Cold start seems a bit...off SD

[Jim Blackwood]

I do have a fuel pressure gage, pretty well instrumented in fact. My supply drops off within a few seconds of being powered down, which is why my Prime Pulse Mass is set to all zeros. No point in it as the pulse would end before FP ever came up and would never be consistent. I use the 1st and 2nd prime pulses instead which if I understand it correctly, fire off after cranking begins. And that does work. Like I mentioned currently the under 70 degree startups light off almost immediately and in fact it does seem to be immediate if I've got the settings just right. Back in early Feb when I first started on the transients problem it had that behavior on startup from 50 degrees up. So I've just spliced some values in from that earlier tune for the upper range of my transients table and my FA multiplier table and will be testing it later today. With a bit of luck that will get me closer.

BTW, those values in the transient table were clearly made up numbers by somebody who was simply guessing, that much is clear from the range and progression. But I only copied 3 or 4 lines in the top right corner at values below .5 so it should be safe enough. I'll see what happens. It's a low performance engine so it can handle a little stress. I'll post the results later today.

Jim

[Jim Blackwood]

Wish I had more to report, results are kinda uncertain. For the most part the engine is starting well but then sometimes not quite as good. Worst case I have to crank the key 2 or 3 times and/or mat the gas pedal but the rest of the time it lights right off. I'm not real sure why the difference but I think as I drive it more I will be able to identify the circumstances when it is harder to start and with any luck be able to tune that out. For now it seems acceptable.

Jim

[Jim Blackwood]

I changed the transients table back to stock Silverado.

At this point it appears that I have two remaining issues.

1) Occasionally I get a throttle hang at about 2500rpm, so far I haven't identified any consistent time when this happens.

2) after warmup and then shut down and restart within 5 minutes it does not start on the first bump. Or second bump. But with choke clear it does start on 3rd crank. This is at about 60-70 degrees ambient. (so far)

Any suggestions?

Jim

[kingtal0n]

Try using the soaktimer to pull around 25-30% of first/second pulse mass fuel interpolated from 0seconds to maybe 10 minutes

[Jim Blackwood]

Thanks for that, I didn't even know that was there. I've attached a screen shot of my Fuel/Open Loop/Cold Warmup/Afterstart/Hotsoak Enrichment and /Initial Adder Soaktime Multiplier. The Soaktime Multipliers are all zero. That doesn't seem right to me. the Initial Adders are .75 up to 540 (s) and 1.0 above 600 (10 minutes).

Would that change your suggestion?

Jim

[kingtal0n]

The afterstart stuff I use to combat IAT heat soaking from when the vehicle sits in the hot sun after being drove for a while, then shut off while hot in the sun. After a good 10 min the IAT will be like 120*F or something under the heat soaked hood while the car is just baking in the sun. The average temp of the hood in sunlight even if you never run the engine at all is still 120-135*F I think. So the underhood temps can easily rise to 120+ after the engine ran + Sunlight here. Thus the engine will start and be very lean because the OEM ECU has no modifier for IAT (you can't access that part of the tuning). So this afterstart table I use to make sure the IAT heatsoak is offset somewhat.

I also have a 500Ohm Variable resistor on the IAT to help with heat soaking once the vehicle is started, as needed. I would use 1000Ohm in the future though.


But all of that is for afterstart. What you want is pre-start. Soaktimer has a couple qualifiers make sure you enable soaktimer temperature thing using ECT. Make sure its working and turned on. Then adjust it so the fuel is less when you recently started the engine.

If it's already working and turned on maybe its pulling too much fuel instead. Check first whether it was already active or whatever

[Jim Blackwood]

OK thanks. Sorry to take a couple days getting back. I had to pull and tear down the engine from our group project car and we just got that finished and off to the machine shop yesterday. I just ran another warmup test to fine tune my ECT warmup curve and think that is one or two test runs away from set but of course when I bump that up I have to bump the FA multiplier down so it still starts easy. Is this consistent with your experience?

Anyway, I'm now waiting on the engine to cool back down for the next test and I haven't decided whether to do hotstart testing this afternoon or wait until tomorrow. I'll be getting into it within the next day though I'm pretty sure. I'll let you know what I find.

Jim

[Jim Blackwood]

Warmup is now very good and cold start is close. Test that again tomorrow and may be done with it until it gets very cold again. That leaves heat soak and transients. I'm going to go back and review again the comments on those areas and make sure I have clear in my mind the steps I want to try. I suspect the heatsoak will need leaned out, that shouldn't be too hard but as I'm not familiar with the new parameters it may take some fiddling before I feel comfortable with those settings. The transients seem to me to be an area where I have muddled about without very much sense about what I've been doing, but in my defense I was just trying to get rid of that injPW spiking behavior which I did, and later trying it on to see if it affected the idle hang which was inconclusive. I never have tried yet to adjust the lean accel spike or the rich decel spike with it. Fortunately those do not presently affect driveability but if I can decrease them by 50% I should be able to lean out the VE map some and get better economy. Here's a scan with a good shot of the transient spikes.

I think I'll leave the idle hang for last. It's annoying, but only happens occasionally now and usually is at less than 1500rpm so in the short term I can live with it. Probably something I did in the adaptive idle before this last round of tuning, so at that time I'll go back and look at the changes I made and see what makes sense.

Jim

[Jim Blackwood]

I guess I didn't post my last notes. Anyway it's in the 20's today and ECT was 45. Fired right up and no hesitation driving down the street so I'm pretty happy with the cold start and warmup.

On the transients: I tried bumping up the Impact and bumping down the Boiling. Went about 10 steps of 5% on each and it made the spikes fatter. Feels like I was going the wrong direction. So I went back to baseline and increased the impact gain 50% and got the following scan but that was with the engine cold. I'll have to warm it up to do an accurate comparison. Looks like the spikes got shorter and fatter though. Not sure what that means.

It will be warm again in a day or two and I'll get back on the hot start.

Jim

[kingtal0n]

Before you play much with transients get the VE table very smooth and finished almost.

The VE table can contain information regarding transient behavior. For ex between 60KPA and 80KPA You will be richer whether its a steady state or a transient. So tune those regions in the VE table as if they are part of transient fueling. It will help you avoid needing to increase the transient fuel after the fact, and rely more on the VE map to become rich.

In fact its a little advanced tuning theory but I think you can handle it- look at my VE table and notice the way I've got extra fuel coming in for 60-85KPA. Those regions are pre-PE mode and yet they enrich leading up until the PE kicks in, after which the VE table flattens back out.
This is because
A. don't depend on rate of change transients to enrich you prior to PE Mode. This allows you to relax the PE mode restrictions, they can be 'late' and you still get enrichment as needed with pressure increase from the VE table.

B. The OEM ecu pretend to use a closed loop leading up to PE- but closed loop is 14.7 and sometimes you want 14.4 or 14.2 and other times you want 13.3 or 12.8. So instead of trying to fine tune the PE mode activation and deal with differences at various RPM (they PE mode "turns on" and sometimes it can miss its mark at 'other' rpms) simply tune the VE map like a stand-alone ECU (whatever A/F @ load point you wanted) and leave PE off until some absolute point of no return where you can be assured the enrichment is worthy of the conditional.

[Jim Blackwood]

Sounds like good advice. My VE table is really close and pretty smooth, I take that pretty seriously. I've seen plenty of way worse OEM tables. Using my AFR% graph I'm showing under 1% in most areas.(fully warmed up of course) But if I understand you, what you are recommending is to enrichen the table for accel like what we did on the MS VE table. I was holding off on doing that, wanting to get the PE and transient tuning dialed in. Actually in terms of driveability the tune is now very good and it's only in the AFR trace that I see an issue where the trace spikes on throttle excursions. If I can get those reduced I think I can lower the VE values at cruise and get better economy but that's more of a long range goal. So I think I really need to get a handle on this transient tuning and understand what things do. Right now it doesn't seem to be working the way I've seen it explained and I don't know why.

But I also think your suggestion is a good one, and I'll give it a shot. Just maybe not in the next few days. I'm going for one more test drive today to get a warm trace on the AFR and look at that. Then inlaws tomorrow and back on transients Monday when it's warmer, then hot start probably Monday afternoon. In the process I hope to get a better understanding of transients and how much I can do with that. Between it and tweaking the VE table I would think I could get a fairly smooth AFR trace.

Jim

[Jim Blackwood]

I think I need to pay more attention to the AFR commanded trace. Maybe clean up the scanner a bit so I can follow that better. It was staying pretty flat before, and now it's moving around a bit. Looks like that could be a solid clue to just what's going on here.

Jim

[smokeshow]

Before you play much with transients get the VE table very smooth and finished almost.

The VE table can contain information regarding transient behavior. For ex between 60KPA and 80KPA You will be richer whether its a steady state or a transient. So tune those regions in the VE table as if they are part of transient fueling. It will help you avoid needing to increase the transient fuel after the fact, and rely more on the VE map to become rich.

In fact its a little advanced tuning theory but I think you can handle it- look at my VE table and notice the way I've got extra fuel coming in for 60-85KPA. Those regions are pre-PE mode and yet they enrich leading up until the PE kicks in, after which the VE table flattens back out.
This is because
A. don't depend on rate of change transients to enrich you prior to PE Mode. This allows you to relax the PE mode restrictions, they can be 'late' and you still get enrichment as needed with pressure increase from the VE table.

B. The OEM ecu pretend to use a closed loop leading up to PE- but closed loop is 14.7 and sometimes you want 14.4 or 14.2 and other times you want 13.3 or 12.8. So instead of trying to fine tune the PE mode activation and deal with differences at various RPM (they PE mode "turns on" and sometimes it can miss its mark at 'other' rpms) simply tune the VE map like a stand-alone ECU (whatever A/F @ load point you wanted) and leave PE off until some absolute point of no return where you can be assured the enrichment is worthy of the conditional.

...no. There is no overlap between VE and transient fuel at all. Everything transient is downstream in the calculation. Do not modify the steady state airflow model to account for a transient, ever.

[Jim Blackwood]

Smoke, could I pick yer brains a bit on what some of these settings for transient do? I've been playing with them a bit but not really getting the results I've expected. For instance, is "Fuel on WallEXP Decay Mult vs Airflow" a gain setting for the fuel evaporation?

Jim

[Jim Blackwood]

I guess it's probably time to close this thread out. Maybe open a transients thread some time.

There's still some fine tuning to do on startup but it's mainly leaning out the FA table and the prime pump curves at this point. The car always starts on the second crank by adding flood clear, often starts on the first crank. So goal achieved, and fine tuning will make it better. Another few sessions of fine tuning the VE table may also help.

Transients may be more of a long term goal with the objective of improving mileage. Right now it doesn't affect throttle response.

Jim

[Jim Blackwood]

I've been driving this for awhile now and, let me know if I'm off base here but I've about come to the conclusion that the starting problem is really a consequence of the fueling system and can be corrected with an in-tank pump. Did the 2002 Camaro/Silverado maintain pressure in the fuel rail between starts? My setup does not. It repeatedly starts on the second crank.

Jim

[Jim Blackwood]

I'm seeing posts that seem to indicate that this business of starting on the second crank is a pretty common problem. Lots of posts about it and lots of guys chipping in and saying theirs does too, asking if anyone has found the solution.

So has anybody? Surely someone somewhere has found the key to this. At least it's good to know I'm not alone.

Jim