kline454

when useing a lt4 knock module in a 1995 lt1 do you need to use a 1996+knock sensor, because 1995 are 3300-4500 ohms and 1996+ are 100k, ?

shoebox

No. The module works with either sensor.

kline454

ok thanks, the problem im having is false knock retared, my knock sensor is 3800ohms running or off witch shows theres no knock, and i have my 5v coming in, but when i plug it in to the sensor it only has 2.5v witch would cause it to retared, it like my knock sensor restiance is to high but it with in spec.

Injuneer

You're way on your analysis. The voltage on the knock sensor circuit has NOTHING to do with whether its sending a knock signal. The sensor is a piezo microphone, and the noise signal is a variable frequency riding on the 2.5V that is normally present on the circuit. The PCM is looking at the level of the variable frequency noise signal, after it has been filtered by the knock module.

Its perfectly normal for the signal wire to have the 5.0V level from the PCM when it is unplugged, and for the voltage to drop to 2.5V when you plug the connector onto the sensor. In fact, the PCM monitors the voltage on the circuit, and sets a code for a faulty circuit if the voltage goes over 3.5V or under 1.5V.

kline454

the GM manual says at 2.5 volt the pcm pulls timing, and it pulls timing at your rite you should have 5v coming to the sensor but the gm manual say 2.5 volt or less will pull timing,

kline454

and my sensor has 3800ohms running or not, unless you tap next to it then ohms go up, so im not getting any knock at the sensor so why is it pulling timing?

sbs

Do you know for a fact that they work together correctly, or are you just observing that lots of people use them together and they appear to work?


I have seen posts on other sites suggesting that the LT4KM is indeed designed to be used with the OBD2 KS. Nothing definitive, however.


Does anybody know if the PCM reads the KS inputs and conditions them into some uniform (i.e., same for OBD1 and OBD2) format before sending them to the KM, or does the KS input go straight to the KM?

shoebox

I have never seen that stated in the manual. It does say that 2.5v is a normal no knock operating condition. Are you using a scanner to tell if timing is being pulled or there is indication of knock? That's the only way to do it. From what I have read so far, sounds like your spark knock system is functioning normally.

shoebox

I know that it does work on my OBD-I PCM (through before and after data logging), though I don't have any schematics of the PCM to tell me how the signal is routed inside the PCM. I don't know if the signal goes straight to the KM or is routed somewhere else first. Since it works with either PCM (and the sensors are different impedances), it might seem the signal is conditioned before going to the KM or the KM does not care and the difference in impedance is passed through it. If the KM was impedance sensitive, you would get an immediate trouble code when installing one with an OBD-I knock sensor.

sbs

How does datalogging show that it the KM and KS are properly matched?

It seems more likely that the PCM is the thing that sets codes, and as long as an OBD1 PCM sees an OBD1 sensor, it is happy, regardless of whether the KM is receiving the signal it was designed for.


Could it be that part of the reason we see less knock reported with an LT4KM in an OBD1 computer is because of the impedance mismatch? Maybe the wrong sensor "deafens" the KM.

kline454

yes im useing the gm tech 2 and it showing 2-3 degrees of knock retart at idle, when my sensor isnt reading any knock because my ohms dont change when i start the car and measure the ohms of the sensor, so how is it working right if its pulling timing with no ture knock being detected,?

Injuneer

Then get a new shop manual (or reread it very carefully), because you are wrong. The STANDARD test to determine if the knock sensor circuit is OK is to:

-with the harness wire removed, measure the KS resistance from the single pin to the block. Should read 4,500ohms on the OBD-I cars. The circuit will function properly with KS resistances in the range os 3,500-5,500 ohms.

-with the harness connector still unplugged, verify that the PCM is supplying a +5.0V reference voltage.

-Connect the harness to the sensor, and measure the voltage from the wire to the block. Should read 2.5V. A reading between 1.5V-3.5V is acceptable. Outside that range and you will get DTC 43. When you get DTC 43, the PCM defaults to an internal calculation of knock retard, based on engine operating parameters.

Hence, if the voltage falls below 1.5V, the PCM will start to retard the timing, NOT based on the signal from the KS, but based on its internal calculations. You can see anywhere from 0-15degrees of knock retard under varying engine operating conditions..... even at idle you can see a few degrees.

The PCM pulls timing when the amplitude of the noise signal on the circuit exceeds a predetermined level.

The PCM can also pull timing when there is NO signal from the knock sensor. Refer to "burst knock" and "knock count".

The mail-order programmers know the answer, since they can modify the PCM's to accept the various KS's.

I bought the LT4KM in 1998. Before I installed it, I did a lot of data logging, including using lower octane fuels to induce true knock. Repeated the same data logs, operating at the same conditions, with the same variety of fuels, and got the same results for retard in response to true knock whether I was running my stock module or the LT4KM. The one exception was the LT4KM reducing the instantaneous spike of knock during lifting throttle shifts on the M6.

I had enough confidence in the results that I continued to run the LT4KM even though I was spraying a 125-150 shot of nitrous at the time. Can I "prove" there is a complete match between the OBD-I KS and the LT4 KM? No.... but the evidence I collected strongly suggests that the function of the PCM knock retard system with regard to true knock was not altered significantly by the use of the LT4KM. I wouldn't have run it if I didn't believe that.

Appears to be a comprehenson problem here....

YOU CAN GET KNOCK RETARD WITHOUT HAVING A SIGNAL FROM THE KNOCK SENSOR. IT CAN BE A RESPONSE TO BURST KNOCK OR TO INCREASING KNOCK COUNT.

IF YOUR KNOCK SENSOR IS DEFECTIVE, YOU WILL GET ALMOST CONTINUOUS KNOCK RETARD, BASED ON PCM CALCULATIONS OF ENGINE OPERATING CONDITIONS.

THE KNOCK SIGNAL IS NOT VOLTAGE BASED OR RESISTANCE BASED. IT IS A VARIABLE FREQUENCY RIDING ON THE 2.5V PULLDOWN OF THE +5V PCM REFERENCE SIGNAL.

That's about as clearly as I can explain it.

Good luck.

kline454

well read over everything and shebox is right, sorry about that. but im still confused on why its pulling timing,

kline454

well my ? is whats going on, is this all normal?

kline454

everything is right with the knock sensor and volts, say it should not pull timing so why is it pulling timing?