[SOLVED]Fixed a fuel leak, and now my RB20 won't idle hot

Discuss the RB20, RB25 and RB26 series engines.
kungfujoe
Posts: 11
Joined: Wed Feb 13, 2013 6:49 pm
Car: 1990 240sx Hatchback - RB20DET

Post

I just bought myself an S13 with an RB20 swapped into it (a considerable upgrade over my old KA). Overall, it runs great, but it has a few issues that I'm trying to diagnose.

The first issue I tackled was that it was having a b**** of a time trying to start cold. My remedy for the first few days was to let the fuel pump prime about 20 times, and then it would just barely start. Once it was started, however, everything was running smoothly.

After scouring the internet for awhile, I came across what I was hoping was the problem: The START signal wire (#43) on the ECU. So a buddy of mine with some experience in this area and myself started diagnosing the issue (splice into the 43rd wire, run a cable to the Starter Solenoid...etc.)

Well, we put the wire to the solenoid and tried to start it. Car fired right up. Which is when my buddy noticed another issue. The engine had developed a pretty bad fuel leak in one of the fuel hoses. $1.50, Phillips head, and 10 minutes later, the problem was resolved.

The plan was that I would try to run the wire when the engine was cold (aka, the next morning) and see if that made a difference. When I went out the next morning, WITHOUT the wire going from 43 to the starter, the car fired up again pretty quickly.

So, I've developed the theory that the reason my car wasn't starting was that the fuel pump wasn't able to generate any fuel pressure to speak of (or something of that nature).

That morning, on my drive to work, however, my car started to die when I'd clutch in, after driving for about 10 minutes. Did it on the way home too.

Here's all the information I could gather. NOTE: I still have a KA tach, so my readings are not accurate...
Leave the car sitting for an hour.
Start the car:
- The car runs without dying, but the idle bounces up to "3500" RPM, then slides down to a bogging point around "1000" RPM, at which point it shoots back up to "3500" RPM. This repeats.
Drive the car around the block:
- While on the throttle, everything seems to be running fine
- When in gear, everything seems to be running fine
- With the clutch in, the idle drops down below "1000" RPM and it slowly bogs down and dies
- Thanks to my Boost Gauge, With the car in gear and me on the throttle gently, the car maintains between 18 and 20 pounds of vacuum with VERY little fluctuation
Back in the Driveway
- The car continues the bouncing between "3500" RPM and about "800" RPM, dying about 1/4 of the time.
- During this bouncing, The vacuum also fluctuates between 20 pounds and about 4. Obviously, when it dies, it drops to 0.


So, I'm thinking vacuum leak I guess. But I dunno. I don't think the addition of fuel pressure would have some negative effect on another part of the engine, but then again, I'm not great at this kind of thing.

Thoughts?
Last edited by kungfujoe on Mon Mar 11, 2013 12:09 pm, edited 1 time in total.


User avatar
Razi
Posts: 28373
Joined: Sun Dec 31, 2006 9:52 am
Car: Moo

Post

Check everywhere for a vacuum leak, and I guess you could try cleaning the IACV and messing with the adjustment.
Make sure the spliced wires are back to normal and covered.

TerminalVelocity
Posts: 552
Joined: Sun Feb 04, 2007 7:44 am
Car: 1996 240sx

Post

Definitely check for vacuum leaks.
Do you have a recirculating or atmoshperic blow off valve? Is your maf in the factory location?

kungfujoe
Posts: 11
Joined: Wed Feb 13, 2013 6:49 pm
Car: 1990 240sx Hatchback - RB20DET

Post

Yeah, totally a vacuum leak. I put an ear to my vacuum hoses, and found some hissing noises near the BOV.

I'm gonna do more diagnosis later (gotta find the hose that's leaking; there's a couple near the BOV), but I'm pretty damn sure that's the issue.

Thanks for the help guys :)

User avatar
Carl H
Posts: 5985
Joined: Mon Aug 04, 2003 4:09 am
Car: 1995 Nissan 240SX SE RB30DET

Post

might want to pull the pump assembly and make sure it has an upgraded fuel pump in there while you're at it.

kungfujoe
Posts: 11
Joined: Wed Feb 13, 2013 6:49 pm
Car: 1990 240sx Hatchback - RB20DET

Post

The fuel pump has been under scrutiny as well. I might just have to pull it this weekend to put the mystery to rest.

Anyway, with the car running, I couldn't tell which hose was leaking, so I pulled all the air related ones of the car in that general area...Do you guys have any idea what I can do now that they are off the car to test them?

I'm halfway inclined to pressurize them with an air compressor. I already tried using water from the sink and plugging one end, but that didn't reveal anything.

On the other hand, maybe I'm going about this the wrong way. I could always put the hoses back on the car if need be.
How would you guys recommend I go about finding the problem hose?

TerminalVelocity
Posts: 552
Joined: Sun Feb 04, 2007 7:44 am
Car: 1996 240sx

Post

With the hoses off the car, plug one end, pressurize with air and dunking in water seems promising. Keep the pressure low. Sometimes though a thorough visual inspection will be all you need to find a leak big enough to give you problems.
Also consider that your hoses may be fine and possibly your leak is at the connections between lines.

To test on the car, Google vacuum leak smoke machine. I've been wanting to try that for awhile.

Edit: improved clarity

User avatar
Carl H
Posts: 5985
Joined: Mon Aug 04, 2003 4:09 am
Car: 1995 Nissan 240SX SE RB30DET

Post

smoke machines do wonders for finding blatantly obvious leaks and some not so obvious leaks but with a turbo car some of them wont show till you have pressure in the system.

best to test 2x whatever pressure you normally run, ie run 14psi, test to 28.

mechanicalmoron
Posts: 790
Joined: Fri Apr 29, 2011 1:04 am

Post

As for the start problem, when they're cold hoses shrink and fittings loosen, once the system's pressurized and warm it will swell a bit. I've had gas leaks that only happened right after starting, until I tightened them up. That's why you had a hard time starting in the cold with the leak. Your ecu was fine and you never needed the wire.

As for your idle problems, don't look at me.

kungfujoe
Posts: 11
Joined: Wed Feb 13, 2013 6:49 pm
Car: 1990 240sx Hatchback - RB20DET

Post

Ok, sorry I've let this stale out a bit (I've been working on other issues with the car). This is what I've figured out:

There is a...well...I don't know what you'd call it. It's an air spout thingy. It's on the top-back of the intake manifold, pointed toward the rear of the car. I have/had a hose connected to it that runs to my boost gauge.

This, however, seems to be causing me all sorts of problems though. When I disconnected the boost gauge hose from it, my car seems to run very well. When I plug the...(damnit, I wish I knew what the thing is called)...spout thing, my engine bogs and dies shortly after.

Now, as far as I know, the important point to remember is that there isn't much difference between me hooking a boost gauge to it, or me plugging it with my finger: air isn't getting past one way or the other, and this seems to make my engine die, or, at the very least, idle for s***.

If I've heard correctly, this spout thing is the default mounting point for a boost gauge, which leads me to believe that I've got air issues elsewhere.

Anyway, I've officially stopped calling this a "vacuum leak", and am instead calling it "air issues", because frankly, something is way off.

In true Seattle fashion, it's raining cats and dogs, so I'll get out there tomorrow and take some pictures. I'm really hopeful that you guys can help me on this one, because I'm at a loss. It also doesn't help that I can't correctly identify the names of things by looking at this engine, so a picture is worth 1000 words I guess.

Thanks for your time guys.

User avatar
Carl H
Posts: 5985
Joined: Mon Aug 04, 2003 4:09 am
Car: 1995 Nissan 240SX SE RB30DET

Post

need to boost leak test it with everything on the car connected.

mechanicalmoron
Posts: 790
Joined: Fri Apr 29, 2011 1:04 am

Post

Have you tried just draining all power from the system, to reset the ECU and put it in self-learn mode? And setting the idle speed? (if rb uses an iacv, i dunno anything about it)

kungfujoe
Posts: 11
Joined: Wed Feb 13, 2013 6:49 pm
Car: 1990 240sx Hatchback - RB20DET

Post

Carl H wrote:need to boost leak test it with everything on the car connected.
Everything's connected, but I don't know how to test for boost leaks. Got any nice guides or w/e?
mechanicalmoron wrote:Have you tried just draining all power from the system, to reset the ECU and put it in self-learn mode? And setting the idle speed? (if rb uses an iacv, i dunno anything about it)
I don't really know what you mean. Drain all power? Please elaborate.

Anyway, here are some photos of my engine. Hopefully the color coordination that I'm attempting will make sense. Does all of this look correct to you guys?

Red: The hose connected to my aptly named "spout thingy" leading into the cabin to my Boost Gauge
Blue: A recirculating hose that we are unsure about.

While diagnosing this issue, we disconnected the blue hose and capped both of the points it connected. This did not seem to help.

Image
Image
Image
Image

If all of this looks correct, I'm up for any other suggestions. TerminalVelocity was asking about the MAF and my BOV. I'll post pictures of those setups as well if all of the above looks good to you guys.

Thanks for your help,
KFJ

mechanicalmoron
Posts: 790
Joined: Fri Apr 29, 2011 1:04 am

Post

kungfujoe wrote:
mechanicalmoron wrote:Have you tried just draining all power from the system, to reset the ECU and put it in self-learn mode? And setting the idle speed? (if rb uses an iacv, i dunno anything about it)
I don't really know what you mean. Drain all power? Please elaborate.
My idea is that the ECU could have become "used" to your previous fuel problem, and been running rich to compensate.

Assuming it's the same as a KA ecu in that everything is stored in RAM (which erases when it loses power) you should be able to either unplug the ECU and wait a few minutes, or disconnect the negative battery terminal and press the brake pedal and then wait a few minutes, to wipe the ECU's learned behaviors, so that it starts off re-mapping how your motor is worn and how it behaves, as opposed to a stock RB.

kungfujoe
Posts: 11
Joined: Wed Feb 13, 2013 6:49 pm
Car: 1990 240sx Hatchback - RB20DET

Post

Tried resetting the ECU. No change.

As for the fuel pump...well...I brought her up to redline this morning on the way to work, and whatever pump is back there sent enough fuel to allow it, so I'm thinking it isn't the stock fuel pump.

I'll play with the IACV next.

Thanks again for your help guys.

User avatar
Carl H
Posts: 5985
Joined: Mon Aug 04, 2003 4:09 am
Car: 1995 Nissan 240SX SE RB30DET

Post

have you tried resetting your tps?

User avatar
Carl H
Posts: 5985
Joined: Mon Aug 04, 2003 4:09 am
Car: 1995 Nissan 240SX SE RB30DET

Post

your blue hose also needs to be capped on the throttle body side and capped on the plenum as well if you're not using the port.

kungfujoe
Posts: 11
Joined: Wed Feb 13, 2013 6:49 pm
Car: 1990 240sx Hatchback - RB20DET

Post

Carl H wrote:your blue hose also needs to be capped on the throttle body side and capped on the plenum as well if you're not using the port.
Ok, so the "blue" hose is a normal hose. It runs from the plenum to the throttle body.

Just to clarify, you are indicating that I should remove that hose and cap the ends that I remove it from (one on the plenum, and one on the throttle body)?

We had tried that at one point and it didn't seem to make things better, but if that is "more correct" than how I currently have it, I will put it back that way.

==============================================================================================

I've been hearing and reading a lot about IACV failure, and I think that it is the next item on my list for scrutiny. Had a nice read about reading Error codes off of my ECU (Yay! I have an LED!), so I'll see if the IACV pops up with an error code. Either way, All eyes seem to be on it as the next logical place to debug. For the sake of potential future visitors, I'll post any/all error codes that pop on my car.

Thanks for the help guys.

kungfujoe
Posts: 11
Joined: Wed Feb 13, 2013 6:49 pm
Car: 1990 240sx Hatchback - RB20DET

Post

Sorry for the long delay, but I did finally manage to get the problem fixed yesterday.

Yesterday, I topped off my radiator (again) despite the fact that it wasn't leaking. Somebody smarter than me told me that I probably need to bleed it, which, incidentally, can cause strange idle issues (when there is air on the temp sensor, it reads hot, and when fluid washes over the sensor, it reads cooler, all of which confuses your ECU because of the constant changing).

To bleed your radiator:
Fill it almost to the top.
Massage the top radiator hose to get it to sort of suction in some radiator fluid: note, this can take a few (say 30) squeezes
Massage the bottom hose too if you can (I couldn't because I have fancy-dancy flex hosing)
Locate and remove the 10mm radiator bleed bolt: It is located on the top-front of the intake plenum (hard to miss, there isn't much else there)
Once removed, gently massage that top radiator hose some more until antifreeze starts to leak out of the bleed port. Antifreeze is cheap. Feel free to make sure you get all of the air out of the system
Finally, put the bolt back on, and top off your antifreeze.
Bonus: If you've got one, add some water to your overflow while you're thinking about it.

Also, I cleaned my AAC (I think this is what actually fixed my problem).
The article that helped me do it is here: http://www.skylinesaustralia.com/forums ... dle-reset/
Granted, that article is specific to an RB25, but honestly they were similar. The task is pretty simple, and pretty cheap. If you are having ANY idle issues, I recommend trying this.

Now my car idles like a dream. Thanks for your help again guys


Return to “RB20DET / RB25DET / RB26DETT Forum”