Either the computers running our cars and their numerous interfaces with sensors and electronic devices are too complex for the average mechanic to fix, or perhaps it's software bugs in the ECU code, but I alone know of 4 cases of new computer controlled cars - (I guess they have been that way since the mid 70's) which have simply died during driving and were taken to the dealer for a diagnostic which returned no trouble codes and were pronounced "OK". The problem, nevertheless, remained.

Case number 1 was neighbor's 2001 Lincoln which frequently died on the highway - he took it to the dealer - they found nothing wrong - the problem continued, they eventually gave him another car.

Case number 2 was my uncles new GM truck. It died on him on the way down from California - it just quit. He had it towed to the dealer, the used their diagnostic equipment on it and turned up nothing. The problem persisted with no help from the Dealer.

Case number 3 - a user on this board - Nismo98 has a 98 Max that chokes and dies when he slows down or stops. He took it to the dealer and their diagnosis found no problems. As far as I know, he still has the same problem.

Case number 4 - My car - A 96 Maxima Auto. It chokes and dies at stop signs/lights, and the tach surges from 1 to 3K rpm when driving. The incidents are apparently totally random, and sometimes it's fine for several small trips. The thing that makes it worse is that the ECU is throwing no diagnostic trouble codes. The dealer checked it with their hand-held computer. They found no stored trouble codes and nothing wrong - except the idling adjusted too high. They reset the idle, and pronounced it fixed. As soon as I drove it off the lot and turned the 1st corner - the car died and I had to restart it 3 times. I hurried back to the dealer and said it's still messed up. I said it needs to be fixed. I drove it with the Service man and he saw it stall and die once. It was fine after that. It's totally intermittent, but if the dealer can't fix it, who can? Should I just replace the AACV - ICV, or replace the MAF sensor (I messed it up when installing a cone filter, but fixed it and it's been working fine for 2 months). Or is this a more widespread problem with an undocumented software bug in the code of the ECU, or many ECU's, or does some ECU manufacturer know more than they are telling? If I could just have someone find out what's wrong with my car, I'd be happy.

Van

 

don't forget that the problem could be mechanical, in which case if there is no sensor on that part, the computer would not give you a error code, also, they might not have an error code for this, and lastly, being mathematician I have to tell you that these cases are fairly RARE, meaning, that if you look at the bell curve, most of the cars fall in the normal ecu range, but then there are a few of those on the extreme that might have a problem, that's just a fact of stats, there is always a probability for that to happen

I know I'm not really giving you the answer that you want, but most cars don't have this problems, and those that do happen once in a while, so those are the black sheep of cars

 

Man, I thought I was paranoid.

 

Perhaps you are blaming the electronics when the real problem is the human element.

Some technicians believe that "No Code = No Problem". This is a mark of laziness or incompetence. No Code means the Engine Control Module didn't notice anything wrong, but the ECM doesn't check everything. The 4Gen Maxima does not have a fuel pressure sensor and consequently the ECM does not check fuel rail pressure. If it did, we would have Diagnostic Trouble Codes for ...
- Plugged Fuel Filter
- Weak Fuel Pump
- Defective Fuel Pressure Regulator
- etc.

Your car may have a fuel problem which the ECM cannot diagnose. Your service professionals are not exhibiting diligence or imagination. Please ask them to make a fuel pressure test. If you prefer to do your own "wrenching" I will provide detailed instructions.

I'll take a wild guess... your Maxima may be suffering from a bad fuel pump relay.

 

Daniel,

Thanks for your timely response. While I was still reading your message, I called the Nissan dealer and spoke with the service tech and actually read him part of your message. I mentioned the fact that there is no fuel pressure sensor and checked on running a fuel pressure test. His reponse was that the computer would have picked up a lean mixture code if it were a faulty fuel pump, clogged filter, or pressure regulator. I didn't argue, but I will have them check the fuel pump relay if they come up with nothing else. They put what the call a "consult" on it - a hand-held computer which connects to the ECU while they are driving the car. They drove it 16 miles and it didn't act up once. At their suggestion, I'm letting their mechanic connect it to the consult, drive it 50 miles to his house, and drive it back the next morning. Hopefully they will be able to get it to act up enough to where the computer will throw some kind of a code to diagnose the problem. If they still can't find the problem, I may just replace the fuel pump relay myself, as it seems the most logical suggestion I've heard yet.

Thanks,
Van

 

Go to
http://web2.airmail.net/scutchen/max_faq/engcodes1.gif and
http://web2.airmail.net/scutchen/max_faq/engcodes2.gif to read the Diagnostic Trouble Codes as they appear in the factory service manual. Try to find a "lean mixture code". I didn't notice one.

The fuel pump relay is mounted behind the plastic kick panel to the left of the driver's feet. It is a Type 1M relay in a blue plastic case. All blue relays are alike, so you may exchange the fuel pump relay with any blue relay found in the passenger cabin fuse block or the two relay boxes in the engine compartment. This will allow you to test the "bad relay" hypothesis without the expense of purchasing a new relay.

 

I think I may know the answer to your friend with the GM Truck.

On his way to Cali, did he go over any mountains?

I had a '92 Camaro, and on my way from Phoenix to Seattle, I drove through Grant's Pass. After coming down off the mountain, the car would not idle...If it idled, it would sputter and die.

I took it to a friend of mine who has worked on GM vehicles for years. He said the problem is that the computer adjusted the air/fuel mixture when it was at high altitude, (more air because the air is thinner) then did not re-adjust when I got back down below 1000ft. This caused too much air in the fuel mixture, bad performance, and lack of idling.

He reset my Throttle and ECU, and it never happened again.

But that's why I bought a Nissan.....GM products suck.
Except maybe Caddilac.

IanS

 

Nice...

I love catching people is a deep pile of BS.

 

Whoa there....

I'm just going by what my mechanic friend told me...From my point of view, there is no BS....I know enough to make sure my car is maintained well, but not enough to know if that was BS.

He said that was the problem, and he fixed it...that's all I cared about....no reason to disbelieve him.

IanS

 

i think mzmtg is referring to the mechanic van talked to, not you. dbm, proved the mechanic wrong. the mechanic is BSing, not you.

 

Oh...heh

Sorry dude.

(Putting foot in mouth)

IanS

 

Daniel,

If you would be so kind as to provide the "Wrenching" instructions for doing a fuel pressure test, it would be much appreciated. Here's the latest scoop: the Nissan Master mechanic drove the car home last night with a $5000.00 computer "consult" connected. It's my understanding, after speaking with the service person, that this diagnostic device has an expanded set of diagnostic codes, of a much wider range than the somewhat limited on-board computer. Hence the "lean fuel" mixture he spoke about. 100 miles later, the car threw no codes which indicated any defective components, except the one thing it did was produce a momentary secondary ignition problem, but apparently not enough to mandate the replacment of any components. Also the first day they drove it, he said the TPS showed a lower number than it should have, as if someone had let off the accelerator, even though they hadn't. Again, though, the reading did not warrant replacement of a defective sensor or component. So we are still up in the air. The price of a fuel pump relay is $15.60 and the price of a fuel pressure regulator is $75.08. What the heck, I have to start somewhere. I may just pick up the car and attempt to fix it myself, unless you can provide a way I can test the fuel pressure system and save the expense of possibly needless new parts.

Thanks,
Van

 

I'm a D-I-Y, not a professional... but I think there has been some misunderstanding or miscommunication. Maybe the dealer technician is oversimplifying his explanations, underestimating your level of understanding.

CONSULT is a valuable diagnostic device, better than anything we driveway mechanics have. However, it has no more and no fewer Diagnostic Trouble Codes. Every Code which exists is listed in the factory service manual. The FSM is chock-a-block full of instructions for using CONSULT and nowhere does it mention any "secret" CONSULT-only Codes. I've read some Nissan technician training publications and nowhere do they mention any "secret" CONSULT-only Codes.

The advantage of CONSULT is that it can make voltage readings while the engine is running, while the car is in motion. The DTC readout we driveway mechanics do is always historical in nature. A Code may tell us (for instance) that at some time in the past there was a misfire in cylinder #4. It doesn't tell us whether that event was last Thursday or today. It doesn't tell us whether that misfire occurred 50 times or 6000 times. It doesn't tell us if the fault is intermittent or solid. The CONSULT allows the skilled technician to watch these things happening, in real time.

CONSULT is a wonderful diagnostic tool. I wish I had one. However, the fact remains that the 4Gen Maxima has no fuel rail pressure sensor. Therefore there is no way for CONSULT to detect a fuel pressure problem. It doesn't matter if you are using a Radio Shack voltmeter, a Fluke 87 DMM, a Tektronix Oscilloscope, or a CONSULT. You can't measure a signal which isn't there!

 

The crux of this problem is that your engine is plagued with an intermittent problem, one which is so brief or so subtle that it escapes the dealer-only CONSULT diagnostic tool. It is a daunting task for the driveway mechanic to succeed where the dealer has failed, but let's try... and at the same time, try to save you from paying for any more diagnostic labor time at the dealer.

The fuel pressure test is a standard diagnostic test. You will find instructions and a diagram or picture in the Chilton repair manual (page 5-3) and the Haynes repair manual (page 4-3,4). The following instructions are a variation on the test shown in the books.

Bleed the fuel rail pressure by pulling the fuel pump fuse and making several attempts to start the engine (it won't start). Install a T-fitting in the flexible rubber fuel line just downstream of the fuel filter. Attach a 3-foot length of high pressure rubber fuel line to the T, and connect a pressure gauge to the end of this line. Route the line in a way which allows you to close the hood. Place the face of the gauge against the outside of the windshield and secure it with duct tape.

Install the fuel pump fuse. Start the engine. You should see the pressure maintain a stable 34 psi. After turning the engine off the fuel system should hold pressure for a long time, 24 hours or longer. If the residual pressure bleeds off in only an hour that is a sign of a problem... bad fuel pump check valve, injectors with an internal leak, bad fuel pressure regulator, or an external fuel leak.

This rig provides some useful temporary instrumentation. It will permit you to watch the fuel pressure as you drive and note whether the choke-and-die symptom coincides with a momentary fluctuation in fuel pressure. As you drive in your normal manner, keep an eye on the fuel pressure. Keep an eye on the road too!

The normal range of fuel pressure is 34-43 psi, depending on engine load.

When the symptom presents, note the pressure gauge reading. Post your findings on this thread and we can go from there.

 

Here's another test I recommend... test your Engine Coolant Temperature Sensor. This is easy enough, requiring only an ohmmeter.

On the 4th Generation Maxima engine the Engine Coolant Temperature Sensor is located in the water outlet tube close to the engine end of the upper radiator hose. There are two sensors in that area. The one nearest the hose is for the dashboard temperature gauge. The ECTS, the one you're interested in, is right next to the the gauge sending unit. There is a good picture of these sensors in the Haynes repair manual on page 3-7.

With the engine cold, disconnect the ECTS and measure its resistance. Reconnect the ECTS, start the engine, run it until fully warmed up. Stop the engine, and repeat the resistance measurement. The "warm" reading should be a much lower value than the "cold" reading.

The manual gives these specs:
Engine coolant temperature 68F, ECTS resistance 2.1 - 2.9 Kohms
Engine coolant temperature 194F, ECTS resistance 0.24 - 0.26 Kohms
(Roughly a factor of ten difference.)

Examine the wiring and connector which attaches to the ECTS. As you can see by the resistance specs, high resistance means "cold engine" and the Engine Control Module responds by calling for a rich mixture. You may imagine what would happen if the ECTS signal were to abruptly and momentarily change from 0.25 Kohms to a much higher value such as might be caused by a loose or dirty connection.

 

Hello All,

I am new here and I must first of all say that I have gotten
a lot of useful info re my Bucking Max, thanks All.

Although my max is a 93 I feel that the problem(sa) I'm having are inherited by the 94's as well. Yes I have the clock problem etc but the biggest headache is the stallin especially when this occours on a highway at speed thing just dies and not startin on demand. I'll cut a long story short.

- Problem Started about 5 months ago - Dealer / no help
- Talked to various auto places - lots of suggestions

Suggestions as to cause(s) of problem:
Fuel Pump, fuel Relay, Sending unit

The reason for these suggestions: the fuel pump can be heard when the key is turned as the line is pressurised following which the car would start however at random times when the key is turned no pump noise is heard and ergo car no start. Cycling(not actually turnin the engine over) the key several times seems to get the Electrical signal to the pump. Ah Ha u say bad fuse, or some such faulty wires etc.

Action Taken:
Changed Sending unit involves removal of rear seat but problem persisted

Changed fuel relay (in 93's the rfuel pump relay is in the trunk and green in colour) no relief.

Then I thought I'd bite the bullet and buy a new fuel Pump, instead however I pulled out all the fuses related to fuel or not checked all electrical relays etc. problem continued. And now the piece de resistance might have been a brain wave or fade, was to swap the Fuel pump relay with the ECU relay. same colour green...BINGO dont ask me why all I can say now MY car starts on demand and had not died since so will get correct replacement ECU relay from dealer as the fuel pump relay get a little warm. The proof in this fix for me was to then swap back the two relays and result was the car did not start and after some words of wisdom to my max I switched em back..vaarroomm.

The ECU relay is located on driver side in engine comp to right of battery.

Van I went long on this but maybe your problem might just be that simple maybe

MaxxedOut

 

Daniel,

Thanks for the diagnostic procedure descriptions. I picked up the car from the dealer on Friday. They had driven it several times that day and it would not act up again. I tried to get further clarification on the ECU codes vs Consult codes issue, the service man told me that only "hard" codes are stored in the on-board ECU and will trigger the check engine light, and the "softcodes" are not stored and will not activate the engine light. The "softcode" troublecodes can only be picked up by the consult if a problem occurs while driving, but I won't see softcodes manifested in the check engine light. I asked the dealer for a service manual. What they said made me drop that in a hurry. It's $150.00 and 6 - 8 weeks for delivery. Obviously they discourage DIY mechanics. I'll have to keep looking if I really want one. Anyway, I decided to begin where I do when I'm at work. As a computer systems analyst, I often ask users when they complain that their PC is "messed up" what the last thing they did was before the trouble started. Well, I asked myself the same question and the answer was I took those lousy screws out of the unservicable MAF sensor. I found a used one in a salvage lot (90 day warranty) for $75.00, which isn't too bad. They range from $50 - $300. Anyway, I replaced the MAF sensor and have been driving the car for 2 days now with no problems. I'll give it a week before I pronounce it fixed. I have printed out your instructions for plan B if this isn't the problem, but I have a hunch that's what it was. Why it didn't kick out an Engine trouble code is beyond me, but then alot of things are, and electronics can be unpredictable. Thanks for all your help and advice. I learned a lot during this experience - like don't make modifications on a car that's working fine if you don't know what your doing.

Van