[MarbleMaker]

Hello Brain Trust.

I need help in fixing an idling issue with my 1991 Suzuki Vitara 2 door. It was purchased with a G16B conversion.

Back story.
It has been running fine for the last few months and had done a few smaller off-road tracks with no issue.
First noticed very poor idling after airing down while heading into a 4x4 track. It seem to get better when I was driving in the higher rpm limit, thought it might be some poor fuel and it would get better when I topped up later. But ended up getting worse as the day went on.

Current symptoms as of today.

Black smoke from the exhaust
Rough idling.
Spark plugs carbon covered.

What have I done?
After inspecting the spark plugs I noticed they were what seems to be carbon fouled, old and ratty so replaced them with new ones and gapped. I'm new to engines so this is a learning curve, I was told spark plugs would lead me in the right direction. the following is a list of what I have done in no particular order, just one thing lead to the other and it generated a list haha.

list of things done.

Fuel

I drained the fuel out and put some fresh petrol in.
changed the fuel filter.
New Injectors.

Air intake.

Cleaned the air filter.
Changed throttle body and Air Flow Meter from my donor car = I did notice the rough idle was not as bad after changing the throttle body as before. Fitted the original AFM and kept the throttle body from the donor installed.
Changed PCV valve with donor cars = no change. so ultrasonic cleaned the old one and refitted.

Electrical.

New pre-gapped spark plugs.
New Ignition and spark leads.
New distributor Cap and Rotor button.

last week then remembered that I had put some radiator sealing stuff in the coolant a few days before it started to play up. So I then focused on the cooling today.

Flushed the radiator.
New thermostat. did notice a lot of rust in the cavity between the thermostat and water pump picture is attached.
New radiator cap.

At this point, I'm at a loss! I feel I am going down a rabbit hole and not getting anywhere closer I am hoping someone would have a better idea of where I should be looking.

The next things on my list are:

The Oxygen sensor as it is still pumping out black smoke I would assume the sensor could be covered in carbon and stuffing up the air/fuel mixture.
I did notice there seem to be an exhaust leak near the oxygen sensor today. video attached

timing
I negated the timing from the start as I thought that would be unlikely as it was perfectly fine one minute then crap the next. But this will be my next thing to check. just dreading taking this part on.

https://youtube.com/shorts/Q7RTOZ02STI?feature=share
https://youtube.com/shorts/eZIBK2q0iio?feature=share
https://youtu.be/0ybsi45y_XA
https://youtube.com/shorts/G93uu3R_Wsg?feature=share

Sorry, I could not get the videos to embed into this post.

Looking forward to hearing people's thoughts on how I should proceed.

[Gwagensteve]

OK, first things first.

As it's a conversion, does this car have a check engine light? If not, start by getting that working and learn how to put the CE light into diagnostic mode. It might tell you what is wrong.
Does it run a stock EFI tank with internal pump? Does it run the stock G16B fuel pressure regulator? Have you confirmed fuel pressure at the rail?

Next: The oxygen sensor has no effect at all on idle fuelling. The computer only references the oxygen sensor at light, steady throttle. You can unplug it and drive the car, you won't notice any difference. (it will put the check engine light on though)

I can see you're throwing a lot of parts at this problem, but that's a very expensive and slow way of trying to fix it.

A common G16B fault is a flogged out keyway on the crank, this will result in the timing retarding. lock the timing using a jumper on the diagnostic plug as per the factory service manual and check the timing with a timing light. if it's way out, you have your answer.

Air locks in the IAC valve will cause the idle to be erratic but it won't run rich or foul plugs. it may have been part of the problem and why it seemed to improve a little once you swapped throttle bodies over.

Tell me more about what lead you to put radiator stop leak in the car? where was the leak? is it possible the car ever overheated or the coolant ran low? G16B's are very prone to head cracks and head gasket failure. I also have a niggling concern about the way the coolant lines might have been hooked up after the conversion. G16B's are very sensitive to coolant flow. It's possible you have a head crack of failing head gasket if the car is using coolant and you have fouled plugs and a rough idle.

[MarbleMaker]

Thank you for your reply.

The check engine light does not seem to be working as this would have been noticed when I had disconnected the oxygen sensor. I will investigate this further when I have a full sunny day free and report back.

The car does not seem to be running a standard EFI tank as the fuel pump is external and connected to a switch in the driver's seat. It is activated by a switch and you can hear it pumping when it is in accessories mode. I did notice when changing the fuel filter that I could not find how the fuel returns to the tank, I would have thought it returns via one of the metal pipes attached to the chassis rail possibly the top? from memory, it ends above the fuel tank and is connected to nothing. I will check this in the next few days as it's an easy check.

I am not sure how to check if it is using a stock regulator I will take some pictures of it and compare it to one on the donor car but other than that I'm clueless.

I had purchased the wrong fuel pressure gauge this afternoon it seems to only go to 16psi for a $100 I thought it would be a bit better than that :/. But just to confirm that I should be connecting it to where the nut is placed on the casting (Closest to the first injector) that holds the fuel injectors? Is this the rail?

Throwing parts at it is indeed a slow way to fix it, but it also gives me a bit of experience in replacing them as well as peace of mind in knowing that they should not fail on me any time soon when out bush. They were reasonably cheap but I do now agree that it is starting to add up in cost.

Quote:

A common G16B fault is a flogged out keyway on the crank, this will result in the timing retarding. lock the timing using a jumper on the diagnostic plug as per the factory service manual and check the timing with a timing light. if it's way out, you have your answer.

This is what I am hoping not to see haha but yes I will try to borrow the Timing gun from work and check this out.

In regards to the Leak in the radiator.
That was a me thing. It involved a pressure washer, no grill in front of the radiator, and me tripping... It was a micro leak in the front of the radiator and I ended up checking the coolant every day while driving it to and from work (30min highway driving) and only topped up the Coolant reservoir once in 2ish weeks. I did a compression test as one of the first things. From memory, it was something like 1=195 2=215 2=215 4=185 this was done on a cold engine as I could not get it to idle long enough to warm up. Would this prove the head is free from cracks and the gasket is in good shape?
As to the history, I cannot be certain is there any tell tails I could use to get a guess?

I can appreciate this is a broad question but do you have any suggestions on how to approach the Check engine light issue? I was just going to find the ECU and see if I can back probe for the dash light and then deal with connecting it to the dash after it's fixed. not sure if this is a thing without researching.

Thanks

[Gwagensteve]

As is almost always the case, it looks like someone has got the G16B running and decided that was good enough. From what you describe, the conversion hasn't actually been completed at all, and whilst each of these things might not, in themselves, be the problem, they will all be making it harder for you to solve your issue. Funny, when people ask about converting a carby vitara to G16B I reply with "it's not worth it to end up with something suzuki already built" and this is the proof.

First a question - what version of the G16B do you have? does it have a distributor or coil packs?

The check engine light will light with the key in the IGN position and should go out shortly after the engine has cranked. If there is a sensor problem, it is recorded by a trouble code and displayed by the check engine light. The coil pack cars have more sophisticated diagnostics compared to the distributor cars, but OBD is hugely valuable. From what you're describing, there will be no way the VSS was hooked up so the car will throw a trouble code after about 30 seconds on the move. That's annoying and will need resolution but for the current purposes it won't matter, it's just another trouble code you will read along with any others.

The fuel system is dangerous and needs to be corrected with urgency. The fuel pump must shut off if the engine isn't rotating or else there is a huge fire risk in case of an accident. Even the earlier mechanical EFI cars used tachometric relays that were triggered by the tach signal to run the pump only when the engine was turning over. The factory ECU has a fuel pump wire which runs to the fuel pump relay and the ECU controls when the pump runs. It primes for 3 seconds on IGN and then only tuns again once the engine is cranking.

I don't know what pump has been fitted, typically Bosch 044 pumps are used in these external applications, which are far too large for the application and I have a concern that they overwhelm the regulator and/or the return line. even a small crimp in the return will be enough to boost fuel pressure and cause the issue you are having.

G16B's do not run returnless. The fuel pressure regulator is on the end of the rail, above the thermostat housing if you have a distributor G16B. Fuel lines run between the chassis and the engine on the RHS Rear of the engine beside the starter motor. Feed is 8mm, return is 6mm. There is a large plug on the very end of the fuel rail, facing forwards above the thermostat. That's where you connect the fuel pressure gauge as per the factory service manual.

There isn't a problem with a 16psi fuel pressure gauge, it's just for a carburettor application, not EFI.

Factory fuel pressure spec is 35.6-42.7psi with the engine idling.

No, a compression test won't necessarily prove a cracked head/gasket. as pointed out G16B's are very sensitive to coolant level/flow. The generated steam pockets in the head if the coolant flow does not mimic factory. I've just compared coolant flow between the carb and G16B vitaras, and it should not be an issue so long as the heater core is still in the system and operational - dies the heater work?
Never, ever put stop leak in an engine you want to keep. The whole cooling system is now coated in that junk.

As for setting up the CE light, this is going to be something of a challenge because I guarantee the wire has just been snipped off at the ECU (along with goodness knows how many others) If you have a stock ECU (and it matches the engine) the check engine light wire is violet with a yellow stripe. that goes to the bulb, which is a generic dash bulb. The other side of the bulb is IGN+ (typically black with a white stripe) you can wire in your own low wattage dash bulb (use a spare brake warning light bulb or similar from another cluster) Of course, the correct way to do it is to use the existing CE light spot in the instrument cluster, but maybe worry about that when you're digging in there to fit a VSS equipped speedo drive. For now, if can just flat float free to use as a diagnostic tool.

Speaking of which you'll need to find the diagnostic coupler. This is a white plug that isn't plugged into anything and it should have a black rubber cap over it. In a stock vitara it lives on the driver's side near the under bonnet fuse block. this is the tool for locking timing and putting the CE light into diagnostic mode.

A note - the ECU doesn't power anything - the power is always on the other side of the load. The ECU switches the NEGATIVE side of the circuit. Don't get this wrong, you will damage the ECU. (same for the fuel pump relay)

[Gwagensteve]

Any progress on this?

[MarbleMaker]

Gwagensteve wrote:

Any progress on this?

Sorry for the extremely late reply

As this happens and life gets in the way, attention gets dragged away from toys and onto other things.

But now I'm back on to this.

I did try your suggestion but I did not reply with my results. After fixing a leak in the exhaust and replacing some hoses and cleaning the intake manifold and replacing the PCV it did seem to do something its not hunting as much now but still hunting enough to be noticeable.

- It is indeed a G16b fuel-injected distributor engine. (not coil pack)

- Check engine light. Well, there is none :/ not that it's not lighting up! but after pulling the whole dash apart to check if the bulb is blown, I only find that there is no actual socket or slot actually there. there is traces on the flexible PCB behind the dial cluster but there is no provisions to actually accept a light. from memory there is no wires populated on the connector otherwise im sure i would have attached a light probe to it. i forgot what the back of the ecu connector is like maybe i could splice a bulb into there?

- Fuel system. I agree this will be sorted out once I get the engine running smoothly.

- I will confirm the fuel pump again but from memory, I did as suggested and checked the pressure at the end of the rail and it was showing consistent pressure and within spec. I will trace the return line when i can to look for any dings or crimps.

- is there a way to get an engine code without the light? can I splice into somewhere?

I was thinking I still have a spare zook laying around I could swap components to and from to promote the engine light showing up on that one. but this is not desired for obvious reasons.