Thought I had this figured out then ???

When I start my modified 383 in colder weather, it can start to see-saw idle speed. I've tried adding more (open loop) fuel, changing the O2 bias point, and flattening the spark curve in the idle columns.

The issue is transient and only happens for a minute or so soon after cold start. If I start the car and let it warm up w/o swings in rpm, it helps a lot. I'd call it normal.

If I tap the accelerator, it sometimes drops and starts see-sawing. If I kill it and restart, it's still there. Weird! Again, it only lasts a minute/so -- if it does -- and only when colder than 50 degrees.

It's a mild cam that can idle at 700-750 rpms if I want. I choose the higher number when it's above 130 degrees. My idle speed when 50 degrees is about 1000 rpms.

I'm starting to wonder if it's an IAC issue as one of the few parts I didn't replace with the build back in 2010? On rare occasions, I've also notice the IAC seemingly lazy? IOW, once in a blue moon, the idle sits down too quickly as if the IAC isn't following engine speed.

Are these both symptoms of an IAC?

 

On a related topic, I'm curious about open loop fueling.....

The open loop fuel ONLY adds fuel at start-up. 9.7 or 2.7 (at colder temps) is the decay rate. What's the scale? Whatever it is, I assume the extra fuel decays fairly quickly and doesn't contribute much to cold performance. Really, I see basically the same enrichment for RPM as you'd see for PE by RPM. It's an increasing addition of fuel (by load) until it reaches a max of 12-15%....just like hot conditions.

I don't really see much evidence that an engine needs much extra fuel (during warm-up) until below freezing. Maybe that's all that happened with a coil choke on a carb? Throttle plates pulled it open even under load when cold? And, quickly opens as the coil heats up.

Maybe this is one area EFI cars are better? .... Reduction of fuel quickly as it warms up? ... And/or quickly after it starts.

 

I would suspect the much shorter fuel path has a lot to do with the amount and time for warm up fuel.

 

In EFI vs carbs?

 

So i assume your tuning and can edit the program correct?

I dealt with this a little, but I'm also using an ebl flash ecm.

First thing is the O2 sensor, if you have headers you have to have a heated O2 sensor, this prevents it from going in and out of closed loop.

Next would be the base idle and idle recovery.

I had to lower my recovery idle and raise my idle in the software to help.

The final thing that got a rock solid idle was to use a seperate timing table at idle vs using the main timing table...

From there it was all messing with fueling and a few learns while idling and low speed low rpm learns.

 

Also one of the best resources i have found so far is over at the third gen, they have a sub forum "dyi prom" or something close to that.

That's also the home of the guy that built the ecm I'm using.

Also re reading your post, you may need to do some logging to see what's happening.

In my ecm there was an option flag to enable the idle sa table.

​​​​​​​What i saw was my timing was going all over the place at idle, and i also edited my knock sensor to not be enabled until a little higher in the rpm.

 

I have this problem only in winter. I have to adjust my TB throttle stop screw and TPS each winter to make sure it gets enough air to not die. I have never gotten around to checking the tune again, since my tuner can never find the BIN file and his other notes on tuning my engine. I will have to buy all the software and do it myself someday.

I have to hold my foot on the throttle lightly in the morning or whenever the engine is very cold to keep engine RPM steady at 1000 or it will cycle up and down until the engine dies. This is even with the idle in the tune set to 900. My logging doesn't show a real problem that I can see, but I'm leaning towards too much fuel. Once it dies it acts like a flooded carb when I refire.

 

Yes, 3rd Gen DOES have better feedback, documentation, and forum for DIY prom. That's where I learned most of what I needed.

I don't believe the 6E prom has a separate (optional) table for idle.

That said...I hadn't considered the KS being the cause. But, it's a great thought. When idle temps get cold enough to raise idle where I have the "Min KS enabled" parm set, that's about when the problem rears it's head. Also, I have forged pistons that may create "knock" at cold temps. The typical 1-2 min interval it sometimes oscillates in might just correspond to the amount of time necessary for the forged piston to heat, expand, fit "well"?

I will try raising my KS min enabled parm from 900ish to 11-1200...and see what happens.

 

Are you setup with a Moates piggy-back chip? If so, I'd be happy to help. TunerPro lets me (anyone) read the BIN off a Moates chip. The BIN wouldn't NEED to be found...though I'd have to physically have the chip for 1 day.

Edit: Just remembered your site has posts about increasing FP. Is yours higher than stock? If so, was it to achieve higher volume for injector flow...or based on observational improvements? It seems relevant because of your description about the car dying "rich". (FWIW, My tuning learning/efforts over the years make me believe the stock tune is already designed to be on the rich side of things...making the addition of fuel potentially problematic. Plus the fact you need to add air....???)

 

You would see IAC instability on a data log wouldn't you Greg?

Something like a surging idle I verify all the mechanicals first like EGR, PCV, fuel filter and pressure. Then electricals like all good grounds, injectors ohms within 1 ohm of each one, O2 data. Last place would be look at and check the tune. I just don't think surging is something tuned into the ECM.

Please let us know what you find.

 

"Tuned into the ECM"? Ehhhh No. Tuning the surge OUT, yes. With modified motors, it's not that uncommon.

In my case, I did my own tuning with the help of a tuner here...and in 3rd gen. I spent hours and hours learning back in 2010. This year was a refresher...with efforts to see what else I could accomplish.

IAC instability would be hard to spot as the IAC "follows" the idle. To me, it would be hard to determine if the IAC was "following" or "leading" if you don't look at other things.

I have the feeling the KS (knock sensor) might be the answer. Today, I bumped it's activation above 1100 rpms (cold, fast idle). I also leaned it a bit in the 46-deg row. And, I offset my primary idle timing rows with the cold offset table. IOW, where the primary table dropped, I added to the cold table and visa-versa. The goal was to have flat timing during warm-up in the idle rows.

First try was after dinner. Cold start-up was consistent. I'm not ready to say it's cured but it's a good sign.

IIRC, I tried 2 of these 3 in 2010. The only thing I'd never thought of was the KS parm -- in conjunction with surging.

When I built my 383, I converted to 4032 forged pistons which don't expand as much as 2618 pistons but both can create mild knock during start-up. (At least that's my understanding -- though I've never heard it. Doesn't mean the KS can't "hear" it though....)

If there is KS-detectable knock during the first minute/two of warm-up, the KS would have been pulling timing above 900 rpms (where I USED to have KS activation enabled.) Then, when dropping below 900 rpms, timing would rise -- as would RPMs. It seems logical this could induce a surge.

I could also have lowered KS activation below my minimum hot idle. I'm not sure which is "better" though it seems logical to avoid accidentally trigger KS retard where it's most likely to occur with forgings. Plus, if it tried to retard timing during COLD warm-up, that's the opposite of what you want to happen. Extra timing is added until it's warm.

Here's crossing my fingers that this was the issue!

 

FYI, This isn't a requirement.

A lot of it depends on the location of your O2 bungs. Mine isn't further away than it was in the stock manifold -- so it gets hot (and stays hot).

I'm not sure if collector-mounted or post header mounted O2's can cool enough to read incorrectly. I was thinking 600-deg is about the threshold. And I would expect temps to stay that high even with headers....though take longer to get there -- especially if headers aren't coated.