Poor Idle is driving me crazy!

[dogfriend]

Originally Posted by vertex

One thing I don't understand, is why is the RPM higher with a leak?

On most EFI engines, idle speed is controlled by regulating bypass air (air bypassing the closed throttle plate). This is usually a function of the IAC valve, which is operated by the Engine Control computer. So, if there is excess air leaking past the throttle plate that is not metered (or accounted for) it will raise the idle speed. My understanding is that for the Prius, the IAC function is integrated into the electronic throttle body. You mentioned that you replaced the TPS. Any chance that it could be misaligned with the throttle plate? How are those two indexed?

[vertex]

Originally Posted by dogfriend

On most EFI engines, idle speed is controlled by regulating bypass air (air bypassing the closed throttle plate). This is usually a function of the IAC valve, which is operated by the Engine Control computer. So, if there is excess air leaking past the throttle plate that is not metered (or accounted for) it will raise the idle speed. My understanding is that for the Prius, the IAC function is integrated into the electronic throttle body. You mentioned that you replaced the TPS. Any chance that it could be misaligned with the throttle plate? How are those two indexed?

The TPS is mounted with two screws and can be rotated to adjust the zero. It has a post that is keyed into the TB plate, so that its mounting cannot cause a leak, but is could be misadjusted. I played with the adjustment, just in case I had it wrong, but it didn't cause any change in operation, but it will cause a DTC if set too low.

[Patrick Wong]

Originally Posted by vertex

...The main symptom is that when the engine idles (that is not shut off, not driving wheels, and not charging the battery) the engine RPM hunts between 1700 & 2000 RPM when in park, and up to 4000 RPM when in drive. This occurs when the engine is in the ultra lean mode, O2 is 0.015 volts. Once the engine warms up to over 158 F, the P0101 DTC occurs. Otherwise no codes...

When the engine is cold, what idle behavior does it demonstrate?

Do you know how to put the Prius into inspection mode so that the engine runs continuously, allowing you to assess engine idle over a long period of time?

[vertex]

Originally Posted by Patrick Wong

When the engine is cold, what idle behavior does it demonstrate?

Do you know how to put the Prius into inspection mode so that the engine runs continuously, allowing you to assess engine idle over a long period of time?

Patrick:
Same behavior when cold, no DTC. I have instruction on inspection mode, but did not succeed in getting it to work.

Howard

[Rokeby]

I hope all-you-all won't mind me asking a connected, but tangential
question. I'm curious about the expected vacuum leak induced RPM
increase... curious minds want to know

As I understand what's been said, a vacuum leak, depending on its
volume, leads to a small but measurable RPM increase. I'm
guessing that the ICE is running just a little lean.

In the thread that I cited earlier, they sprayed small amounts of carb
cleaner on vacuum hose joints suspected of leaking. This method also
apparently produces a noticeable change in RPMs. When I read it, I
presumed it was an RPM increase.

Now however, I suspect it would be a decrease as the when the
flamable liquid is drawn in below the throttle body, the mixture is
somewhat richer than before.

Am I even close to understanding what is going on here?

[jayman]

This is a head scratcher. In the past, I've found the propane method to *very* reliably find vacuum leaks

Seems we've covered all the usual bases here. A cracked intake could very well cause these problems. Patrick and Jason have covered the relavent diagnostics, nothing I can add except some photos

[Patrick Wong]

Originally Posted by vertex

...Same behavior when cold, no DTC. I have instruction on inspection mode, but did not succeed in getting it to work.

Hi Howard,

I started my 2004 (which was warm from having been driven ~100 miles or so earlier in the day) in inspection mode. Then I tried removing the oil dipstick. Very little change in rpm.

Then I removed the oil filler cap. Engine rpm increased slightly. This illustrates that the idle speed control logic is able to compensate for a relatively large vacuum leak. Therefore I'm less confident now that a vacuum leak is causing your driveability problem.

Have you had a chance to perform the recommended tests that are shown in the repair manual regarding DTC P0100, P0102 and P0103? This would validate that correct voltages are reaching the MAF.

In particular, note that if the hand held tester thinks that the air flow rate is excessively high (which may be the case since you have DTC P0101), check the resistance between terminal EVG (E5-32) and body ground. If the resistance is above zero, then the engine ECU is at fault. See for example page 05-97 of the 2005 repair manual or the analogous page of the Toyota repair manual that you are referring to.

The manual goes on to show how to check continuity of the wiring harness that connects the MAF to the engine ECU. Since your car was in an accident, that wiring harness damage is a reasonable possibility to consider.

[dogfriend]

Originally Posted by Rokeby

I hope all-you-all won't mind me asking a connected, but tangential
question. I'm curious about the expected vacuum leak induced RPM
increase... curious minds want to know

As I understand what's been said, a vacuum leak, depending on its
volume, leads to a small but measurable RPM increase. I'm
guessing that the ICE is running just a little lean.

In the thread that I cited earlier, they sprayed small amounts of carb
cleaner on vacuum hose joints suspected of leaking. This method also
apparently produces a noticeable change in RPMs. When I read it, I
presumed it was an RPM increase.

Now however, I suspect it would be a decrease as the when the
flamable liquid is drawn in below the throttle body, the mixture is
somewhat richer than before.

Am I even close to understanding what is going on here?

What better way to get information about leaks than to get it from the experts: Ford owners

help finding vacuum leak... - FFCobra.com - Factory Five Racing Discussion Forum

I believe that the second poster in that thread is correct: the propane will raise the idle slightly when sucked into the intake because it is a source of additional fuel.

Here is an article covering the different techniques, with the addition of the smoke method which may be the best way of all if you have the equipment:

Tech To Tech

[vertex]

I checked again tonight each vacuum fitting again by removing 1 at a time and blocking the port. No change. I carefully looked at the manifold, and see now sign of cracks. I checked it when I had the car apart months ago, and did not see anything. Short of taking it out, the only test that I can think of is the propane, which did not work last time I tried it.

Is a vacuum leak the only possibility, or is there something else? I can try the propane approach one more time.

Thanks again for the help!

Howard

[vertex]

I like that VacuTech machine. I will see if I can find a shop that has one.