battery voltage to ECU question....is it safe to.....?

[the t3d]

ok here's my problem. i am only getting mid to low 12v to the computer. when the car has been on the road for a while, it will be around high 11's to low 12's at idle. so i'm losing like 2+ volts by the time it gets to the computer. at night is the worst. when i put the headlights on it drops about a volt, which is NOT good. so i can't beat on this thing in fear of running low voltage to the fuel pump and leaning it out. Alternator is good...even swapped out alternators. same thing. battery is less than a year old. i have cleaned and made new grounds for everything. the fusible link checks out. everything passed. so there must be resistance somewhere in the wire and when it warms up, it causes more resistance causing less voltage. which is gay. it's driving me crazy! so...my question is, instead of ripping apart my freshly taped up, blue engine harness, is it safe to run a wire from the battery to the B+ wire that goes to the ecu on top of the existing wire, so the ECU can get the proper voltage? maybe with an inline fuse or something... i am doing all the monitoring with my SAFC 2. i have checked power at the SAFC 2 and the battery wire just make sure the SAFC 2 is not sucking the power. it's not...it's reading exactly what the voltage is.

i know it's long, but read everything i wrote so you understand exactly what i'm asking, please. i know a fair amount about cars, so i'm not a noob at this. i just don't want to fuck up anything on my beast :evil2:

THANKS!

-t3d

 

[Supracentral]

Have you eliminated the possibility of a systematic problem that you need to address? In other words, how are the voltages at other critical systems? Do you have ample power to the fuel pumps? What about the coils?

Provided all of that is ok, and you've isolated it to just the ECU:

I'd use a relay to trigger a home run directly from the battery to fix the issue, but be sure to fuse it as well.

I doubt the ECU pulls very many amps. I'm sure 3p141592654 can tell you what the maximum draw that the ECU is capable of. I'd set the amperage of the fuse just hair above that.

While a home run can be a quick/cheap/easy way to address a wiring problem, it can also be a curse. Without the proper fusing, you risk anything from component damage to burning your car to a crisp.

Also, be aware that you may be ignoring a larger problem in the harness. It's rare that one wire gets damaged in a harness, and you may be ignoring potential future problems by not tracking this down right now.

 

[the t3d]

Have you eliminated the possibility of a systematic problem that you need to address? In other words, how are the voltages at other critical systems? Do you have ample power to the fuel pumps? What about the coils?

Provided all of that is ok, and you've isolated it to just the ECU:

I'd use a relay to trigger a home run directly from the battery to fix the issue, but be sure to fuse it as well.

I doubt the ECU pulls very many amps. I'm sure 3p141592654 can tell you what the maximum draw that the ECU is capable of. I'd set the amperage of the fuse just hair above that.

While a home run can be a quick/cheap/easy way to address a wiring problem, it can also be a curse. Without the proper fusing, you risk anything from component damage to burning your car to a crisp.

Also, be aware that you may be ignoring a larger problem in the harness. It's rare that one wire gets damaged in a harness, and you may be ignoring potential future problems by not tracking this down right now.

i see what you are saying. i was trying to see what things are tied to that wire, so i know what to check but i couldn't find anything on the TSRM. i know that when the voltage drops below 12v, i can see the a/f ratio go lean because the fuel pump isn't getting enough voltage.

 

[Supracentral]

I know you said you checked your alternator, battery and grounds, but this really does sound like a systematic problem.

I'd suspect grounds before anything else. Double check everything before you start messing with the ECU. I'll almost guarantee you've got a larger issue that you need to deal with.

 

[the t3d]

grounds are perfect. metal to metal and new wires. what exactly should i check so i know what each thing reading? i want to eliminate as much as possible before i do something like this

 

[Supracentral]

Get out an ohmmeter, unplug your harness at the ECU and do a resistance test on the B+ wire to start with.

Do the same on your grounds - I've seen grounds that "look" perfect but fail a resistance test.

 

[the t3d]

alright. i'll give that a shot. if anyone has more input, jump in

-t3d

 

[jetjock]

Your problem is easily resolved using proper measurement technique. As I've stated many times voltage drop testing is likely the most valuable automotive electrical skill a person can learn. Not a perfect tutorial but a good place to start:

http://tinyurl.com/yfbldut

 

[Supracentral]

Great vid JJ (even if it is paced for someone who is developmentally challenged.)

 

[TurboStreetCar]

Turn the key on and check voltage from the Ground on the ECU to the chassis.

Then check voltage from the Battery + Terminal, to the B+ terminal on the ECU.

Both should read Zero. If anything above .1-2 volts you found the problem.

 

[jetjock]

I'll admit that vid is "drop testing for dummies". There are a few other things wrong with it too. One is his method of testing for drop across both sides of the circuit. A much better technique is testing for drop across each side independently, the way it's done on higher current paths like those used for starters and alternators.

Edit: What chunks said except you'll never see zero. There's always some resistance and therefore always some drop. A rule of thumb is less than 100 mv on high current paths and proportionality less on low current paths.

 

[the t3d]

Turn the key on and check voltage from the Ground on the ECU to the chassis.

Then check voltage from the Battery + Terminal, to the B+ terminal on the ECU.

Both should read Zero. If anything above .1-2 volts you found the problem.

are you talking about the ecu ground on the intake manifold?

check the voltage on the battery terminal itself and then the B+ terminal?

 

[jetjock]

i know a fair amount about cars, so i'm not a noob at this.

Err, ok...

Put one meter lead on the ECU ground pin and the other on battery negative. Set the meter to it's DC millivolt scale and turn the ignition key on. If you read more than 100 mv move the battery meter lead forward along the ground path until the drop is acceptable. The problem lies between there and the last place you were. Move the ECU lead to that place and the other lead back to the battery and start over, moving either lead towards the other until the area of resistance is pinpointed.

If you don't see more than 100 mv move to the positive side of the circuit and do the same...

 

[the t3d]

i meant more than the average person. i don't know alot of things, which is why i'm askin. i get what you're saying now, but i'm running out of daylight. the tsrm says the battery voltage is supplied to that wire through the EFI relay when the ignition switch is turned on. and current flows through the ignition fuse and then to the ecu. so, i'm just curious why no one is mentioning that possibility

 

[jetjock]

The possibly was inferred. Any one of those things could be at fault. That's why the positive side is a bit more difficult to troubleshoot than the ground side. Still, using a meter allows you to split the circuit in half and then in half again, thus zeroing in on the culprit. The only alternative is to swap stuff out or jumper around it. Again, you can either jumper around all of it or split the jumpering up until the problem goes away.

The meter will tell you where it's at though, and quickly, once you understand the concept. If more current is needed (circuits must be active to do VD testing) additional loads can be applied. I use an electronic load but a head lamp (or two) also works pretty well. Just don't load a circuit beyond what the lowest ampacity device in it can handle.

Fwiw all this is *the* solution to finding the real cause behind the starter relay mod...

 

[the t3d]

well, i'm using the Power Probe 3. which is an awesome tool. i'm still getting familiar with it. all this is very frustrating. i'm going to taking you guy's advice tomorrow, i'm just trying to gather all i can before i get workin.
...and what's the starter relay mod?

 

[jetjock]

While the Power Probes are nice tools for simple stuff (I own a PPII and it is handy) they can't beat a full featured DMM for serious work.

The starter relay mod involves sending current directly from the battery to the starter solenoid via an added relay piloted by the original starter signal. Proponents claim the stock wiring develops high resistance from age, too high of a resistance to deliver the handful of amps needed to power the solenoid. It treats the symptom and not the disease which, contrary to popular belief, isn't bad wiring. At least I've never found it to be bad wiring.

 

[the t3d]

yea i guess that's true. i have a POS craftsman DVOM that i guess i'll bust out.

 

[the t3d]

so me and a ase master tech went to down on this thing. checking all grounds and power supplys. we jumpered grounds and power to notice any changes. and there wasn't no change with the jumpered grounds and little to no change with the jumpered powers. i have like completely killed this battery twice and had to re-charge it with a battery charger. i am reading like 8-9v through the battery casing about an inch away from the neg. terminal on the battery. the battery is under warranty, so i'm probably going to swap it out. since i'm running more than the average electronics, i'm going to get a higher amperage alt. (the 140A one from DM.com). i'm going to get an amperage reading at work on what it's putting out now. so, now my question is...would i have to upgrade the alt. wiring with this alt? and would i have to change out the 100amp fusible link that is stock?

THANKS!

-t3d

 

[grimreaper]

totally forgot about this on my car. When i had the safc i noticed similar readings. Night time with the head lights and fog lights on made it worse! looks like ive got my project for the week!