ONe21

I have done some quick research and have found information regarding the issue. Here are a few of the threads I have found:

https://www.civicforums.com/forums/1...-civic-hx.html
https://www.civicforums.com/forums/2...rror-code.html
and then a continuation of the above thread...
https://www.civicforums.com/forums/2...163-again.html

My ECM is part #37820-P2N-A31 (unfortunately). I purchased the car last week and it just threw the engine code P1163 today. Since I bought it, I have noticed hesitation during acceleration once the engine is warm. It doesn't seem to be isolated to a particular part of the RPM band. I'm not sure if it is related, as TSB 00-028 states that there are no observable drivability issues caused by the DTC.

The only code being thrown is the P1163. I'm not getting a code for misfires, nor does it feel like a misfire. It feels more like if you were in a lower gear and started pulsing the gas pedal. For the DTC I would suspect the manifold, but the previous owner replaced it fairly recently along with the upstream o2. I'm not sure about the downstream, but it looks new.

I kind of suspect the TPS for the hesitation problem, but I'm not sure. I tested the connector with a multimeter and it is getting a solid reading for power and for ground. I am not sure how to test the sensor itself without cutting the wire (any suggestions?).

I'm going to do a tune up (plugs, wires, dizzy cap/rotor, ignition coil...maybe) because I checked the plugs and they are due for it. I also removed the EGR and looked at the passages--there was a good build-up of soot, but it wasn't blocked.

Also, back to the P1163, I heard some people say they will keep a code reader and clear it out whenever it comes up. Is this just because they hate seeing the light, or because it throws the ECM in limp mode or something? I honestly don't care about the engine light if it doesn't have adverse affects. The previous owner said that he was getting 36mpg, which is less than I would expect. I'm wondering if this is due to the P1163 or whatever is causing the hesitation.

Any help is appreciated.

ezone

P1163: A slow response problem in the Primary Heated Oxygen Sensor (Sensor 1) circuit.

Got freeze frame data? Does it match the info given in the bulletin?

You can have the code without having the ECM be the cause.

AND the code could be totally unrelated to the symptoms you feel!

#1) Scantool with live engine data list. Invaluable tool. Saves a crapload of time and effort.
#2) Backprobe the circuit. Google it. http://www.enduringautomotive.com/ho...robe-a-sensor/

Do your tune up, see if it helps, sure.

D16Y5 engine? You can't evaluate that by simply removing the valve.



Here's 2 hints for ya, if the tune up does not help. (I would do this before trying the tune up)

Disconnect the battery for a few minutes to clear code memory and adaptive values. Reconnect the battery.
Disconnect the front O2 sensor (ignore the CEL).
Go drive the car in whatever manner you felt the hesitation before, see if the hesitation feeling is gone now.

Did it change anything?
Reconnect the sensor.

Now unplug the EGR valve connector (ignore the CEL).
Drive the car again.

Did it change anything?

ONe21

Alright, I didn't disappear--it just took me a little longer to get to it today.
I did not get freeze frame date, because apparently AutoZone's code readers don't do that!?!

I did the tests on the O2 and EGR. Here are the results that I saw (Note: I reset the ECM before each test):
-Both O2 and EGR disconnected, I didn't really notice the hesitation during this test. The check engine light came on upon start-up
-EGR connected/O2 disconnected, no hesitation. Check engine light came on upon start-up
-EGR disconnected/O2 connected, noticed hesitation
-Both connected, notice hesitation

I'd like to get an OBDII reader for my Android tablet, but maybe later.

ezone

Lack of relevant information sells more parts!

That tells me it's likely a fuel control problem, not an EGR problem. Disconnecting the sensor sort of puts the computer into a base fuel map, it can't use the sensor to make corrections.
At this point, I'd be using a good scanner to evaluate live engine data before pulling the trigger on a new sensor. I'm not sure what to tell you about this part....I'd hate to tell you the wrong thing..

Ooh that's a5 wire wideband sensor, isn't it? Not cheap!
NTK? Denso? Use whatever brand sensor was installed from the factory.

ONe21

Hmmm...alright. The previous owner replaced the o2 sensor a little over a year ago (he kept pretty good records). I couldn't figure out the brand of it, but considering that the shop only charged him like $70 for it, I'm assuming it is a universal. I'm reluctant to purchase a new one as of yet.

I think I'm going to get a bluetooth OBDII adapter and download Torque on my tablet. I'll see if I can get some good data pulled and go from there. I'm not sure what the OEM brand is, but you mentioned NTK. I know they make a lot of OEM parts. I did a search using the honda part number (36531-P2M-A01) and came up with the part in the picture. If it comes to it, would this be a suitable replacement?

EDIT: http://www.oxygensensor.net/widebandsensors.php

ezone

Link fail.




EDIT: Nevermind, the link finally showed up. Go for it.

36531-P2M-A01, retail is about $550 USD through a dealer. Cheaper through dealers online.

I wouldn't trust a $70 O2 sensor at all. For only $70, it's something that is not right for that car at all! I hope the harness hasn't been hacked.

NTK and Denso are both OE suppliers to the manufacturer. The parts catalog does not tell which one made the original sensor though.

I have no idea what sensor you found, the link was bad. EDIT: Nevermind.
There is a reason I say to use only original equipment: it works.

But I don't know which brand was original to that car.

I found a cross reference chart that gives a Denso equivalent of 234-5052
Link: http://oskin.ru/pub/chrysler-dodge/manuals/Catalogues/DENSO_Oxygen_sensor_CROSS_catalogue.pdf
Page 467

And the NGK website gives NTK sensor #24300
Go for it.



Just think: After you replace with a good sensor, you will find out that the ECM bulletin DOES apply to your car, and another shop stuck in that crappy sensor because of the same code instead of addressing the true cause of the code. Plus they installed a discount part that is causing problems!

ONe21

I posted the link for the part in the thread with the picture. Sorry for jacking up my posting, I'm more of a lurker so I'm pretty terrible at this stuff.

The harness is fine, it actually has the OEM style connector.

Yeah, I figured the bulletin applied. I just wasn't sure if the ECM problem would be causing any of the symptoms I observed. I am now leaning toward the O2 sensor. One question though; does the ECM glitch affect fuel economy to any measurable degree? What exactly is going on in the ECM? Is it reading the A/F ratios wrong?

ezone

No big deal, it just didnt show right away and I didn't wait very long for it. By the time I typed the reply and went back, then I could see it.

Amazing.
I still think that's the cause of the hesitation and poor running, but I have no other way to prove it than to unplug it to disable fuel trim corrections in the ECM, and that's still just a hunch from where I sit.

The bulletin says the ECM issue does not cause any driveability symptoms, only the code. " and the customer does
No clue here. Probably not, since the bulletin states "the customer does not have any driveability complaints."

The code definition is "A slow response problem in the Primary Heated
Oxygen Sensor circuit".

Malfunction Threshold
The A/F sensor inversion cycle is 2.0 seconds or more.

The updated ECM probably is less sensitive to the code set criteria or the timespan allowed was widened from what was stated in the original threshold (copied above).

HELLifIknow, I'm not a programmer or an engineer.

It's hard enough to actually fix the headache cars sometimes.
My motto: I HATE working on cars, I like to fix them!






Copied from that other post, with a few corrections:

TSB 00-028
Applies To:


SYMPTOM
The MIL is on with DTC P1163 (O2 sensor slow
response) stored in the ECM, and the customer does
not have any driveability complaints




DIAGNOSIS

1. Use the PGM Tester to retrieve the Freeze Data.

TPS - between 3% and 7%

Yes - Go to step 3.

No -Disregard this Service Bulletin, and follow normal troubleshooting procedures.


4. Find the part number on the current ECM, and write it down.

5. Locate the part number of the current ECM under

PARTS INFORMATION, then order the appropriate
replacement ECM.

6. Inform the customer that he or she will be notified
when the replacement ECM arrives.

Current ECM/Replacement ECM:
37820-P2N-L31/37820-P2N-405

37820-P2N-A31/37820-P2N-406

37820-P2N-L32/37820-P2N-L33

37820-P2N-A32/37820-P2N-A33

ezone

Also, a scanner that can read some live engine data would at least show what some of the sensors are doing, like the ECT, MAP, and TPS, all of which are major players in fuel calculations and could possibly cause odd codes-- without setting codes of their own.

ONe21

Yeah, I just loaded Torque Pro setup on my tablet. I'm playing around with it and will hopefully be able to get some readings soon. I'm not sure what all it can do right now. I know it gives live engine data though.

ezone

Ok, good deal.
Hook it up to the car in the morning before you start it up, check the readings from the ECT and IAT. They should be within a couple degrees of each other and nearly the same as ambient temp.

Then watch what ECT does as you drive it, and watch fuel trim values (both STFT and LTFT) as you drive it. What are each doing when the car starts acting up/hesitating?

If you erase the 1163 code, will it act up before the code resets, or does it act up only after the code already came back?

ONe21

This is long over-do, but here it is. At start-up:
ECT = 52*
IAC = 32*
Ambient air temp = 53ish*
STFT = 3.1%
LTFT = -4.7%

While driving:
STFT = 7.8% (while accelerating)
= 0.08% (cruising at 35mph in 3rd gear)
LTFT = -4.7% (did not change at all)

I didn't pay attention to the ECT while driving but I'm pretty sure that it increased. If it's really important I'll do another run tomorrow. It appeared that the IAC remained at 32* the entire time.

The check engine light hasn't come back on since I originally reset it, but the car hasn't stopped hesitating. During the test run the car was hesitating the entire time.

I hope you can draw some conclusions from these numbers.

ezone

IAT= Intake Air TEmp
IAC= Idle Air Control
Very different.


I hope you were looking at IAT, not IAC. If it is, then it's 20 degrees off.
Doubt it's the cause of the problems as this input is a minor player in fuel calculations, but probably should be corrected anyway.

ECT looks accurate for the initial check if ambient was 53. But is it accurate throughout the entire operating range?


Yeah it should increase with coolant temp as the engine warms up and reaches operating temps.
The whole point of reading that was to make sure it is actually giving accurate coolant temp information to the ECM at all times. ECT IS a major player in fuel calculations.

What is the operating temperature:
What is the temp when flying down the road?
What is the temp when stopped and the radiator fan kicks on?

--------------------------------------

I still think the front O2 sensor will be the main cause of the hesitation issue though. Front O2 is the main player in fuel correction.

ONe21

I had a 4 year old in the car who was reciting the entire script from the Polar Express from the back seat, so it was hard to concentrate. I did another run tonight. As for Intake Air Temperature, the only thing I see in Torque is "Intake" and it stayed at 34-35* consistently.

The STFT was a little different tonight. While cruising at 35mph, it was fluctuating between -25% and -16%.

Also the ECT never exceeded about 90* while driving, but crept down whenever at a stop. I have a Honda thermostat laying around from a previously owned vehicle (same part number) that I could throw in.

ezone

Too low, should be near ambient at cold startup.

Bad. Taking away a lot fuel. BUT Fix other issues first.
But it is supposed to fluctuate constantly, and needs to have LTFT considered when looking at STFT.
Fahrenheit? That's dang near stone cold. You have problems. Either that engine is running ice cold, or the reading is a lie. Which is it?



Could also be false readings from your tools too, that's sometimes an issue with aftermarket scanner stuff. Just a thought.

ONe21

Yeah the "Intake" temp changed when I started the car again, it was a constant 42* during the drive subsequent to the test run.

The only LTFT reading I've ever seen is -4.7%--not sure how normal that is.

As far as ECT, it may be accurate. The dash board temp gauge never reaches the normal operating temp level. It hangs out at about 1/4 from the bottom during a long duration drive. That is why I assume that the thermostat is stuck wide open. Does this sound right?

When you say other issues are you referring to the abnormal temperature reading? I guess my first order of business would be to replace the thermostat? If that doesn't correct it, should I test the thermo reading sensor thing (not sure of the exact name) that sends info to the ECM?

ezone

Curious, what does the temp read if you unplug the sensor?
What does the temp read if you short the 2 terminals in the harness connector?

Have you tried another IAT sensor yet?

I can't tell you at this point. A number such as that is ok, but I can't tell you if it is right or not given the rest of the issues here.
All of this is easy to verify.

Grab the top radiator hose after a drive.
Can you hang on to it? Does it FEEL like 90 deg F?
Or is it way hotter than 90F?


Also, I specifically asked what the temp reading is if you let it sit still and run long enough to have the radiator fan come on.
At what (data)temp does the fan kick on?
Grab the top hose now. Is it HOT? How hot?

Is that sensor reading a lie or not?

Yes. Both of them.
Not my first move, I test first.

I've been trying to get you to do this with the data readings vs. reality here.
Figuring out if it is accurate IS the test.



I do as much diagnosing as I can--- while sitting in the comfort of the drivers seat, with a scanner. The scanner is a powerful tool, if one can put it to good use.

ONe21

I apologize for leaving out info and not being super thorough. Thanks for your patience. I'm pretty busy as of late. This car was meant to be kind of a fun project/hobby, and I'm finding less and less time for it lately.

-40*
I assume you mean, jumping the connector? If so, it was 215*
No

I drove for approximately 20 minutes today and checked the ECM Coolant temperature reading and it was 91*. After letting it sit at idle for a while, it read 83* I felt the top radiator hose and it was definitely hotter than 83*. This is just a guess, but I would say it wasn't too far off of normal operating temperature. The dash gauge was sitting at a little over 1/4 of the way up from the bottom, maybe 1/3 of the way up. The radiator fan never kicked on, but I unhooked the thermo-switch connector and jumped it and the fan kicked on.

ezone

Ok, then that reading is definitely incorrect. First try a new sensor.

That false ECT reading (far too low) makes it run super rich (likely cause of hesitation and bucking?), thus causing the negative short trim numbers as the O2 sensor tells the computer it's got too much fuel.

IAT too. Replace that too.

(Suggest you use factory sensors, someone else here had issues with low quality aftermarket.)

If there are still any issues with the readings after you replace the sensors, it would be nice if you can get an actual thermometer reading in the top of the radiator to double check the accuracy of your data readings.

Operating temp is controlled by the thermostat for the minimum, and the fan for the maximum. Stat is usually 180F, fan is around 200 or so on that engine IIRC (varies by year and models)

See what happens. That car is old enough to have flaky electronics in the ECM.
Hopefully just the sensors themselves will take care of it all.
The computer and the gauge use 2 different sensors.
Gauge uses a single wire sensor, the computer uses a 2 wire sensor. Both are in the end of the head near the distributor (If this is wrong, let me know.)
IAT is in the air filter box.

It should come on at around 200*F, give or take a little bit (like 197-212* at the fan switch). It can take a loooong time for it to come on, because any air moving through the radiator might keep it cool enough to not need the fan.

ONe21

Refer to the link for the numbers called out below.

**EDIT: I just updated my previous post with the Intake test results included.

I see several sensors, but I'm unsure of which one to replace. The one at the thermostat housing can be eliminated, because I know it is the fan switch. Then there is one underneath the VTEC solenoid to the rear of the distributor (closest to the firewall) and one that is on the opposite side of the distributor closest to the radiator (#13)--both are two wire sensors. Additionally there is one located directly underneath the distributor, but from what I can tell, it is the single wire sensor for the gauge (#11).

So I am replacing the IAT sensor and thermosensor that sends readings to the ECM? Should I also replace the single wire sensor that controls the dash gauge? I thought normal operating temperature should register slightly above halfway on the gauge, but what do you think?

ezone

That's the oil pressure switch for the VTEC.


Random pic of the temp sensors


^^^That's a lotta white space.


Erm, ECT and CTS (Engine Coolant Temperature and Coolant Temperature Sensor) are terms that can be interchangeable. Pay attention to the wires instead of the names.

You could watch your data readings as you unplug the suspected coolant temp sensor for the computer. It should also go to -40 when disconnected.

Each car seems to be a bit different. More important to know what is normal for your own car and be aware if something ever changes from your normal reading.



Your edit: Yes I did. That reading seems unusually low to me.

Can you do the same check using the coolant temp sensor wires? Does it give different readings?

There might be an underlying ECM problem here, 2 temp sensors that are skewed.
Can anyone else do the same data check on a "known good" 6th gen car and post the result?

ezone

Another check:
What is the TPS reading at closed throttle, and what is it at wide open throttle?

If you go back to jumpering the temp sensor wires, does the reading go up if you ground the jumper wire while bridging the connector terminals?

ONe21

9.4% at closed
right around 50% for wide-open

Not sure if I did this correctly, but I stuck a paper clip in and jumped the connector then I touched the paper clip to the engine block. Nothing changed when I did this, assuming that is correct procedure.

ezone

Can you get a voltage reading instead of %?
I'm looking for 0.49V closed, and somewhere around 4.0-4.5V at wide open.

Backprobe the TPS wire connector, use a real voltmeter?

50% at wide open is clearly wrong.
Yeah that was what I was looking for.


So now I gather you have 2 different skewed temp readings, and skewed TPS reading.
Did you check that TPS reading with a hot engine?

What other readings are skewed?

Get out that voltmeter!

Pull the connector off of the TPS, check voltages on all 3 terminals. Should be a Ground, a 5V, and one other for signal return.
Green/black is ground,
Yellow/blue is 5v (reference voltage)
red/blk is the signal return, or TPS voltage to the computer. (Could be either 5v or 0v when unplugged.)

I'm mostly interested in whether or not the 5v ref is actually 5v, or if it is much higher or lower than 5v.

Do the same check with the MAP sensor wiring next.
Grn/wht is ground
Yel/red is 5V
red/green is signal return.




How are the ground wires at the thermostat housing? Clean shiny aluminum? Clean shiny bolt and terminal?
That is the "GOD" ground for almost all of the engine controls and the computer. You might run an extra ground wire to it direct from the battery?

ONe21

EDIT: I was actually giving numbers for the MAP sensor. Here are the correct readings.

Ground: 12.35v
Yel/Red: 5.05v
Red/Green: 3.86V

Alright here is the TPS:

Ground: 12.35
Yellow/Blue: 5.05
Red/Black: .49 (Closed) 4.54 (Open)

I didn't get a chance to check the grounds. I'll look at them tomorrow.

Yes the engine was hot when I checked yesterday.

ezone

So far so good.
Was that check done with the key on?
Was this done with the TPS connector unplugged or connected?

Ohm and volt check it too using the Neg batt terminal to same wire. (it will change with key on or off ..... Should stay at 0 volts, and maybe only a few ohms resistance with the key off, and usually stay under 100 ohms with key on.)

Something is screwy with this if the connector was still plugged in to the TPS.
If the connector was unplugged from the TPS, then moving the throttle can have no effect and the 2.92 might be ok, but it's not what I was expecting to see. I was expecting 0 or 5v.. (I wish I could check a similar known good car to compare)

If connected, the red/blk is supposed to be the signal return, it should have corresponded with your 9%-50% TPS readings on the scanner, and it should have varied when you changed throttle position.

I guess I'm trying to figure out if the scanner is displaying what is truly going on at the sensor, or if this could be a flaky ECM issue that isn't reading inputs correctly.


Sounds good. Was it connected or not?

God invented flashlights for this very reason LOL



I was hoping to be able to check while it is acting up, since you indicated it is worse when hot.

ONe21

God invented the flashlight?

Anyway, I think you were already responding when I edited my last post. I was actually giving you the MAP sensor numbers... yeah. I corrected everything though.

ezone

Oh yeah, God invented errthang.
Sort of a Mad Libs thing. "___ invented the ___." Fill in the blanks.

It's kinda something I used to tell the kid, and I still say it at work all the time.

You must have been, and I must have gone to sleep immediately after.

Ok, you were using the pos batt post when checking the ground?
Engine not running, this looks about right.
Engine idling, hot, no loads, MAP is usually somewhere between 0.80-0.90v on a good running engine @ sea level.

This output voltage appears perfect.

If the ECM can't see the same voltage, it may have a problem.
I'm thinking 50% opening would be somewhere around 2.5v as seen internally at the processor.
This coincides with the skewed temp sensor readings.

Got any way to see this data on a scanner that can display readings in volts instead of percentages? I'd really like to know what the computer is seeing, if it doesn't agree with your voltmeter then we know that's a big problem..

I'm curious if you can watch data and see anything change when a redundant ground is added.

I'm starting to wonder if one can manipulate the temp of the ECM and see if that affects any readings.

ONe21

It's incorporated into a wire loom, while the part that isn't in the loom has black heat shrink plastic over it. The very end looks pretty clean--no apparent corrosion. It is fastened tightly to the Tstat housing.

I didn't see anything anywhere that would let me change it in Torque, but I did get a better reading. I don' think it updated fast enough for me to catch it last time. I put the car in 4th gear at about 20mph with wide open throttle for about 3-4 sec. It held steady at 85%. That seems a lot more accurate.

All tests were done with TPS and MAP unplugged, except for the ones that required back probing. Also, car was off and ignition in II position.

ezone

Yes, there are several ground wires attached to the same bolt.. It's a really important ground.
That's why I mentioned running a redundant ground to it.


You'd probably need to take the thing apart and/or volt drop check it under a load. Aluminum corrodes and isn't always visible, and the stat housing is pretty darn far away from real ground points on the engine and trans.

Rats.
Anyway, 85% sounds more reasonable. Does it give the same reading with the engine not running (key ON) and wide open throttle?

Any way to get hold of a different scanner or anything that can give engine data in a different format? Sure would be nice to be able to compare apples to apples.
There is no real accurate conversion table for % to Volts.

Thought: When you floor the gas pedal, does the throttle actually go completely wide open?

If you operate the throttle cable by hand, does it give any higher reading?
I'm still stuck on the skewed temp sensor readings. I don't suppose you've checked new sensors yet?



/forest--->trees