LTx knock module differences
#1
LTx knock module differences
Yes, I've searched and didn't find anything concrete. I am curious about 2 things when it comes to swapping knock modules and am looking for personal experiences.
Does OBD matter? (can OBD2 modules be used in OBD1 pcms and vise versa)
Does # of knock sensors matter? (F cars have 1, B/Y/D have 2)
A lot of guys here are using the LT4 module in OBD1 F-bodies despite it coming from a OBD2 vette with 2 knock sensors. I know not everyone is satisfied with the LT4 KM, but it is a fairly common mod.
Resistance across knock sensors is 4k ohms for OBD1 and 100k ohms for OBD2.
Does OBD matter? (can OBD2 modules be used in OBD1 pcms and vise versa)
Does # of knock sensors matter? (F cars have 1, B/Y/D have 2)
A lot of guys here are using the LT4 module in OBD1 F-bodies despite it coming from a OBD2 vette with 2 knock sensors. I know not everyone is satisfied with the LT4 KM, but it is a fairly common mod.
Resistance across knock sensors is 4k ohms for OBD1 and 100k ohms for OBD2.
#4
F-bodies only have 1 sensor.
I know B-bodies have 2 and the sensor wiring changed between OBD1 and 2 in addition to internal resistance. The knock sensors are wired in parallel and share the same pin in OBD1 vs. having separate PCM pins in OBD2.
The B-body and F-body sensors share the same part number for OBD1. The B-body knock circuit would have less resistance than the F-body because of its parallel wiring and I'm wondering if there are any differences in the knock module because of this. The PCMs are identical.
I know B-bodies have 2 and the sensor wiring changed between OBD1 and 2 in addition to internal resistance. The knock sensors are wired in parallel and share the same pin in OBD1 vs. having separate PCM pins in OBD2.
The B-body and F-body sensors share the same part number for OBD1. The B-body knock circuit would have less resistance than the F-body because of its parallel wiring and I'm wondering if there are any differences in the knock module because of this. The PCMs are identical.
Last edited by Catmaigne; 08-01-2016 at 01:14 PM.
#6
TECH Fanatic
I have only done the LT4 module on B Bodies, and that was on a 1995 Caprice. So, there was no issue using the module from an OBDII car in an OBDI PCM.
I have no experience with the number of sensors, but I doubt it matters.
I have no experience with the number of sensors, but I doubt it matters.
#7
Hmmm... I just thought it was strange that there are no knock modules that overlap with different OBD or knock sensor #s. There probably is a difference... doesn't seem like anybody knows how much effect it has. I guess I'll just experiment with them.
http://lt1swap.com/lt1_bcc.htm
http://lt1swap.com/lt1_bcc.htm
Trending Topics
#8
TECH Veteran
Hard to tell how the 1 or 2 sensors interface with the PCM without a schematic of the PCM, itself. The module may be farther down the circuit and not care about how many sensors there are, since it seems only to be engineered for listening. Perhaps that is why it works across both OBD types.
#9
12 Second Club
iTrader: (3)
It would be so cool if GM engineers would come play on the forums, get REAL answers about the sequence of operations of the PCM, ranges of sensors, everything. I don't get why you don't ever see them on any of these platform forums, I get this LT is old, but how cool would it be if the guy who programmed and designed the new LT1 would join forums? Millennials have to be good for something in this age of twittering and pokemon and new fangled technology. We did a project called Ecocar in college with GM and the DOE, 16 schools including Viginia Tech, Georgia Tech, WVU (let's go mountaineers!) Ohio state, penn state, etc designed a hybrid for GM using the Saturn Vue, we got to meet the guys who designed the 5th gen Camaro, Z06, and 2 mode hybrid transmission and I'm sure some people from that competition went to work for GM. How cool would it be if someone could walk into work at the GM engineering office and say "hey, some LT1 guys want to know what they can use for the now discontinued LT4 knock module, any ideas?" And someone chime in "a couple of X ohm resistors and blah blah and your golden!"
Last edited by bufmatmuslepants; 08-05-2016 at 05:29 AM.
#10
TECH Regular
I do recall seeing a post from the GM engineer that had a hand in designing the LT1 and LT4 knock modules. It was a long time ago, though. It went into very specific detail about what the differences were, and why. I will see if I can find it again.
#12
12 Second Club
iTrader: (3)
Im guessing, based on my programming experience on other things, that the PCM is looking at readings and has a time average on the readings, say over 5 seconds stock, if the 5 second rolling average is (Just using random numbers here) 40 counts, then that's a knock. Maybe the LT4 has a 10 second rolling average, so it isn't as succeptable to spikes. Or maybe it's as simple as scaling the sensor, maybe the LT1 knock module has a range of 1-5 volts being 0-100 counts, and the LT4 is 1-5 volts being 0-50 counts, so it reads half of what the LT1 sensor does, making it less sensitive.
#13
This is my understanding of the whole thing, correct me if I'm wrong...
LT1 knock sensors have a piezoelectric material inside of them similar to that in certain microphones, acoustic guitar pickups, etc. Vibration/sound waves cause the piezoelectric element to deflect which in turn generates a voltage difference across the sensor with AC current. The PCM somehow detects the voltage/current changes and filters them using the knock module to discern between normal operation and detonation so it can pull timing accordingly. I would imagine that the knock modules themselves act as noise gates and only command the PCM to pull timing after the knock sensor current/voltage exceeds a threshold (which could be a fixed value or variable with signal, we don't know). Think of it like a noise gate on a microphone that's used to filter out buzz and ambient noise when not being spoken into. Even though there is an input signal when the mic isn't used, all output is cut until the input signal reaches a certain threshold because everything below it is deemed undesirable.
What concerns me is that the PCMs are said to be identical internally and only really differ in tune/modules. Differences in sensor amount (1 or 2), wiring (series/parallel), and impedance of the sensor(s) would all affect the signal that the computer sees. This has to be accounted for somehow in the tune (like knock fast attack rate table) or in the module itself. I think too many people are jumping on the LT4 KM bandwagon without OBD2 B/Y bodies and results could be better, worse, signal could be nonexistent... The truth is we don't know. I will try to come up with an experiment.
LT1 knock sensors have a piezoelectric material inside of them similar to that in certain microphones, acoustic guitar pickups, etc. Vibration/sound waves cause the piezoelectric element to deflect which in turn generates a voltage difference across the sensor with AC current. The PCM somehow detects the voltage/current changes and filters them using the knock module to discern between normal operation and detonation so it can pull timing accordingly. I would imagine that the knock modules themselves act as noise gates and only command the PCM to pull timing after the knock sensor current/voltage exceeds a threshold (which could be a fixed value or variable with signal, we don't know). Think of it like a noise gate on a microphone that's used to filter out buzz and ambient noise when not being spoken into. Even though there is an input signal when the mic isn't used, all output is cut until the input signal reaches a certain threshold because everything below it is deemed undesirable.
What concerns me is that the PCMs are said to be identical internally and only really differ in tune/modules. Differences in sensor amount (1 or 2), wiring (series/parallel), and impedance of the sensor(s) would all affect the signal that the computer sees. This has to be accounted for somehow in the tune (like knock fast attack rate table) or in the module itself. I think too many people are jumping on the LT4 KM bandwagon without OBD2 B/Y bodies and results could be better, worse, signal could be nonexistent... The truth is we don't know. I will try to come up with an experiment.
#15
TECH Enthusiast
i to would be interested in knowing how to desensitize our knock sensors. this would come in handy for the guys who cant find the lt4 knock modules and and get them desensitized through a tune
#17
TECH Fanatic
iTrader: (7)
Straight from the GM manual.
"The Knock Sensors produce an AC signal that is sent to the KS module. The about of AC signal is proportional to the amount of knock."
And..
"If the KS module determines that excessive knock (detonation) is present, the PCM will retard the spark timing based on the signals from the KS module. When knock is present, the KS module voltage input signal to the PCM goes low. The PCM will then retard timing until no knock is present."
From this, I would assume all the module does is scale its output voltage to the incoming AC signal from the knock sensor. This would probably mean the LT4 module has a lower scaler, indicating less knock than an LT1 module.
If you wanted to test that, all you'd need is an O scope, comparing the input and output signals of the knock module (although I imagine it wouldn't be easy, unless you crack open the PCM and prod around). Unless the input value can be viewed by a scanner, like a Tech 2. Anyone have any knowledge on that?
"The Knock Sensors produce an AC signal that is sent to the KS module. The about of AC signal is proportional to the amount of knock."
And..
"If the KS module determines that excessive knock (detonation) is present, the PCM will retard the spark timing based on the signals from the KS module. When knock is present, the KS module voltage input signal to the PCM goes low. The PCM will then retard timing until no knock is present."
From this, I would assume all the module does is scale its output voltage to the incoming AC signal from the knock sensor. This would probably mean the LT4 module has a lower scaler, indicating less knock than an LT1 module.
If you wanted to test that, all you'd need is an O scope, comparing the input and output signals of the knock module (although I imagine it wouldn't be easy, unless you crack open the PCM and prod around). Unless the input value can be viewed by a scanner, like a Tech 2. Anyone have any knowledge on that?
#18
Quick update...
I tried swapping a OBD2 F-body module into a OBD1 B-body for ***** and giggles. The datalog showed that this module is definitely more sensitive than the stock one in this config. Currently trying to source an OBD1 F-body module to try again.
I tried swapping a OBD2 F-body module into a OBD1 B-body for ***** and giggles. The datalog showed that this module is definitely more sensitive than the stock one in this config. Currently trying to source an OBD1 F-body module to try again.
#19
i get this hole thing for the sake of knowledge. but to be honest i havent had a knock sensor in my car since.... i dunno when. just dyno tuned it. can see pre ignition on the graph with the correct smoothing selected. Done. No need for modules just turn em off?