Low voltage, Knock Sensors, or ?

[fastenuf]

With apologies if something like this post is already up, my searches have not revealed it.

I have a 94 Q45T with relatively low miles (125k) in near perfect condition. It has become increasingly sluggish in the last couple of years and of course I am seeking a solution. Took it to the dealer who told me that consult could not read the ecu. Bought a used ecu on ebay and the car ran better for a while but eventually got to about the same place performance-wise. Looking for a reminder on re-setting the ecu, I came across one of this forums articles on knock sensors, and the symptoms fit me to a "T" so I replaced them. Car is definitely running better but still returning a '34' code on the ecu. Today, in some hot stop-and-go driving, I noticed some very strange behavior - at slow speed the power steering would come and go, and eventually my aftermarket radio shut down on low voltage and the indicator lights on the dash (trip meter, AC settings) were dim. After a bit of highway driving the radio came back up and the lights were brighter, and I got home with no further issues. I had no CEL or low battery warning on the dash and the car ran fine. No code other than '34' is given by the ecu. The continuity is tested fine from the KS to the ecu and the KS's both measure in spec with a DMM. Could all of this be a bad battery or alternator? The battery read 12.4v with power off and 13.8 at idle which would not normally give me cause for concern.

Thanks in advance for your thoughts.

[Infinitiguy19]

Check the battery voltage at idle and at 1500 RPMs with a full electrical load (high beam headlights, rear defogger, side mirrors defogger, radio all should be on).

Your battery and Alternator do appear to be operating fine thus far though (According to what you reported).

Is code #34 (Knock Sensor(s) circuit faulty) the only code stored on the ECU?

[tfitzs01]

just my two cents, but were the knock sensors OEM? Aftermarket ones could throw some interesting results.

Was the harness replaced with the sensors?

[lino]

I went to 3 dealers that said the consult wouldn't work on my car but eventually managed to get a consult in my hands and found out that it does work. Maybe try another dealer?

[fastenuf]

Thanks for the full-load test recommendations, Paul - that's a good idea and I am embarrassed not to have thought of it. Results: Power off, battery voltage 12.2, immediately after startup, full load, 14.1 @ 1100 RPM (engine cold), tapering down to 13.9 as RPM dropped. Then the A/C condenser fan kicked in, voltage dropped below 12 within 30 seconds. Turned off A/C, and the alternator cycled on and off like it was and A/C compressor. Seemed like it did not cycle when RPM was held at 1500, but it would be harder to tell as the RPM drop and sounds were different. This seems to point to a failing alternator, but first 2 questions: 1) can anything else cause the alternator to cycle in ant out? and 2) is a reman Hitachi an OK choice on the replacement?

And yes, 34 is the only code.

tfitzs01, the replacement sensors are not from Nissan, but they measure the same as the ones I took out which measure good. I did not replace the harness but I did test the harness continuity from sensor to ecu and it tested sat.

Thanks all.

[Q451990]

to NICO! It sounds like your alternator is toast. We have found that the OEM alternator is the most reliable replacement - available from www.everythinginfiniti.com

The reman. Hitachi - if it's remanufactured by Hitachi might be a good option as well.

There is some speculation that the ECU actually analyzes the signal from the knock sensors as well as just looking for the correct resistance - so an aftermarket part may not be sending the correct information to the ECU. Or maybe it's a stored code?

Heath

[fastenuf]

I have now replaced the alternator and the low rpm/high load cycling and low voltage are gone (whew). However, the 34 code returns essentially immediately after clearing the ecu and starting the car. The failure cannot actually be a knock problem as the code is set before any load is on the engine. As previously noted, the wiring continuity checks out and the knock sensors measure the right resistance. Can this be an error in the ecu itself? I have a spare (exactly the same part code) but the code indicator light does not function on it. Will the ECUTalk software and cable give me any additional information, or reveal the codes on the unit with no light? Thanks.

[Infinitiguy19]

Use the ECU with a bad LED (First time seeing a burnt LED thats used to check the codes on a ECU) and the plug in your laptop and start up ECU Talk and see if any codes come up.

[fastenuf]

OK thanks, will try that, will be on travel for a bit but will post a response when I can.

[Q451990]

As previously noted, the wiring continuity checks out and the knock sensors measure the right resistance.

Have you checked the ohm reading at the harness? Are you able to clear the codes and then get a code 55 for at least a few seconds?

Heath

[fastenuf]

Yes, I checked the harnesses from the KS to the ecu according to the service manual instructions while the car was apart, and checked the resistance of the new KS's both before the install and since they are installed (the latter including the lower part of the harness). Yes, after clearing, the code is 55 until I start the car again. I have not checked the KS resistance from the ecu connector, but I suppose I could....