Fuel pump on/off

I searched for an hour trying to find a similar problem/fix. Its a 2000 Malibu. Just replaced my fuel pump 2 days ago thinking all problems pointed to fuel pump. Was losing power especially up hill and would turn over and over before starting like it lost its prime or pressure. Yesterday I tried turning it over and the pump wasn't coming on everytime I turned the key to power the pump. I did this more than a dozen times with the Bu starting 50% of the time. When I heard the pump come on it would start first turn. Had fuel pressure checked at rail(43psi) new filter and pump replaced with a used one from an 2001. Where do I go next?

Sounds to me like your contacts are worn out in the fuel pump relay. Replace your fuel pump relay.

Not sure if everything else that is supposed to cut on with your key in the on position is coming on either though. Usually an intermittent problem like that leads to a switch, anything basically with contacts, so a relay as well. Could also be your ignition switch. Just check first that you are getting power to the coil inside the relay with the key in the on position. You should have four connectors in to your relay. 2 for the control side of the circuit, there is a coil between these to connectors, one of them should have 12 volts to it with the key on. The other two are for your power circuit witch actually sends 12 volts to your fuel pump by closing two contacts. What happens is this.

You already have 12 volts on one side of the power circuit, the wire going to the relay, and there is a open set of contacts b/t that wire and the wire going to your fuel pump.

When you supply 12 volts to your coil side by turning on the ignition, the coil energizes and by magnetism pulls the set of contacts close, completing the power circuit, supplying 12 volts to your fuel pump.

If those contacts are bad, some times they may get a good connection and sometimes they may not, and sometimes they never get a good connection.

Hope that helps.

I have not got a wiring diagram specific for your model. A typical GM system works like this :-

Turn ignition on. Pump relay gets energized via ECM and contacts on relay feed power to pump. Pump runs approx. 3 seconds, primes fuel rail and relay switches off. Crank engine and engine starts. Engine runs and contacts on oil pressure switch close which now feeds power to pump. This is a safety interlock to prevent the engine running with no oil pressure. In the GM drawing I have, the pump relay can be removed with the engine running and the engine will continue to run. The relay is only used to prime the fuel rail.

Maybe your oil pressure sensor has defective contacts?

I have not got a wiring diagram specific for your model. A typical GM system works like this :-

Turn ignition on. Pump relay gets energized via ECM and contacts on relay feed power to pump. Pump runs approx. 3 seconds, primes fuel rail and relay switches off. Crank engine and engine starts. Engine runs and contacts on oil pressure switch close which now feeds power to pump. This is a safety interlock to prevent the engine running with no oil pressure. In the GM drawing I have, the pump relay can be removed with the engine running and the the engine will continue to run. The relay is only used to prime the fuel rail.

Maybe your oil pressure sensor has defective contacts?


Thanks for adding the extra help there, my Nissan is like the way I told him. I did look at my haynes manual for my buick and it is exactly like you said with the oil pressure switch. But I'm still going with the fuel pump relay seeing how he said it's when he turns his key on half the time it works and half the time it don't. Either way it should turn on when he cuts the car on. So I'm still thinking something in his ignition switch or his fuel pump relay.

Turn ignition on. Pump relay gets energized via ECM and contacts on relay feed power to pump. Pump runs approx. 3 seconds, primes fuel rail and relay switches off. Crank engine and engine starts. Engine runs and contacts on oil pressure switch close which now feeds power to pump. This is a safety interlock to prevent the engine running with no oil pressure. In the GM drawing I have, the pump relay can be removed with the engine running and the the engine will continue to run. The relay is only used to prime the fuel rail.

Maybe your oil pressure sensor has defective contacts?


I just looked at your wiring diagram- there is NO oil pressure switch in the fuel pump circuit.... I used to think like Jetex does- and it was true in my 77 Vega- if oil pressure went low, car stalled.. but ever since about 1988 or so here is what happens.

The Oil Pressure Switch in the EFI System

There seems to be some confusion in the auto repair industry today regarding the role of the oil pressure switch in today's fuel injection systems. Many people think that the oil pressure switch turns off the fuel pump in case the engine loses oil pressure. This in turn shuts down the engine to protect it from internal damage. This is not the case in most vehicles.

Many vehicles today have electronic fuel injection and many tie in the oil pressure switch to the fuel pump circuit. However, the switch does not act as a safety at all. Instead, in most cars, it serves to continue operating the fuel pump in case the fuel pump relay fails. This would enable the vehicle to reach it's destination. The symptom of a failed pump relay, in this case, would be hard starting or prolonged cranking. The reason the car would start at all with a failed fuel pump relay is that after cranking for several seconds the oil pressure may be high enough to close the oil pressure switch and connect voltage to the fuel pump. The pressure needed to close the switch contacts is around 4 PSI.

Also, many manufacturers do not use the oil pressure switch at all in conjunction with the fuel pump circuit.

Hope this helps.. a wiring diagram is always the best thing to reference...

I know I'm getting off the subject now about his question, but this got me looking at both haynes manuals now. It doesn't show anything in the one for the nissan I have for oil pressure switch in the electrical diagrams, or even in the index does it mention the words oil pressure like it does for my buick. Is there a way to add a pressure switch to the engine somehow if it doesn't come with one? Should there already be a plug somewhere that I can take out and add in a pressure switch or would you have to drill through and thread somewhere? It's a 95 sentra.

I've Tee-ed in an extra switch on a few of mine- they use standard plumbing and pi[e threads- so it is easy to do- just nack out currentsensor- put in a small pipe with a tee- and the use a known oil pressure switch- like off of a GM Safari or Astro van- the contacts close at 4 psi= then wire your fuel pump through that switch.-- I saw 2 Astros burn up their engine due to low oil pressure- 16 years ago- is when I discovered the way the circuit really works= then I wired the circuit "my way" as to turn off engine if no oil pressure-

Seems the car manufacturers don't want you stalled in the middle of the freeway- they want the engine to always run to get you to safety- even if there is no oil pressure....they probably got sued once too many times....

I have a 2000 Cavalier circuit diagram on hand and the oil pressure switch is used to supply the fuel pump after the fuel relay drops out. So it seems as if GM was using that system at least up to then. The Power Control Module energizes the relay and after 3 seconds drops it out. I strongly suspect my 2004 4cyl. Malibu is using the same system. If I switch on the ignition and don't crank the pump stops running after 3 secs. I don't think they are using fuel rail pressure to switch it off. It's easy enough to check. When I have time I'll start the car and pull the pump relay and see if the engine stops.

Coming back to the suspect pump relay. Swap the relay out with another on the relay panel that has the same part #. Be sure to double check the number. There are normally quite a few relays with the same part #. Just mark your suspect relay with a marker. It's easy to get confused.

Hate to be a PITA but...There seems to be some confusion here- I have online access to the OEM GM factory service manuals for 1998 to present... I just looked up your 2000 J-car-(both engines are the same) the fuel pump relay drives the injectors and fuel pump- the fuel pump has no connection elsewhere...What reference are you using?? Maybe if you look at your oil pressure sensor we can completely figure it out...

Also I looked up the oil pressure sending unit- it has only one terminal- to run the gauge or switch

Here is how the OEM service manual describes the fuel pump circuit for a 2000Cavalier

An electric fuel pump (2) located in the modular sender assembly is used to provide fuel pressure. To control fuel pump operation a fuel pump relay is used. When the ignition lock cylinder is turned to RUN position, the fuel pump relay activates the electric fuel pump for approximately two seconds to prime the injectors. If the PCM does not receive reference pulses after this time, the PCM opens the fuel pump relay power circuit. The relay will reactivate the fuel pump when the PCM receives reference pulses

Here is the wiring diagram for your 2000 Cavalier fuel pump

http://idle-hourranch.com/2000cavalier.JPG

Here is what I have for a 2004 Malibu fuel pump circuit....sorry so big- but it is easier to read that way....

http://idle-hourranch.com/2004malibu.JPG

Here is the description for an 2004 Malibu

The fuel system is a returnless on-demand design. The fuel pressure regulator is a part of the fuel sender assembly, eliminating the need for a return pipe from the engine. A returnless fuel system reduces the internal temperature of the fuel tank by not returning hot fuel from the engine to the fuel tank. Reducing the internal temperature of the fuel tank results in lower evaporative emissions.
The fuel tank stores the fuel supply. An electric turbine style fuel pump attaches to the fuel sender assembly inside the fuel tank. The fuel pump supplies high pressure fuel through the fuel filter contained in the fuel sender assembly and the fuel feed pipe to the Fuel Injection System. The fuel pump provides fuel at a higher rate of flow than is needed by the fuel injection system. The fuel pump also supplies fuel to a venturi pump located on the bottom of the fuel sender assembly. The function of the venturi pump is to fill the fuel sender assembly reservoir. The fuel pressure regulator, a part of the fuel sender assembly, maintains the correct fuel pressure to the fuel injection system. The fuel pump and sender assembly contains a reverse flow check valve. The check valve and the fuel pressure regulator maintain fuel pressure in the fuel feed pipe and the fuel rail in order to prevent long cranking times

The fuel pump is mounted in the fuel sender assembly reservoir. The fuel pump is an electric high-pressure pump. Fuel is pumped to the fuel injection system at a specified flow and pressure. The fuel pump delivers a constant flow of fuel to the engine even during low fuel conditions and aggressive vehicle maneuvers. The control module controls the electric fuel pump operation through a fuel pump relay. The fuel pump flex pipe acts to dampen the fuel pulses and noise generated by the fuel pump.


Also the engine oil pressure switch is a "one" terminal switch- used only for oil pressure gauge or light...

Hello brcidd

I've sent an attachment in PDF form in reply to your e-mail notification. It's a scan from my Chiltons manual that covers the 1995 to 2000 Cavalier. I do not have access to factory manuals so must depend on what I can lay my hands on.

As you can see, according to Chiltons, the pump is wired through the oil pressure switch. If I am wrong, then I do apologize. However, I feel this is a worthwhile discussion and opens up more avenues in trying to locate the problem of the original poster.

Thanks for all your comments. I did switch out the relay but to no avail. I asked a co-worker who claimed he had similar issues with a Dodge and he said he end up changing the ignition assembly. The key was a loose fit causing the said problem of fuel pump working less than 100%.

I would make sure you have a good ground connection a loose ground will cause these problems.

For what its worth- My Mitchell-on-Demand also agrees with the OEM GM Service manuals in both the 2000 Cavalier and the 2004 Malibu wiring diagrams- the oil pressure sending swith is a single terminal in both.

So why don't you just look at your oil pressure switch- and count the wires coming out of it- this will easily prove which wiring schematics are correct. I don't have a J-car handy...or I'd look myself.

I will be checking the oil pressure switches. I don't want to labor the point but it is still possible that even with a single connection on the oil pressure switch, by using logic circuits, an interlock can still be achieved. The discription of the operation of the circuit you posted says the pump relay drops out after 2 secs. after ignition on. What conditions must be met before the relay is energized again? Altough the oil pressure switch has only 1 contact which is no doubt used to turn off the low oil pressure light, the same contact could also be used as an input to a logic circuit. This might then allow the circuit to energize the pump relay again when the oil pressure is correct.

Note that pin 8 of the PCM is named "fuel pump relay control" and although it depicts a simple switch it might in fact be a rather complicated circuit which depends on its operation for other logic inputs. Modern car electronics tend to integrate as much as possible into low cost components. By using a cheap integrated circuit the manufacturer has saved on a set of contacts and terminals on its oil pressure switch.

Please note that this is only wild speculation on my part and I don't want to be drawn into a lengthy argument about this. If I am talking crap you may beat me to death.

Technically you are right, however I have no clue how GM would do this or any other car company. But you can have supply 12 volts to as many components you want to with that one set of contacts, whether you have two wires attached to the one contact that split off going somewhere else, or one wire that goes to a component that also has another wire there that leaves and goes somewhere else.

As long as you don't exceed the current rating of the contacts and also the wire going to the contacts and leaving the contacts of coarse.

The only real reason there would be to have two sets of normally open contacts on a really to control two different things would be if you have two different objects come on at the same time that operate at different voltages. For example, say in industrial wiring or any kind you can think of, you need an object to come on that runs off 24 volts dc and also another one come on that runs off of 208 Volts ac. and your control wiring for example lets say will be 120 volts ac.

You would wire from your control switch, the hot wire to one side of your coil, the other side to your neutral. That would be your control circuit. When you flip your switch the coil energizes and the contacts close.

You would have two power circuits in this case, the 24 volts dc circuit wired in with one of the sets of contacts. And you 208 volts ac circuit wired in with the other set of contacts.

As long as the contacts are rated for the volts and how many amps you are pulling from each set of contacts you are good.


That was just an example to explain that you don't need two sets of contacts to control to electrical components.

Most industrial devises as well use an extra set of contacts as well to keep the coil energized after pushing a start button, not a flip switch, in series with a stop button, as a safety feature.

For example, you have a supply voltage going to your stop button which is normally closed, then from there to your start button which is normally open, but one set of your relay contacts is in parallel with your start button. then from there you go to your relay coil. When you press the start button, the coil energizes, closing whatever normally closed contacts you have, the one in parallel with your start button keeps the coil energized so it stays engaged after you let off the start button. When the stop button is depressed everything shuts off. The other set of contacts could then be hooked to your power circuit. That set up is a lot safer than a normal flip switch in line with a coil. If power goes out, for example, someone doesn't think about it and starts messing with a piece of machinery, and the power comes back on they could get killed. But using the relay contacts to hold the coil energized keeps the machine off until the start button is pushed again.

I just wanted to add that, has nothing really to do with cars I just added that b/c I said industrial wiring up above for an example and wanted to make sure I didn't get corrected on that even though I was just using it for an example.

I don't think there is any problem of current overload with logic circuits. The input impedence to logic circuits is high hence current draw is extremely low.

The question still remains. What conditions must be met for the pump relay to switch on once again after it has dropped out after 2 seconds? That is after the ignition has been switched on without cranking. Ponder on that.

I will be checking the oil pressure switches. I don't want to labor the point but it is still possible that even with a single connection on the oil pressure switch, by using logic circuits, an interlock can still be achieved. The discription of the operation of the circuit you posted says the pump relay drops out after 2 secs. after ignition on. What conditions must be met before the relay is energized again? Altough the oil pressure switch has only 1 contact which is no doubt used to turn off the low oil pressure light, the same contact could also be used as an input to a logic circuit. This might then allow the circuit to energize the pump relay again when the oil pressure is correct.

Note that pin 8 of the PCM is named "fuel pump relay control" and although it depicts a simple switch it might in fact be a rather complicated circuit which depends on its operation for other logic inputs. Modern car electronics tend to integrate as much as possible into low cost components. By using a cheap integrated circuit the manufacturer has saved on a set of contacts and terminals on its oil pressure switch.

Please note that this is only wild speculation on my part and I don't want to be drawn into a lengthy argument about this. If I am talking crap you may beat me to death.


In the description I posted- the conditions are that the ECM must be seeing crank sensor pulses in order to re-energize the fuel pump (and injectors) relay.

The oil pressure schematic shows the sensor going straight from the oil switch (where it seeks a ground with less than 4 psi pressure) to the instrument cluster. Pull your oil pressure switch wire off and ground it-- your car will run and the oil pressure warning icon on instrument cluster will come on. The car should not stall. If in doubt, I always test- so that I can fully understand the logic- why guess what is going on?... things change, engineering philosophies change. I always thought oil pressure was part of the equation, until my sister in law burnt up her Astro van with no oil-15 years ago- in that application the oil pressure switch is in PARALLEL with the fuel pump relay- to act as "a back-up" as per the fuel pump description on that vehicle....

Ok brcidd,

That's fine. I'll be changing the oil soon on the Cavalier and will check out the oil pressure switch. I'm too lazy to go out into the cold now. It's actually a 1998 Cavalier that my wife uses. I just happen to have Chiltons that covers 1995 - 2000 models. The chances are that the fuel pump wiring is as per Chilton diagram. I might be wrong. Anyway it was an interesting discussion and makes us think outside of the box.

I learned 20 years ago not to buy Chilton's - That's why I stick with Mitchell-On-demand and use OEM manuals as much as possible. I am currently selling all my paper back OEM manuals in lieu of the electronic on-line which are much more handier. I may even still have the J-car manual for '98- I'd have to look...

I don't think there is any problem of current overload with logic circuits. The input impedence to logic circuits is high hence current draw is extremely low.

The question still remains. What conditions must be met for the pump relay to switch on once again after it has dropped out after 2 seconds? That is after the ignition has been switched on without cranking. Ponder on that.


My bad, lol. I thought he asked about connections to a relay, and overlooked where he was talking about splitting off going to a logic circuit. I seen it again he was talking about his oil pressure switch. Either way I was right about the relays. It's still the same for a switch too yes you can control more than one thing with it. And yes you are right it going to a logic circuit it's not going to draw many amps. I was just plainly letting him know that you can have as many components hooked to the output of one switch or relay contacts as your wires and contacts will allow for simply because he asked, not particularly talking about his circuitry in general.


Too answer your last question though, where you said ponder on that, that's real easy if a GM is like some other cars I worked on.

When talking abut the control circuit, (when you turn the key on it sends 12 volts to your coil in your fuel pump relay). The other side of that coil is not hooked to just a regular ground on the vehicle. It is hooked to the ecm, and gets ground pulses as needed, to energize the coil, which then closes the contacts completing the power circuit to your fuel pump.

Ok, well, my buick is different, when you turn the key you do not have a wire from your ignition that feeds 12 volts to your coil inside the relay. One side of the coil in the relay is grounded permanently. And the other side gets it's positive from the ECM, well, PCM for buick, same thing. But either way that's what energizes the coil inside the fuel pump relay on either vehicle. Which is how your fuel pump can then cut back on after it has dropped out.

Of coarse you have to crank the vehicle though as well.

But his problem was it not even cutting on for that 1st two seconds. Some times it does, and some times it doesn't. So the question is what won't allow that. So I don't even know why everyone here keeps talking about the oil pressure switch. Sounds like something loose in his ignition switch, causing the PCM to not send the votlage to the coil inside the fuel pump relay. Or a problem with PCM itself. We already discovered his fuel pump relay was good right?

The oil pressure switch was brought into the discussion by me because in many cases it forms part of the fuel system wiring and it's good to know how things work or don't work. We know that the pump relay is not directly activated by the ignition key. The original poster has swapped out the pump relay with the same results. That leaves 3 suspects. The ignition key, a bad wiring connection or, I hope not, the PCM.

The above fault conditions might cause the dash warning lights to flicker. It might be something to look out for.

On reflection about why the oil pressure switch was dropped out of the equation in later models, it might be that in the event of a collision, you would want the fuel pump to switch off immediately to prevent a fire. It often takes several seconds for an engine to lose oil pressure once it has stopped.

There seems to be some confusion here- I have online access to the OEM GM factory service manuals for 1998 to present...

Hey brcidd how did you get the service manual CDs? I think they are like $500 if you buy them from GM and I think maybe you have to register as a tech with AC/Delco. I've seen them on ebay really cheap, but the ones I got once worked for a few months then quit working because they et deactivate when they go online or something maybe because they're not authorized copies or something. Anyway, I was wondering where/how you can get them and that will keep working.

The original poster needs to follow a schematic and check at the fuel pump relay for voltage from the PCM when switched to run. then check for voltage from the fuel pump fuse to relay, then then jumper fuel pump circuit fuse to fuel pump circuit, etc. This would tell if the problem was in one of the fuel pump circuits and wich one or whether it was a problem wih the ignition switch or PCM not energizing the relay. Does turning the ignition switch to RUN activqte the instrrument warning lights when the fuel pump problem is happening?

Now that I come to think of it, the OP might also want to check the relay socket itself. I would spray something like WD-40 into it. I've also found that this trick works for flat pinned relays. Give the flat pins a slight twist with a pair needle-nosed pliers. This lets the edges of the pins cut into the socket connection below giving a much better connection.

Does turning the ignition switch to RUN activqte the instrrument warning lights when the fuel pump problem is happening?

I was trying to get to the same thing there by asking him to check power to his ignition coil with the key on. But then I got to thinking, that maybe other things that are supposed to work with the ignition key on might still work, if possibly there are more than one set of contacts for those things when the key is on the on position. I'm not sure though, I haven't ever took out and replaced a factory ignition switch, but I've added in universal key ignitions on dashes before and those only have one set of contacts for run, one set for accessory and one set for start. So some help there if someone knows would be greatly appreciated, does the factory ignition switches only have one or could they have possibly more than one set of contacts for the key in the run position.

If they have more than one set of contacts for the run position then it's still possible the ignition switch is bad even if getting power to coil, instrument panel, etc. If anyone knows please reply.

I was trying to get to the same thing there by asking him to check power to his ignition coil with the key on. But then I got to thinking, that maybe other things that are supposed to work with the ignition key on might still work, if possibly there are more than one set of contacts for those things when the key is on the on position. I'm not sure though, I haven't ever took out and replaced a factory ignition switch, but I've added in universal key ignitions on dashes before and those only have one set of contacts for run, one set for accessory and one set for start. So some help there if someone knows would be greatly appreciated, does the factory ignition switches only have one or could they have possibly more than one set of contacts for the key in the run position.

If they have more than one set of contacts for the run position then it's still possible the ignition switch is bad even if getting power to coil, instrument panel, etc. If anyone knows please reply.

Well there's a couple connectors and about 9 wires related at the ign switch not including passkey. But several of them run to the BCM and a problem in one of them would set a trouble code. The rest are basically just battery voltage, starter, access. RUN. Lots of things are powered on when in ON and RUN but these circuits arent wired through the ignition switch if that's what you're asking. They are spliced into the wiring upstream. A problem in the ignition switch would affect several circuits. That's why I asked him if most everything else is working.

Anyway, all that matters in this case is whether or not he is getting power to the fuel pump relay terminals where he should be and if power is getting to the fuel pump by following the wiring schematic.

Late model cars use the Engine Control Module to switch just about everything on or off. The ignition on wire runs to an input of an integrated circuit. Outputs from that IC will turn relays on or off. So one wire can control as many outputs or relays as the manufacturer wants. You'll most probably find that the low oil pressure light is switched on or off by the ECM and not directly by the contacts on the oil pressure switch.

Logic circuits also convert analog signals to digital. If you take the oxygen sensor for example which gives a fluctuating voltage output, and connect it to the input of a logic circuit, the logic circuit will see 1volt for example as on and, a voltage below that as off. The manufacturer can decide what the threshhold voltage should be. So the oxygen sensor is either working or it is not in which case a check engine light will come on.

Logic circuits can also be used as and gates in which case this input, and that input, plus that input, must be present before a relay can be switched on. We have seen in this discussion that the fuel pump relay can only come on again if signal pulses are received from the engine to indicate that the engine is running.

I've tried to simplify things here so I hope this helps a bit. I did write this before breakfast.

Changed the ignition with one from a '98 Malibu from a junk yard and solved the problem. It's been working without one false start since I changed it. Thanks for all your comments.