97 Elantra, over 200kmi. I noticed decreased MPG’s over the fall, not a lot, but some. Then, it started running rough and throwing a random misfire code and misfire in cylinder #4. Replaced the plugs, and the wires appear good (less than 40kmi on them). Swapping out the coil pack didn’t help. I unplugged injector #4 and plugged it back in. It ran fine for a couple of hundred miles. Shortly after that, one cold night it started and ran rough throwing the same codes. I was going to turn around, but after a few minutes it ran just fine again. Drove it another couple of hundred miles with no problems. It sat for a day or two in the cold, and now won’t start at all. It acts like it will almost start, and there seems to be some combustion taking place, but not enough to get the engine going.
I replaced injector #4 and cleaned out the fuel rail. It had deposits of a tan/white chalky substance. I figured a piece was binding up the injector. Put everything back together and the same thing… wants to start, but won’t. Compression is 180 to 220 in all cylinders. I shot about 2 seconds of starting fluid into each cylinder and it ran for about 1 to 2 seconds. I have spark…now I’m thinking maybe the fuel pump has gotten weak (I don’t have a pressure gauge) or perhaps a clogged filter. I tried banging on the bottom of the fuel tank a few times.
Is there anything else I should be checking? It ran rough before but smoothed out once the engine warmed up. I’m wondering if whatever it is, is related to the cold? It hasn’t been above 30 for some time here in upstate NY.
Thanks!
Were the new spark plugs properly gapped?
Have you checked the timing with a light?
[i] Compression is 180 to 220[/i]
Someone will correct me if I am wrong, but that sounds like a rather large spread.What you really need is to figure out what the error codes were.
And I wouldn’t do another without getting a fuel pressure gauge on it. Can you at least hear the pump energize?
Plugs were correctly gapped. Haven’t checked timing.
Compression from #1 to #4 was 180, 210, 220, 200. Is that a problem? I thought a difference of 40 was high, but about acceptable. I wouldn’t doubt that the engine is showing its age, but it should at least start.
P0300, P0304. Random engine misfire and misfire in cylinder #4.
I’ll check the pump.
I picked up a pressure gauge for $40. I don’t use it much but its so nice to have around to slap on anytime I have reason to wonder. Well worth the purchase.
Update: I could not hear the fuel pump run when applying 12vdc to it. I needed a new sending unit anyway, so I replaced both. Still won’t start. I definitely have spark, but I suspect that something is not allowing the injectors to fire at the right time, or at all. After trying to start, the #4 plug is always wet and the others are dry and I did get it to run after a shot of ether into each cylinder. Sitting cold, the crank angle sensor is about 476 ohms (tolerance is 486 to 594 at room temperature). I put a hairdryer on it and got it up to about 510 ohms… is this a normal reaction? Should I replace it? Unfortunately, I think to test for the camshaft position sensor it needs to be running. I’m going to try and check the main fuel relay next. Slowly reaching the end of my rope here. Thanks.
Resistance increases with heat (which is why the spec comes with a temperature range), so heating up the crank angle sensor and finding higher resistance is normal.
Is the fuel pump actually priming up when you turn the key (2-3 second hum when key is turned to on)? If so, is the pressure within specs?
Hmm… the pump should prime up for starting, at least a bit when the key is turned on. However, I don’t hear that at all. While cranking, its hard to hear anything due to the noise of the engine cranking, but after cranking stops I can hear the pump run for a second before it kicks off. It WANTS to start… it almost catches and chugs, but it won’t. There is a “main control relay” which feeds several systems on startup. Its cheap enough where I think I’ll try replacing it.
I must admit that I have not gotten a fuel pressure gauge yet. I know its a key piece of test equipment, but every dollar counts at the moment. If the relay doesn’t fix it, I’ll get a gauge and check the pressure.
Just turn the key to on without cranking it - the position just before it would crank when all of the dash lights come on. Listen for the pump then. Did you ever toss a new fuel filter on it?
Fuel pump won’t come on with the key in that position. I can hear it run and kick off when I have stopped cranking. The old pump did not do that. I wonder if the old pump was fried, and it fried the relay associated with it. I do know there is 12vdc going to the pump when cranking as my multimeter makes a beep when it detects a voltage.
The fuel filter has about 50kmi on it. I’m considering that too, but one day it ran and the other it didn’t, so I haven’t replaced it. I did find a fair amount of dust and fine particles in the tank which I suctioned out with a hand-held suction pump. There is a bag/screen on the old pump which was kind of dirty, but it may have been the original pump so perhaps that is normal.
The relay clicks when voltage is applied and there is no resistance along the switch. I’m assuming that means the relay is good? Time to replace the fuel filter… though if that works, it would be the first time a clogged fuel filter has ever been the source of a problem for me. This is a tough one.
Lets go back a couple of steps:
Since you can get the engine to run for a little bit when you spray in a starting fluid, that means you aren’t getting any fuel.
Since you say you can’t hear the fuel pump running, but you just replaced it and the sender unit, lets run a test to see if the pump is actually getting a signal to run.
Unplug the connector to the fuel pump, and hook up a test light to it.
You obviously want it connected to the wires that provide power and ground to the pump.
Now, when you switch the ignition to on, you should see the light turn on.
Also, when you crank the engine, the light should come on.
If it isn’t, then you need to trace back why the light isn’t coming on.
Probably a fuel pump relay, or the computer’s ecu.
You can jumper the relay to test to see if it is the cause.
If your car starts with the relay jumpered, then you have the solution, and need to buy a new relay.
You should also be able to hook up a fuel pressure gauge to the fuel rail under the engine bay, to see if the fuel pump is providing any fuel pressure at all, and if it is providing enough.
If you are getting fuel pressure to the rail, then you need to find out if the fuel injectors are firing. Hold a long screwdriver to each injector, and the other end press your ear against, and see if you hear a clicking noise while someone cranks the engine over for you.
Do most or all of these tests, and let us know what you get for results.
BC.
pull filter and energizes pump with hose on outlet check flow should be about 1/2 a liter in 15 sec. The regulator maybe vaccume operated if so pull vacc hose and plug. the lost in vac will bump pressure. I have seen a bad injector fry the driver circuit. Can you here injectors clicking?
OK, there is power to the fuel pump when cranking, and for a second after I shut the key off. There is no power to the pump when the key is simply in the on position (the position just before the cranking position). I don’t know if that’s normal for this car. There is 12vdc to all of the injectors when the key is on and when cranking (I don’t know if this is normal). The fuel pump produces about 1/8 to 1/4 quart of fuel per 7 seconds of cranking. This was measured after the filter.
I jumpered the relay that controls the fuel pump so I could run it independently of the car starting. That made virtually no difference, but I was able to seemingly push a lot of fuel into the rail this way and the car was the closest to starting it has been, but it chugged or “dieseled” for a few seconds before quitting. Some combustion did take place. The exhaust manifold goes into a twin pipe before the converter and the passenger side pipe was warm where the other side was not. This leads me to believe that not all of the cylinders are getting fuel. Its too hard to hear if the injectors or working or not over the noise of the engine cranking.
The specification resistance on the crank position sensor is about 486 to 594 ohms at room temperature. I’ve got 476 ohms at 20 degrees, but it increased to within spec when slightly heated. When cranking, the resistance dropped to about 120 to 140 ohms. Does this seem normal?
I’m thinking there is either a bad sensor or a bad computer causing this problem. I may try replacing the crank position sensor and go from there unless anyone has any other ideas. Thanks for all of the suggestions!
Replaced crank position sensor. Still won’t start. #4 plug is always wet and the others aren’t. Is it possible that the rest of the injectors are bad too? The fuel rail had chalky deposits in it, but I only replaced injector #4.
You are not believing your own diagnostic test. If the fuel pump is not getting power in the “run” or “on” position but only gets power in the crank position of course it won’t run when you let go of the key. That is the problem you have to address.
Yes, I thought that was strange. It seems to me that the pump should run when the key is in the “on” position. Perhaps when the car isn’t running, something in the computer is telling the fuel pump to shut off? I did jumper the pump relay to be on all the time and it still wouldn’t start.
I have an idea… I’m going to pull the fuel rail again and make sure I’m getting spray from all of the injectors. If I’m only getting spray from #4, then I guess the rest of the injectors are bound up.