Air flow sensor test

[Silent]

Is there a test for the air flow sensor?

I've been having problems of late with my '07 FJR. It's running like the choke is still on while taking off from a stop. Even after the idle has dropped and theres 3-4 bars on the temp display. It hesitates and sputters badly and occasionally dies. It's to the point it's not rideable. I pulled the fairings off and checked all the connectors I could for corrosion and tightness, and poked around for any obvious issues. None of this had any impact on the issue.

Today I adjusted the fit of the intake boots since they seemed a bit far from the throttle bodies for my liking, and

I pulled the air flow sensor and wiped it down with acetone. I sprayed the throttle body linkages with WD40. I had a small leak from the thermostat housing, so I fixed that as well. The bike ran OK on the test ride, but I can still feel some of the hesitation and sputtering on occasion when taking off. I'm thinking that maybe the air flow sensor is going bad, and I'd like to test it to find out for sure.

Yea, I know, it's been awhile. I'm still alive tho

 

[A41Billy]

My '06 is having this same issue. I was planning on checking the connectors under the tank for corrosion. I am hoping someone has some insight on this. Please report back if you find anything out and I will do the same. I am curious about the airflow sensor as a culprit.

 

[Silent]

The air flow sensor is mainly just a thought from what I've been reading. The bike doesn't throw any codes at all, and it makes no difference if the PC3 is connected or not. Low milage (30 MPG +/-), and a general feeling that it's running rich. Plugs look good tho, and she runs fine when she decides to run.

 

[Denver_FJR]

There's no "flow" sensor but there is a MAP sensor and you can check it by going into DIAG mode, code 3. With engine off, it should display ambient pressure in kilopascals. I'm in Denver and code 3 typically runs about 80 to 83 (or about 25" Hg). Lower elevations should read higher value. Per manual, the test is to actuate starter and watch for fluctuation in reading of code 3. Yeah, not exactly scientific, but I would say if it's reading ambient and fluctuates during the test, it's good.

 

[Surly]

Yea, I know, it's been awhile. I'm still alive tho

 

[Silent]

Diag 3, 101cobstant, down to 94-95 while cranking.

 

[Silent]

Now that I've found my manual, apparently I cleaned the air temperature sensor, not the air pressure sensor

Once I locate the pressure sensor, I'll check that out as I still think it may be the culprit.

 

[JamesK]

IIRC, the Air Pressure Sensor is the little rectangular gizmo bolted onto the fuel rail, it will have a vacuum hose plugged into on the bottom and a single electrical plug from the side facing the back of the bike.

If this sensor was dodgy or the vacuum hose kinked or leaking you should be seeing something like an error code 14 and/or 32 from memory. It would also light up the yellow engine light.

As Ionbeam explained it in another post, the bike should/will run fine with this sensor not working since the ECU would automatically switch to a stored table of values. If this is not happening than you might be experiencing the problem I'm still trying to find where this circuit works intermittently, hence the ECU does not simply switch over to use default values table and the bike will not idle or not idle smoothly and will tend want to stall until it gets up to operating temps.

In searching for my problem I have confirmed the wires/plug are ok, the vacuum hose is not kinked or damaged and even eventually replaced the sensor. It seemed to fix the problem for a couple of weeks and then it started again.

I personally have run out of ideas of what else to do to track this intermittent problem down, so will likely be installing a small switch (to temporarily cut the signal wire to the Air Pressure Sensor) to manually force the ECU to use the default look up table. As soon as the bike is up to operating temps just flick the switch of to reconnect the circuit.

 

[Donal]

There was another thread recently (don't ask me where) about sensors and I seem to remember that the OP found a tiny bead of 'gloop' in one of the vacuum hoses.

When you check these hoses the thing to do apparently is to actually push something through the hose. Not sure if you can get anything like a 'pipe cleaner' (used to be available in all tobacconist shops) but something like that should do the trick....................

 

[ionbeam]

On the Gen I, the Intake Pressure Sensor (IPS) at the left end of the fuel rail and the Ambient Pressure Sensor (APS) under the seat are the same part and can be swapped. The Gen II IPS is in the same location as the Gen I, at the left end of the fuel rail.

On any Gen, unplugging the IPS will cause the ECU to use a look up table. If you unplug the IPS and your driveability issue gets better your problem is the sensor or hose going to the sensor.

As Donal says, people have been finding the vacuum hose gets plugged up with goo. NEVER: poke anything in the opening in the IPS, or spray anything in the opening, the sensor element is extremely delicate and will be damaged.

The diAG values in posts #4 & #6 are correct.

The engine will run fine with the sensor unplugged, but it will not run OK if the IPS has an erroneous output. It is like the old question, "Which is more accurate, a watch that is slowly losing time or a watch that's stopped?" The watch that is stopped will be correct twice a day (sensor unplugged), the watch that is losing time will never be correct (erroneous sensor).

 

[Silent]

Good to hear the values are OK for the sensor, I was wondering about that. I did find the air pressure sensor last night after RTFM for awhile, exactly where JamesK said it would be BTW. The sensor looked OK with no visible cracks or damage. The hoses were replaced not long ago when the timing chain and tensioner were replaced. I did check them for tightness and kinks, and I shortened a couple for a smother flow. I'll hit a smoke shop for some pipe cleaners and poke them thru the T connectors since those are the original bits and they have smaller holes. Worth a try anyway. I did add a missing clamp for the hose that runs from the crankcase to the air box since it was missing.

The bike was having stalling issues before the timing chain replacement, where the RPM would drop from 1200 to about 500, with occasional stalling, at stop lights. This was when the bike was running 3-4 bars on the temp gage. That issue was elevated by switching back to the pre-recall ECU. Switching back to the post recall ECU did not correct the current problem.

The only code the bike has thrown has been a tip sensor, and that was from pulling the fairings. I wish the bike would just throw a code so I could have some direction on what sensor to look at.

I'll ride the bike to work today as a test run and see if anything has changed since yesterday.

 

[Denver_FJR]

The engine will run fine with the sensor unplugged, but it will not run OK if the IPS has an erroneous output. It is like the old question, "Which is more accurate, a watch that is slowly losing time or a watch that's stopped?" The watch that is stopped will be correct twice a day (sensor unplugged), the watch that is losing time will never be correct (erroneous sensor).

I find it maddening that the FJR engine can run with a sensor disconnected yet not run well with all sensors connected and no sensor faults found!

Or another way to put it... If the ECU is so sophisticated that it can accurately infer the output of a disconnected sensor, it should be cake for it to manage fuel calculation when everything is connected. And yet it can't.

 

[ionbeam]

I find it maddening that the FJR engine can run with a sensor disconnected yet not run well with all sensors connected and no sensor faults found!Or another way to put it... If the ECU is so sophisticated that it can accurately infer the output of a disconnected sensor, it should be cake for it to manage fuel calculation when everything is connected. And yet it can't.

Running with the sensor disconnected using a look-up table: Sufficient.

Running with a good sensor connected: Optimized.

 

[HYCLE]

Silent, with a digital multi meter you can measure the output of an intake manifold pressure sensor. On most engine management systems, they have three wires. Power, Ground and signal. The power side is usually reference voltage of 5 volts with the signal wire showing aprox. 5 volts with no vacuum and moving toward 1-2 volts at idle, depending upon altitude.The voltage drop on the ground side should be somewhere less than, point 2 volts.

With the wires connected and key on, backprobe the terminals and see what readings you get. If you can't find specs from Yamaha look at the test procedures for a MAP sensor on a car.

Check this website for testing procedures of many vehicle sensors.https://www.wellsve.com/ds_map.html

 

[DennisJ]

There was another thread recently (don't ask me where) about sensors and I seem to remember that the OP found a tiny bead of 'gloop' in one of the vacuum hoses.
When you check these hoses the thing to do apparently is to actually push something through the hose. Not sure if you can get anything like a 'pipe cleaner' (used to be available in all tobacconist shops) but something like that should do the trick....................

See this thread for the referenced: https://www.fjrforum.com/forum//index.php/topic/124576-intake-air-pressure-sensor/, post #13

 

[ionbeam]

Silent, with a digital multi meter you can measure the output of an intake manifold pressure sensor...

Yes, the sensor can be tested like this, but... when you go to diAG you are getting a better test than this. IMO, there is no need or benefit using a volt meter to test the sensor.


The ECU sends the sensor +5 volts, sensor ground and a readback wire. The +5 volts powers the sensor. The sensor generates a voltage that is proportional to the pressure and sends the voltage to the ECU on the readback wire. Inside the ECU a circuit converts the analog signal to a digital value that is equal to the analog signal. The ECU takes the digital value and performs some math on it and converts the voltage to a pressure value. The pressure value is used as one of the many 'terms' in the fuel injection trim calculation. When you go to diAG and read the value on the meter assembly you are seeing the exact value that the ECU is using for fuel trim calculations. If the value on the meter is correct then everything in the Intake Pressure Sensor circuit is working correctly.

 

[Silent]

Update:

I rode the bike to work this morning (40 degrees ambient temp), and it ran just like it should. Felt like my old bike again. On the way home however, it acted up again and started stumbling on take-offs (70 degrees ambient temp). I got home and checked diAG 3. 101 dropping to 94 on cranking, same as before, I also checked the TPS, and it was clean from 15 to 98. I popped the tank and disconnected the electrical to the air pressure sensor. Started the bike and it ran like total crap. The RPM dropped below 1,000 (set at 1200) and I could hear mechanical lashing in the engine. Not a good thing. So I shut it down and reconnected the electrical and on restart it went back to 1200 RPM and sounding decent. I shut it down, pulled the hose and plugged it. When I started the bike, it sounded the same as when I unplugged the electrical connection to the air pressure sensor.

Since all of these episodes seem to be when the bike has been ridden at least once for the day, and during the warmer temperatures of the afternoon, I unplugged the electrical to the air temperature sensor. Bike started fine, and idle was good. Took a test ride, still stumbling. I disconnected my PCIII, and plugged the O2 sensor in. Same results. Air temp sensor is reconnected.

I'm at a loss of where to look next.

I need to remove my left faring lower just to put coolant in. Seems a big ol connector is in the way now, and needs to be relocated

 

[HYCLE]

Silent, is your problem engine temp related? I have seen coolant temp sensors give faulty readings at warm engine temps, causing intermitant problems without setting codes.

Do we know if a reading of 94 and 101 is correct for your map sensor. The range of that sensor is 10-200mmhg? (taken from a previous ionbeam post) https://www.fjrforum.com/forum//index.php/topic/9612-diagnostic-codes/

Post #4 states 80-83kilopascals KOEO = 25"hg

94mmhg = 3.7"hg and 101mmhg = 3.976"hg so the change at the map sensor was less than 3"hg. I could be wrong with the math, not much of a change though. You would have to see what this reading is at Idle and do the math. You could also attach a vacuum pump to the sensor and watch the reading change. That would give you a wider sweep of the MAP sensor range.

I disagree with Ionbeam, for the same reasons he states you should trust the diag on the dash. The diagnostic codes will only set if the reading is outside the range of the sensor under specific conditions. It may not catch an intermitant issue. The cranking test only tests the sensors ability to send some type of signal at cranking rpm. The difference in readings while running at vaired rpm, ie: varied manifold pressure can only be tested at the sensor with a DMM or Scope. Infact if you have access to a Digital Storage O-scope, connect it directly to the sensor to see if there is any noise in the signal. Noise is voltage spikes in the signal from the sensor that can cause driveability problems without setting codes. The onboard diagnostics can't test the sensor to this accuracy.

The fact that the signal at the diagnostic monitor is computed by the ECU leaves the computer in the circuit and is like getting the information second hand from an interpretor. At school, all of our test vehicles have test points at the sensor that the student use to verify the information they get from the onboard diagnostic connector. We have multiple faults that will give readings within normal parameters at the control unit, but will cause drivabiltiy problems.

For example a high resistance at the TPS connector will show a 20% throttle opening with a closed throttle. Since 20% is a normal reading if the throttle is open to that value, the computer sees no problem. With this starting value the reading reaches 100% throttle at an actual 75% opening.This throttle opening will cause an increase in automatic transmission pressure and a very hard shift at all throttle openings. This type of problem can also occur with a MAP Sensor.

 

[Silent]

I have neither a vacuum pump, nor an oscilloscope (Not that I"d have a clue how to use one if I did

). Tho I do have a friend who may have a vacuum pump.

I'll get back on that in a few.