In 2010, I spent several months scanning and tuning. Mostly, I was doing small tweaks to the timing table to see the result. With a hogged-out L98 TPI, their may/may not be a difference in strategy, but just wanted to post it for clarification.

Basically, I tried everything from retarded to advanced. I tried it in the lower, middle, and higher rpms. My changes were more/less in percentages -- so lower loads got some bigger shifts than the WOT loads.

I hit the dyno after a few months and confirmed I was close. (I ended up using some calculations that proved reliable).

For startup, I varied the spark (add-in table) from 9-20 degrees. That's because AYPY used 20 and ARAP used the lower amount. I tried combinations in-between -- and a graduated slope (rather than the factory drop-off method.

I also converted the timing table to a single "entity" by adding total timing and putting that in my new table. Then, I smoothed the transitions until I could see a pattern (3D picture) taking shape. From there, I varied as described above -- always sticking to the general shape theme.

My smoothest timing curve was nearly a direct percentage (drop) from stock to the AFRs. I considered: if stock total timing was 36 at WOT, that AFRs averaged 30, I should try 20% less across the board. That strategy wasn't all bad. It drove with linear acceleration from idle to redline.

Partly, I added more timing on the low end to keep heat out of my new headers. Then, I liked how it was snappier off the line (around town). I added more timing in the middle loads and it became even more responsive. When I went back and compared it to a stock LT-4, mine was significantly better in low-mid response.

Because low-mid performance was SO good, it made the top-end feel less impressive (good but not great). That's why I kept "playing" for quite awhile. I was looking for the best combination of smoothness and response.

In the end, I settled for maximum response -- since I don't push it past 4500 rpms very often. Don't really need to since it pulls from idle to that rpm so fast. (If I was a lightening fast shifter, I'd probably get under 4 secs for 0-60mph. And, coming out leisurely of turns, I can get to 60mph in 2 seconds.) That makes it a fine street car. Highway response isn't too shabby either -- especially with a ZF6 and that extra gear.

My issue now is a mild shudder as it sits down to idle. Also that it runs a bit too hot at idle. The shudder is like an exaggerated "cam" wobble before it drops off the IAC. The heat issue is how it climbs at idle -- even with a HV water pump and high capacity all-alum radiator.

I know a bigger CFM fan would help or cure the idle "problem", but I wonder if my idle timing is a bit too hot. I'm 3-degrees hotter than factory/ARAP tables show -- but I liked how well that performs. I find that approach more satisfying than bumping idle up to 900 rpms. Again, I also hoped it would keep the header coating from "cooking". (You see that when timing is too retarded.)

I'm not sure if there's even such a thing as too much timing -- if the KS isn't engaging. (Really, the only place I saw it in scans was the upper rpms of light load and the middle-upper WOT region. I took timing out of those areas in the very early stages of tuning.)

I could put it back to the lower (smooth) timing I tried for awhile. But, I can't remember it being significantly cooler with more retarded timing. (25 vs. 28 @ 800rpms). If anything, it might have gotten hotter faster. I don't recall.

Again, the issue is if there's such as thing as too much low-mid range timing -- if the KS says I'm OK? My thought is no.

If I'm correct (and my timing is OK), I either need a bigger primary fan, aux fan, or both. My first try will be a better aux fan with a 205-deg turn-on. (My compression is 10.5 -- with a small cam -- so it's better for me to run lower temps to avoid detonation. On the dyno, I saw KS activity when the temp got much above 210.

Mostly, my questions are about max timing and that weird shudder when the moter settles down.

I'm also not terribly happy with very low cruise performance. It has an occassional miss between 1300-1600 rpms, but that might be a slightly sticky injector. If TPIs weren't a PITA to pull the plenum/injectors, I'd have sent them out for cleaning way before now.

None of this makes the car run poorly. Really, it's pretty darn good. I'm just looking for perfection (because I'm kinda O.C.). I'm looking to get the last few imperfections gone.

Any ideas would be appreciated. (BTW...I really can't tell much else from the scans I've done over time.)

Thanks!
gp

 

Gregg - you're a way ahead of me on perfecting the timing tables, I admire your persistence! While I cant add any wisdom to the timing question, I have found a couple of things on tuning my 1992 383 related to low speed cruising / traffic.

First, if any part of my tune is lean, I can almost watch the temperature gauge increase - I can see 20 degrees more coolant in less than a minute.

Second - if the ecm is pulling too much fuel on one side due to cam reversion you'll see higher temps on that side plus it might give you that herky-jerky sensation at high vacuum / low throttle points. I'm not sure if L98's have cam reversion issues with bigger cams.. if not then please ignore me on this point!

I have switched my tune to open loop and slightly richer at low throttle / high vacuum. Low-speed driveability has improved enormously.

Good luck!

 

One thing I think might be relevant here is that with a stock engine you can put in low octane fuel and not get any knocking. That tells me that there is plenty of room to add advance if you always use premium fuel.

Your shudder going to idle could be the IAC algorithm taking out steps too fast. When you mash on the gas pedal the ECM adds IAC steps (throttle follower mode) for later when you let off on the gas so the engine won't stall when the throttle blades slam shut, basically making the airflow suddenly zero. There is an IAC decay factor that determines how fast the extra IAC steps are taken out when you go to closed throttle (or even close the throttle slightly). Sorry, don't remember the location, but at least this should give you a hint on what to look for.

 

Thanks for replying guys! I'd almost forgot about this post.

Open loop all the time, huh? I was never convinced that was the best course since the ECM ought to be able to compensate much better than my calculations. And, I'd think you'd burn more fuel, with higher emissions.

Part of my "confusion", is the issue of maf tables. It's my understanding that they're a starting point (much the same as the injector constant) where O2 sensor data with "trim" the real world results to the correct needs of the motor at any given millisecond. (So, in one regard, tuning should be easy).

In reading my plugs, they appear a smidge hot -- though just barely. When the motor was new, new, they looked at bit more lean. That was also when I experimented with injector constants trying to the resolve the difference between BLMs and INTs. While experimenting with this, I could swear I could tell the difference as I made the constant smaller (which would increase the injector PW). But, if the O2 was really compensating, I wondered if my perception was imaginary.

Obviously, a wideband would have told me what I was feeling.

That said, I deliberately set the lowest MAF values a bit low. Just in case it could run a hair leaner on idle, I thought it might be worthwhile to save fuel. (Pay attention to the amount of time a city driver sits at stoplights. Plus, published ranges I find on the net indicate there's a fairly high margin for running lean -- especially when not under load.) But, that doesn't account for hotter cylinder temps and faster rising coolant temps. Since I opted away from the EGR, that too, could raise temps above the designed norm. (Not to mention my higher-than-stock compression).

Some have suggested that a Mezziere's water pump would be the solution and I can't disagree. But, I wouldn't mind better mpg's either. I'm averaging about 15mpg -- though it's mostly city and shorter trips. I've also measured it around 13.5 when my right foot starts doing the happy dance!

I'm intrigued by your comment about the ECM pulling more fuel "on one side" though. Off the top, I assume you've got right/left O2 sensors WITH independant right/left PW for engine balance?

L98's don't have that. The entire fuel rail runs at a single pressure and all the injectors fire every revolution -- in batch mode.

I could conclude that causes a bigger imbalance under certain conditions (e.g., different lag times before fuel is ingested). But, IDK.

I did see the IAC slope parm and experimented a bit with that. I tried to get it to let the follower close a bit faster since the high-idle sit-down is a bit annoying from my perspective. Mine's not so fast that you'd call it out of the ordinary though. (I think I changed the slope from 125 to 150.)

Cliff:
I disagree about the room for more timing. That knock sensor pulls timing with lower octane. With higher octane, it's "allowed" to run at the intended timing. If the tables were set higher, I think you would see timing being pulled ALL THE TIME -- regardless of fuel being used. That's a bad idea in the event your knock sensor (or ESC) fails.

All:
Another thing I considered lately is the connection from my motor to the frame via my sidepipes. Though they are isolated from the rail by rubber bushings on the bolts, there are still bolts connecting the exhaust to the frame. With an engine having a more pronounced lope -- like a 383 with cam, it might be the case that I feel the engine "shuttering" down to idle -- because of that "hard" connection.

Just as I tried to tweak all circumstances, I took out timing in a narrow range of 1600 rpms. That's where the exhaust resonance was loudest -- to I tried this to make the sound more linear. It did just that. That's why I'm considering a similiar approach where I removed timing in the "shutter" range to see if that reduces it's effect. I could also try removing the exhaust pipe-to-sidepipe connector....then drive it around. If the phenom went away, I'd know the reason. I never noticed it before the 383 (and with sidepipes) but I didn't have a "cam" then either.

 

Timing: you need to look at knock counts to see how well you're doing (assuming the knock sensor isn't picking up false knocks because of your engine configuration). If you're not seeing knocks then you have room to add more advance. I see a few knocks from time to time, but they're transient events and happen rarely so I ignore them.

The problem with open loop is that the BLMs are not adjusted. If anything changes you're not running with the optimum values.

INT is used to tweak the BLMs. Theoretically they will be equal if enough time passes. I think of the BLMs as a "running average" of the INT with a lag filter. My experience is that the INT changes with many parameters and never really settles to a constant value. That's why it's name is "instantaneous". The BLM table is limited in so many ways... I find I'm either in cell 0 (idle) or cell 15 (driving).

One problem in my opinion is that the (stock) L98 has only one O2 sensor. You have no idea what the other bank is doing. The 1227165 ECM has no extra inputs and it would take major mods to add a second O2 sensor (do we really need to measure the fuel pump voltage?).

 

If you're not seeing knock counts, there may be more to add timing. If you add it and they start, you should take them back out. The idea is to have it "naturally" timed where you leave room for running w/o the KS if the ESC or KS fails. It's how I timed mine.

Of course, I did time mine for the highest octane available. So, if either sensor failed, it would ping on the lower quality fuel. That's the way I think it should be set.

I also think that's where stock units are set -- which is why you get the recommendation to run premium.

In my case, my O2 sensor is sitting in a junction of two tubes. My ECM doesn't get a chance to see the other six. (That's due to the Tri-Y header config I run). There is a bung on the other side. It was placed there for use on LT cars. I react to that limitation by reading the plugs to help insure the other six are performing as desired. Plus, everything was new when I installed the 383 -- which should make for "even" operation.

As Jon pointed out once, I didn't start with complete new fuel lines and a new, clean gas tank. So, it's certainly possible that my injectors don't deliver exactly the same amount of fuel any more (especially at the lowest PWs).

 

The spark advance calculation is pretty complex in my BUA firmware for the 1227165 ECM. It adds all these factors:

ALDL spark advance (8 degrees -- not sure what this means (didn't look into it))
Spark advance for error 43 test (20 degrees -- only at startup)
Hot restart retard
Highway mode spark advance (I heard that highway mode is disabled)
Coolant compensated spark advance
Main spark advance
Startup spark advance
Base spark advance (6 degrees for the mechanical distributor advance -- could be modified, depending on your setup)

Then it's all limited to 42 degrees. I read somewhere that the physical layout of the distributor limits the advance to 45 degrees. I have always wondered how far away from the distributor cap contact the rotor contact is at 42 degrees. Seems like there would be a big gap to jump. Obviously the engineers figured all that stuff out because it works just fine.

I had an intermittently bad plug wire on the passenger side and my stock L98 couldn't see it. It ended up ruining the cat on that side and I had to buy a new front Y pipe. The smog guys here in California are ridiculously thorough in checking to make sure everything is there and hooked up. I had to buy a California-legal Y pipe just to get by this. The state sends "ringers" out to the smog stations to try to sneak by the rules and then they nail the stations with a HUGE fine, so they can't afford to take a chance and let you slip through. The last smog check I went through they used a new machine and spent about 15 minutes testing the gas tank to make sure the venting and all was working.

I don't have the plug-reading skill myself. They all look the same to me and I don't have enough experience to be able to do that.

I think the fuel filter will get enough junk to prevent plugging up the injectors. The holes in the injectors are relatively large compared to the size of the particles that the filter will take out. My tank was perfect inside. I have seen pics of severely rusty tanks, so there is obviously a range of "badness" there. I don't know what causes the rust. Maybe the relatively mild climate here in So. Cal. (and my car is in the garage most of the time) makes the difference. I have heard of plugged up drains in the gas filler rubber piece causing water to get into the tank. The drain hole is pretty small in my opinion.

 

Not sure what the info above is for/directed. Regardless, it should be noted that successive years have new/better control over prior ones. For example...

Yes, there is a hwy mode. On my 1989 $6E bin, you can activate it by adjusting the parms. I tried it and believe it does get better MPGs. (jumping the 383 hwy MPGs from 25 to 27-28) HWY MODE is a combination of add'l spark advance and slightly leaner mixture.

For whatever reason, my total advance can exceed the 42-deg max total timing parm. I've seen it in scans. It's probably that the HWY mode module doesn't look at that parm?

Start-Up spark advance decay rate is another parm you can control. (I never tried negative numbers to see if you can affect retarding for hot conditions. But that might be worth a try. I do notice slightly richer fueling for restarts...though it's hard to say that really helps.

Anyone doing tuning also will notice considerably more advance in the low load areas. With a beefier cam and lots of timing, that MIGHT be were I'm getting some shutter on idle sit-down. Next time I drop the PS trim panel, I'll probably get the chip out and lower those values to see what happens. (I did have lower values in at an early stage in my tuning but know that it ran smoother near idle. That was too long ago!)

Actually, I don't recall the shutter during the first 6-mos/year. That's why I wondered if it could be related to injector balance. (Which can not be controlled via ECM parms in an L98.)

 

I posted that because you were talking about adjusting the advance. The calculations are relatively complex with all those parameters and it's not just a matter of tweaking a table. I got the maximum of 42 degrees advance from this (I just happened to remember it -- the firmware adds all the other parameters and then compares the number to this parameter and uses whichever is lower):

I haven't looked at what "relative to lead" means.

I heard that highway mode is disabled by setting the parameters to impossible numbers. This is the kind of stuff that drives me crazy because it all happened 25 years ago and nobody knows why the programmers did what they did. Like the 10,0031 RPM rev limiter. What's up with that? I read somewhere (probably thirdgen.org) that somebody lowered the rev limiter. He said, "you won't like the result". I don't know what THAT means.

I wish GM would make the firmware source code available, with comments, obviously. I speculated a few years ago that they probably couldn't find it now even if they wanted to.

It sounds like you're burning EPROMs. Have you looked at the Moates Ostrich? It uses flash RAM to replace the EPROM and you can make changes while the engine is running.

 

Keep in mind the $6E bin is different. For cruising in closed loop, hot running conditions, the timing table really is THE only thing I need to look at (except hwy mode or WOT conditions). (All those other parms you mentioned are present in mine...they're just irrelevent after warm-up. The slope of the IAC is related to air input.

Probably refers to the mechanical lead of 6-degrees -- or whatever you choose to set.

The programmed it, and it worked. The EPA disallowed it as a tool to "enhance" MPGs. (That's been posted here before.) I suspect it also had something to do with emissions -- since deliberate leaning could affect them -- and/or the converter life. Changing the parms to make it impossible to engage was the easiest way to get rid of it.

I think the rpm limiter was mostly a tool for sales. It might be that Camaro's had it set lower? IDK. But, it won't hurt to set it at an rpm where it'll prevent over-revs.

 

It's been a while since I looked at the firmware (I spent about 6 months in 2010 going through it -- lots of interesting, weird, wrong stuff in there). My BUA code was the first year for the 1227165 ECM. It's a $32 bin.

What do you need to do to enable highway mode?

 

With the $6E bin, you:

• lower minimum temp for hwy fuel mode,
• lower miminum temp for hwy spark mode, and
• lower minimum speed for hwy fuel mode.

I actually have a $32 bin for comparison but won't load it and look unless what I posted above doesn't help.

It looks like the hex locations are: 43D, 440, & 17B respectively

 

This is what I found in the BUA firmware for highway mode:

It looks like lowering the LV8 and RPM will do it. Coolant temp looks OK as is.

I thought highway mode also leaned the mixture, but I didn't find anything related to that.

 

FYI, enabling hiway mode as noted above gave me over 10% better mileage on a road trip last year.