P J Moran

I'm making runs and adjusting my spark table based on the spark retard reported on my AutoXray. It tells me (at 28 one second intervals) what the spark advance is. It also tells me the knock retard.
Is the displayed spark the actual timing at that time, or the value read from the table? In other words, has the retard been subtracted from the table to give the result shown? How does the initial timing figure into the value shown?

I have tried two chips, each with less (up to five degrees) timing in the "trouble area", and am still seeing spark knock detected by the ECM. I just burned a third one with some more (up to three degrees) timing removed, but am beginning to wonder, when will this end?


------------------
'87 IROC 5.7l TPI - original owner!

Grim Reaper

NO! The "Total Spark Advance" is strictly what the ecm tried to give you. You have to subtract the "knock retard" to get your effective.

But the BEST number to use is your Relative Spark Advance. This is what the ecm REALLY added to your base timing. Generally, TOTAL_SPARK_ADVANCE - KNOCK_RETARD = RELATIVE_SPARK_ADVANCE + BASE_TIMING.

There are a few situations where the Total Spark Advance will read much higher than the relative (deceleration generally). That is why I prefer to just use the "Relative + Base".

The ODB

you forgot to mention..
the computer has no idea what your base (initial) timing is set at..
so you have to add that to whatever you end up with after retard.

spark = good

retard = bad

Willie

iTrader: (1)

It should be noted that on some years, the Xray reports only relative spark advance (i.e. 1987 TPI). Total advance = base timing + relative advance - knock retard.

------------------
Willie

Supercharged 1987 305 IROC-Z, Daily-Driver, Emissions-Legal.
Paxton (6-psig): 12.57 @ 111 mph.
Paxton (6-psig) & 50-hp nitrous: 12.04 @ 114 mph.
ATI D1SC (10-psig): ?

http://willie.camaro-firebird.org/

1987 "20th Anniversary Commemorative Edition" Z28 Convertible -- Super Chevy Show Class Winner, 1998.

Grim Reaper

This may be dependent on the year/type of ecm. I can only confirm for the SD TPI 8D eprom, the relative spark advance already has the knock retard removed and it represents only the amount of spark the ecm has added. Thus Total Actual Spark Advance on my car = Relative + the base timing.

And OBD is correct, that when we are talking base, we are talking the physical base of your ignition timing. There is a base constant inside the ecm which SHOULD be changed to match the actual base timing, but only those people that can burn an eprom would be able to change that. When a scan tool displays Total Spark Advance it uses the base constant inside the ecm. Thus if you have adjusted the physical base, the Total Spark Advance displayed will be incorrect.

That is why I watch the relative spark advance on my SD TPI 8D car. Using Diacom, Relative Spark Advance does account for the knock retard (total spark advance does not), and unless you have set the base constant in the eprom to match your physical base, the total spark advance will not be correct.

There are also situations where the Total Spark Advance is completely wrong. On deceleration down long grades, my relative spark advance will drop to 0, but my Total Spark Advance will continue to display 30-40* spark advance. That is why I believe "Desired Total Spark Advance" is a more correct term for the Total Spark Advance.

As a footnote, if you can reprogram the eprom, there is not reason to play with the physical base timing as you have control to all the tables. I just leave my physical base (and constant inside the ecm) to 6* and I control all the spark within the various tables. There is about 6 tables on the SD TPI 8D that conrol spark.

The 2 main spark tables based on rpm/load, one for additional spark during startup based on coolant temp, one for additional spark during power enrichment based on rpm, one to control total max spark @ WOT, and one for Highway Mode.

Plus a constant for the base timing and total overall spark. Along with the Knock Retard attack/decay rates and maxiums based on MAP and RPM.

This "jacking up the base timing to 10*" is something from the 80s. We are in the 21st century now. Jacking up the base timing by 10* can introduce a variety of problems and does not really address the largest problem with spark controled systems; the amount of retard pulled out the system.

Being capable of 38* total spark and having 14* of knock retard taken out is not as effective as only have 34* total spark advance and only 3* taken out.

FYI, on a stock 8D SD TPI car, the maximum spark advance with the stock base timing of 6* is just under 30* total. Increasing it to 10* will only bump this up to 34* and does nothing to address the knock retard situation. I found prior to getting inside the ecm, that just jacking up the base timing tended to trigger more knock retard, but also made the part throttle far to high and introduced knock retard there.

The real solution is to do it inside the ecm through these various tables I've mentioned and playing with the knock retard's attach/decay rates. I found that a knock "incident" at 3,200 rpm can linger for the rest of the powerband and penalize throughout the gear even though there were no further occurance of knock "incident".

But I believe either this Monday or next monday, Tim will be addressing that subject as this is an area that when controlled properly, can give noticeable power increases and throttle response at all loads/rpm and even increase fuel economy when done correctly. I call this "the fun part" of eprom programming and testing as you do notice the effects (unlike VE tables where they just make the car seem to run smoother though they set you up to properly take advantage of other features).

Grim Reaper

I should add, all the above information is based on the 8D SD TPI car 7730 ecm using Diacom. There will be similar tables (for the most part) in most of the other ecms, but I cannot comment on which (if any) might be missing, how they interact or how the relationship between total, relative, knock retard or base works as they might function a little differently or if a particular brand of scan tool displays differently than another.

Grumpy

False Knock?
Old serpentine belt and exhuast rattles can be rpm sensitive.
Plugs show any detonation?.

This in addition to previous posts.

The ODB

Glenn,
I love reading your posts. You may be the only person so far that has even brought back the remote possibility of me ever using fuel injection or an ECM again. I'm a control freak, one reason I love my carb.

Tell me about this timing deal a little more. With the stock ECM on my 86 I would only get about 26 - 28 total with a base of 6*. This suk'd. I would run my best times with a base of 15 - 17 depending on the air temp outside & how rich I set it. I never had knock retard because I never ran with the sensor hooked up. (I know, more neanderthal tactics)
My car seemed to run super like this. Then I switched to the old DFI (also suk'd) so I set my base back to 10, and did my own curve. It was a bit easier this way even though I had to carry a laptop and had massive spaghetti in the passenger floor.

I am an oldschooler for certain.. but. Tell me this.
Could I use a 1992 ECM in my car (using a MAP, speed sensor, and knock sensor) to control the timing, TC-lockup, and give me an indicator (light or readout) of detected knock, with my non-CC MSD distributor ?

I would want to program totally dynamic activation of TC-lockup, and spark curves based on RPM AND speed for each gear... plus have a heads-up indicator for knock.

If this is possible then I MIGHT let another ECM back into my car. I still have my old knock-sensor, speed-sensor, 1986-ECM, and wire-harness stuffed away in my closet like a murder victom. I live very near a TPI-freak who can make custom wire-harnesses cheap. I have a laptop, and enough knowledge to be dangerous. I also know a couple of electronic wiz's that could do a minor frankenstein job on the ECM for trigger-relays, etc.
Please give me good news!

Grim Reaper

ODB, funny you should mention that. I am looking at the possibility of how a "minimum" ecm/eprom could be setup to control limited functions (like the TCC).

This would be highly useful for a person who wanted a non-computer controlled distributor and a carb. Also, it may allow a solution for people that may have an older cccCarb that can't find any info on their eprom and just want the distributor controlled.

Many possibilites I am currently twirlinig in my head. Basically, you pick the level of control you want from the computer. You just disable "code" in the eprom. I am an "old school" guy too, but I also am an "old computer school guy" that is used to "bit twiddling" machine code.

I am just heading out of town on a 1 week business trip and will be review various information while I am away. Some of the places I go to only have a bar and receive a couple of TV channels, so I need something to occupy my time with when I am tired of work. I find all this stuff very relaxing.

Yup, there MIGHT be a "limited" ecm in your future. Computer controlled TCC lock-up has always been the "bain" of people wishing to convert to carb/non-ccc distributor.

Grim Reaper

ODB, I almost forgot, it is a pity that this "knowledge" wasn't more common place when you abandoned EFI. But regardless, playing with the Knock Retard "attack" and "decay" rates are the best ways to control the knock sensor in my opinion.

You can choose the various rpm levels to determine how quickly you will let knock retard kick it (that's the attack rate) and then choose how quickly you want it pulled out (that's the decay rate). Basically, through software you can keep your knock sensor working at the lower rpms for part-throttle driving with load conditions (like going up a hill), but virutally eliminate it for higher rpms (like at the track). Or you can have two eproms so you have a "street" eprom and a "track" eprom. Swapping an eprom is a heck of a lot easier (and cleaner) than physically disabling the knock sensor.

Oh, and you can set the "maximum" level of retard based on MAP and based on RPM when in "Power Enrichment Mode" too. Lots of control over that beast.

FYI, I currently have my knock sensor settings such that I have the MOST spark possible and it works as a "safety" when driving at part-throttle on the street, but basically disables itself @ WOT.

I like the "safety" of the knock sensor because I run 17.2-17.4:1 when in "Highway Mode" with 45* of Spark Advance. It helps pull a little (I never let it pull too much) when I go up a slight grade and I want to avoid detonation. If the grade is steep enough, the Highway Mode just disables itself and runs "normal". Once I "crest" the hill, Highway Mode and Highway Mode Spark Advance both kick back in.

This is how I am getting V6 gas mileage out of an L98 when cruising at part-throttle. And then make it "pollute good like a car should" when I boot it. I get the best of both worlds with this setup.

The ODB

Thanks Glenn,

The TCC was no problem at all for me. I used a pressure switch in the tranny and electronic solenoid triggered by ingnition and an RPM-switch (two modes). Would have been easier with ECM for race-mode though.
Would the ECM be able to lock and unlock it based on speed & RPM. I mean during WOT have it lock at 4600-RPM & unlock at anything below 4200-RPM... this means unlocking after every shift & then locking again.

You are right, there are big possibilities in tuning a carbureted engine with an ECM. Is this easier done with aftermarket modules or by modifying the factory ECM?

I programmed in assembly between 84 & 86, so I am familiar with hexidecimal at least. I might be able to follow what you're saying. I have never been inside a factory ECM though... only monitored with diacom and tricked the sensors.

ODB

P J Moran

Looking at a chart with my readings on it, it looks like the timing displayed is strictly that read from the table, but it's hard to confirm due to interpolation. I say this because the timing values don't "move" much, but the amount of retard is as much as 9*.

TunerCat has not made available the ability to change the initial timing constant on my $32B, so I have to keep that in mind. Looking at my timing tape while the scanner is monitoring spark advance (car parked, no load, manipulating throttle by hand), I have found a 2.5* difference between the two. This makes sense since I set my initial to about 8* versus the factory 6*. This tells me that the scanner is trying to display total timing (considering initial), but there was obviously no knock retard while I was making these underhood comparisons, so I don't know how it would be reflected.

Since the engine can idle somewhat poorly while trying to set initial timing, I think it is a good idea to use a higher initial value (say 10* or 12*), just so that it will idle better with the bypass disconnected. I agree, however, that this should be compensated for within the prom. If I had access to the constant, I would change it. Otherwise, one could simply subtract four or six degrees from the entire spark table. BTW, TunerCat only gives me access to two spark tables. I don't know if others exist on my EPROM.

I also pulled the plugs this weekend and couldn't see any problems. The rim was sooty as was the recessed area, but the exposed ceramic and electrodes were absolutely clean. I'm no plug reader, but I don't like the sootiness.

------------------
'87 IROC 5.7l TPI - original owner!

TRAXION

PJ,

Consider going to ECMGuy's site and downloading the $32 hac. Although it is not the $32B hac they are very similar. You can use TunerCat's TDF Editor to ADD any paramter that you want to change. No need to wait for TC to add this parameter to the TDF. You can do it yourself. Also - TC is very good at incorporating other parameters in the TDF files. After your addition is working just e-mail TC and tell him what you added and he often adds this information into the next release of the TDF.

Tim

------------------
TRAXION's 1990 IROC-Z
Best Time = 12.244 @ 112.51mph (1.778 60' / 7.819@88.32mph in the 1/8)
All Natural. No Force. No Drugs. Stock Bottom End. Stock Body Panels.
Gunning for NA 11's with bigger cam, bigger stall, and bigger exhaust.
-=ICON Motorsports=-
Moderator: PROM board at thirdgen.org