PDA

View Full Version : 2-Bar Woes



NAH
December 5th, 2006, 11:38 AM
Ok, here's the thing. I have an LS1 supercharged & intercooled setup. It is running very badly with a 2.5-bar MAP sensor & custom OS. I have followed the Custom OS Tutorial & applied all the relevant mods (including the MAP sensor calibration, which reads correct against a barometer - we are approx sea level here in the UK).

It idles ok & pulls away fine but when I floor the accelerator pedal, it coughs & jerks until I back off the throttle :bash:. I can accelerate fairly smothly if I go easy on the pedal.

It is pulling timing when the problem occurs (as far as I could see with ScanTool).

I am going to try a few things tomorrow (having read some of the threads in this forum). I have added the GM.DYNCYLAIR_DMA & GM.CYLAIR_DMA PIDS to the dashboard for logging & will do a run before making any mods to the tune file.

Has anyone got any pointers or suggestions (or even had a simialr experience)? Any help gratefully received.

Ta!

SS2win
December 5th, 2006, 01:24 PM
I haven't read the tutorial in a while but when it first came out it recommended you scale the booost tables 300% which I did. Found out very quickly that 300% is an extreme CYA move and it really only needed 15% or so - bad memory. So check your AFR to see where it's at and adjust accordingly. before you try 15% see what 100% or 50% gets you. go slow and be safe.

onfire
December 5th, 2006, 02:39 PM
What's the wideband log when it bucks?

ringram
December 5th, 2006, 09:43 PM
Yes and check what your timing is doing too. It could be burst knock even etc.

NAH
December 5th, 2006, 10:08 PM
What's the wideband log when it bucks?

Erm, don't have one fitted. Was thinking about doing it but only if I really have to. Would you recommend one? (That sounds like a dumb question)!!

TAQuickness
December 6th, 2006, 01:04 AM
MUST HAVE A WB!!!!!

A wideband is a very cheap investment to ensure you tune your motor correctly. You might as well be blind-folded driving your car off a cliff with out one. PM black02ss or visit www.greenlightmotorsports.com (http://www.greenlightmotorsports.com) to purchase a WB.

SS2win
December 6th, 2006, 01:05 AM
Erm, don't have one fitted.
danger will robinson!!

ringram
December 6th, 2006, 01:17 AM
Yep deffo a sensible idea especially with forced induction. In fact almost mandatory.

onfire
December 6th, 2006, 02:41 AM
MUST HAVE A WB!!!!!

A wideband is a very cheap investment to ensure you tune your motor correctly. You might as well be blind-folded driving your car off a cliff with out one. PM black02ss or visit www.greenlightmotorsports.com (http://www.greenlightmotorsports.com) to purchase a WB.

I would not drive the car anymore until you have a wideband to tune the VE tables, etc. I like the Innovate LM1 basic. Chad at GreenLight can get you one quick.....and you need one quick!

NAH
December 6th, 2006, 03:00 AM
I have had the original tune done on a dyno, the AFR is not lean. I have sent a mail/PM to black02ss so hopefully can get one soon.

onfire
December 6th, 2006, 03:26 AM
Did it run well after the dyno tune or did it always buck at wot?

NAH
December 8th, 2006, 05:33 AM
No, the original dyno run was with a 1-bar sensor fitted.

Something is pulling timing when loading the engine. The High Octane fuel table shows a timing of 20° for a given rpm/cylinder-grams/air. The actual timing recorded using scan tool is 4°. What would cause this?

ringram
December 8th, 2006, 06:54 AM
Some cells in your tune have the low octane table reading higher than the high octane.
(eg) 3200rpm at 0.64g shows 22* in low and 18* in high.
Without knowing the exact algorythm that the PCM uses to calc final spark Id suggest that sanity checking the entire tune would be a good start. Weird stuff like that might throw a spanner in the works.

As suggested starting from a known good stock tune and following the guides and tutorials on here and in the docs is always recommended.

NAH
December 9th, 2006, 08:26 AM
Some cells in your tune have the low octane table reading higher than the high octane.
(eg) 3200rpm at 0.64g shows 22* in low and 18* in high.
Without knowing the exact algorythm that the PCM uses to calc final spark Id suggest that sanity checking the entire tune would be a good start. Weird stuff like that might throw a spanner in the works.

As suggested starting from a known good stock tune and following the guides and tutorials on here and in the docs is always recommended.

I have constructed a tune based upon the stock 2004 Monaro tune from Holden Crazy but using the custom OS *03. The values are very different in places & I'd like to know more about what the calibrations do. The whole ETC area is very 'grey' as mine has a cable throttle. For some ETC calibrations, the folder in the list is RED & I assume that these are not relevant to my PCM.

Anyhow, what I am attempting to do is co convert the current tune (which has been constructed using the 2001 Commodore Custom OS *03) to the one mentioned earleir as you suggested. Then I can compare the differences between the the stock 04 Monaro tune & the current tune. I will have made the changes to the stock tune as described in the custom OS tutorial.

BTW, sorry to not have called you today but had some serious s**t with an ongoing site - work gets in the way of everything.

ringram
December 9th, 2006, 11:15 AM
Do you want to convert the old tune? Depends on what you are trying to do.
You can convert it to see if the OS level helps. But you might want to stage the migration so you can observe what happens as you bring over each table.

NAH
December 10th, 2006, 11:42 AM
Success, the problem was with B6216 Burst Knock Retard. Setting this to zero cured the problem of the pulled timing. Now I am seeing 140+kpa manifold presure but the AFR (indicated by the sensor voltages) a bit rich.

I need to tidy up the tune for the boosted region A0009 & the AFR settings, but need some dyno time to make it right. I will also fit (hopefully) a wideband next week.

I am a happly chap. Thanks Richard for your invaluable help, top job.

dc_justin
December 10th, 2006, 04:49 PM
Success, the problem was with B6216 Burst Knock Retard. Setting this to zero cured the problem of the pulled timing. Now I am seeing 140+kpa manifold presure but the AFR (indicated by the sensor voltages) a bit rich.

I need to tidy up the tune for the boosted region A0009 & the AFR settings, but need some dyno time to make it right. I will also fit (hopefully) a wideband next week.

I am a happly chap. Thanks Richard for your invaluable help, top job.

Ah, I had that problem when tuning. It means that the column to the right of the highest column you're seeing in the boost VE table is too high. Look for that "shelf" in the boost table and reduce it manually to something reasonable.

Example. You may see 179kPa at WOT. Scanner will log that data in the 175 column, and your AutoVE stuff will only alter that 175 column. In actuality, it's interpolating between the 175 and 185 columns. If your 185 column is drastically larger, BKR will be triggered.

ringram
December 10th, 2006, 09:03 PM
Actually it was being triggered at quite low map, even in the stock OS.
Adding the new 2bar sensor obviously put bigger numbers into the equation making it worse, but it was there in the stock OS too.
All sorted now, most people zero out BKR anyway. As long as the spark map is done properly it shouldnt be needed. We logged no KR under load and only a small one or two degree tick on taking off from idle.
I think its for designed for granny who refuses to shift down from 6th gear to overtake around town.

dc_justin
December 11th, 2006, 04:20 AM
Actually it was being triggered at quite low map, even in the stock OS.
Adding the new 2bar sensor obviously put bigger numbers into the equation making it worse, but it was there in the stock OS too.
All sorted now, most people zero out BKR anyway. As long as the spark map is done properly it shouldnt be needed. We logged no KR under load and only a small one or two degree tick on taking off from idle.
I think its for designed for granny who refuses to shift down from 6th gear to overtake around town.

If I zero out BKR, I get real knock BAD when rapidly going from light throttle to heavy throttle. Roots supercharger will go from 10" of vacuum to full boost almost instantly, so BKR is a lifesaver there. :)

ringram
December 11th, 2006, 04:24 AM
Cool, NAH can keep an eye on it and maybe put a little bit back in if required. It was set to 14* before zeroed. I guess the delta airflow values could also be increased to move the trigger level up too.
Lots of fun to be had tweaking :D

dc_justin
December 11th, 2006, 04:26 AM
Cool, NAH can keep an eye on it and maybe put a little bit back in if required. It was set to 14* before zeroed. I guess the delta airflow values could also be increased to move the trigger level up too.
Lots of fun to be had tweaking :D

Yep.

I'll typically reduce the amount of effect BKR has by 50% on a N/A motor, but even that will trigger big knock with mine. Probably also has to do with the fact that the gasoline here is pathetic as well.

NAH
December 11th, 2006, 05:06 AM
Yep.

I'll typically reduce the amount of effect BKR has by 50% on a N/A motor, but even that will trigger big knock with mine. Probably also has to do with the fact that the gasoline here is pathetic as well.

Well presumably I can add some BKR back in until the problem re-materialises, then back off a little.

Can someone give me some typical values for a s/c tune for B6210 Burst Knock Delta Air Mass & B6212 Burst Knock Retard (for reference purposes)?

I tried to open a boosted tune from Holden Crazy but it flashed up a PCM warning & the calibrations were empty.

In B6210, can the change in air mass be negative or positive? If so, what is the reasoning behind this?

Presumably I could log rpm/DYNCYLAIR & establish the delta air mass which could then be used to retard when boost was higher than normal?

Justin, is your Roots a centrifugal? I think the Vortech gives a more steady boost than screw type. Whith your BKR setting, to you notice the timing being pulled when you gun-it?

dc_justin
December 11th, 2006, 05:29 AM
Justin, is your Roots a centrifugal? I think the Vortech gives a more steady boost than screw type. Whith your BKR setting, to you notice the timing being pulled when you gun-it?

Roots is a type of compressor, the way centrifugal is. I have a Magnacharger radix, and yes, your vortech builds boost in a much smoother manner.

I do have timing being pulled when I mash the throttle, as much as 20* or more being pulled. Without that, I'll have big KR spikes. One instance was 8* of KR that stayed peaked out for about 3/4 of a second before starting to fall.

When I was tuning the VE table, I experienced much of what you described in your original post at heavy throttle. When my map value would increase ever so slightly and start to interpolate with the next column of data, the PCM would calculate a large change in airflow and my timing would end up looking like a saw blade, bouncing between 15* and 10* or so...

NAH
December 12th, 2006, 09:10 AM
Been looking at the Boost VE table. The settings were as 105kpa (Main VE table) to +300% for the last colume.. I have reduced the value to +150%. I have also added some figures to the BKR (max 4°).

Logging today showed KR recorded at less than 0.5 under acceleration (3.9 under throttle lift-off). If I set the BK Delta Air Mass with a higher number to a lower number rather than a lower number to a higher number, will this affect how the table is applied?

Also, I logged some runs today & didn't get above 140kpa manifold pressure but got 2.10 grams/cylinder. Is this right/possible as the Octane Tables only go as far as 1.20?

Anyway, gonna flash the modded tune & go for a run. Ho ho ho!

ringram
December 12th, 2006, 11:58 AM
Probably yes. 2.10 sounds ok. But it is likely to be affected by your high VE table values at present. When you tune them right the value is likely to come down. Timing will be either the last row on your timing table. Or look in the COS3 additional tables there is an extra table there to tweak timing. A0010 is what you are after.