Ghost in the Machine - Trying to Fix for 3 Years Now

This is a last attempt to save my 1999 SR5 - I've been working to solve this problem for 3 years, and mechanics (and mechanic friends) have not made much headway.

Problem: Starting 3 years ago, every 4th or 5th use, my car would be slow to start. Keeping the key turned and letting the engine crank, combined with a bit of gas petal action and turning off the ac (how much these help I’m unsure about), will usually get it to start after 5 labored seconds. Should that fail, waiting 45 mins - 1 hr without an attempt gets it to start.

It's degraded over the past few months. Now, I might have to crank the starter for 4 or 5 rounds of 15 second attempts. One thing I'm sure about: it has something to do with the engine having run recently. 98% of the time, it starts right up in the morning, the first time I use it for the day. 90% of the time, it starts up right away after having sat in the parking lot of my workplace for 9 hours. However, if I stop for groceries and return to the vehicle after 30 minutes, or even stop for gas and start it 4 minutes later after fueling - it's in real danger of cranking over and over until the battery is dead. My band-aid solution: shop for 30 more minutes and try again. Often that works.

Observation: After running a few minutes, it tends to idle high when in park - around 2k rpm. Dropping into the drive gear with the brake applied put it into the normal 700 rpm range.

Observation: Once running, the car never dies on the road.

Codes (a few months old - my OBDII reader is in storage right now as I just moved):
P0120: Throttle pedal position sensor
P0136 O2 sensor circuit bank 1 sensor 2
P1133 A/F sensor circuit response bank 1 sensor 1
P0136 O2 sensor circuit bank 1 sensor 2

O2 Sensor test:
Bank 1 sensor 1 "does not support"
Bank 1 sensor 2
Min 1.275 V fail
Max 0 V fail

Regular maintenance:
All fluids are maintained
Battery and alternator voltage output nominal
Spark plugs are about 3ish years old - I did them myself. The old ones were badly corroded. Installed new spark plug wire harness.
All belts are only a few years old. They do squeal a bit upon startup.

Replacements I've made in the last 6 months:
New starter
New fuel filter - [Hard to tell, but this might have helped? I replaced the starter at the same time.]
New throttle response controller
New O2 front and rear sensors. - [Note: The front sensor might have been for manual version of the 4runner. My experienced mechanic said this should not cause a significant problem (at least not the problem I'm experiencing now)]
New battery

My not-very-hopeful next steps:
1. Replace the knock and water sensors - perhaps they send bad signals when hot?
2. Fuel: Drop the fuel tank and replace fuel pump. Check for rust - if present, clean fuel injectors. Note: I REALLY want to be confident this is a fuel issue before dropping the tank. It looks like a massive pain (esp with rusty boltheads), and I don't want to drop it for nothing.
3. ECS computer. Remove, get refurbished through Module Master, get software re-flashed.

If anyone can suggest a winning solution, I will personally fly out to shake your hand.

 

I remember working on a few company ford cars 20 years ago. They would consistently do this because of vapor lock. Your fuel pump could be the culprit. Just a thought.

 

Do another code read, it's hard to make suggestions if I don't know which of those codes have been resolved by the work you've done. Did these codes appear all at once and did the starting problem begin when these codes appeared?

Also what do you mean by throttle response controller and ECS computer? A throttle response controller is normally something like a pedal commander, and I've never heard of an ECS computer (maybe you mean ECM).

Your not going about diagnosing this in a good way (no offense), a lot of the things can be tested without needing replacement or be ruled out entirely. You want to fix the starting issue first sense it could cause the high idle.

It's not a knock sensor, that wouldn't cause starting issues.

A fuel pump can cause unreliable starting, but it shouldn't be affected with the passage of time because it's caused by the physical orientation of the commutator in relation the the brushes (don't worry about what that means). If you think it's a bad fuel pump have someone crank the engine while you hit the tank with a rubber mallet, if it doesn't help the engine start your fuel pump is likely fine. If it does reliably help the engine start then replacing it is warranted.

Clean the MAF sensor and check that the throttle body closes properly and that it's not jammed full of carbon. Inspect basic things like air filter, maf sensors, and throttle system if you haven't already (I'm saying this because these are overlooked often). Finally does spraying starter fluid into the intake make the engine start running?

 

Good suggestions Trekker,

I'll do another read on the codes as soon as I can. Timeline on the codes is fuzzy. I got the codes when I used the reader for the first time, and thus I learned the check engine light had burnt out (I then replaced all the dash lights, and found the battery light to be on as well). Upon replacing the O2 sensors I didn't get a change in codes - although perhaps the codes needed to be cleared. When the battery has been disconnected, it usually takes a day or so for the check engine light to come back on.

By throttle response controller I really meant throttle position sensor - it sits under the carburetor. The old carburetor was pretty choked with carbon, so I put in a new one. I tried spraying carb cleaner into the engine air intake with the carburetor off, but it didn't have an effect.

No offense taken. Some of the parts (like the starter) I figured were close to end of life anyway, so I was glad to replace them. But yes, not best practice.

I'm going to inspect the MAS and see where that gets me.

 

I did not get continuity on any of the MAS pins, so that points to a failed sensor. Found an OEM one on ebay - hopefully this will help - we shall see.

 

This sounds like the same issue I had with my '85 Supra. Oddly, a jump would help sometimes when it didn't want to start after refueling, or a short period of being shut off.

I originally suspected it was something with the fuel pump, and replaced it. This helped for awhile, and then the issue returned. Unfortunately, I sold it before I ever figured it out. I chocked it up to electrical demons.

Good luck! Sorry for the useless anecdote.

 

Don't buy sensors off from ebay or Amazon there are too many fakes anymore.

Get them from an online discount Toyota dealer or from a big box store.

Your codes and erratic idle suggest that you need to clean your entire throttle body and replace your TPS and maybe your IAC. MAF cleaning too.

The starting issue: For the ECU to start it needs good data from the crank\cam sensors as well as fuel pressure. You should check the crank and cam sensors and run a fuel pressure test. If it were me, I'd just replace the fuel pump with a DENSO pump...it's due. You have spark since when the ECU wakes up, it runs fine.

 

Carburetor?

 

Any codes to report? Also as Daddykool suggested these 4runners don't have carburetors, they are all fuel injected

 

Update:

I pulled out the MAS and tested it. The power pins weren’t reading continuity and the bulb looked dirty, so I ordered the new one. I installed the new one, and started the car three times. Performed great. Victory, I thought.

Not the case. It started right up in the morning, and then again on my drive back from work. But upon attempting to start it 30 minutes later - nothing but cranking, and this time gas petal pumping doesn't even work. Back to square one.

Also, a bit of unrelated backstory - after changing the MAS sensor, I tried starting it a fourth time, and got a nasty grinding sound - it turned out to be the sound of the starter solenoid dying. No cranking after that. Sure enough, the starter ground wire was not as secure as it could be, and I saw a black mark from where electricity had arced. I replaced the starter (this time with a new one instead of a reman) and it started right up.

Next step? Not sure. Maybe buy another code reader to get codes. And to try starting it without to MAS and see if that affects anything.

 

Isn't the 99 4Runner Fuel injected?

Not sure how pumping the pedal is helping.

I'd put a fuel pressure gauge on and see if its supplying enough fuel before dropping the tank.

And since you changed the plugs what did they look like when pulled.

 

Update 2:

Sounds like the 4runner is fuel injected - thank you for the correction. The "carburetor" I worked on was really the throttle body.

Got the codes today using a reader - there only seemed to be one:
P1133 (Marked three times as Current, History, and Pending).
1. A/F Sensor Circuit Response Malfunction Bank 1 Sensor 1
2. Power Take-off accel position sensor circuit high

Readings of note:
Fuel Sys 1: CL-Fault
Fuel Sys 2 N/A
ECT 32 degF (Note - it was about 60F outside at sunset, so 32 is definitely low)

I looked up the O2 sensors I put into the 4runner last year. Both turned out to be the CA compliant ones, and meant for automatic transmission (This 4runner is the CA compliant version, automatic, so that checks out).

Fuel pressure check is next on the list.

 

Good recommendation, that's next up on the to-do list.

When I changed them 3 years ago they actually didn't look too bad. I replace them with higher end ones. I can pull one or two out and see how they look.

 

So now the TPS, IAC, and MAF have all been replaced within the last year, to no effect. Throttle body has been cleaned.

I'll look into the DENSO pump. After all, the one thing that seemed to have any effect on this problem was the fuel filter replacement.

 

Well one thing is "pumping" the Gas petal can make starting hard if you do it while cranking over - and give you Code PO120 because that is the Throttle Position sensor being moved and the ECU expects it at a "home position" when starting and doing that alone can make hard starting Not saying that is your problem however it certainly is Not helping.
Pumping the gas petal is something I do on my 31 Packard due to it having a Carburetor and it needs some gas in the intake manifold to get started and Toyota's have not had Carburetors since 1991

 

You say your starting problems started 3 years ago and that you changed from OEM plugs to higher end plugs 3 years ago. I would be tempted to go back to a new set of OEM Denso plugs to see if that helps.

 

Finally - a conclusion.

When looking through the engine stat readouts on the OBDII sensor, I noticed the temp was reading at 32 deg F. I had just run the car (across the desert no less), so obviously not correct. I replaced the engine block coolant sensor and - perfect! The car runs like new, starts like new. All the engine light errors have not come back after I cleared them, and I couldn't be happier. Thanks for your help.

This explains the very strange behaviour. When the engine was cold - anything under 100deg F, the faulty 32 degF reading wasn't a bit deal because it wasn't too far off. Once the engine got hot and rose to 180degF+, the computer kept trying to start the hot engine like it was freezing cold, and the air/fuel mix just wasn't right. How I was still able to start it all this time while it was still hot is a bit of a mystery - perhaps the sensor worked for a brief second, or perhaps fuel built up in the engine which made the fuel/air mix right by sheer luck.

Tested the fuel pressure with a fuel pressure reader - all good. Fuel pressure dropped by a few psi every couple of minutes so a new pump should be put in eventually.

Lesson learned: Look through your data on the OBD sensor, and apply a sanity check to the numbers you see. I never got a code for the temp sensor being bad - but should have notice the low temp and registered it as bogus right away.

 

Awesome. You definitely were smart and got it figured out. Those codes wouldn't have screamed temp sensor to me.

A low temperature reading SHOULD throw a code. I just dealt with a Mazda 3 that was throwing a low coolant code due to a bad thermostat. Easy fix since the code pointed to the T-stat or sensor.

 

@forestlink glad you found your problem & I’m hoping you solved mine also!
I have similar starting problems for about a year or so. Replaced many parts to no avail.
Is there another name for the “engine block coolant sensor” that you changed?? Because I’m having trouble researching this part for my 4runner.