[Kingfysh]

My wife and I just bought a 2007 Element SC with 9,500 miles. After driving it for a few days we noticed that the car would some times lose power momentarily while accelerating. This culminated in the car stalling out while pulling onto a major highway in town, which while exciting was less than desired behavior. The car also had difficulty starting at times, battery seemed strong, but engine would just crank and crank. Eventually, the engine would start, but it would take about 45 secs to a minute.

So, we took the car into the local dealer. They thought that both of the above problems were connected and diagnosed the problem to be the computer, which they then replaced. Here is the exact wording on the work order: Retrieved DTC code PO 685 internal fault in PCM replaced ECM.

Well, today the problem has come back. I sat in the driveway cranking the car for over a minute until it finally started. Then later in the day as we pulled into traffic the engine seemed to lose power for half a second. The RPMs dropped and an audible click could be heard from the gear selector (sounded like a relay). This repeated multiple times throughout the day. It happens only when accelerating and it seems to occur at about the time when the transmission shifts from second to third, but I am not sure.

Initially I thought the problem was water in the gas, so I put some Dry Gas in the tank. Unfortunately, this did nothing.

I have already made an appointment with dealer for tomorrow, but what do you guys think? I have not been able to find another problem like this on here. Any suggestions? Anyone had to deal with this before?

Thanks.

 

[wipeout97]

Anyone think this could be a bad knock sensor? If it read a misfire it would cut power immediately to prevent engine damage? The problem is if this is the case and a dealership can not recreate the problem I doubt they will fix it. If they can not get the problem sorted for you, it might be worth your while to have it replaced elsewhere to see. If it fixes the problem at that point I would send the invoice to honda corp and attempt to be reimbursed.

 

[slimerdogs]

It sounds like a fuel delivery pressure issue to me. The fuel pump will not get enough pressure up to get it to start and when you step on the gas the fuel pressure is supposed to go up and its not. If it does not want to start again turn on the ignition foe a couple seconds and then shut it off. Do that a few times before cranking the engine and see if it will start after that. On some vehicles the pump is a variable speed type compared to having a return regulator and I do not know if the Honda is that type.

 

[Kingfysh]

Slimerdog, that is what I end up doing to get it to start. Does that point in the direction of a problem in the fuel system? Clogged filter? Dying fuel pump?

 

[wipeout97]

If slimmerdog is correct it would more than likely be a faulty fuel pump or clogged fuel filter. Fuel filters are supposed to be swapped every 15k-30k anyway so the $20 dollar part would never hurt to replace. On the E however it is located in the fuel tank so while you may be eating labor to replace one you might as well be on the safe side and replace both.

A nice bottle of lucas fuel treatment can often do the trick on clogs. Either insert a complete 5.5 ounce bottle for around $6 or buy the 32oz bottle for $10 dollars and place in 6 ounces to start with followed by 3-4 ounces each fill-up until the bottle is gone.

I am just shocked the car never bucked you bad considering it is fuel delivery but cars are fickle at times.

 

[MikeQBF]

Similar problem I had with a before-the-E turned out to be a loose connection in a critical engine cable, this one going to the crankshaft position sensor. Under hard acceleration, the engine would twist slightly under the torque load (like it's supposed to), tug on the cable, and the connector would lose contact.

So... if I was looking at your situation I'd be all over the place re-making every connection I could find and checking that nothing is pulling on the wire bundles. That the original problem popped a "bad ECM" code leads me to suspect this might not be too far off the mark - intermittent connection at or near the ECM.

 

[xequar]

Similar problem I had with a before-the-E turned out to be a loose connection in a critical engine cable, this one going to the crankshaft position sensor. Under hard acceleration, the engine would twist slightly under the torque load (like it's supposed to), tug on the cable, and the connector would lose contact.

So... if I was looking at your situation I'd be all over the place re-making every connection I could find and checking that nothing is pulling on the wire bundles. That the original problem popped a "bad ECM" code leads me to suspect this might not be too far off the mark - intermittent connection at or near the ECM.

Agree ^^ When the camshaft position sensor went out in the BF's Neon, it would basically accelerate to 2,300 RPM, then would cut out. Basically, the sensor was thinking 2,300 was actually red line and was killing the engine accordingly. If your connector's losing contact, it'll flake out in all sorts of interesting ways.

FWIW, I had an old Bonneville that would sometimes take a minute of cranking to get running if the engine was cold. Turned out the computer had flaked out and always thought the engine was warm, so it would only have the fuel pump give 13 PSI at startup instead of the 46 it required. Of course, this led to a mis-diagnosis, as it looked like the fuel pump had gone bad. I know you've had your computer replaced already, but is there any kind of sensor that could flake out and cause a similar scenario?