When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
The OP needs to learn how to tune or take it to someone who does. There is nothing wrong with learning but it takes time and the actual work at hand is maybe only an hour to get 90%+ correct.
.
Actually, I've been tuning for much of the past decade, since I have an NA with MS2 that's my year-round, daily driver in Alberta winters that sometimes get down to minus 40. But on that, I've never had a problem with these weird idle spikes like I have right now. I'll have the idle nearly stationary in both VE and spark advance, and nestled in a perfectly homogenous group of cells, and then every 20 seconds or so I'll get another annoying spike. So I'd really like to know what is prompting the ECU in these cases.
What your saying and what I have seen in your tune don't make sense. I can develop an adequate afr and timing map for these engines in 30min. Is it the engine requirement that you lack knowledge in? Put down a conservative timing map and the afr requirement for an NA application is very straight forward. Max power is between 13.5-14:1, lean cruise with a .040" (1mm) plug gap can be 15-16:1 and the best idle is in the 13.8-14.2 range. Whats the hold up? You need the basics to be correct in order to really problem solve your issue. It might be a mechanical or an electrical issue but your so out of tune its impossible to make a definitive diagnosis.
Originally Posted by quixotic
Actually, I've been tuning for much of the past decade, since I have an NA with MS2 that's my year-round, daily driver in Alberta winters that sometimes get down to minus 40. But on that, I've never had a problem with these weird idle spikes like I have right now. I'll have the idle nearly stationary in both VE and spark advance, and nestled in a perfectly homogenous group of cells, and then every 20 seconds or so I'll get another annoying spike. So I'd really like to know what is prompting the ECU in these cases.
What your saying and what I have seen in your tune don't make sense. I can develop an adequate afr and timing map for these engines in 30min. Is it the engine requirement that you lack knowledge in? Put down a conservative timing map and the afr requirement for an NA application is very straight forward. Max power is between 13.5-14:1, lean cruise with a .040" (1mm) plug gap can be 15-16:1 and the best idle is in the 13.8-14.2 range. Whats the hold up? You need the basics to be correct in order to really problem solve your issue. It might be a mechanical or an electrical issue but your so out of tune its impossible to make a definitive diagnosis.
Uh, we're talking about an NB (first sentence in the thread). My afr table is a solid 14,7 in the idle range. Spark at 10 degrees in the idle range. And VE at 42 in the idle range, is directing the dash gauge to display mid-14's...except when the cursed afr/rpm spikes occur I'm not sure why you're bringing max power and cruise into this discussion.
NA is a common abbreviation for naturally aspirated as in not forced induction sometimes referred to as FI. The timing table is off and severely in the idle area. Ignition timing is to allow the BMEP break mean effective pressure to occur at 18 degrees after TDC top dead center. The flame propagation rate will vary with mixture and speed but will not have nearly the effect on timing as the change in time from the change in engine speed. If your running high quality 93 octane you can run 33BTDC in the 100kpa row but in these engines they make the most HP horse power around 30-31. It sounds like you have good handle on things good luck with figuring things out.
Status 6 is telling us you turned on the fan. When that happens your idle up setting adds 300 rpm to your closed loop target. Your dumpster fire idle timing curve immediately adds 19 degrees of timing. This is absolutely not the only thing wrong with your tune but it's why the car spikes lean.
You're doing a really good job of pissing off the strangers that are taking time out of their day to help you for free.
I should add that the timing jump wouldn't even be that big of an issue if the engine got more fuel to go with it but since you set all of your idle ve cells to the same value it's not getting any more.
I didn't turn anything on or off during the logging process. However, at least once in the past (with the hood open and a stethoscope in my ears, trying to figure out where the clicks were coming from, at the same time as the spikes), I noticed the radiator fan switching on at the precise moment as a spike occurring. Could that be what you're referring to?
I can now see how my post #15 could have been misinterpreted, and I apologize for that. But at the time, the analogy I was thinking about was a fire marshal telling a home owner that their house was dangerous, and then saying nothing more.
I didn't turn anything on or off during the logging process. However, at least once in the past (with the hood open and a stethoscope in my ears, trying to figure out where the clicks were coming from, at the same time as the spikes), I noticed the radiator fan switching on at the precise moment as a spike occurring. Could that be what you're referring to?
Defrost was on right? I assumed the fans triggered because you had just turned on defrost to simulate the dip but it will obviously still cycle occasionally if you had defrost on for that entire log.
The dip is definitely caused by the fans turning on and your tune not be configured correctly to handle it.
Originally Posted by quixotic
I can now see how my post #15 could have been misinterpreted, and I apologize for that. But at the time, the analogy I was thinking about was a fire marshal telling a home owner that their house was dangerous, and then saying nothing more.
I don't think I've heard that one before, but I can now see your intention.
Defrost was on right? I assumed the fans triggered because you had just turned on defrost to simulate the dip but it will obviously still cycle occasionally if you had defrost on for that entire log.
The dip is definitely caused by the fans turning on and your tune not be configured correctly to handle it.
Yes, defrost would have been turned on before the logging started.
Yes, defrost would have been turned on before the logging started.
Good. No mystery here then. Defrost will turn the a/c compressor on just like turning on the a/c will. When the a/c compressor comes on both of your fans are set to come on which increases your closed loop target rpm to 1250. Your timing correction curve then adds what is probably twice the amount of timing you want to add, and the engine doesn't get the increased fuel it needs so it spikes lean.
Make your idle timing correction curve less aggressive and tune your VE properly and that issue will go away.
Good. No mystery here then. Defrost will turn the a/c compressor on just like turning on the a/c will. When the a/c compressor comes on both of your fans are set to come on which increases your closed loop target rpm to 1250. Your timing correction curve then adds what is probably twice the amount of timing you want to add, and the engine doesn't get the increased fuel it needs so it spikes lean.
Make your idle timing correction curve less aggressive and tune your VE properly and that issue will go away.
Many thanks for that. Succinct and very informative.
I see now that the base map comes with rpm designations on the x-axis (for afr, VE and spark) which don't match, so I'll synchronize them for easier tuning.
For spark, would you recommend that I turn off either the idle advance table or the idle RPM timing correction table for easier tuning? Or would the extra precision be necessary for what I'm trying to accomplish?
Actually, I may not be tuning for while, since my NB goes into hibernation tomorrow (our first blizzard hit today). My NA then comes out, due to its heated seats, winter tires, VLSD, and slightly better ground clearance.
My recommendation would be for you to turn off idle VE and idle advance and rely on a less aggressive idle rpm timing correction curve to keep the idle smooth when things like heavy electrical loads are applied. I think that strategy will be the easiest to tune and still get you the result you're after.
My recommendation would be for you to turn off idle VE and idle advance and rely on a less aggressive idle rpm timing correction curve to keep the idle smooth when things like heavy electrical loads are applied. I think that strategy will be the easiest to tune and still get you the result you're after.
This is how I figured it out. I'm still not convinced idle VE is all that, as a well tuned VE table in the idle area works really well.
Another vote for not using Idle VE. I recently disabled AC idle up completely on an NB. I made sure it ran well at 900rpm and my planned AC idle of 1000rpm. Even if you get the delay/idle up perfect, it'll still dip with the extra load once the compressor and fan kick on. Use idle RPM correction curve to keep the idle stable as AC kicks on. Once you're happy with it, then try enabling AC idle up just to bring the idle up to 1000rpm.
Apologies for not returning to the sooner, but I haven't been using the a/c much, since I have a rally roof vent for city driving, and covid has curtailed my vacations (and thus highway driving).
Anyway, I've followed the above advice and zeroed the timing correction curve, and turned off the idle advance and idle VE. Now the idle and AFR hunting has completely disappeared. Thanks very much to all of those folks who helped. When I stop at a light, the AFR tends to sit around 15.2 or 15.3, but compared to the hunting, it's a minor annoyance. And once I'm in the city, I tend to turn off the a/c and open the roof vent anyway.
10-15-2020, 06:08 PM
-1
Thanks very much to all of those folks who helped. When I stop at a light, the AFR tends to sit around 15.2 or 15.3, but compared to the hunting, it's a minor annoyance. And once I'm in the city, I tend to turn off the a/c and open the roof vent anyway.
