sorry, didnt realize this was in the DIY Prom Board. Anyway, I havent messed with the ECM, it's on factory tune. Everything is 100% stock. The car sat for six years, and I replaced FP/sender/filter/fuel tank, etc, as I said, and just had the transmission rebuilt. I have had some problems with over pressurized fuel tank and the pump overheating and whining to a halt. My BLM's according to tuner pro with the super8 are rich, and the INT is leaning accordingly (in the low 130's), but the Knock Count rolls straight up to 250+ and resets, the graph in tunerproRT while datalogging shows these knocks, and is zigzagged forever. I'll attach the datalog files so you can see what's going on if anyone cares to see. It's a 91 z28 5.7 TPI and I dont see a ESC module over by the brake booster relay bracket. Where is the filter for the sensor?? The wire appears ago going to the knock sensor, just cant find the module. I've even tried to get the SES light to come on with 2 consecutive WOT runs so it can check itself. I asked the dealership to check BASE timing and look at the fuel pump/over pressurization issue, and all they did was "set timing", and gut the air inlet valve hanging off the sending unit so it vents out of the gas tank--fuel pump starting running quiet again, but I couldnt deal with the fumes, so I installed another inlet valve (fresh air) and it's starting to pressurize and get mildly loud--I'll test that later. They also said the catalytic converters (dual cat optioned car) had 0.00 back pressure and that the fuel pressure was 43 and some change. I havent verified any of this. When I had the transmission rebuilt, they put in a 2000 stall converter (is this okay on stock 350 tpi for now?? feels like it's slipping until it winds up, but the computer is pulling too much timing out for me to get any performance and test the new stall converter??) here's what I'm thinking, somone please review the datalog with tuner pro and see how I attack this. also, the a/c compressor clutch locked up and burned up my belt, so I unplugged it and using it as a dead pulley for now (bearings are a little noisy on the "dead pulley")

Loose something in the transmission rebuild causing 250+ knocks in 2 seconds??

A/C compressor clutch causing rolling knock??

Idiots at the dealership didnt know how to set timing (EST by pass not used)??

I've lost a cyclinder/valve, etc, and need to pull plugs and sniff around??

Fuel pressure dropping, or too much, causing runaway knock?? stuck FPR, Clogged return line??? gawd!! hope not

Injectors are whacked and causing issues?? (have audible clicking, but not equal on each injector)???

Some part of the spark system isnt working?? (module, coil, pickup in distributor)??



Anyway, just wondering if anyone has ever seen this before......??????? Or can help with where to start FIRST, cause all of these issues are on my mind. Having sat undriven for 6 years I'm about to slap a L31 in there with the Scogg'n Dickey TPI intake and new injectors/sensors.

I have no getty'up...and it feels like when you get your spark plugs wet on an old dirtbike. Exhaust smells normal like before, but she just want get up and put it down like before. Seems like it happened when I was driving on the interstate about 75 mph and romped it one time and it felt weird after that!

sorry for so freaking long, but the devil is in the details.

thanks for all your suggestions. I'm going to verify timing myself, and the fuel pressure. Just seems like something broke or failed somewhere as the knocks seem fake, but the timing being retarded is what I'm feeling. So where is the ESC module?? I've replaced the knock sensor to no avail, but I kinda figured it wouldnt have changed anything.

HELP BROTHERS!!

 

well, s'ppose I posted here b/c of the relevance. Good info. So, I've been watching these datalogs trying to make a diagnosis...other than I've had 4 cyclinders faster than what my tpi is now. Knocks seem artificial, timing is getting stolen from me too, cause it sure feels like she's holding out on me starting at around 1600 rpm, all the way to the firewall!!

see attached, 3 logs for tunerpro, 1 log for TTS datamaster

TTS logs have a 90mph run, start from cold motor to warmup
tunerpro has an idle, a brief run, and a few weak pulls on the large file.

Knock, knock! 3400 knocks in 60 seconds?? retarding 20 degree at anything above 25%TPS... just wanted someone else to see this.

 

change to 1st.uni and open in tts datamaster, that should be a good enough sample set.

 

well, i like talking to myself. haha.

anyway, here's some followup.

I've been after this thing for 24 hours in the past 2 days and here is what I've done: replaced distributer, ignition coil and set base timing to 6 degrees BTDC (was at 8 BTDC by the dealership...finally bought my first timing light and its the coolest thing ever), removed failed a/c compressor and installed delete pulley, removed plugs and had oil on 2, one was almost bridged (cleaned all that needed it, but PreTTY sure it's time for the L31 swap or rebuild in regard to hardparts), removed the knock sensor and grounded outside threads and extended 5volt wire away from motor (used the old/good one and kept it uninstalled to eliminate true physical knock possibilities). and the results are in... still have rolling knock runaway knock. every 5 seconds I get 250+ knocks and it resets (upper threshold maybe?) and it pulls the default 20 degrees of timing out at anything above 1800 rpm, and roughly 4 degrees are removed under steady load below 1800rpm.

So, essentially, I have runaway knock without the sensor even installed. The voltage to the knock sensor is 2.4 volts to ground. I read somewhere it should be roughly 5 volts?? I'm off to trace the blue knock sensor wire and go get a new oil pressuer sender since I broke the original while removing plugs (cant say its the first time).

Therefore, the ESC control is integrated into the ECM, correct?? I purchased a new one from autozone for $80 and had to swap the memcal to the new unit. That still didnt fix the problem with the knocks. My question is, is the ESC a physical component on the ECM board, or the MEMCAL??? I've either got a bad/shorted wire (unlikely), or a ESC on drugs...but where is it exactly?? I know speed density machines dont have an external module, but rather, it's built in to the. . . . . . ??? ECM or MEMCAL itself? survey says, ???, ???, ???

I'm 100% sure that putting a new motor in wont solve this problem at all, seeing as how the the original L98 is "seeing" knocks without the sensor even installed in the motor, but laying on the garage floor wired up properly. cripes!!

Also, I did adjust timing to greater extremes than what was on the dial indicator on the balancer, and it smoothed idle a little, but absolutely no change in the amount of knocks. This is crazy. I wouldnt be concerned about 4, or even 40 knocks, but this is whacked. Welcome to the Phantom Knock Thread!!!!! If anyone wants the datalogs from tunerpro or datamaster, hit me up and I can email them!! they wouldnt post correctly.

. . . please help, I want my TORQ Monster back!!!

 

the knock board is in the memcal which is next to the prom inside the blue chip that plugs in the ecm. that's why I've seen guys with rolling knock counts have luck making sure the chip is seated right. they are very hard to find aujp chips these days.

 

I think it's time to replace the MEMCAL, unfortunately. I stopped by the dealership today so they could provide me some information. I received some flow charts and a knock sensor schematic. Both sides of the flow chart end in either replacing ecm and/or memcal, or the knock sensor. Basically, what I did was verify 5volts DC to the sensor, Ohmed connector for grounding or continuity. Other than that, it would have to be in the ECM or MEMCAL. (the other test was impossible to complete: check simulated knock by rapping on the engine hanger or block and measuring for atleast 2-3 Volts AC.) All checked out okay, so I removed the MEMCAL and checked the white chip standing off on 6 or 8 legs--it wasnt loose and I couldnt remove it or change it's "seating" in anyway (1R3282 in peephole on the left in the MEMCAL picture below, correct??). Who freaking knows. I'm going to attach some screenshots of datalogs. Where can I find another 012227730 ECM AUJP (860571, M011097435)?? GM wants nearly $600.00--part number 16151348. 350 tpi 1991. cant i buy the burner and chip for a lot cheaper and do my own?? GMPartsdirect has them for $300. Described as: "DESCRIPTION: EPROM ELECTRONIC CONTROL MODULE (ECM)This is the right part number if your ECM has 4307ANJF, 4312ANJ, 1350AUJP" . . .is this right for me??

I'll attach the flow charts as well when I get them scanned for someone else's treat if needed. for 91 ecms it was the only dk blue wire on terminal 9, yellow connector at the ECM.

Basically, my sensor is good, the ECM registers 255 knocks without the sensor installed (but wired up), the wire is good... I dont think this can be anything other than the part of the memcal.

I'll follow-up with what happens. hope not $300 for a new memcal.

RBob, you've been a lot of help in some of your posting!!! People like you make this site awesome. Thanks for the input, also, ???!!!!

 

you may be able to find a used one over on the forsale section.



when you buy the stuff to turn you car, your changing the values in the .bin which is on the eprom not the memcal. so you would still need a new memcal to get the knock sensor working right if it is that, that's bad causing this.

what you could do in the bin is turn off the knock sensor so it doesn't use the crazy knock courts to retard the timing. but its your call if you want to tune without the safety net of the knock sensor. some do, some don't.

 

thanks...i'll consider myself warned on tuning w/o the knock sensor. Seems like a vital sensor not worth turning off.

I understand about the memcal, that's what I thought after reading through one of RBob's posts with a memcal torn apart.

I would need the tuning software first to be able to turn of the knock sensor, havent made it that far yet---feel like I'll hafta cough up a pretty penny for a new memcal. I've seen them new for $165, $275, $300, and even $577.58. haha, expensive little deal huh??

thanks for the help, I'll finish this thread once I get a new memcal to try out.

ya'll been a lot of help!!!! and yes, it is BAD causing this... runs like a 3cylinder geo metro!! haha

just want one today to get it to try and work right, but i'll keep looking for the 1350 AUJP somewhere. patience is important sometimes. damned delco parts, geeesh. Not an easy part to find I guess.

 

Call Brian at TBI chips. Just have him send you a tuned chip for your application. It will cost in the $150 range if I remember right. That way you will have a new memcal that is programmed for more perfomance - better than stock, lower price.

 

I called BRIAN, he's kinda local, but couldnt do anything for me. Apparently the SD MEMCALs for 350's with the AUJP are kinda scarce.
They dealerships claim to have them available, so does an email from gmpartsdirect.

I found an ACDelco AUJP brand new from a guy who bought up old dealership stock. Cost me $175.

Replacing the MEMCAL is eventually what fixed the problem (today, finally). I'm now running without the rolling knock counts, and timing retard is next to nothing (I saw 2 degrees max) at any given time. The TORQ is back, and it feels like a 350 tpi should. I ended up trying the 3.9k resistor wired between chassis ground and the ECM instead of the sensor, just to say I did it, and the knocks continued. I knew a couple of days ago, however, that the knocks were artificial, and that my wire was good, and the sensor was working properly--which left only one thing: the knock board on the MEMCAL. Fluke Voltmeter, The Invaluable Validator. heh.

See the following posts/pictures...

Anyway, this might keep some people actually using the stock knock sensor, or save some time. I couldnt find this info anyway, so I'll assume I have a pretty isolated case... or a case of wanting to know too much. Let it be found! This is from beginning to end... testing spark, and hunting the knock that wacked out the timing. Fuel pressure, timing, and ignition all checked good, too. Internals of the motor have seen better days (165k to the wind), but functional, and the new tranny is hitting good.

Thanks, ???, you nailed this one, sort of... it was the board, but not that it was loose on the MEMCAL, something physically on the board...

 

Here are the pictures for testing ignition coil

The start of testing pickup coil is on the last picture

 

last of the pickup coil test

 

And, the long awaited Knock Sensor Test or Flow Chart

Holla'

 

glad you found it.


nice feeling getting your car running right again.