nissan s14 sr20de, 34volts at injectors while running.

hello

I've recently made a custom loom and fuse box for my Nissan s14, i followed the haltech elite 1000 wiring diagram. powering my solenoids (vtc and iac) from the same relay as ecu and ecu controlling the ground side. The car runs and idles normally with no signs of issues, until it revs to about 3k rpm and basically dies from lack of fuel.

Upon further testing while the car is running and my VTC solenoid is plugged in, i get 30v to the injectors. unplugging the VTC reduces the voltage at the injectors back to 12 volts. I've got stable 12v at battery, alternator and ecu, coil packs. I'm currently wiring the solenoids onto their own relays as I've been recommended by my tuner. im just wondering if anybody's experienced this issue or a similar one? what could cause this?

the 30volts was confirmed with 2 different voltmeters and an oscilloscope.

Hi Caleb,

I am not familiar with internals of aftermarket ECU's but a fair bit factory .

Not sure if you could share a recording of the oscilloscope pattern or screenshot of fuel injector measured offset and differentially ? Below 3000 rpm and above before dying ?

Ive included a pic of what a normally working injector looks like on a scope

hey Hamish, thanks for the reply!

unfortunately it was my tuners oscilloscope and the car was only there for the morning. so i don't have access to it anymore, Haltech has an integrated oscilloscope in the software, but I'm not sure it will replicate/show this issue. if i was to compare it to the file you added, it had no flatline. it just spiked up and down rapidly, like 50-100 peaks across the screen, and I'm unsure of what the axis scale was.

I should of mentioned its showing 12 volts when the engine is turned off, so emf noise is quite likely, i was more thinking it was a back feeding issue, but im only just beginning in auto electrical, so its all just google and guesswork for me haha. but yeah i would of thought 30v feed to 12v injectors would fry them. If it is noise, how would i go about locating and removing/reducing it?

I also rewired both the solenoids this afternoon onto their own relay circuits and moved the coil pack grounds to the other side of the head away from the ecu ground, the issue is still present, with my multimeter still bouncing between 9-37volts at each injector plug (while unplugged from injector).

On a side note that might be of use. i was also having interference issues with my wide band, causing it to keep restarting and unable to get further than boot mode. This issue was solved by re routing the wire away from the main loom. this suggests to me that there is definitely some noise.

Hi Caleb,

Yeah it will be electromagnetic noise from collapsing magnetic fields in high current consumers most likely. Factory wiring would be ground shielded to contain these fields from interfering in other wires possibly close by.

It is likely an integrated oscilloscope with a very low sample rate in ECU software ( I am predicting , question for Haltech ). You need 50 000 - 100 000 measurement samples per second to get a decent reproduction of fast moving electrical signals, with no filtering of signals , particularly at higher engine speeds. A multimeter does around 2 samples per second so is very slow sample rate, also averages readings often over this time ( RMS ), stuff gets missed in between these measurement gaps. High voltage spikes will probly be effecting multimeters averaged reading.

When you disconnected your timing solenoid , did engine still stall over 3k revs?

Have a look at fuel pressure at 3k .

I think you need to see what is happening to your injector pulse width over 3k revs when stalling with one of the scopes channels and another scope channel monitoring crankshaft speed sensor ( crank vs injection ). This will be a scope job

Ive used shielded cabled for all the sensor/trigger wires, that would be shielded from factory. cas( trigger+home), tps signal, knock and IAT grounding the shielding at one end only to the ecu main ground. could changing injectors to shielded cable reduce noise/ help the cause? kinda like a band aid fix haha

fuel pressure was our first thought, as the wide band showed the car leaning out once the vtc comes on (around 3k) we thought maybe there was a leak inside the tank, so pulled the pump and changed the lines with new hose clamps, i added an inline pressure sensor and fuel pressure is constant 40ish psi, regardless of rpm. so we've ruled that out. My friend is bringing round their oscilloscope tonight help me do some further testing. ill do the crank vs injection test and post the results up here.

i believe the issue still occurs with the vtc unplugged, but ill confirm as well tonight

Nice, the part of injector pattern where injector is switched to 0v is when injector is open and is pulse width. If noise on crank sensor pattern, ecu can count these as tone wheel teeth, if voltage goes to similar thresholds as tone wheel teeth. It will stop injecting when it notices illogical tone wheel tooth counts.

Have you changed exhaust same time as ecu+ wiring ? Done any changes around bell housing- removed brackets etc?

I Have changed exhaust system, it still has a factory de exhaust manifold and down pipe to a custom 3in, but all the same mounting points. but it was running this no worries when i had a stock ecu until late last year. When randomly i stopped getting spark, after further diagnosis i found the loom itself had been pulled/ stretched near the ecu and then wire had been damaged causing it to not give a constant signal. then in the process of repairing the loom, i somehow the 5v transformer on the stock ecu. at which point i decided to put in my haltech ecu from my other car and build a full new custom loom following either factory nissan or haltech wiring diagrams.

the engine has also been fully rebuilt, this was done at the same time as as the exhaust. (just over a year ago) Its a bit of a Frankenstein motor.

oem DE block and crank, intake, throttle body, ehxaust and down pipe

oem DET pistons and rods, rings, valves, springs,retainers and head gasket.

tomei 256 pon cams

Acl bearings

Arp head studs.

new water pump, oil pump, oil pick up and hardware etc

oem 370cc det injectors and fuel rail

oem CTS, IAC, VTC, TPS, knock,

since doing the new ecu i converted to cas and coil packs instead of a dizzy. I've also added a IAT and deleted the MAF as haltech has an onboard map s sensor.

i haven't removed any brackets from the gear box, or motor. at least not since i've done the ecu and loom upgrade. essentially the car was running fine and saw events all of last year. while doing the loom. i increased the thickness of all my ground straps, and cleaned the connection points on the drive train and the chassis. so i have 5 ground straps in total, 1 on box, 2 on block, 1 on head and 1 on intake.

but yeah the way the car was behaving on the dyno we even had a suspicion that the timing chain had jumped a tooth. but we took of the rocker cover and all the marks lined up and we had 20 pins and 11links between the cam marks. i'm also running a solid chain tensioner. so it would of had to fail in order to jump a tooth.

Ive added some photos of the engine bay/ my custom fuse box. all the engine wires have been sleeved since these photos were taken.

Hello Caleb,

I'm curious to see what that waveform looked like, hopefully you can post a screenshot of the scope.

Due to the nature of a coil (an inductor) that is triggered on and off with pulsed DC, a single spike is totally normal when the field is collapsing, as explained by Hamish. For this reason, multimeter are pretty useless here... good thinking to use a scope.

I don't think I've seen a case of the injector circuit causing noise in another circuit, and in the OEM Nissan world, the injectors wiring is not shielded. So I doubt that shielding the injector wires would solve the issue. Was there other symptoms other than the wideband leaning out? Was the engine stumbling / misfiring? What happens if you try to richen the mixture? A misfire will be read as a lean mixture by the wideband (no combustion = full oxygen in the exhaust), so a common mistake is to interpret an ignition issue for a fuel issue.

I don't want to steer you in another direction, but if you have access to a timing lamp, I'd check how the ignition react when your issue happens. The trigger system in the 90's Nissan (if you're still running it) is prone to creating wierd issues when paired with aftermarket ECUs. You could also check the trigger scope in the Haltech software.

*EDIT* I've seen reversed fuel line (feed and return inverted) causing some similar issue, just saying!

Let us know how it goes!

Frank

I'm fairly confident its not a trigger error causing the motor to do this, as the rpm is stable/synced with the motor, and trigger/home signals are what give the ecu this reading and the ecu gives this reading to the oem cluster, which also works fine displaying the same rpm as the ecu. trigger is also full and synced when the car is running, no trigger errors are recorded, (unless the car stalls, in which case a p1300 is thrown out by nsp, but that's normal and expected) .

ive added a video from the weekend of the nsp software while the car is idling. showing the trigger synced and full with 0 errors counted. the wide band error in the video has also been solved now as mentioned previously.

but yeah at the same time the wide band leans out, the engine stumbles and almost dies as if fuel has been cut completely. no amount of adding fuel to the base map/tune was able to change how much it leaned out. ill try get a recording of this and upload it later, how ever for the dyno i had a second wide band connected further down stream for more accurate results.

we did take a timing light to the car yesterday at the tuners as well and the timing was all over the show. not giving a consistent result, even at idle. but after moving the coil pack earth to the other side of the head, it seems to have settled down and be more consistent with the timing mark now. ill take a video of this later on aswell and upload it along with the oscilloscope readings and nsp report.

I'm also confident the fuel lines aren't backwards, as i had them backwards when i first installed the new motor last year and obviously fuel wont back feed through the regulator so i wasn't getting any fuel to the injectors at all, i changed them around back then and solved that issue. i haven't taken the lines off since.

but i appreciate any input, so thank you :)

I've resolved the issue!!

so first the car wasn't starting today, unlike yesterday, even though I hadn't changed anything since turning it off last night. we connected my friend's oscilloscope to try record the issue, and the injectors showed one pulse/wave form at the begining of cranking and then flat lined, we then decided to recheck over basics in which i had found the main battery to chasis ground connection was only hand tight and had corroded underneath. so i sanded/cleaned the both sides of the connection, then while checking over the fusebox i found the injector relay power feed had also pretty much come out of the back of the plug. so i re pinned it into the relay plug. after everything else seemed to okay, we tried to run the car and it turned on and with the issue now resolved, happily reaching set 7000rpm while in neutral and under load.

I just wanna say thanks to Hamish and Frank, your input definitly helped my understanding of what was going on. sorry i dont have any recordings or readings as I wasn't able to replicate the results. ill see if the tuner logged the recordings somewhere, as i also would be intrested to see it again now ive got a slightly better idea of what im even looking at

Hi Caleb,

Nice work ! Late at night you might have seen sparks. Its pretty excellent when you find these problems( good enough to post at almost midnight ). Just for reference , your earthing problem would show up on an oscilloscope every time something was being grounded ( such as your injector ) as not going down to near 0 volts ( could be pretty much any below system voltage except 0 volts ) .

This is provided you earth your oscilloscope earth lead when measuring offset ( from earth ) on the only true earth of vehicle( the battery negative post ). Every other part of the earth is part of the circuit and could potentially have high resistance. Don't earth measuring equipment on engine blocks/chassis etc or you could miss problems