240 no start - ICU vacuum?

[jkior]

I've been struggling with a non-starting 88 240 (B230f + LH2.2 Volvo/Chrysler ignition) for a while now. I've either tested or replaced (or swapped in with a known working part) almost everything involved in the ignition system, with no luck. Anyway, one of the manual's ignition system checks is the vacuum line running from the intake manifold (past the throttle) to the forward ICU. This line must be intact, apparently, for the ingnition to fire the injectors and fuel pumps (my problem). But when I put my finger over the port in the intake manifold and have a helper crank the starter, there is no discernible vacuum whatsoever. Throttle closed.

Is this supposed to be this way? I'm getting a little frustrated with this problem.

Thanks,
mike

 

[n xntrx volvo]

please list everything you've replaced/ tested w/results. could be something simple.

 

[Three Fat Tigers]

↑↑Ditto ↑↑

This line must be intact, apparently, for the ingnition to fire the injectors and fuel pumps (my problem)

This vacuum line does not need to be connected to the ICU (ignition control unit) in order to start or operate the engine, nor does the ignition system fire the injectors or operate the fuel pump relay, that is the sole domain of the LH 2.2 ECU. An rpm signal is supplied to the LH ECU via a wire connected between the ignition coil terminal #1 and the ECU's connector terminal #1, make sure that is in good order.

The test for the vacuum line is to run the engine, disconnect the hose at the ICU and at about 1500 rpm verify there is a strong vacuum present. If not your vacuum hose or throttle body is suspect. If there is no vacuum the engine will not run at it's best but it will still run.

 

[jkior]

Darn. I was hoping. Anyway, starting with the simple stuff:

battery is charged (terminals cleaned and greased)
fuel in the tank
spark is strong
timing belt was replaced very recently (timing is confirmed good)
fuel/main relay is confirmed good (swap with known good)
ecu is almost certainly good (swap with known good)
ecu grounds are good (I'm reasonably certain. the behavior of the problem seems like ecu grounds, but I've checked and rechecked the ground point on the intake manifold.)
fuel pumps are certainly good
fuel filter was replaced (ran fine after)
air filter was replaced
AMM was swapped with an almost new one, known good
checked function of the ignition switch by the manual
icu is good (swap with known good)
replaced intake and camshaft cover gaskets (incidental, but its's been part of this project) as well as injector o-rings
replaced fuse panel and all fuses
checked underhood 25A fuse

And now, when I'm re-testing stuff this morning before I open my big mouth, it seems that the ecu grounds are NOT good. Wtf? They all tested perfect before. . .

The manual (Bentley) says to check terminals 5, 11, 19, and 25 continuous to ground. (and continuity between 11 and 19, which does exist at .1 ohms) Terminals 5, 11, and 19 are testing absolutely zero continuity. Where do these ground? If it's that one point on the intake manifold, I've checked and cleaned that, unfortunately. Does anybody have a clear idea of where those grounds go? 5 and 19 seem to merge into one large green wire, that looks remarkably like the oxygen sensor wire that comes out of the firewall. Is this true? I wouldn't think so, but. . .

Help!

 

[Three Fat Tigers]

5 & 19 are related to the O? sensor. 5 is the shield for the O? wire. 11 is a critical ground for the ECU, if it doesn't show ground it wont run. The ground points for the LH are usually the bolts that attach the fuel rail to the intake manifold, and or one of the bolts that attach the throttle spool to the intake.

So is the car not starting for a lack of fuel or a lack of spark or both?

 

[jkior]

5 & 19 are related to the O? sensor. 5 is the shield for the O? wire. 11 is a critical ground for the ECU, if it doesn't show ground it wont run.

I was just down there again, and it turns out that it's 5 and 20 that merge into the oxygen sensor wire. Terminals 11 and 19 lead to two black wires which, in both of my LH2.2 systems, terminate in what seems to be a simple jumper a few inches away from the ecu connector. What's this? The manual says these terminals are supposed to be continuous to ground; is this test supposed to be performed with the ecu plugged in? That will be a little difficult, if so.

The ground points for the LH are usually the bolts that attach the fuel rail to the intake manifold, and or one of the bolts that attach the throttle spool to the intake.

Yeah. There's two wires that attach to the fuel rail. One of them is actually four wires that merge at the crimp connector, while the other is a single black wire. Do you know what that wire leads to? The four-wire one is continuous to ECU terminal 25.

So is the car not starting for a lack of fuel or a lack of spark or both?

The pumps are not firing, and neither are the injectors. The spark is fine; I even checked it while turning the engine by hand - strong blue arc.

 

[jkior]

*bump*

I'd love some help, if anyone has any ideas. . .

 

[jkior]

What exactly would be the ignition/injection behavior if the ecu grounds were bad or non-existant? Would I have spark but no pumps or injectors or main relay (since the main relay grounds come from the ecu)?

Thanks,
~mike

 

[Three Fat Tigers]

The terminal 11, as far as I know, must ground or the ECU will not function. The ECU is not supposed to be plugged in for this test.

The single black wire on the intake, well I did just recently see a picture and a description for that exact wire, but I forgot what it was and where I saw it. Still looking...

I'm going to head out to the Volvo shop here shortly and see if I can find a green book for an '87-'88 240. Meanwhile things to check:

→Check the main LH fuse, 25 amp blade type, located out next to the battery on the fender. Then verify that there is 12V+ reaching the terminal #30 of the fuel pump and the LH injection relays under the dash, there should be power there at all times.

→Make sure the signal wire from the coil to terminal #1 is making good continuity to terminal #1 (-) of the ignition coil.

→Try starting with the fuel pump relay bypassed by jumpering the #5 & #7 fuses together.

 

[Three Fat Tigers]

What exactly would be the ignition/injection behavior if the ecu grounds were bad or non-existant?

The ignition system would still work normally, the ECU would not turn on the fuel pump relay or fire the injectors.

 

[Three Fat Tigers]

→Check the main LH fuse, 25 amp blade type, located out next to the battery on the fender. Then verify that there is 12V+ reaching the terminal #30 of the fuel pump and the LH injection relays under the dash, there should be power there at all times.

Gee I'm making all kinds of mistakes today your car only has a single relay, aka fuel pump relay, aka Relay Fuel Injection. Internally it contains two indipendent relays, one for the fuel injectors and one for the fuel pumps. Both are controlled indipendently by the ECU. Still verify there is power at the #30 terminal.

→Make sure the signal wire from the coil to terminal #1 is making good continuity to terminal #1 (-) of the ignition coil.

Wrong again The rpm signal that the ECU sees at terminal #1 comes from terminal #8 of the ICU, aka igntion control unit.

→Try starting with the fuel pump relay bypassed by jumpering the #5 & #7 fuses together.

Wrong again fuse #4 & #6 are jumpered to bypass the fuel pump relay on '85 or newer 240s.

 

[Three Fat Tigers]

So far I have confirmed: terminal 11 is supposed to be grounded, I believe at the intake manifold, it is also looped to terminal #19. Terminal #5 is grounded, again I believe at the intake manifold, and is also the shield for the O? sensor wire (#20). The ICU also grounds to one of the bolts for the fuel rail.

If the ECU is missing the RPM signal from the ICU it wont fire the injectors . When you first turn on the ignition you should still be able to hear the fuel pumps run for a second.

 

[Three Fat Tigers]

Internally it contains two independent relays, one for the fuel injectors and one for the fuel pumps.

*cough* umm, wrong. One relay supplies power to the fuel pumps + fuel injectors + IAC valve + heater element inside the O? sensor. The other relay provides a power feed back to the ECU & power to the MAF sensor.

Sorry for the mistakes, I guess maybe I didn't get enough sleep last night.

 

[jkior]

First, I have to say - you are really, really awesome. I much appreciate the amount of time you've put into relying.

I've been checking a few more things, and. . .

The terminal 11, as far as I know, must ground or the ECU will not function. The ECU is not supposed to be plugged in for this test.

Huh. I have the wiring harness apart at the ecu connector and 11 really does seem, as far as I can tell, to do a simple loop around to terminal 19. Maybe the linking to ground is ECU internal?

→Check the main LH fuse, 25 amp blade type, located out next to the battery on the fender. Then verify that there is 12V+ reaching the terminal #30 of the fuel pump and the LH injection relays under the dash, there should be power there at all times.

That fuse is good (the contacts were corroded, but I cleaned them up, and replaced the fuse), and yes, there is battery at terminal 30 (of the single main relay, as you mention later.)

→Make sure the signal wire from the coil to terminal #1 is making good continuity to terminal #1 (-) of the ignition coil.

I'm sorry, I don't understand. . .

→Try starting with the fuel pump relay bypassed by jumpering the #5 & #7 fuses together.

Yeah. The pumps run when the fuses are jumpered, but the car doesn't start (no injector firing?) Without the fuses jumpered, the pumps do not run when the starter is cranked. BTW, I've swapped in that same relay from two other known running 240's, so I'm pretty confidant it's good.

The ignition system would still work normally, the ECU would not turn on the fuel pump relay or fire the injectors.

Seems to describe my problem.

Wrong again The rpm signal that the ECU sees at terminal #1 comes from terminal #8 of the ICU, aka igntion control unit.

How would I check this? The hall effect sensor that was in this car was definately on the verge of dying (bent trigger wheel, I have no clue how long) so I put in one from the j-yard, and it definately works. (turned the engine over by hand and watched a removed spark plug, which does fire at the proper time. . .TDC, right?)

I'm gonna go check terminal #5 again now, I don't think it was grounded before. . .

 

[jkior]

Nope. No continuity to ground.

Well, *ahem*. I think this might just be my fault. At one point, I thought the error was a short in the injectors wiring harness. So, I took the intake manifold off, opened up the wiring harness at the junction between the four injector leads, and found . . . nothing. It was a measurement error. So I re-wrapped the harness and put it all back together, but now I've noticed that there's a ground wire (that single black one, that grounds on the forward fuel rail bolt on my other '87 245) that is missing. Uh-oh. I guess I must have wrapped it into the harness and lost it then, since it certainly isn't hanging out loose anywhere. So my next step will be to take the intake manifold off again and check that.

Even so, I don't honestly think that this is probably the cause of my problem, since it is most likely the oxygen sensor ground (#5). The ECU ground at terminal #25 is fine, grounding at the rearward fuel rail bolt, and the other two grounds (#11 and #19), can't possibly be connected, as they - as far as I can tell - only loop around to each other.

BTW, do you know if the oxygen sensor is the only thing grounding into this missing wire? If so, it would be considerably easier for me to just run another wire than to rip into the harness again.

magnas gratias,
~mike

 

[Three Fat Tigers]

The terminal 11, as far as I know, must ground or the ECU will not function. The ECU is not supposed to be plugged in for this test.

Huh. I have the wiring harness apart at the ecu connector and 11 really does seem, as far as I can tell, to do a simple loop around to terminal 19. Maybe the linking to ground is ECU internal?

The factory diagram I looked at today specifically showed a black wire running from #11 to ground and also another black wire running from #11 to a small connector where a loop of wire connected it to a wire (also black) coming from terminal #19.

→Make sure the signal wire from the coil to terminal #1 is making good continuity to terminal #1 (-) of the ignition coil.

I'm sorry, I don't understand. . .

This I latter corrected (see the next item ↓as this information was inaccurate, this was related to the RPM signal that the ECU requires to operate, however there is no direct connection between the ignition coil and the ECU on your car. Sorry for the confusion.

Wrong again The rpm signal that the ECU sees at terminal #1 comes from terminal #8 of the ICU, aka ignition control unit.

How would I check this?

The Bentley manual specifies checking between the ECUs Terminal #1 and ground, with ignition on you should see battery voltage (12V). If not then with the ECU unplugged, confirm there is continuity between terminal #8 of the ICU connector and terminal #1 of the ECU connector. (see page 280-9 for the pin layout of the ICU connector)

I'm gonna go check terminal #5 again now, I don't think it was grounded before. . .

The factory manual indicates a black wire connecting #5 to ground, and another wire connecting #5 to the shield of the O? sensor.

BTW, do you know if the oxygen sensor is the only thing grounding into this missing wire? If so, it would be considerably easier for me to just run another wire than to rip into the harness again.

I really am not sure which ground wire is or maybe missing, the factory manual for the 240 does not indicate precisely where each ground point is.

I would go by the electrical tests outlined in the Bentley, and correct any missing or weak grounds. If all of those tests pass you likely need a new ECU.

 

[jkior]

WOOOOOOOOOOOOOOOOOOOOO!!!!!!!!!!!!!!!!!!!!
FINALLY! THE! CAR! IS! RUNNING!

man. . .

It was, indeed, that last, missing, ground wire. With the help of my brother's tone generator/line tracer, I found the thing, pretending to be an injector lead. Or something. Connected it up, and the pumps started running when I turned the starter (YES! for the first time since it died.) Anyway, the car runs fine now. I know that one of those things I fixed was, indeed, the problem, because the car didn't run before with this wire attached.

The factory diagram I looked at today specifically showed a black wire running from #11 to ground and also another black wire running from #11 to a small connector where a loop of wire connected it to a wire (also black) coming from terminal #19.

I guess this must be an error, because mine is absolutely and certainly a simple loop:

The wires are continuous through the connector cover.

The factory manual indicates a black wire connecting #5 to ground, and another wire connecting #5 to the shield of the O? sensor.

I really am not sure which ground wire is or maybe missing, the factory manual for the 240 does not indicate precisely where each ground point is.

Yes. That single black wire that attaches to the forward fuel rail bolt is the terminal #5 at the ECU. (Confirmed) It is absolutely necessary for this to be connected for the ECU the function.

I would go by the electrical tests outlined in the Bentley, and correct any missing or weak grounds. If all of those tests pass you likely need a new ECU.

Problem solved! Having the opportunity to go back and forth with someone else on this has been unbelievably valuable. I'll always be grateful for the amount of time you've been willing to spend to help me with this.

Thanks!
~mike

 

[Three Fat Tigers]

<img src=http://home.att.net/~saher/FunStuff/RecommbCheer.gif align=left>Awesome!




<br clear=all>That tone generator/line tracer sounds interesting, I'll have to look into getting one.
<img src=http://www.threefattigers.com/Protocore/FunStuff/My580spacer.gif>

 

[jkior]

Yeah, my brother uses it in his work as a network sysadmin. It can be a real pain in the neck to use, and only really practical for wires that are disconnected at both ends (it works by putting a "warble" signal into the wire, and using the other tool to trace the wire by inductance (?). So it can't be used on an attached ground wire, for example. )

Saved a huge amount of time in this application, though.
The one I used was #115 3685 on this page:
http://www.cyberguys.com/templates/SearchProducts.asp?search=1to%2D2&child=&sdesc=Test+Equipment&curpage=4&ureq=y