Baro Update

[aaronc7]

Hey guys. Running COS3, 2003.5 C5 Z06 Corvette. Pretty standard A&A corvette supercharged setup, SD tune.

I'm at 4300' elevation and typical atm pressure is 87 kPa. I was using Baro to log a custom "Boost" PID and noticed that the baro parameter was increasing after getting into boost and then it will stay at that "new" baro reading.

My concern is that once baro is updated, it's going to change the MANVAC value in the IFR table in particular and I don't want to end up chasing my tail tuning fuel.

Searched around here and hptuners forums and found a little info on here and almost nothing on hptuners.

I initially changed the various RPM, TPS and time values in the tune to avoid this update scenario per the EFI COS help guide.

Based on my observations and few discussions I found on here... there is a separate logic that sets a new baro if MAP exceeds baro under any condition, with a max of ~104 kPa. This checks, my baro always jumps from 87, to ~104 kPa.

Here in 2019, is there any known workaround for this? I set B0304 to 1.00 across the board based on a recommendation from a thread from 2005. Does B0304 play into this at all or am I SOL? A 20 kPa jump in baro isn't the end of the world....but if I can fix it now before I tune VE anymore, that would be ideal.

[aaronc7]

Nothing?

Here's a log, but it's pretty straight forward. When MAP > Baro, New Baro = MAP, with a limit of ~104 kPa.

Around cells 750 and 1600

22725

[statesman]

You're lean in boost and rich everywhere else... and I don't think your baro should be updating when it is.

Post your tune file.

[aaronc7]

Yeah I changed injector data and "started over", actively working on the VE table right now.

I *think* I had B0304 set to all 1.00 when I did that log (based on a recommendation in an old thread). I put it back to stock values to see if that made any difference.

22733

[statesman]

Have you tested if your current settings actually disable the baro update?

[aaronc7]

I haven't changed the "other" baro update settings in the last few revisions, only B0304. I will do a quick test this afternoon/evening and see if it behaves any different with the stock B0304 table settings.

[statesman]

That's odd because all the other settings you have 'should' effectively disable any baro updates.

[aaronc7]

I agree, that's why I am here lol! These discussions are very old, but all I could find. I'm guessing that 99 percent of people never even pay attention to this, who usually logs baro anyways?

In this thread, GMPX mentions that if MAP > Baro, then New Baro = MAP. Sounds like maybe there is a separate logic vs the normal "baro update' window provides by the baro update tables.

https://forum.efilive.com/showthread.php?850-BARO-creep-in-2-bar

In this thread, he recommends setting B0304 to 1.00, which is why I started with that.

https://forum.efilive.com/showthread.php?518-2-bar&highlight=b0304

I still don't understand what the actual math/equation and how B0304 plays into it. Maybe I can put in some different values in the 105 kPa cell so that when it updates, it will update back to ~87 kPa.... or maybe setting 0.00 across the board will disable any updates all together. IDK, just speculating. I should try this out next.

[statesman]

Interesting problem. Maybe try allowing baro update at very low rpm, say 1200 to 1400 with a TP of maybe 20%... see if that brings baro back down after it has creeped up.

[statesman]

You could also try setting B0304 to 0.87 across the board... see what happens there.

[aaronc7]

I will try all of the above and report back, thanks for chiming in and bouncing some ideas around with me.

[statesman]

Hmmm... you've got a 512kb operating system. I'm wondering if you can run the 12212156 OS. If you can, then you could upgrade from that to a COS5, which I think handles boosted engines better.

[aaronc7]

I actually tried that OS a while ago and the throttle was closing on it's own/hitting some sort of limiter at 6500rpm. Made a thread about it, never got resolved.

https://forum.efilive.com/showthread.php?28573-Hitting-rev-limit-too-early-COS5

I also noticed that trans fluid temp wasn't working. IIRC COS5 only adds a couple nitrous related tables. I tried it initially because I wanted to see if it could add timing when an input signal was triggered (water/methanol injector), but it was not coded in a way to support that. Those couple oddities deterred me away from and made me wonder what other quirks I would run into along the way. But either way I think all boost/SD stuff should be identical.

I will still do the above stuff and report back with results... should have time this weekend.

[aaronc7]

Statesman,

I tried B0304 set to stock settings, 0.83 (104.4 * 0.83 = ~87 kPa) across the board and 0.00 across the board.

I didnt take logs, but all settings had identical results, and all resulted in baro updating to 104.4 instantly once that MAP was reached.

From what I can tell hptuners doesn't even have these tables, so most must just ignore or accept this.

I will be doing VE tuning only after baro is set to 104 to avoid chasing my tail. I've really only noticed a difference at idle. My fuel trims were swinging back and forth and I was trying to figure out wtf was going on. It should be only about a 5% difference but I was seeing more than that at idle for some reason. Now the car runs lean on cold startup, open loop, I'll just have to manipulate the targeted AFRs I suppose.

[statesman]

The way you have things set up, this will always be a compromise tune. You can have a good tune... but you'll need to make some physical changes. This would involve converting your fuel system to return style (which I have seem people do successfully) and fitting a boost referenced fuel pressure regulator. I would also recommend that you install and run a MAF sensor. I don't know what size pulley you're running on your A & A unit, but most engines I've tuned with fan boosters have been delivering around 3-4lbs boost, so if you're boosting around that level you should be able to run a MAF sensor without pegging it. Of course, if you have a cam and other mods it may tip you over the MAF limit even at modest boost.

If you just want to try to tune it out, then your constant problem will be cold starts when the baro resets to actual baro. I suppose what it all comes down to is how long you want to keep this car and how much effort you want to put into making it good.

[aaronc7]

Yeah good point. I opted to do a drop in fuel pump because it was going to be sufficient for my needs and not spend an extra grand. I am OK living with some compromises.

The car made ~450whp SAE before the supercharger and I hit 14-15 psi at redline (180-190 MAP)... I think I hit 512 g/s at around 5100 rpm, so I think I am SOL there.

Looks like I'll just have to tune around it a bit, no the end of the world. thanks for your comments per usual

[joecar]

If you will run boost, you must convert to manifold reference FPR with return line (as statesman said)... your IFR table will now be flat.


AND you will have to scale IFR/VE/MAF (and shift/squeeze all airflow referenced tables) to overcome the MAF airflow limitation (of the PCM).

[aaronc7]

So just to follow up on this, I think I found a pretty good solution. I'm almost hesitant to post up about it because I'll probably be burned at the stake for changing "good injector data", but I traced back the issue(s) at hand and seems to be the best solution/compromise.

The biggest problem was actually not the IFR table, but the dead time / voltage compensation table (forget what it's called exactly, don't have the software in front of me).

When GM.BARO updates from 85 kPa to 104 kPa, the GM.MANVAC parameter is now wrong for all vacuum conditions. If you were at sea level, it would only be a 3-5 kPa delta and probably not enough to really matter. I'm not eevn at super high elevation, but even here, the 20 kPa MANVAC change caused fueling issues and inconsistencies.

I did some maths to see the change that would occur from this... IFR table changed fueling approx 5%, not a big deal. The voltage compensation table accounted for something like a 0.3ms difference, which is significant with my injectors that idle right around 1.6ms.

Long story short, I made my IFR and voltage offset table "flat" like you would for a rail regulated/rising rate setup. Such that the values do not change with changing MANVAC. I copied the 0 MANVAC values down, so this change would not affect my WOT fueling which was already good to go.

It basically just required a full VE table retune and changed the overall shape quite a bit the lower you go in MAP...but my VE table was never "realistic" anyways with a contant pressure fuel system anyways.

So far everything has been working great and consistent....the consistent part being a huge plus. I hated getting everything tuned up just right. Then next time I start the car, fuel trims are 20-25% until I went WOT to 'update' the Baro/MANVAC.

Here's some uncorrected vdyno numbers at 4300'. I really haven't messed with initial tune WOT timing yet and I've seen zero knock so far, so I'm hoping/shooting for mid 600s. Planning to do some final WOT spark timing and part throttle VE tuning on the dyno to finish it off.

https://i.imgur.com/f8ckKTxh.png

[statesman]

I'm almost hesitant to post up about it because I'll probably be burned at the stake for changing "good injector data"

I'm not going to burn you at the stake.... that's joecar's job. :laugh:



So far everything has been working great and consistent....the consistent part being a huge plus.

Getting a good and consistent tune is what's really important. Your tune was always going to be a compromise, so bending things to make the better tune definitely gets the thumbs up from me. :cheers:

[joecar]

lol, I don't have any stakes or matches...:grin:




Good job :cheers:

You might be scaling and not be aware of it.