Nissan Forum banner

41 - 47 of 47 Posts

·
Registered
Joined
·
35 Posts
Discussion Starter #41
****************************
* FIXED! I think,
****************************

Well looks like the new Hitachi OEM MAFS fixed things!!

…at least most things, there are still a couple of things that look at little abnormal, but I’m not sure if they are transient and might go away as the ECM learns the new MAFS.

So, Friday morning I installed this third MAFS, this one an OEM Hitachi, which I had just got from eBay for $200 (the dealer wanted $550).

The idle got better immediately, both during engine warmup and when engine warmed up. ECM entered closed loop normally after 2.5 minutes of idling. I immediately noticed the STFT was not doing the 0 to +25% ramp any more, though I did have a strongly positive STFT (+15 to+20%) with an LTFT of 0%. I let it idle for about 6 minutes past the point it entered closed loop and the situation remained the same STFT= +15 to+20% and LTFT = 0% while idling. There was actually a bit of STFT ramping, but much narrower, between +15% and +20% and not nearly as periodic and regular as the one I had with the previous MAFS.

Then I went out on some traffic, and eventually cruising some on the Freeway. When cruising, I had a LTFT of +9.3% and a negative STFT between -7% and -14%, so total net trim was close to 0%. After about 15 minutes of cruising, both STFT and LTFT went closer to 0%.

Also, the idle trims changed eventually. Idle STFT started oscillating closer to 0% (between 0% and +3%), but LTFT at idle adjusted to a rather strong +9.38%

There is no crazy MAFS signal after 30-40’ and the associated rough idle. The idle is very smooth. Revving up under no load seems healthier too. The O2 sensor is doing its rich-lean-rich-lean cycling as one would expect. The engine’s low RPM torque seems to have improved too. Transitions from deceleration to idle and then reacceleration are smoother too.

Unlike before, deceleration seems to be running open loop, as you suggested is normal jdg, with STFT going to 0.00% on deceleration. Full power (full open throttle) seems to be running open loop too.

All in all, drivability is good again!

The only thing I’m still a little unsure about: Why do I have this rather strong positive LTFT of +9.38% at idle. LTFTs at other engine loads seem to be closer to 0%.

I’ll wait a few more days, see what other observations I can make, and then I’ll write a summary post, so that the next unlucky chap that is tormented by two successive heat sensitive MAFS does not have to read through all 40 posts of this tread.

Meanwhile, I’ll enjoy my newly rediscovered smooth ride for a few days!!!

Regardless of what other problems might remain, this is a significant improvement!!

Thanks!!
 

·
Registered
Joined
·
3,193 Posts
Well looks like the new Hitachi OEM MAFS fixed things!!
Hate it when that happens...Or do I like it? I forget... :D
2 MAFS's with the same stupid problem. Who would've thought.

ECM entered closed loop normally after 2.5 minutes of idling.
Does your vehicle have heated O2 sensors? If it does, it should go into closed loop within a minute easily, even on a bone cold engine.
meh...let it ride. Good enough. My old '95 Grand Prix without heated O2 sensors goes closed loop within a minute in the middle of winter. But, an apples and oranges comparison here.

Unlike before, deceleration seems to be running open loop, as you suggested is normal jdg, with STFT going to 0.00% on deceleration. Full power (full open throttle) seems to be running open loop too.
As they should. I think the numbers go to zero's because there aren't really any numbers for it to crunch in that situation.
STFT is generally what the ECM is thinking and adjusting for right now and added in accordingly. The LTFT is more permanent and is added in (or subtracted out) even under open loop situations...which is probably the only reason why a vehicle that drops an O2 sensor can still function fairly well.

The only thing I’m still a little unsure about: Why do I have this rather strong positive LTFT of +9.38% at idle. LTFTs at other engine loads seem to be closer to 0%.
Tiny vacuum leak, slightly plugged injectors, fuel pressure a tad bit on the low side, altitude, temperature, gas quality. A hundred different things for that one. Although, I'd be more willing to go with the vacuum leak since at idle a small vacuum leak is a larger proportion of all of the air going into the engine, whereas under load, that small vacuum leak becomes an insignificant part of the whole. Regardless, until the ECM flags a code, I wouldn't worry about it. That's why they have fuel trims...to compensate for small variations in this, that, and the other thing, wear and tear.
 

·
Registered
Joined
·
1 Posts
Thanks so much!

I just bought a 98 Nissan frontier with the same problem although I have had no loss of power at speed. I found this thread this morning and just read through every post. I'm incredibly impressed with the level of troubleshooting that was implemented! You guys have given me all the tools (hopefully) to tackle this problem.

After owning a 95 Volvo 850 turbo for two years I can attest to how incredibly helpful good forums can be and since I don't think I've owned a car that doesn't have a good forum to follow. Seriously stuff like this keeps me from just banging my head against the wall on my own.

Thanks guys, keep up the good work.
 

·
Registered
Joined
·
35 Posts
Discussion Starter #44
=================================================================
* SUMMARY

98 Frontier 4WD with two successive bad MAFS
=================================================================

Now more than a week after new (third) MAFS, truck is still running fine, so, as promised, here is my final summary…

If you are unlucky to have a bad MAFS (Mass Airflow Sensor) and have the further misfortune of buying another one (especially a non OEM) that has a similar problem…

…then you may start pulling your hair out, and in desperation may find this thread…

So to spare you having to read all 40+ posts on this thread, or at least to put those 40+ posts into context, here is the summary of what happened and how it may apply to a similar situation:



My truck’s idle had been deteriorating a bit, but nothing that someone would not be tempted to attribute to an aging vehicle (98 Frontier 4WD with 160k miles).

Then after a few months, finally, after a four hour drive, the engine started occasionally stalling on idle when warm. The idle also got more noticeably and obviously pathologic. Also, once in a while I would experience engine “hiccups” short ¼ second loss of power at highway speeds, but it was infrequent. A couple of hiccups and then ok for another ½ hour or so.

What had happened, is that the MAFS (this was the original OEM Hitachi MAFS) signal had become occasionally unstable. The signal was spiking with noise when pressure was applied to the MAFS housing. The noise became more pronounced as the MAFS got warmer, hence engine idle was getting worse as the engine was getting warmer and warmer, and that is why it probably got to the point of stalling after a long drive on a hot day. Of course I did not know that at the time. This is a diagnosis I came up with later. Hindsight is 20/20.

So I took the truck to a local garage. It was 4th of July, I was 250 miles from home and this one man mechanic shop was the only open. Very nice fellow in retrospect, I owe him some money. He correctly diagnosed that the MAFS signal was noisy, but attributed the issue to the MAFS electrical connector, because pressing on the connector would cause the engine to stall, but sometimes only. He tried to fix the connector, and since finding a new connector on the 4th of July was out of the question he did not even want me to pay him. As I said, when I go back I’ll give him a gift card because although he was not able to fix the problem his initial diagnosis was 90% correct.

I futzed around with the connector myself once I got home, spending lots of time on it (something the garage cannot afford) and came to the conclusion that the noise in the MAFS signal was coming from the MAFS device itself, not the connector. This is because when I applied pressure to the MAFS housing I could see the noise (with a voltmeter initially and then confirmed with a borrowed scope) and after many tries I discovered that the pressure sensitivity got worse with rising temperature (engine compartment warm). When I mean pressure, I mean a lot of pressure, probably 10-20 pounds of pressure on the housing. When pressed, the MAFS would create signal spikes, even when the engine was not running, simply with the ignition on.

I’m saying this because you may also have the same Hitachi MAFS, and it could be that this is one of the main modes in which they fail (speculation here, I’m only one data point). There is a lot of circuitry inside these MAFS and so it is possible that some connection starts going bad and becomes sensitive to mechanical and thermal deformation. BTW, the circuitry inside is submerged in some sort of gel, so attempting to fix such a device seems rather impossible (I tried).

Was I sure at this point that MAFS was the culprit for all my issues? No. Why? Because a typical test in these situations is to disconnect the MAFS, so that the ECM flags the device as bad and goes into safe mode operation. If that fixes the problem then it’s a bad MAFS. I did that, and still had a rough idle, and more or less the same rough idle that I had before, so I was skeptical that MAFS was the true culprit. In retrospect what was happening (I think) is that the MAFS was bad and so its bad signal had altered the ECM’s parameters (ECM attempted to adjust), so when I disconnected the MAFS, the ECM continued to run with those bad parameters and the idle was still rough (with a very lean mixture in retrospect).

But I did not know all this at the time and, more importantly, the OEM MAFS at the dealership was $550, so changing it on a suspicion seemed like a costly gamble.

So, I reckoned, I’d just buy a cheap after market Chinese MAFS for about $30 and if that seems to more or less fix the problem, then I’m sure it’s the MAFS and then I’ll buy the OEM part, possibly online where the OEM device was apparently selling for around $250.

Well, trying the cheap $30 MAFS turned out to be a bad decision. To make this story a little shorter, this cheap MAFS …WAS ALSO HEAT SENSITIVE, albeit in a slightly different way. Its signal never spiked (signal was normal and responsive to RPM changes) until the engine compartment got warm, after about 30-50minutes of driving, depending on traffic. Once it started misbehaving, the signal spiked a lot. But the overall behavior was similar to the original bad MAFS: The erratic behavior when the device finally warmed up would teach the ECM bad habits (make it alter its learned operating parameters in an attempt to adjust). Then, even when the device was operating correctly, the bad ECM parameters learned would still cause a rough idle (rough in a different way though).

At this point, I gave up and took the truck to a local but reputable garage close to my work. They were puzzled since they could not explain the behavior. In retrospect, they tested the truck as engine was warming up and then also when it warmed up, but did not wait the full 30-50’ that it took for the MAFS to misbehave (who knew at that time that it took a whole 30-50’ for this second cheap MAFS to start misbehaving). So what they saw was the messed up ECM parameters (eg. Long term fuel trim) and an ECM unwilling to adjust. Little they knew that these were the parameters the ECM was adjusting to when the MAFS was eventually going crazy after 30-50’.

I then took the truck to the dealership, where they apparently first drove the truck, giving the heat of the engine plenty of time to reach and warm up the MAFS. Then, when they took measurements, they saw a crazy MAFS signal and wondered how does the truck even run?

Hearing three different stories from three different garages with two different MAFS, I decided to test the MAFS myself with a scope. Unfortunately I did that before the 30-50’ warmup period and saw a normal signal. Just like the second garage did.

I continued to debug and finally, two months later I came to the final conclusion that both the original OEM MAFS had become heat and pressure sensitive and that the new MAFS was just heat sensitive. You can read the history details in the previous posts if you want.

I bought an OEM Hitachi MAFS online for $200 and all the idle problems went away. The overall engine behavior seems to have also reverted to 50k miles younger, better torque at low RPM and smother transitions from acceleration to deceleration and vice-versa.

Especially,

Thanks a lot to all who helped with the problem, especially jdg who had the patience to hang around until the end! I might have given up, or would have at least taken me much longer to figure out without the guidance I got from this forum.


========================
P.S. I have one more problem turns out (small manifold crack seems to be affecting O2 sensor readings) but that is a completely different problem, so I’ll post another thread.
 

·
Registered
Joined
·
3,193 Posts
here is my final summary…
Summary? How about "Short Story" ? :D

Thanks a lot to all who helped with the problem, especially jdg who had the patience to hang around until the end! I might have given up, or would have at least taken me much longer to figure out without the guidance I got from this forum.
It also helps out a bunch when the guy asking the questions (that be you) is willing to do the legwork, not to mention go way above and beyond, to troubleshoot the problem and has a decent understanding of "things" in general vs. trying to fit what "a friend says" or "he read somewhere on the intardnets that" to his/her particular situation.

P.S. I have one more problem turns out (small manifold crack seems to be affecting O2 sensor readings) but that is a completely different problem, so I’ll post another thread.
Exhaust manifold cracks will do that. Those same pulses that mess with the MAFS of course propagate thru to the exhaust side of the house and can easily draw fresh air thru those cracks and try to pull the O2 sensor lean.
 

·
Registered
Joined
·
1 Posts
=================================================================
* SUMMARY

98 Frontier 4WD with two successive bad MAFS
=================================================================

Now more than a week after new (third) MAFS, truck is still running fine, so, as promised, here is my final summary…

If you are unlucky to have a bad MAFS (Mass Airflow Sensor) and have the further misfortune of buying another one (especially a non OEM) that has a similar problem…

…then you may start pulling your hair out, and in desperation may find this thread…

So to spare you having to read all 40+ posts on this thread, or at least to put those 40+ posts into context, here is the summary of what happened and how it may apply to a similar situation:



My truck’s idle had been deteriorating a bit, but nothing that someone would not be tempted to attribute to an aging vehicle (98 Frontier 4WD with 160k miles).

Then after a few months, finally, after a four hour drive, the engine started occasionally stalling on idle when warm. The idle also got more noticeably and obviously pathologic. Also, once in a while I would experience engine “hiccups” short ¼ second loss of power at highway speeds, but it was infrequent. A couple of hiccups and then ok for another ½ hour or so.

What had happened, is that the MAFS (this was the original OEM Hitachi MAFS) signal had become occasionally unstable. The signal was spiking with noise when pressure was applied to the MAFS housing. The noise became more pronounced as the MAFS got warmer, hence engine idle was getting worse as the engine was getting warmer and warmer, and that is why it probably got to the point of stalling after a long drive on a hot day. Of course I did not know that at the time. This is a diagnosis I came up with later. Hindsight is 20/20.

So I took the truck to a local garage. It was 4th of July, I was 250 miles from home and this one man mechanic shop was the only open. Very nice fellow in retrospect, I owe him some money. He correctly diagnosed that the MAFS signal was noisy, but attributed the issue to the MAFS electrical connector, because pressing on the connector would cause the engine to stall, but sometimes only. He tried to fix the connector, and since finding a new connector on the 4th of July was out of the question he did not even want me to pay him. As I said, when I go back I’ll give him a gift card because although he was not able to fix the problem his initial diagnosis was 90% correct.

I futzed around with the connector myself once I got home, spending lots of time on it (something the garage cannot afford) and came to the conclusion that the noise in the MAFS signal was coming from the MAFS device itself, not the connector. This is because when I applied pressure to the MAFS housing I could see the noise (with a voltmeter initially and then confirmed with a borrowed scope) and after many tries I discovered that the pressure sensitivity got worse with rising temperature (engine compartment warm). When I mean pressure, I mean a lot of pressure, probably 10-20 pounds of pressure on the housing. When pressed, the MAFS would create signal spikes, even when the engine was not running, simply with the ignition on.

I’m saying this because you may also have the same Hitachi MAFS, and it could be that this is one of the main modes in which they fail (speculation here, I’m only one data point). There is a lot of circuitry inside these MAFS and so it is possible that some connection starts going bad and becomes sensitive to mechanical and thermal deformation. BTW, the circuitry inside is submerged in some sort of gel, so attempting to fix such a device seems rather impossible (I tried).

Was I sure at this point that MAFS was the culprit for all my issues? No. Why? Because a typical test in these situations is to disconnect the MAFS, so that the ECM flags the device as bad and goes into safe mode operation. If that fixes the problem then it’s a bad MAFS. I did that, and still had a rough idle, and more or less the same rough idle that I had before, so I was skeptical that MAFS was the true culprit. In retrospect what was happening (I think) is that the MAFS was bad and so its bad signal had altered the ECM’s parameters (ECM attempted to adjust), so when I disconnected the MAFS, the ECM continued to run with those bad parameters and the idle was still rough (with a very lean mixture in retrospect).

But I did not know all this at the time and, more importantly, the OEM MAFS at the dealership was $550, so changing it on a suspicion seemed like a costly gamble.

So, I reckoned, I’d just buy a cheap after market Chinese MAFS for about $30 and if that seems to more or less fix the problem, then I’m sure it’s the MAFS and then I’ll buy the OEM part, possibly online where the OEM device was apparently selling for around $250.

Well, trying the cheap $30 MAFS turned out to be a bad decision. To make this story a little shorter, this cheap MAFS …WAS ALSO HEAT SENSITIVE, albeit in a slightly different way. Its signal never spiked (signal was normal and responsive to RPM changes) until the engine compartment got warm, after about 30-50minutes of driving, depending on traffic. Once it started misbehaving, the signal spiked a lot. But the overall behavior was similar to the original bad MAFS: The erratic behavior when the device finally warmed up would teach the ECM bad habits (make it alter its learned operating parameters in an attempt to adjust). Then, even when the device was operating correctly, the bad ECM parameters learned would still cause a rough idle (rough in a different way though).

At this point, I gave up and took the truck to a local but reputable garage close to my work. They were puzzled since they could not explain the behavior. In retrospect, they tested the truck as engine was warming up and then also when it warmed up, but did not wait the full 30-50’ that it took for the MAFS to misbehave (who knew at that time that it took a whole 30-50’ for this second cheap MAFS to start misbehaving). So what they saw was the messed up ECM parameters (eg. Long term fuel trim) and an ECM unwilling to adjust. Little they knew that these were the parameters the ECM was adjusting to when the MAFS was eventually going crazy after 30-50’.

I then took the truck to the dealership, where they apparently first drove the truck, giving the heat of the engine plenty of time to reach and warm up the MAFS. Then, when they took measurements, they saw a crazy MAFS signal and wondered how does the truck even run?

Hearing three different stories from three different garages with two different MAFS, I decided to test the MAFS myself with a scope. Unfortunately I did that before the 30-50’ warmup period and saw a normal signal. Just like the second garage did.

I continued to debug and finally, two months later I came to the final conclusion that both the original OEM MAFS had become heat and pressure sensitive and that the new MAFS was just heat sensitive. You can read the history details in the previous posts if you want.

I bought an OEM Hitachi MAFS online for $200 and all the idle problems went away. The overall engine behavior seems to have also reverted to 50k miles younger, better torque at low RPM and smother transitions from acceleration to deceleration and vice-versa.

Especially,

Thanks a lot to all who helped with the problem, especially jdg who had the patience to hang around until the end! I might have given up, or would have at least taken me much longer to figure out without the guidance I got from this forum.


========================
P.S. I have one more problem turns out (small manifold crack seems to be affecting O2 sensor readings) but that is a completely different problem, so I’ll post another thread.
I registered to be on this sight to thank you so much for your thorough troubleshooting and accounting of same. I was able to align your symptoms with what my father and I were witnessing on our truck. We then opted to purchase a new OEM MAF and the truck now runs like new. It is rare to see a thread of any kind where follow through is so thorough and complete. Thanks again for helping reduce our downtime!
 
41 - 47 of 47 Posts
Top