Injector no. 2 spraying fuel when fuel pump primes

[gropula]

I've got an issue with my no. 2 injector -  specifically it's wiring or ECU. I noticed that sometimes when I turn the ignition on and prime the fuel pump the injector no. 2 sprays fuel, or leaks slightly. It doesn't always happen, but it does. I've replaced all 4 injectors and the injector sub harness since I've noticed that happening. The problem persists. What should I do to diagnose and fix this? 

 

Video of the issue

[gropula]

The enigne runs, starts, but with a bit of a struggle sometimes. Sometimes it fires up instantly. Rides normal. When changing the sparkplugs I've noticed that no.2 wasn't getting enough fuel, it was white. So I investigated, noticed this spraying and concluded that injector is faulty. Replaced all 4 injectors and the injector subharness. Still happening. Can't see exactly whats happening when starting because it can't start at full throttle, but after starting it runs fine. My rapid bike says that one OEM o2 is alwas reading richer than the other. I also have the MTB wideband o2 and it runs rich at idle and low throttle openings. The autotuner corrected to around -60% in the 5% and 10% throttle to run at about 13.4 AFR. Something strange happens when I open the throttle alot then close it. When I reopen the throttle slightly the engine stumbles like it needs more fuel. For example, 6th gear accelerate hard to 150, close throttle to slow to 120 and open the throttle slightly to maintain speed. That's when it stumbles. But if I accelerate with similar low throttle from lower RPM through that 5000-5500 rev range it doesn't stumble. It seems that the big vacuum produced by closing the throttle at 6-7k RPM has something to do with the stumble, because it affects the fuel pressure in the FPR.

[Grum]

Wow, never seen that before!

 

If the Rapid Bike intercepts and alters injector duration etc. Then I'd remove it and restore original operations.

This will then identify the problem as being ECM or Rapid Bike. Perhaps those that use the Rapid Bike system can offer better suggestions, I have no experience with these tunning devices.

Sorry can't help much.

[gropula]

I think this was happening before I installed rapid bike. The mechanic that was servicing the bike said not to prime the pump without starting the bike. He said it can flood the cylinders. At that time I just believed what he said. Later I found out why he said that. Now I'm doubting his expertise because he didn't point it out as an issue. I can remove the rapid bike just to test but i doubt that's causing the issue.

[bmart]

Unemcumbered by knowledge, I suspect that it pressurized the system only. 

 

I still remember when I was very young. A mechanic let me watch him bleed my car brakes. He removed the fluid from the master, put in new fluid, pumped the brakes and told me it was done. They're out there...

[ShipFixer]

Rapid Bike does not touch injectors or timing below 3,100 RPM.  It has nothing to do with start behavior or idle.

 

If your injector is leaking by, your injector is leaking by.  You should be able to prime the pump prior to start without any issues.

 

Among other things...I got a used set of injectors off eBay and had them cleaned and rebuilt.  This is way cheaper than new injectors and also eliminates downtime for waiting.  Recommend something like that vs. replacing one injector, where you would have one new one and three old ones which are at a minimum dirty.  The stock ECU and RapidBike are limited to what they can do with two narrowband O2 sensors and a few other things, they cant reasonably do single cylinder tuning.

 

(Although Rapid Bike can...if it has the inputs.)

[gropula]

Thanks for the input. As I've said, I've eliminated the injectors as a possible source of the problem because I bought 4 new ones. Chinese special not OEM. They work fine but the problem persists, again only at no. 2 injector. It has to be the ECU or the wiring loom. I'll try and narrow it down. I have an idea to depin the connector of the ecu, just the pin for no. 2 injector. Then use a new wire and connect it to the injector. That way the wiring loom is totally isolated from the injector. If I were to just backprobe the pin it would still get voltage from that unknown source in the wiring loom (if that's the case).

 

If I understand this injector system is always under 12V, but the ECU sends a pulse of higher amperage to open the injector, right? That's why I can read 12V on the connectors, but the injector is closed. Is there a way to know if the ECU sent the signal, without the injector spraying fuel? Or is the simplest solution to just stuff a rag in throttle bore to keep it from flooding while testing?

[DannoXYZ]

7 hours ago, Grum said:

The ECM sends varying duration pulses to ground. So if you were to measure the two pins/sockets on the injector plug Black/White wire being positive and your other lead to the other pin/socket you should only see 12v for the duration of the spray, it seems a long enough spray for your voltmeter to pick it up.

Actually you'd measure +12v fulltime, then it would drop to 0v during spray interval. Pulsewidth of 5ms not long enough for voltmeter to register anyway.

 

+12v is applied to ALL injectors when you turn key ON and start/stop to RUN. You would measure +12v on BOTH terminals of injector connector.  ECU's job is to ground individual injectors to make them spray. When injector is grounded and fired, current-flow causes voltage to drop to 0v on ECU's trigger wire (would stay +12v on blk/wht supply wire). Signal of ECU trigger line looks like this on oscilloscope

 

[DannoXYZ]

7 hours ago, gropula said:

Thanks for the input. As I've said, I've eliminated the injectors as a possible source of the problem because I bought 4 new ones. Chinese special not OEM. They work fine but the problem persists, again only at no. 2 injector. It has to be the ECU or the wiring loom. I'll try and narrow it down. I have an idea to depin the connector of the ecu, just the pin for no. 2 injector. Then use a new wire and connect it to the injector. That way the wiring loom is totally isolated from the injector. If I were to just backprobe the pin it would still get voltage from that unknown source in the wiring loom (if that's the case).

 

If I understand this injector system is always under 12V, but the ECU sends a pulse of higher amperage to open the injector, right? That's why I can read 12V on the connectors, but the injector is closed. Is there a way to know if the ECU sent the signal, without the injector spraying fuel? Or is the simplest solution to just stuff a rag in throttle bore to keep it from flooding while testing?

That's not how injectors operate. They have power full-time on blk/wht input wire, ECU grounds trigger wire on other side. Which then closes circuit and allows current to flow. Current-flow then opens injector. Injectors are wired up similar to this light-bulb circuit with full-time +12v power from battery (engine-stop relay)) and ground-switched trigger side from ECU. No power comes out of ECU injector lines.

 

 

Wire up this test circuit yourself with battery, bulb and switch to understand how it works. Then measure voltage at point E. :

- 1st, connect power-input line to battery, you'll measure +12v on input-side AND ground-side terminals of bulb/injector (point E. ECU line)

- 2nd, next close ground-trigger switch and bulb/injector turns ON, you'll measure 0v on ground-side terminal (ECU line)

 

It's current-flow that makes things happen, not voltage, which just pushes current. With open-circuit, full-voltage is present on input-line and output-line because energetic +12v electrons goes to end of line and just sits there waiting, no flow occurs. When circuit is closed by flipping switch, then current flows. Voltage (energetic electrons) going through filament/injector-solenoid, dumps their energy and does work to light up bulb, fire injectors, and gets drained to 0v. Then they leave ground side terminal and you'll measure 0v because their energy has been sucked out to do work. *

 

Use V=IR to determine voltage-drop across work-load (filament, injector solenoid).

For example, we have 24w bulb, and you measure current-flow of 2 amps

Then use V=IR to find resistance of bulb-filament, 12v=2amps*R, R=12v/2a=6 ohms

Plug back into V=IR to find voltage-drop, Vdrop=2amps*6ohms = 12V drop across bulb-filament,
12v goes in, 0v comes out when circuit is closed and current is flowing (work being done).

 

DO NOT replace ECU wire with new one, you will be introducing TWO additional variables into system and possibly add additional errors! Also extremely difficult to extract terminal pin from ECU connector without causing permanent damage. Do passive non-interference testing and measuring only! Do not change system in any way!

 

1. to determine if ECU is problem, unplug it from connector and turn on bike same as before, key ON/start=RUN.
Does no. 2 injector still spray unintentionally?

 

2. if NO spray with ECU disconnected, then you've got ECU problem

 

3. else if YES still spraying, then you've got short on #2 trigger wire to ground. Verify by leaving ECU disconnected and unplug all injector connectors. Key ON/start=RUN, Measure resistance between chassis-ground to each injector connector's ECU trigger wire.

  #1 pnk/blu ohms to chassis-ground = ???

  #2 red/yel ohms to chassis-ground = ???

  #3 pnk/grn ohms to chassis-ground = ???

  #4 pnk/blk ohms to chassis-ground = ???

 

 

BTW - put original injectors back in. Don't replace quality OEM parts with lower-quality aftermarket parts. Do not modify system for troubleshooting by replacing perfectly-working parts with brand-new hopefully perfectly-working parts. You'll introduce additional unknown variables into system. Many aftermarket parts are bad right out of box and needs to be tested to verify proper functionality (send to https://mrinjector.us for proper solenoid and flow testing). Leave all parts as-is and only do passive testing and measurements to positively identify issue from looking at numbers only. 

 

Could've tested original injectors by swapping #2 and #4 injectors. If problem goes to #4, then injector is issue. If problem stays on #2, it's something else and injectors are fine (put injectors back into original positions).

 

* I'm using traditional flow model for circuits here. There are other more-accurate models where electrons actually don't flow at all and it's alternating electric and magnetic fields that carry energy (not needed to understand how injectors work).

[gropula]

Thank you very much for detailed replies. I will do the testing. It's frustrating because it doesn't happen every time the fuel pump primes. Only sometimes so it's hard to recreate the problem as it only happens intermittently. From what I've gathered it's more likely to happen on the first prime after waiting for a while. If priming a few times in a row it likely won't happen again if it happened first time.

 

I bought these injectors because I've already had OEM ones cleaned and later one turned out leaking. It was a quick conclusion is faulty. It turned out that it actually wasn't a faulty injector but hey, they were cheap so it's not a big waste of money. When I diagnose this I can put back the original injectors. I hate doing that, emptying the tank, fuel leaking from the rail everywhere...

[gropula]

On 5/7/2023 at 2:31 AM, DannoXYZ said:

1. to determine if ECU is problem, unplug it from connector and turn on bike same as before, key ON/start=RUN.
Does no. 2 injector still spray unintentionally?

 

2. if NO spray with ECU disconnected, then you've got ECU problem

 

3. else if YES still spraying, then you've got short on #2 trigger wire to ground. Verify by leaving ECU disconnected and unplug all injector connectors. Key ON/start=RUN, Measure resistance between chassis-ground to each injector connector's ECU trigger wire.

  #1 pnk/blu ohms to chassis-ground = ???

  #2 red/yel ohms to chassis-ground = ???

  #3 pnk/grn ohms to chassis-ground = ???

  #4 pnk/blk ohms to chassis-ground = ???

 

Tried doing these tests:

1. If ECU is disconnected the pump wont prime. If the injector is disconnected it won't spray fuel. The injector gets grounded somehow on the yellow red wire. Took a few tries but got it on video. Connected the multimeter to red/yellow and battery negative with continuity test. It sounded the signal for a short period.

 

 

2. Inconclusive still, but my ECU looks like this 😞

 

 

3. When ecu is disconnected there is no continuity to ground from any of the wires. When ECU is connected the resistance is 63k Ohm on each wire.

 

I know a guy that has a crashed 2002. Should I ask him to yank out the ECU and the ignition key barrel and try to hook it up to my bike?

[Grum]

- Does your bike have the HISS system? If not you won't need the other Ignition key barrel 

- Why does your ECM look like its been hacked into then filled with silastic?

- Seems conclusive that only the ECM could be firing that injector, always at the same time for the same duration!

- Is there a part number on the other side of the ECM?

 

Good luck hope the crashed 2002 ECM works for you.

[gropula]

-It does have the HISS system. That's why I mentioned the key and barrel. Will need the tank cap and seat key barrel as well if it turns out to be the solution.

-I honestly don't know why would anyone hack into the ECU, was like this when I bought it. Noticed it long after buying, didn't occur to me that it might be a problem until recently.

-That's what I think as well. It's just that it's intermittent issue when priming, which is weird. Although it seems to be a persistent issue while riding as the sparkplug in no.2 cylinder is white. Not enough fuel while running. 

-There is, didn't take a picture to be 100% certain but I remember googling it, I think it was a stock 2002 HISS unit. 

[Grum]

23 minutes ago, gropula said:

-It does have the HISS system. That's why I mentioned the key and barrel. Will need the tank cap and seat key barrel as well if it turns out to be the solution.

-I honestly don't know why would anyone hack into the ECU, was like this when I bought it. Noticed it long after buying, didn't occur to me that it might be a problem until recently.

-That's what I think as well. It's just that it's intermittent issue when priming, which is weird. Although it seems to be a persistent issue while riding as the sparkplug in no.2 cylinder is white. Not enough fuel while running. 

-There is, didn't take a picture to be 100% certain but I remember googling it, I think it was a stock 2002 HISS unit. 

 

Just wonder if it's worth checking every Ground wire of the ECM has good continuity back to the battery Negative? Green and the Green/Pink wires.

- Are the 2 ECM plugs fully seated? Silly question maybe!

[gropula]

Yes, both ECU plugs click into place. I thought about checking the ground wires. Can do it in a few days.

I'll have to unstick this foam off the ECU where it was hacked open, just to see what's going on in there.

[gropula]

I borrowed the ECU and ignition barrel and plugged them in. The problem persists. Still spraying fuel sporadically. Rapid bike wideband reports idling at 10 AFR, subtracts more than 50% of fuel in 5, 10 and 20% columns. I guess I'll have to borrow the main harness now. I'll take mine out and inspect it for damage. Injector sub harness has been eliminated as the source of the issue, I've replaced it when I tried replacing the injectors. Put OEM back in. No difference, Chinese ones seem to work well.

[Grum]

I really think you need to Isolate the Rapid Bike System! Revert fully back to OEM and re-evaluate the situation!

 

In your fault situation, it doesn't matter Jack Shit! that the "Rapid Bike does not touch injectors or timing below 3,100 RPM.  It has nothing to do with start behavior or idle." !

The fact is the Rapid Bike does alter injector signals! 

If the unit was faulty, or being mis triggered by interference, poor power or ground connections etc It could well cause the problem.

 

You've isolated......The Injectors, The wire harness, and The ECM. Even a short to ground somewhere doesn't clock up with the spray duration being consistent with Ignition Switch On.

So the only thing left that interferes with the Injectors is the Rapid Bike System!

[Grum]

On 5/7/2023 at 1:11 AM, gropula said:

I think this was happening before I installed rapid bike. The mechanic that was servicing the bike said not to prime the pump without starting the bike. He said it can flood the cylinders. At that time I just believed what he said. Later I found out why he said that. Now I'm doubting his expertise because he didn't point it out as an issue. I can remove the rapid bike just to test but i doubt that's causing the issue.

Think the Mechanic is only guessing or spinning BS!

Injectors will not fire without the ECM seeing valid engine rotation info from both Cam Pulse Generator and Ignition Pulse Generator. Different story if you have leaky Injectors!

[gropula]

Yeah, I'll unplug the Rapid bike system and see if it will do the same. Fortunately I don't need to have the engine in running condition or the injectors plugged in, it can be detected on the injector connector.

[DannoXYZ]

On 5/10/2023 at 12:12 AM, gropula said:

 

Tried doing these tests:

1. If ECU is disconnected the pump wont prime. If the injector is disconnected it won't spray fuel. The injector gets grounded somehow on the yellow red wire. Took a few tries but got it on video. Connected the multimeter to red/yellow and battery negative with continuity test. It sounded the signal for a short period.

 

12 hours ago, gropula said:

Yeah, I'll unplug the Rapid bike system and see if it will do the same. Fortunately I don't need to have the engine in running condition or the injectors plugged in, it can be detected on the injector connector.

 

Looks like you found problem. Trigger wire to injector is getting grounded outside of ECU. So with just key ON and power to injector harness, that unintentional ground will fire injector.