briholt

Howdy,

I'm in the process of checking my ECU on my 91 22re. I get no codes, but I'm getting about 13 mpg.

For my truck at the ECU, I should get a standard voltage for the Vs - E2 of 0.5v to 2.5v with the ignition switch on.

I've checked this twice and measured 2.88v the first time and 3v the 2nd time, so technically, it's out of the specified range.

According to this troubleshooting guide, page 27/55 (which is labeled EG1-148), the "trouble" is if there is no voltage at these connections.

Does it mean anything that I'm getting a higher voltage for this connection?

FYI, I've checked the TPS and they are in specs. I didn't have time to get resistance at the VAFM.

Guidance appreciated. Cheers.

scope103

While I've never delved in quite so far as you are now, I would not worry about that 0.38 volt "out of spec." I would try pushing the VAF vane open to see if the voltage transitions smoothly to "about" 10v. (Ideally, you'd use a needle-type voltmeter so you can see if it is smooth, but good luck laying your lands on one. As a second choice, there exist digital voltmeters with an "analog bar" for that same purpose.)

briholt

So I got the MAF open and did some testing.

The volt readings at the ECU when the MAF plate is closed is just a bit higher than spec at 2.8 - 3 volts, but as scope13 mentioned, probably not a problem.

And when I open the plate to Wide open, it is also in spec, about 7 volts.

I did notice that when I opened the plate there was lots of bouncing around of these volts. Dead spots.

So, I did some resistance tests and while I had my analog volt meter connected, I took a video of its behavior:


The scale to read is at the top, in green. It's a bit hard to read but with the plate fully closed, it was reading in specs at about 150 ohms. When it was completely open, then too, was in specs at just above 200 ohms. In between, though, it seemed to bounce around 600 ohms.


My curiosity is whether the bouncing is a problem. I was pretty smooth opening and closing the plate. I think if I could time it correctly, I would get resistance levels below 20 (the bottom floor of acceptable specification) and perhaps down to zero ohms.

But maybe all that doesn't matter.

I know if I disconnect the VMAF, it doesn't run at all--starts but immediately dies.


The dog barking in the background is my boston terrier, buster, chasing his tail. poor guy was bored.

I am sort of hoping that the VMAF isn't working because The TPS and all of the ECU troubleshooting specs are okay, meaning to me at least the ECU is working right and is receiving data within specs.

I'll chip away at other EFI troubleshooting as I get the time, but at this point I know that TPS, O2 sensor, and probably MAF are working okay. I get no trouble codes but get about 13 mpg.

Gamefreakgc

How old is your O2 sensor? Just because it's within specs doesn't mean it's working right. If it's older that 90k miles then replace with a DENSO sensor.

I just finished reading the long-term fuel trims on my truck. It's on a 0.5V scale from 0 to 5. Mine came in at 3.5V. I checked it by setting the multimeter to voltage and jumped between VF1 and E1 on the diagnostic port (correct me anyone if I did this wrong?). No change is 2.5V, 5V is overly lean, 0V is overly rich.

This will at least tell you if the ECU is trying to compensate for some condition (such as a vacuum leak) by adding more fuel.

briholt

I replaced the o2 sensor less than 4k miles ago. Actually, I probably didn't need to change it, I was grasping then.

The rest of your post is food for thought. Thanks

briholt

When you say "jumped" do you mean that you short circuited them with a paper lip or similar tool, or do you mean you measured those two connector points?


I found this page and I think this is what you are describing:

http://personal.utulsa.edu/~nathan-b...42oxygense.pdf


http://alflash.com.ua/vf1.htm

Gamefreakgc

Yes, that's it. The top link isn't too helpful, since it's only for diagnosing check engine codes (only says it's 0V or 5V) but the chart from alflash.com is what I'm talking about. Looks like it's not in .5V like I suspected but close enough. It's not gonna solve your problem but at least it'll tell you what the ECU is trying to do.

Just put one prong of the voltmeter in VF1, the other on E1 with the engine running. It will give a voltage reading that doesn't move.

briholt

Okay, I just did that test with VF1 and E1. At idle and after warming up, the reading was mostly at 2.25 v.

Sometimes it would wiggle between that number and 3.55.

Interestingly, I used that link to the FSM in my earlier post and followed the decision tree down to the step where it instructed me to remove the PCV hose.

Assuming I pulled the correct hose (see pic below), two things happened: 1) the truck nearly stalled out, as if it got no fuel whatsoever. I figured it would die, but it never did, and 2) the volt reading jumped to 3.6 volts.

According to that decision tree, anything other than zero volts with the PCV valve disconnected means I should 'repair' something because it's too rich.

But it doesn't say what needs repairing.

My ignorance is bare: #1, why does disconnecting the PCV hose lead to the truck almost stalling, #2, why does that hose relate to the volts at the VF1-E1 terminals, and #3, what must I repair?

####
The PVC hose I disconnected was just the upper one labeled 'a' in this pic:

Gamefreakgc

Wait, stop!

That first link is for diagnosing a code 25 or 26 or other issues with a bad oxygen sensor. You'll notice that there's no outcome that says "OK", all lead to a "Repair/Replace" option. You did say there's no check engine lights, right? If not, then you shouldn't be doing those checks beyond the VF1 to E1 volt check.

briholt

Well actually it says that it applies to "normal code" as well.

scope103

You only get to THAT point if the needle fluctuated less than 6 times (in 10 seconds) on the O2 sensor test. What did you get in that test?

If the system is otherwise working, a VF1-E1 reading of 0v (no jumper) means the ECU thinks the engine is over-rich. Is it really? Detaching the PCV valve hose allows extra (unmetered) air into the induction system, and therefore should lean out the over rich situation. If the VF1-E1 voltage does rise with that extra air, it means you really do have an over-rich situation (and you should fix it.)

In your case, you had a "normal" fuel trim number, and when you pulled the PCV valve hose it just leaned it out so far it couldn't keep running.

Gamefreakgc

I think your poor gas milege isn't an ECU issue, the last reading you got says that it's pretty much right where it needs to be a 2.5V. I think something is robbing the power, causing you to have to give it more of the "skinny" pedal than it really needs.

Maybe just slowly but surely start checking things, mainly sensors for starters. I was in your same spot 3 years ago with my rig. Just start at one end of the engine and make your way across. If any sensor seems damaged and out of spec, replace it (within reason). Use the FSM as your guide.

I discovered a bad TPS, bad ECT, bad wires, bad plugs, three vacuum leaks, mal-adjusted valve, bad O2 sensor, and other broken parts by doing that. Not a single check engine light. Runs twice as good now as when I first got it. Feel free to post back as you go, lots of people can help out.

briholt

When I ran that test, I got about 1 fluctuation per second, so more than 6 in the 10 seconds. So, it's true I didn't need to go further in the flow chart, but I had some other nagging observations:

1. When I was testing the VF1-E1 (no jumper) it was usually at 2.25v but sometimes it would bounce to zero, sometimes up to 3.55. But overall it was mostly steady at 2.25v. The more it warmed up the more stable it was (though not perfectly stable).
   
   It was at that time that I saw in the FSM that a normal code would lead to the disconnecting of the PCV hose.
   .
2. Also, after I tested the o2 sensor by jumping Te1 and e1, I took out the jumper and retested the VF1-e1 circuit and the needle just bounced at the same rate as it did in the o2 test, about 1 bounce per second. So that seems very odd to me, that maybe there is short there.

That, I totally understand--thank you for clarifying what is happening there.

But here is a naive question: Assuming the engine is working as it should, no codes, no gas mileage problems, etc, if you were to disconnect the PCV hose while testing the VF1-e1 during idle, what should the volt reading be? My expectation (could be wrong) is that the volt reading should drop to zero and anything above zero indicates a rich condition. No? Because when I disconnect that hose the volt reading goes to 3.55volts, which suggests a rich condition.

If I am correct and I do have a rich condition, is it the PCV valve that is suspect or am I back to square one in trying to sus the location of the problem (which is where I'd be with Gamefreak's suggestion to just start systematically checking everything)?

Thank you Scope and Gamefreak for your help.

scope103

2.5v (approx) means the ECU thinks "everything is working fine." Less than 2.5v means "uh oh, when I go by the book, the engine is running RICH. I'll try to compensate." Pulling the PCV hose adds unmetered air leaning it out, the ecu detects that (through the O2 sensor), and the the fuel trim number goes UP from 2.5v. (which "means" the engine is stretching the injector open-time, to add more fuel, to overcome the lean condition.)

So you're back to Gamefreakgc's advice: systematically checking everything.

briholt

I guess what I don't understand is why should my volt reading be above zero when the PVC is disconnected. The manual says that if normal codes (and a few others) should disconnect the PVC you should read 0 volts. So why does mine jump to 3.55v?

Are you saying that is what a healthy truck would do? I guess I can't reconcile that with what the fsm says.

Again, thanks for your insights.

scope103

Are you referring to http://personal.utulsa.edu/~nathan-b...42oxygense.pdf p.213? I don't read it that way.

What I read is "Your fuel trim reads 0v, which means over-rich, but is that correct? If the O2 sensor is not working, it's all baloney. So lean out the mixture by pulling the PCV hose. IF the O2 sensor is working, that will lean out the mixture and raise the trim voltage. If the trim voltage stays at 0v, then the O2 sensor must be giving bad information, so replace it."

Gamefreakgc

I think we're hung up on something here that really is not an issue. The PCV valve has no impact on fuel trims... it just creates a massive vacuum leak that should be reflected on the O2 sensor readings. You register a move, which means O2 sensor is working fine. Awesome, you've got a functioning O2 sensor. But that's not going to solve your issue at hand.

briholt

Thanks for your help. I've been busy so this is the first day I've had to think about this. I started looking at the wires and noticed that there is a wire connected to the cold start injector, but then another wire taped up out of the way, as if it used to be THE wire:





Is it possible that this wire had a purpose, say in a california rig? If the wire connector was replaced at some point, I can't figure why the previous owner wouldn't have just removed this old one.

This has me thinking how to check whether my cold start injector ever shuts off. Could that explain the poor gas mileage? If it didn't shut off (ie, the cold start timer or the coolant gauge says it should keep putting in gas) wouldn't their be a check engine light?

scope103

I'm not real familiar with the 22re, so I can't help you with the unused connector (and, I don't recognize it, for what that's worth).

I doubt your CSI is stuck open, as I would expect that to flood the engine. The CSI Timer is easy to check; just unplug the connector from the CSI and see if it makes a difference.

BUT: you could have a leaky CSI; not enough to flood the engine, but enough to wreck the mileage. Since this would make you run rich (and you're got fuel trim saying you're fine) I doubt that's the case. But you can check that by removing the CSI, leaving it connected to fuel and signal. Start the fuel pump by jumpering. The CSI should not drip anything.

briholt

I just checked the ohms spec for the the cold start injector (3.3 ohm, which is between 2-4 according to the '93 manual) and the ohms for the cold start timer (80 ohms, which is appropriate for both temperature and is within range).

Is this the test where you put the cold start injector into a jar and add 12 volts to see it work? If not, I'm not sure what you mean by the quoted statement above. If it is that test, are you saying that if the CSI isn't working, then it should "not drip anything".


And, yeah, I sure wish I knew what that extra connector was about.