At 60-70 kpa in the 2000-3000 rpm range my spark advance goes up to 42.9 and causes a lot of spark knock, and the backfiring/sputtering that goes with the knock retard. I'm using a stealth ram intake with the gm 1227730 ecm on a 9.5:1 350 with iron heads. I'm also using a modified AXCN bin file with the spark logic and temp settings from a hypertech 122352. (160 degree thermostat 5.0l, manual tranny). I can't figure out why this is happening. I don't have that much spark anywhere, and at 73c at 60-70 kpa the coolant comp is 20.04. I'm under the impression that this would give you what is in the spark advance main table? I did notice that at 71 kpa and 3000 rpm where this happened once it was in cell 4, but the cells change most of the time. The other time was at 62 kpa snd 2200 rom, but it was in cell 10. Another thing I noticed is both times the SA table called for 37.97. I lowered it to 35.86 to test, but I don't think this is gonna work.

if i had to guess..... check out highway spark settings?

I thought highway spark was turned off in axcn, but I'll check again. The scalars have 151 for min temp and 255 for min speed. The table Highway mode spark advance vs MAP is zero'd out. Could it be the ESC in the ECM is going out, or is there something I'm missing??

Check the PE adder table too.
Address 1B3 is the start location of the table.

Could it possibly enable PE spark with 28% throttle? The lowest is 60% in the table Enrich Power % TPS. I can send the datalog, but it's too big to put on the site. It's a datamaster log, and it's long. It happened twice in about a 45 min log. I haven't had this problem before, or at least I just thought I was doing something wrong if it did. I just noticed that the Advance was higher than anything I had programmed though and it made me a little confused.

i wonder......

i know with some of the older masks, to test the knock sensor, the ECM will purposelly add enough advance to trigger knock....

$8D has this and it is enabled in AXCN...

RBob.

Will it just do it randomly in the middle of a drive down the highway? I think it does when you first start it to test the knock sensor because it always shows 12 on the knock count when you start the car while logging. I assume there is some way to turn it off?

Yes & no, it depends upon the load on the engine. Needs to be above a certain level. This is so the knock will occur when pushed. Difficult to get knock under a decel or light load.

When the forced knock routine is enabled, the SA will stair-case upwards. Can see this in the data log.

To disable, set the CTS enable high, at: $29C

RBob.

I have a knock enable min temp in tuner pro, but will upping that value disable it from seeing real knock, or is that just for the forced knock routine? Why would they want to retard the timing if there wasn't real knock. Also, why would this just start plaquing me? It runs great when it isn't doing something stupid like that, but break out to pass and it may sputter and backfire, or it may push your head back and go. Is it enabled in aujp?? Maybe I should use a different bin. That's a lot of changes though. There's no spark added for accel enrichment is there?

I would stay with the calibration you are been working on and just disable the forced knock test. If need be use the hex editor to change the value, I gave the location in my post above. Use $FF as the value.

As for why it happens some times and not other times, one is the coolant temperature required to be above 95* C (203* F).

Seriously, just disable the test. I still recall the first time a forced knock test was done on the first engine I converted to EFI. Quite the surprise when it added 12 degrees all at once while under WOT. It was most definitely an eye-opener.

RBob.

Thank you very much RBOB. I'll give it a try this weekend. Does it only do the test when it reaches the temp assigned? I mean each time it reaches that temp??

Most likely it does it once per engine run cycle. Once the test passes it won't occur again until a key-off/key-on cycle.

RBob.

It's doing it again. It spiked to 2920 degrees of advace once and over 700 several times in a 30 minute datalog. Does this mean my ECM is going out? I know 3000 degrees of advace isn't called for anywhere in my program.

Either the ECM or whatever is being used to hold/retain the calibration. ZIF sockets are unreliable, long ribbon cables and so on are also unreliable.

RBob.

I'm only using a moats memory adaptor to bypass the memory on the stock memcal. It does the same thing with just the stock memcal in place and no adaptor between them. Another thing I noticed is that the ignition timing doesn't match what is programmed in any map above 40. I think I'll trade my junkyard ecm for a reman and see if my problems go away.

i've seen similar crazy numbers pop up in some A1/DF datalogs as well, either a temporary(1 sample) incredibly high or incredibly low value, but without that value actually being used to trigger the coils(i'm DIS), since even for just one cylinder to run that would be easily detectable and i've never felt it happen before.

Well, mine bucks and backfires and generally doesn't want to go, and it triggers knock. I've seen 39 degrees of knock retard when I was only commanding 28 degrees of advance. I think the spark logic in my old 730 is on the fritz. I hate to throw parts at it in hopes, but I don't see anything else it could be. I may be posting this in the wrong forum, but I thought it was a programming error until it did the same thing with a stock memcal.

I had something similar going on and never did figure it out.
https://www.thirdgen.org/forums/diy-...k-advance.html

Very close to the same thing, but a new (reman) ECM and it hasn't bothered me today. If it continues at a later date I'll see what else I can find out. I really didn't think it was gonna be that easy. No big problem like this is fixed with 15 minutes and a single bill.

my ECM has also been replaced recently, but i dont know how long the problem has been going on.. Did replacing yours fix your problem?

I haven't noticed it bucking and backfiring when I accelerate since I got the new ECM, but it was very erratic, and may not show up for weeks. I'm sure it's sometning in the ignition though as everything else looks normal except for huge spikes in timing and then huge knock retard. Do you feel the problem in your car, or just see it in a datalog??

i dont really feel it per se but I can hear a mild misfire at idle, that doesn't match up with the spikes from what i can tell...
btw i am using AXYC on a '730 ecm, 5 speed L98

5-speed, LB9. There isn't any 5-speed L98 factory calibrations.

RBob.

I'm using AXCN on a 730 because I have a 350 on my 5-speed as well. Maybe it's the corvette bin. I'm not sure they used a 730. Seems like they used a computer under the hood, but I'm not sure if the bin is laid out exactly the same. I read once that the AXCN didn't work very well, but that was the closest to my setup.

AXCN is a y-body calibration used in the '7727 under-hood ECM. That ECM is the exact same as the '7730 just shoved into a different box with different connectors. Of course the y-body's used 6-speed transmissions.

AXCN is $8D, so the calibration layout matches others such as AUJP.

RBob.

Well, it's still doing it with the new ecm. Ignition module? It couldn't cause the computer to show timing that the ecm didn't command? Is the really hign timing a glitch in my laptops connection to the ecm and not connected to my knock and rough running?

The ICM could cause spikes (up or down) in the reported RPM. Which in turn can cause the wrong SA to be used.

Check the data log in the areas of improper SA. Check the CTS, MAP, and RPM. They are the inputs the ECM uses for SA.

RBob.

Almost always the datalog from datamaster shows the problem between 2800 and 4000 rpm anywhere above 60 kpa with a data error. It's usually around 70c because that's where my 160 stat runs. It never gets hot. I was looking at my coolant compensation though and it had some pretty high numbers here and there. I changed them all to 20.04 and will try again tomorrow. Can I turn off the coolant comp? Also, datamaster shows an error when ccp starts. I thought I turned that off to. I guess I should read the sticky about that again.

It quit after I replaced a burned up spark plug. Did that plug just firing when it wanted to make the knock sensor see a knock and retard the timing until it was in the negative which made it look like it spiked up that far?? The plug finally gas fouled and there was a dead miss. Easy to find then with a temp probe. The plugs weren't over 3 months old. Does anyone make a decent shorty plug with .460 reach? Do I even need .460 reach on iron eagle ss heads?? They are recommended, but I have no issues so far with standard reach plugs. R43TS. I would rather have the 45s, but they are much longer and burn up my plug wire boots on the header tubes.