95 civic ecu prob? what conditions must be met to start
 

Hi, i have a 95 civic ex that is not starting right. at first i thought it was the fuel pump. replaced it. then i thought it was the main relay, which also runs the fuel pump. replaced it. twice. The car turns over fine, but if i don't hear the fuel pump prime for 2 sec., i don't bother turning it over, because i know it wont start. when i first get in to the car, i have to turn the key to start, and wait up to 5 or 10 minutes for the fuel pump to kick on, and then it starts right up and i could drive until i ran out of gas. so after the pump primes, i can turn the key off and then back on, and only wait about one minute until the pump primes. key on/off again, and again, the wait keeps getting cut in half until the pump primes instantly. If i let the car sit over night, i have to start over. SO, does anyone know what conditions have to be met before the ecu allow the car to start? I have done the troubleshooting from the honda manual for the main relay and pump, and the problem is not there. I think something is not getting satisified to let the ecu complete the circuit to start the fuel pump and start the car.

I think the 2-second prime is unconditional. Every time the ECU powers up and resets it's going to do the CEL bulb test and fuel pump prime. Check that all the power pins on the ECU are getting power with the key on, and all the grounds are grounded (common to have trouble with the ground wires on the engine at the thermostat housing). Confirm there's no short in the 5 volt feed to the sensors under the hood. If all that is OK, replace ECU.

i will check this tonight and report back tomorrow. which sensors are you refering to under the hood?

How many volts should the STS (blu/wht) wire have at the ECU. I'm getting 0.3V, then it jumps to 0.55V while the pump primes, and then back to 0.3V. Also, IGP1 (#1 on the B ECU harness) is supposed to be a power ground, but it shows 4ohms before and after the pump primes, and I believe 0.55 while it primes.

MAP sensor

Focus on the ECU wires that get voltage from fuses 24 and 31 via the main relay.

Also clean and tighten the G101 ground on the thermostat housing, as mk mentioned.

ok, i cleaned the grounds (which looked fine), and I have 5V exactly on the MAP. I have found that as soon as the key is turned on (II), I have 12V on both the yell/blk power to the ecu, and the green/yell fuel pump relay from the the ecu. When the pump kicks on for 2 sec. (after waiting for about 20 seconds), the volts on the green/yell wire go very low for the 2 sec, while the yell/blk wires stay at 12V. Can you explain how/why the contact for the pump is not closing if there is power on the coil for it? even if there is a problem with the ecu, shouldn't the pump contact close regardless??? The diagram can be found here: https://www.hondacivicforum.com/foru...6/?ref=esp-img
I'm checking these voltages on the relay, but the power is the same at the wires going into the ECU. Does this mean that the green wire is not getting grounded somewhere? If so, does this happen inside the ECU (ie I'm screwed) or is there some place else I can look?

https://www.hondacivicforum.com/foru...6/?ref=esp-img

The ECU drives the bottom side of the coil to ground (green/yellow wire). That should happen immediately upon turning the key on, not after 20 seconds. It really looks like your ECU is bad.

is the ecu sink or source? maybe this would explain why i see potential on my voltmeter (12V) but it may not actually be grounded though the ecu. i'm trying to make sense between what i am seeing with the multimeter and what is says here: How the Main Relay works

i found that I have the same problem that Riqand had at the end of the thread: https://www.hondacivicforum.com/foru...rs-more-73972/
but i do not know if he ever fixed it

The ECU sinks terminal #8 of the main relay, during the 2 second prime and whenever it senses that the engine is rotating. If the engine is not rotating the ECU quits sinking terminal 8 of the main relay and the fuel pump stops. In order to make these decisions, the microprocessor in the ECU must be up and running. It sounds like yours is slow to start up.

The fuel pump section main relay coil is sourced either from the other section of the main relay, or from the starter circuit (terminal #6). Sourcing from terminal 6 supplies full battery voltage to the coil (bypassing the diode and resistor) to make sure it will pull in despite battery voltage drop while running the starter.