Not Motorhome Related - Toyota 5SFE Question

[Ctgriffi]

So this is a pretty random question for the forum, but I know a lot of you guys/gals have tons of mechanical experience, and I'm having a tough time with an old 5SFE (common Toyo motor in the 90s and beyond) in my daily driver, a 93 Celica GTS. At a stoplight, in neutral, the idle bounces constantly between about 500-900, and it's been driving me bonkers for many months. I've tried replacing various small parts (including the IAC valve, VSV, TPS, EGR modulator, and throttle body gasket), looked everywhere for vac leaks, searched the internet high and low for answers, and I'm pretty much stumped! Aside from this issue, the car runs well, has good pickup, and fuel mileage is consistent with EPA figures. (Note: Bouncing idle seems to worsen in very hot weather.)

Where would you start looking? Any suggestions appreciated. Thanks!

[payaso del mar]

does having the A/C on affect it, or not?  most vehicles have some way they kick up the idle when the A/C is energized to compensate for the load of the compressor.  the reason I ask is that the 500 figure sounds like hot idle and the 900 sounds like the "A/C on" idle....wondering if something is wonky in that realm and the ECM is getting an intermittent "A/C on" signal?

secondly, my 02 Taco is programmed so the idle doesn't really settle down to the "lowest" hot idle RPMs (the hot idle figure given in the shop manual) until after it's been idling for 15-20 sec.....i'm about 100% sure that this was something Toyota designed in, altho i'm not exactly sure on the purpose.  again, not sure how you might check this but.....

I doubt it's the ECM, but at the prices I sometimes see em for on CL, it almost might be worth trying the "just a throw a part at it" approach if you don't find anything else?

[Ctgriffi]

Yeah, A/C on does change things a bit because, as you mentioned, it does raise the idle speed like it should.

It's strange because I can coast up the road in neutral, foot off the gas, and the motor idles just fine, very steady and quiet. Then, I get to a stoplight, apply the brakes and sit... 2 seconds later, idle starts bouncing: vroom vroom vrooom. Because of this, I have to think that some sensor is giving the ECU bad info, causing it to react erratically. (maybe?)

[payaso del mar]

I think we have another data point here.  wondering if you might have a vacuum leak in the brake booster diaphragm or line.  high, erratic idle is characteristic of vacuum leaks. 

try temporarily disconnecting the vacuum line to booster and plugging it (also cover nipple on booster to keep crud out) and take it for a drive.  be ready for Brit-sports-car braking force requirements with no vacuum assist....doesn;'t affect the stopping, but a hell of a lot more pedal effort so be careful til you get used to it.  if the problem disappears, find a new booster diaphragm and you're set.

[Ctgriffi]

I like your train of thought here, and I will definitely try this! Muchas gracias

[Ctgriffi]

Update: I didn't completely follow the scenario described earlier, but I did do some more extensive testing in my detached garage.

With the motor warmed up and idling, I removed the brake booster vac line from the intake, plugged the nipple on the intake and... after letting it run awhile longer, noticed no change whatsoever. I then started pulling off all the small vac lines, one by one, plugging each orifice and observing any changes in the idle. Obviously, the idle would ramp up quickly when I disconnected a hose, but once I plugged the opening on the intake, the "hunting" idle would resume, exactly the same.

I took a video of the tach and may post it later. Right now, it typically just bounces between 700-800, and it's more of a minor annoyance than anything (vroom, vroom, headlights dim/brighten). I know it'll probably worsen in the summer heat, like it did last year, so that's why I'm motivated to fix it.

[payaso del mar]

OK.  that;'s the story of troubleshooting, eliminating possibilities, yes?

the headlights dimming is a sign of something else problematic....even at idle, your alternator should still put out enough to run your headlights, and your battery should provide enough of a reserve that even if alternator didn't, it should run the headlights for a  traffic light cycle without them dimming.  EFI systems don't like low voltage, won't usually even run below about 10.5 volts, and sometimes act erratic before they get that low.  check your battery voltage at rest and your system voltage when the engine is idling and see what you see.  if this checks out OK, maybe have your local otto partz check the alt amperage output and battery condition.  or you could swap in a known good, fully charged battery and see how it idles.

[Ctgriffi]

Good info, thanks. It had crossed my mind that electrical problems might be involved. The (rebuilt) alternator is only about 1.5 years old, purchased from a reputable seller on eBay, so I hope that's not the problem. Could a bad ground (or two/three) be contributing perhaps?

(And, yes, I'll get my multimeter on there and see what I can determine.)

Edited April 14, 2016 by Ctgriffi
addition

[payaso del mar]

bad grounds can always be contributing and are a common source of electrical weirdness on old vehicles.  good news is they're easy to fix....just assume they're bad and remove/clean every one you can find, and add a redundant spare for critical connections.  but I think a bad ground would show up more as occasional misfires rather than idle fluctuation.

I was thinking about this again......i'm also wondering if you might be getting some signal fluctuation from either your airflow sensor (the big flapper valve thing in the sorta crescent shaped housing just downstream of air cleaner housing) or the throttle position sensor.  assuming you have a Haynes manual.  maybe also test the output from both of those.....I think it's either variable resistance or voltage output change that gives the ECM the info, and if it's sending psychotic messages to the ECM, that could cause an idle fluctuation.  oops, just saw you did the TPS so probably eliminate that.

I've also seen coils that worked pretty well when cold but when they got heated up, were iffy.  do you have a known good coil you could swap in to test?  doubt this is it since this also would more likely cause rough running than just wobbling idle.

finally, there's the old school mechanical possibilities.....intermittently sticking valve, etc.   How many miles on the engine?   A vacuum gauge is cheap from HF and will tell you this sort of info, once you learn to read it (the Haynes should have some info on that, IIRC)

[Totem]

does this motor have a mass airflow sensor? if so check and clean or replace.

Edited April 15, 2016 by Totem

[payaso del mar]

that wouldn't explain the low voltage at idle, although that could be an unrelated problem. but you're right, I think an engine of that year would have gone from the VAF flapper vane to a MAF hot-wire sensor.

[Ctgriffi]

On 4/15/2016 at 11:01 AM, payaso del mar said:

bad grounds can always be contributing and are a common source of electrical weirdness on old vehicles.  good news is they're easy to fix....just assume they're bad and remove/clean every one you can find, and add a redundant spare for critical connections.  but I think a bad ground would show up more as occasional misfires rather than idle fluctuation.

I was thinking about this again......i'm also wondering if you might be getting some signal fluctuation from either your airflow sensor (the big flapper valve thing in the sorta crescent shaped housing just downstream of air cleaner housing) or the throttle position sensor.  assuming you have a Haynes manual.  maybe also test the output from both of those.....I think it's either variable resistance or voltage output change that gives the ECM the info, and if it's sending psychotic messages to the ECM, that could cause an idle fluctuation.  oops, just saw you did the TPS so probably eliminate that.

I've also seen coils that worked pretty well when cold but when they got heated up, were iffy.  do you have a known good coil you could swap in to test?  doubt this is it since this also would more likely cause rough running than just wobbling idle.

finally, there's the old school mechanical possibilities.....intermittently sticking valve, etc.   How many miles on the engine?   A vacuum gauge is cheap from HF and will tell you this sort of info, once you learn to read it (the Haynes should have some info on that, IIRC)

Yeah, I've tried cleaning grounds, and have even added another ground, with no improvement.

Yep, TPS was replaced with new: no difference.

Coil could be a problem I guess. It's internal on this distributor, so I don't know if it's replaceable. Might have to research that more.

I do have a vacuum gauge but don't know much about how to use it in this scenario.

(252K miles on the car and original engine/tranny.)

Totem: This motor does have a MAP sensor... it's about the only thing I haven't replaced because they're about $270 new.

Edited April 18, 2016 by Ctgriffi

[payaso del mar]

wow.  it does sound like you've really exhaustively attacked this.  the # of miles on engine are high enough that you might have some internal issues like sticky valves.  compression is good and even?

i'm far from an expert at reading vac gauges, but I understand that the old-line experts can tell you a lot from those.  pretty sure Haynes shows info on vac gauge needle behaviors and what they indicate.  but for your purposes, you want to see if there's any difference between the reading (and needle behavior) between when it's idling high and idling normally. 

[Totem]

17 hours ago, Ctgriffi said:

Yeah, I've tried cleaning grounds, and have even added another ground, with no improvement.

Yep, TPS was replaced with new: no difference.

Coil could be a problem I guess. It's internal on this distributor, so I don't know if it's replaceable. Might have to research that more.

I do have a vacuum gauge but don't know much about how to use it in this scenario.

(252K miles on the car and original engine/tranny.)

Totem: This motor does have a MAP sensor... it's about the only thing I haven't replaced because they're about $270 new.

Mass Airflow sensor was dirty on my celica. Made it psychotically ramp RPMs in early morning like .5 to 1.5 k .. drove me  nuts ...carefully very carefully opened MAF and found it filthy, carefully carefully clean sensor with alcohol on cue tip and put back. problem never returned. be careful they are very expensive; cleaning them is a tried and true way to fix if they are not failed and just covered in dirt

Edited April 19, 2016 by Totem

[Ctgriffi]

On 4/18/2016 at 5:24 PM, payaso del mar said:

wow.  it does sound like you've really exhaustively attacked this...

Yeah, I've done about everything I can think of (although, sounds like cleaning the MAP sensor is my next step). I get a little crazy like that on these kind of repair projects; once I start something, I just hate to "get beat" on it, makes me miserable. You'd probably be amazed/appalled to hear how much time I spent repairing the cruise control system on this same 23 year-old car... just couldn't give up on it. Nice to have it now though, for highway driving.

[Totem]

must be a v6. my 4 banger... not sure cruise would even work; Original owner removed it

[Ctgriffi]

On 4/20/2016 at 1:44 PM, Totem said:

must be a v6. my 4 banger... not sure cruise would even work; Original owner removed it

I think there might be some confusion... this thread is (randomly) about my 4-cyl, 93 Celica. Although, my V6 Warrior does have CC, and it seems to work fine (on very level highways!).

Back on topic: My car does have a MAP sensor, but it's not user-serviceable, just has a tiny vac port. Also, I can't imagine it needs much cleaning; seems like any object that has no connection to the outside world, save a small, negatively-pressurized hole, wouldn't encounter much debris.

I definitely see the idle worsening as temps increase. It's bouncing hard between 400-800 now, at every stop light. Ugh.

[payaso del mar]

I think Totem was talking about a "hot wire" type mass air flow sensor (MAF), not a manifold absolute pressure sensor (MAP) like you have....MAP sensor is just sort of a deluxe vacuum sensor that reads pressure above as well as below ambient.....that's where the "absolute pressure" part comes in.  it's used in what's called a "speed density" controlled EFI system, where the primary ECM inputs are from the vehicle speed sensor and the MAP sensor rather than from a measurement of intake air flow or mass.....most of the manufacturers use this type now due to the reduced flow restriction and effects on mpg (and thus CAFE). 

from Wikipedia:

The manifold absolute pressure sensor (MAP sensor) is one of the sensors used in an internal combustion engine's electronic control system.

Engines that use a MAP sensor are typically fuel injected. The manifold absolute pressure sensor provides instantaneous manifold pressure information to the engine's electronic control unit (ECU). The data is used to calculate air density and determine the engine's air mass flow rate, which in turn determines the required fuel metering for optimum combustion (see stoichiometry) and influence the advance or retard of ignition timing. A fuel-injected engine may alternatively use a mass airflow sensor (MAF sensor) to detect the intake airflow. A typical naturally aspirated engine configuration employs one or the other, whereas forced induction engines typically use both; a MAF sensor on the intake tract pre-turbo and a MAP sensor on the charge pipe leading to the throttle body.

MAP sensor data can be converted to air mass data using the speed-density method. Engine speed (RPM) and air temperature are also necessary to complete the speed-density calculation. The MAP sensor can also be used in OBD II (on-board diagnostics) applications to test the EGR (exhaust gas recirculation) valve for functionality, an application typical in OBD II equipped General Motors engines.

*

Vacuum comparison[edit]

Engine vacuum is the difference between the pressures in the intake manifold and ambient atmospheric pressure. Engine vacuum is a "gauge" pressure, since gauges by nature measure a pressure difference, not an absolute pressure. The engine fundamentally responds to air mass, not vacuum, and absolute pressure is necessary to calculate mass. The mass of air entering the engine is directly proportional to the air density, which is proportional to the absolute pressure, and inversely proportional to the absolute temperature.

************

Common confusion with boost sensors and gauges[edit]

MAP sensors measure absolute pressure. Boost sensors or gauges measure the amount of pressure above a set absolute pressure. That set absolute pressure is usually 100 kPa. This is commonly referred to as gauge pressure. Boost pressure is relative to absolute pressure - as one increases or decreases, so does the other. It is a one-to-one relationship with an offset of -100 kPa for boost pressure. Thus a MaP sensor will always read 100 kPa more than a boost sensor measuring the same conditions. A MaP sensor will never display a negative reading because it is measuring absolute pressure, where zero is the total absence of pressure. Vacuum is measured as a negative pressure relative to normal atmospheric pressure. Vacuum-Boost sensors can display negative readings, indicating vacuum or suction (a condition of lower pressure than the surrounding atmosphere). In forced induction engines (supercharged or turbocharged), a negative boost reading indicates that the engine is drawing air faster than it is being supplied, creating suction. The suction is caused by throttling in spark ignition engines and is not present in diesel engines. This is often called vacuum pressure when referring to internal combustion engines.

In short, most boost sensors will read 100 kPa less than a MaP sensor reads. One can convert boost to MaP by adding 100 kPa. One can convert from MaP to boost by subtracting 100 kPa.

 

 

interesting that Toyota was using this in 93......my 02 Taco and MR2 both have gone back to the theoretically less efficient MAF sensor approach

fact that it's dropping to 400 rpms indicates something about your issue, I think.  isn't this WAAY too low for that engine?  i'd guess 800 is about normal idle.  ??? 

[Ctgriffi]

15 hours ago, payaso del mar said:

...fact that it's dropping to 400 rpms indicates something about your issue, I think.  isn't this WAAY too low for that engine?  i'd guess 800 is about normal idle.  ???

Yeah, I think the exact idle spec is 750. Like I mentioned, the warmer the outside air, the worse things get. When we eventually get back up in the 90s, it'll be nearly-or-actually stalling at intersections.

[payaso del mar]

well, this is a useful bit of data.  to me, this indicates that issue is something with an erratic malfunction dragging your idle down, not something glitchy kicking it up above normal.  obvious, huh?  but a whole different realm of possible causes to investigate.  a timing light (remember those?) can be used to isolate an erratic miss to one cylinder, but it sounds like you're not hearing any missing, just the idle dropping?

[Ctgriffi]

I don't think I'm hearing any misfires. I do have a timing gun (had to have one, after I did the t-belt and water pump on this thing a few years ago). Actually, I can't even verify that timing is dead-on anymore (10deg BTDC) because the idle is so erratic that you can't get a decent read with the gun.

[Totem]

On 5/4/2016 at 6:08 PM, payaso del mar said:

I think Totem was talking about a "hot wire" type mass air flow sensor (MAF), not a manifold absolute pressure sensor (MAP) like you have....MAP sensor is just sort of a deluxe vacuum sensor that reads pressure above as well as below ambient.....that's where the "absolute pressure" part comes in.  it's used in what's called a "speed density" controlled EFI system, where the primary ECM inputs are from the vehicle speed sensor and the MAP sensor rather than from a measurement of intake air flow or mass.....most of the manufacturers use this type now due to the reduced flow restriction and effects on mpg (and thus CAFE). 

from Wikipedia:

The manifold absolute pressure sensor (MAP sensor) is one of the sensors used in an internal combustion engine's electronic control system.

Engines that use a MAP sensor are typically fuel injected. The manifold absolute pressure sensor provides instantaneous manifold pressure information to the engine's electronic control unit (ECU). The data is used to calculate air density and determine the engine's air mass flow rate, which in turn determines the required fuel metering for optimum combustion (see stoichiometry) and influence the advance or retard of ignition timing. A fuel-injected engine may alternatively use a mass airflow sensor (MAF sensor) to detect the intake airflow. A typical naturally aspirated engine configuration employs one or the other, whereas forced induction engines typically use both; a MAF sensor on the intake tract pre-turbo and a MAP sensor on the charge pipe leading to the throttle body.

MAP sensor data can be converted to air mass data using the speed-density method. Engine speed (RPM) and air temperature are also necessary to complete the speed-density calculation. The MAP sensor can also be used in OBD II (on-board diagnostics) applications to test the EGR (exhaust gas recirculation) valve for functionality, an application typical in OBD II equipped General Motors engines.

*

Vacuum comparison[edit]

Engine vacuum is the difference between the pressures in the intake manifold and ambient atmospheric pressure. Engine vacuum is a "gauge" pressure, since gauges by nature measure a pressure difference, not an absolute pressure. The engine fundamentally responds to air mass, not vacuum, and absolute pressure is necessary to calculate mass. The mass of air entering the engine is directly proportional to the air density, which is proportional to the absolute pressure, and inversely proportional to the absolute temperature.

************

Common confusion with boost sensors and gauges[edit]

MAP sensors measure absolute pressure. Boost sensors or gauges measure the amount of pressure above a set absolute pressure. That set absolute pressure is usually 100 kPa. This is commonly referred to as gauge pressure. Boost pressure is relative to absolute pressure - as one increases or decreases, so does the other. It is a one-to-one relationship with an offset of -100 kPa for boost pressure. Thus a MaP sensor will always read 100 kPa more than a boost sensor measuring the same conditions. A MaP sensor will never display a negative reading because it is measuring absolute pressure, where zero is the total absence of pressure. Vacuum is measured as a negative pressure relative to normal atmospheric pressure. Vacuum-Boost sensors can display negative readings, indicating vacuum or suction (a condition of lower pressure than the surrounding atmosphere). In forced induction engines (supercharged or turbocharged), a negative boost reading indicates that the engine is drawing air faster than it is being supplied, creating suction. The suction is caused by throttling in spark ignition engines and is not present in diesel engines. This is often called vacuum pressure when referring to internal combustion engines.

In short, most boost sensors will read 100 kPa less than a MaP sensor reads. One can convert boost to MaP by adding 100 kPa. One can convert from MaP to boost by subtracting 100 kPa.

 

 

interesting that Toyota was using this in 93......my 02 Taco and MR2 both have gone back to the theoretically less efficient MAF sensor approach

fact that it's dropping to 400 rpms indicates something about your issue, I think.  isn't this WAAY too low for that engine?  i'd guess 800 is about normal idle.  ??? 

YES thats why i said MAF and not MAP and even spelled it out.

and you CAN clean a MAF and they DO get dirty as thats the air intake going to engine from the filter... swipe your finger in that housing if you do not believe me.

And BTW when the MAF sensor is dirty and your 93 celica does indeed have one; you get the exact symptom the OP described in here at stop or idle.

Its ok if you find this all out later with a bill from a mech; just dont say i didnt tell you so first. Your cars never had a new one; its dirty , guaranteed at that age.

Edited May 6, 2016 by Totem

[Totem]

http://www.theautopartsshop.com/auto+car-usa+parts+year/toyota+celica+mass-air-flow-sensor+1993.html

 

^^^^ THIS

[Totem]

On 4/20/2016 at 9:52 AM, Ctgriffi said:

Yeah, I've done about everything I can think of (although, sounds like cleaning the MAP sensor is my next step). I get a little crazy like that on these kind of repair projects; once I start something, I just hate to "get beat" on it, makes me miserable. You'd probably be amazed/appalled to hear how much time I spent repairing the cruise control system on this same 23 year-old car... just couldn't give up on it. Nice to have it now though, for highway driving.

MAF MAF MAF... not MAP ; MAF is located in air intake circuit most likely after air filter in the neck tube

Edited May 6, 2016 by Totem

[Ctgriffi]

Thanks, Totem, but no MAF on this vehicle; MAF is moot.

[payaso del mar]

the plot thickens.  Totem, I just checked Rock Auto and I think he's right....they don't show a MAF sensor for the '93 Celica with the normally aspirated 2.2, they list a MAP sensor, and the pix are definitely of MAP sensors, vacuum port and all.  check it out:

 thttp://www.rockauto.com/en/catalog/toyota,1993,celica,2.2l+l4,1274251,fuel/air,manifold+pressure+(map)+sensor,11207

 

 

[Totem]

weird they also show CARDONE 7420107 {#2225074200}; for the 1993 all trac celica....

 

 

probably a different motor offering.in my case it was a malfunctioning MAF and not a MAP.

My celica was a 2000 as mentioned.

Edited May 12, 2016 by Totem

[Ctgriffi]

In other news, my AC compressor is now becoming reluctant to engage when triggered on the dash. I spent several long nights last summer—replacing the compressor, many other AC components (all from RockAuto), flushing and evacuating the system. Grrrrr..... fun little car, but getting tired of all the repair work.

[payaso del mar]

I hear ya.  i once gave away a '91 Mazda truck because neither i nor my carb-whiz buddy could get it to run right....overnight went from years of running perfectly to getting 15 mpg and no one could determine why, even after replacing every vacuum line on the damn thing (used a whole 50 foot roll!)...one of the last years of carbs, when they were computer contolled.  guy i  gave it to slapped a Weber DGV on it (no emissions inspections where he lives) and is probably still driving it