure, but yes, the engine should run at that pressure.

Chilton's manual # 28282 specifies 40-47 psi primed pressure, with a 3-10 psi drop upon starting.

It is pretty much all over every documentation I've ever read on GM multi/tuned/sequential-port injection systems, and that matches with my experience on these systems as well. I have often seen ~51 psi at high throttle settings with a healthy system.
Where are you getting max pressure of 43.5 from? What told you fuel pressure rises precisely 8 psi from 'idle' to 'max'? None of the operating ranges in ANY book I've read on these gives 8 psi difference, so I don't see it.

Try unplugging the FPR vacuum-line to see these readings, sorry. My point was that 30 psi is quite low -hasn't anybody else noticed that little detail??

yellow3800, no, the fuel pressure isn't relative to RPM (at least not on a healthy system), it is dependent upon manifold vacuum (or pressure on a forced-induction engine). Yes, I think I've heard that accumulator referred to as the 'pulsator' (one of my old F-Body tech books called it that, I think), but I don't recall if it is possible to install it backwards (flow direction may not matter on this part), but check for an arrow stamped on the unit. Its' job is to help take out peaks of pressure-spikes in the fuel system as injectors open and close, relieving fuel pressure for split-second intervals.

Not to jack your thread, but I am very interested to see some pics of your car -there aren't many of us with 3800 Berettas on here...

[yellow3800]

I'll post an update when I go back into it. I think the pressure does increase when the regulator is unplugged. not sure, so I'll try that and then take the tank out, and look for a directional arrow on the pressure accumulator.

Thank you for spending time on this! I now have some direction when taking the fuel tank out.

I'm definitely be waiting for warm weather, so expect at least a couple weeks for response.

andy
lemon
89gt<---just sucked the fuel pres. reg. diaphragm up last week, and is okay now.

[3X00-Modified]

43.5 is the pressure most use during flow tests because it's the most common pressure. I've never seen a 3100 or 3x00 reach 51 psi at high throttle because that's way over spec. The regulator is 43.5 and the 2000+ one is 55. I will verify my numbers against my service manuals but I've never seen a car run at those pressures... Nevermind at idle where they should be reduced due to the vac on the regulator.

Edit
Checked my GM manual and yes it states 41-47 for prime pressure which is one, not limited due to VAC as well as not reduced due to injectors firing during running.

If you do any asking on 60degreev6.com most will ask are you running 43.5PSI, or 55PSI

Also the regulator specs for a 96

http://www.rockauto.com/catalog/moreinf ... cc=1028887

listed 44PSI operating pressure.

[1988GTU]

I thought 47 was 2000+ ?

[3X00-Modified]

2000+ regulator info... Same brand to keep consistent..

http://www.rockauto.com/catalog/moreinf ... cc=1358748

55PSI

Everything I'm reading in the manual keeps saying it should be BETWEEN 41 and 47... and NOT exceed 47, as well as not exceed 41 at idle because that will cause a rich condition.

[Rettax3]

It is pretty much all over every documentation I've ever read on GM multi/tuned/sequential-port injection systems, and that matches with my experience on these systems as well. I have often seen ~51 psi at high throttle settings with a healthy system.
Where are you getting max pressure of 43.5 from?

Try unplugging the FPR vacuum-line to see these readings, sorry. My point was that 30 psi is quite low

I feel that a lot of what I am saying is being over-analyzed, and selectively pieced back together. Several of the cars I have worked on that I mentioned were TPI cars, NOT JUST 60-degree V-6s, so take that into account. While the specs are the same for most of the earlier GM systems, I have seen those higher pressures on some of them too, and they were running great. Please also bear in mind that I have also worked on 3800s, not all of which are N/A engines, as well as some of those newer 60-degree engines you mentioned... Since I wasn't sure about what engine he had in his car (despite his name, I wasn't certain and didn't want to assume), I went with the "47" that is mentioned universally throughout all of the manuals I have read.

My point (aside from the main one I was trying to make, which is that 30 psi is quite low, particularly for a pump that was just installed -which I assumed was new until I saw otherwise) was that there is a wide range of pressures that can be found on 'correctly running' engines, and most of these Port-injection systems work great at high 40+ psi. I did correct my statement to say that "~47 psi" is typical with the vacuum-line disconnected from the regulator, so please also read that again. If you are happy with 43 or 44 psi, that is close enough to the ~47 I initially mentioned not to argue about it, it certainly is within spec.

I have not read ANYTHING that states that the idle pressure cannot exceed 41 PSI, but whatever. In neither case does anyone seem to be saying that 30 psi is correct. Also, if the system is operating properly, it shouldn't matter at idle (very low fuel-consumption) if the injectors are drawing fuel from the lines, because the regulator is ultimately what sets the pressure in the system, providing that the pump can maintain the supply. Of course, that works in theory...

[yellow3800]

My haste had induced me to forget to check fuel pressure when returning from a drive...and having 1hour to work on it before it gets dark.

I just pulled the tank out, expecting to find the pressure accumulator [inside the tank, in line with the output line 1/2" above the fuel pump motor] with directional arrows.

NO ARROWS and not needed. This piece is a passthrough, with what looks to be 2cu in of pressure space to absorb pressure spikes. This makes sense.

I then, in accordance with epa regulations, blew into the fuel pump motor to simulate back pressure, or to see if pressure in the line can go back to the pump. IT DOES. It leaks the air I blow into it at my measly lung pressure of probably 1psi. This must be wrong.

I'm sorry again for not giving pressure readings, and I understand this is now more confusing by me. After blowing on the output tube of the fuel pump motor itself, in/out, in/out, and after 5 times, it finally layed down its internal one way valve. "FLAP!" and then it MOSTLY restricted my blowing... as if to simulate a resistance to pressure leak down. seems to me that anything intermediate like this is a problem.

Do I replace the fuel pump motor?

I think I will, as this seems to be the problem, and will look for a 3.8 pump, cross ref part #s and all that jazz while awaiting an answer here.

rettax3, you have a sc 3.8 on a manual trans? are you coming to bfest?
Yes, I've been living under a rock!

andy

[Rettax3]

seems to me that anything intermediate like this is a problem.

Do I replace the fuel pump motor?

I think I will, as this seems to be the problem, and will look for a 3.8 pump, cross ref part #s and all that jazz while awaiting an answer here.

rettax3, you have a sc 3.8 on a manual trans? are you coming to bfest?
Yes, I've been living under a rock!

andy

Good test idea, I don't recall anyone going to that level of detail. I think you have found at least part of your problem, yes, I would put in a new F/P. I really don't think you are likely to have a problem running with a stock 2.8/3.1 Beretta fuel pump, but since you are replacing it anyway, it isn't a bad idea to go with the 3800 unit, if it is actually different for your year 3800. I ran the stock '89 GTU 2.8 pump for a year or two on my 3800SC, and it wasn't a problem until it finally burned-out, but I think it was the original pump from '89 with something like 160k miles on it, so I really can't say that the added stress of running the bigger engine had anything to do with its' demise. I did replace it with the 'correct' '97 Bonneville SSEi 3800SC pump, but that was mostly because I already had the entire fuel sending-unit from the car that donated the engine's top-end, super-charger, wiring harness and PCM. I also wired-in the two-speed fuel-pump's control module at that time, making the only permanent wiring alteration to the GTU in relation to the 3800 swap by cutting one ground-wire from the tank's sending-unit.

Yes, my 3800 GTU has a manual trans, it was initially converted from auto to a Muncie/Getrag 282 sourced from another GTU, a red '88 which also donated its' Z51 sway-bars. I have since upgraded it to the F40 six-speed manual trans from a G-6 GTP, making it one of only two 'documented' six-speed Berettas on this site (that I've noticed), and I managed to beat Shawn to it by only about one year. Bfest would be awesome. The Colorado one would have been doable for me if it had happened two years earlier. I don't see making it out to one any time in the near future, unless it comes farther West, which seems doubtful considering that most of the site's members seem to be from back East.

[yellow3800]

part #s are the same. 350kpa [why they don't give this in psi is beyond me] is the fuel pump spec, for a '89 gt 2.8 and for mine. I actually took that pump, because I started out with the ~10psi 2.0 I4 engine. no one cares about that. sorry for wasting bandwidth! I'll post next week with the update though after reading about back flow prevention on the fuel pump motor when buying it, I think thats the culprit.