Any idea of what min idle airflow at the higher rpm's do?
Ie 4000rpm?
Is that for coastdown or what?
Any idea of what min idle airflow at the higher rpm's do?
Ie 4000rpm?
Is that for coastdown or what?
Get EFILive in europe (http://www.efilive.eu).
2007 Escalade ESV L92 6.2L VVT.
2014 VF SV LS3 Maloo.
[QUOTE=hymey;84988]Here is some info on PID controllers, You plug in a commanded number into the calculus equation. It looks difficult but is relatively simple calculus.
http://www.engin.umich.edu/group/ctm/PID/PID.html
I have read through this site 3 times now and the lightbulb still hasnt gone off
I will have to read it some more.
Cheers Ian
Yeah min idle airflow at higher rpms, controls coast down, the rate of speed at which engine rpm drops.
Its good to use to control the rate of speed at engine rpm drops especially at 1500 > 800 rpm. If it drops to quickly in this area it is most likely to hunt.
Don't look into it to far Ian, It basically describes the relationship of Proportional and Integral control.
Reducing the proportional tables by around 20% and increasing the integral tables is a good place to start. But with a stock cam leave it alone. Holden have spent enough time setting this up and it is only when you fit an aftermarket camshaft that they will require tweaking.
Thanks Joel,Trying to get my head arruond it before I do a cam. Have been in maffless 4 some time now but seem to have a problem with coast down. It sounds like it is dropping revs to quick and then recovers. hard to explane, it is if as the auto changes down a gear it sruggles 4 a sec under load.Thinking reading this might give me some insight into the problem.
Cheers Ian
More on this...It appears that min. idle airflow is not used at higher rpm. B1651 and B1652(in autos) is the max idle area, The stock setting is 18.8% TP%. In the tune this is around 1.89. So as soon as you lift the throttle you will see 18.8% in the log which is around 12g/s in min idle airflow. Only if min idle airflow is less then max idle area it could command less throttle at higher rpm. It will control the rate at which rpms drop on coast down and clutch in.
In a cammed car or stroker it would be a wise idea to raise max idle area starting at 2.00 and go from there, you will see how much TP it commands when you back off at cruise. Then you can raise min idle airflow to close the gap between min. idle and max area. For eg set min idle airflow at 11.00g/s at idle. Then set B1651 to 2.00. The PI controller will then learn between around 17 and 19% throttle. This is an example on how it works and unless you raise the max idle area from stock you may run into trouble with hunting issues. By raising min airflow and max area (narrowing the difference) via experimentation, the PI controller has enough room to work and correct idle speed error it will adjust in steps of around .4%. So if idle is hanging and timing is low it is worth lowering max idle area slightly or if the idle dips to much raise min idle airflow.
If min idle area is raised to high(ie 14g/s) and max idle area is stock setting, max idle area will overide rendering the PI controller useless.
I have two tunes, one like above and the second with zero learning. The factory PI controller works well although a PID controller would work better and be easier to setup. By using the standard PI tables and getting the TP% idle limits sorted is an excellent starting point.
Using an "open loop" setup for e38 idle is simple. Max idle area will overide min idle airflow if the later is commanded higher then max idle area(B1651). For eg, set min idle airflow to a high number like 15g/s, so max idle area takes control of the TP, it is locked in at what ever you command. Experiment with idle area until it sits where you want it. Factory setting is a good start, then work out roughly how much timing you need. ie 18 deg idle timing, and 1.30 idle area will give you 18.4% TP. This is how my car is setup and works great. If I tap the throttle now I get no annoying surges associated with having max idle area(to much airflow) to high in a closed loop arrangement.
For an automatic increased airflow is required once in Drive and foot is on the brake. In this circumstance B1652 comes into play. Again if min. idle airflow is higher then B1652(max idle area 2). The later will overide it and the pcm cannot command higher then B1652. So if I command 1.30 at idle in neutral and 2.20(eg) in gear, when I shift from N to D idle TP% will increase from 18.4% to x amount(eg 21% or what u desire) and if the brake is released the car will drive forward. To get this right again would require experimentation but tuning my car in open loop has given more more control and it runs a lot smoother and economy has been better in stop start traffic.
The best way to go is to tune in open loop first and get familiar with the tables, see what the engine likes(ie 18.3% TP idle). Then add a percent on either side for idle TP% limits, For eg 17.3% to 19.3% TP by manipulating B1829 and B1652. Then reapply the factory PI controller values in B1843 and B1845. The idle should be pretty decent and then these tables can be experimented with to help with hunting.
Nice thread. Im going to throw a little hypothesis in.
Im thinking that transient is the new follower and coastdown the new cracker equiv.
So the idle tables with the 3 columns relate to static idle, coastdown/cracker and transient/follower
Ive been messing with these tables for weeks, though only one or two changes a day.
They can drive you a little insane, I do think the stock P settings cause some nightmares, a look into the hsv r8 tables shows some columns to be zero'd out as are different ones in the Z06 tune.
If you get pissed wipe out the P tables, then as Hymey says add them back once you get the rest done.
Perhaps we can come up with a 10 point plan for tuning idle like the old Desired airflow days.. It may also be the case that the E38 is actually a PID controller and we just dont have access to the D function. Or do you know its actually not there Mr H?
Get EFILive in europe (http://www.efilive.eu).
2007 Escalade ESV L92 6.2L VVT.
2014 VF SV LS3 Maloo.
Good idea Ringram. These E38's can be tricky at times, but probably no more tricky than the LS1 in the early days. If we can get some basic processes down on paper, it should make things a lot easier.
In the L76 tunes, the settings are a little different than the VE L98 tunes. For starters, MAX Idle area is already at 2.2 which according to my TPS graph is about 21% real. I've also noticed in some latter tunes, that the "Transient" column is not displayed at all. The second point also means that there are no transient "limit" settings to adjust ( B1835 - B1842, depending on OS ). These limits can also be used to control how much effect the integral and proportional controls make.
The theory here is that you can alter the maximum amount of correction each component can command, both positive and negative. I'm guessing that these will act more like a soft limit per PID rather than the hard limit of B1651. You can use these to restrict the effect of say, normal idle, but allow coast down or transient to have a wider range of control. So, I'm guessing that if you alter min idle air flow, you should probably check these limits as well as they may mask any increase in B1829.
Simon.
On the VE OS the max idle TP% is 18.8% and on L76 it is 21%. Which the later is more ideal for a cam swap.
There is no desired airflow which is a commanded TP% and then the pcm trims the idle from the desired position.
The E38 has a low limit and upper hard limit. Any min idle airflow figures above b1651 are ignored so it will command 18.8%(factory L98 settings). I have my idle perfect now. The trick is to use B1651 to find a desired airflow whilst zeroing the airflow correction tables. Run the min idle airflow tables to around 25g/s so they are ignored. And then manipulate b1651 until a steady idle is reached with timing in the correct region(say 20 degrees). Once you have found a desired value for b1651 "eg 2.20 area which is 21%" then turn on the aircon and see how it reacts. At this stage leave timing correction on. 2nd stage is to enable correction tables. Then set max idle area so it is slightly higher then the desired amount which may raise TP% upper limit to 21.4%. Also reenable to standard min idle airflow tables. Start the engine and watch the commanded TP% on coast down (which will always be max. idle area). Take notice of how rpms dip, also taking into consideration of timing correction and settings. Check it a/c on and off as it will dip slower a/c off.
Once you are satisfied with the upper limit find the lower limit. Which can be done driving a/c off and with stock correction tables enabled. Start the engine then log after the idle has settled, Go for a drive and then experiment to find what the lower limit is. You dont want it to low as if it overshoots on correction TP% will only drop so far. You only need a narrow gap in TP% limits. An example would be to have a lower limit of 20.0% and upper limit of 21.5%. Which is plenty to correct small changes in idle speed and eliminate hunting.
Last is to setup airflow correction. I take about 50% away from the stock proportion tables. And then as simon suggested, use a linear arrangement in the integral tables. The factory integral table will cause overshoot..... a linear arrangement is more effective. The upper parts of the integral table can be .05-.06 which is roughly .5% of correction then tapering to zero with a perfect linear graph. If you take notice with the stock table it isnt linear at all so slight errors in idle will cause larger fluctuations.
You can have excellent control with the e38. Using the above techniques it is possible to have a very choppy idle or very smooth idle both well mannered. Manipulating the airflow limits and timing can give you just that. More airflow-less timing is a choppy idle and vice versa.
It is very much trial and error. But as ringram has suggested small changes daily will reap rewards and once familiar with the tables it is a simple process.
It is hard to put it in steps but generally first off...
1. Find desired TP% using b1651 max idle area and zeroing correction tables.
2. Once you have found this by testing with a/c on and off reenable correction tables and stock airflow tables and then commanded a slightly higher max idle area then desired and test rpm dipping revving the car in neutral again a/c on and off. Find a happy medium.
3. Next we try to narrow the gap between min idle airflow and max idle area so there is a difference of approx 1.5-2% between the upper and lower limit.Set min idle airflow to 10g/s as a starting point at desired rpm. The low setting is easiest found driving with a/c off, higher then usual timing taking notice of the lowest idle limit achieved in the data tables. (please start logging once idle is settled as it will record a lower limit.) We end up with for eg a bottom limit off 19.5% and top limit of 21.5%.
4. Everything between these limits is learn't and idle steadied via the PI controller on the electronic throttle body and also timing correction. Timing correction will be minimal, and then adjust proportion and integral tables. Proportion is reduced and a linear integral table is built.
5. Automatic cars only. B1652 is max idle area for D/R. Select drive and watch what TP% is commanded adjusting the value in B1652 to a desired airflow is reached. Please note that B1651 max idle percentage can only be tested whilst free revving in neutral or park.
6. Once tables are set up effectively, the timing correction tables can be dulled down. At 16 deg over idle it could be 0 or -0.5 instead of -1.5(around stock). This way we are relying on the airflow correction to trim the idle. Timing correction is fast and if it pulls timing to quickly to drop idle rpms to commanded idle speed, airflow will stay high at the top limit until it settles for a few seconds which then it will fall midway(eg 20.5%). After the timing tables are redone and dulled somewhat the airflow and timing correction will both share an even amount of correction in obtaining a desired idle speed with consistent TP% and ignition timing. It is easy to notice if the timing is doing to much as the idle will be a little choppy till it settles. You want to see the idle settle midway between the upper and lower limit. If timing correction is to aggressive it won't do this, the whole purpose is to let the ETC do its job without over compensating with timing correction which is largely a common mistake.
All in all the purpose is to maximise control over the e38s settings(via the limits) whilst still being able to retain a certain amount of the factory learning capabilities for minor airflow corrections. I hope I have shed some light on the topic. If I have wrote anything confusing(from my poor grammar) please just pm me.
cheers
Joel
Last edited by hymey; January 16th, 2009 at 01:14 AM.
hmmm, I'll try it soon. and get u the result.
Regards,
Odd Boy
Since this has been made a sticky, I have decided to follow on here. Basically I went in deep initially and maybe made things to confusing.
I have decided to simplify this. After helping Ian "alian" with his L76 idle today I thought it would be best to outlay some points. There is only a handful of steps and less then ten flashes of the pcm to obtain a rock steady idle.
The only complaints I ever get are that the 230s cam sounds like a 220s cam or 220s cam idles like stock.
The e38 has an amazing ability to control idle. Once idle is in place and correct everything else can fall into place. I know most people on here know how to tune afr and timing well. Idle is tricky but really its simple to obtain.
The first thing is to think old school. I am unsure how many of you guys have tuned an old small block with a dizzy. I can remember how good they idled for a basic carb and a set timing figure. Basically I am pointing out that very bland simple adjustments are to be taken in order to get the idle 99% right. Then altering some of the tables for drivability can be taken.
The purpose of this thread is not to setup throttle followers or crackers. The e38 is missing these. But in getting the idle correct, these characteristics fall into place and then the tables can be slightly changed in order to get a desired result.
Step 1.
Drill a small hole in the throttle blade, 3 to 4mm, depending on the size of the cam. For stock cams bypass this step.
Reason for this is that the ETC TB only trims 0.4% at a time, this can be heard with the engine idling and in some cases a 0.4% change is to much and causes hunting of up to a difference of 100rpm. The min adjustment is 0.4% regardless of integral or proportion adjustment.
This procedure isn't done to correct a bad idle , increase idle air or for plain sillyness like some do with LS1 TB but its to create a air bypass, A buffer ie if the throttle trims it's an abrupt airflow change and with a small hole some air will go through this and some will pass the edge of the butterfly it smooths airflow changes in TB movement and somewhat decreases the trimming effect of airflow and buffers the air movement which in our case is a good thing.
Step 2.
B1651 and B1652 these are both max idle area tables. One is for automatic transmissions when in gear the other for in neutral. I set these the same values. Having a higher max idle air for autos in N isn't required, min idle air covers this and also timing correction, The hard limit in the e38 is much lower then both the factory values anyhow.
We have to adjust this, it allows the correct idle airflow to be obtained.
In order to do this "min idle air" needs to be maxed out to 63g/s so that the pcm uses max idle area as its Throttle position %. So set B1829 "min idle air" to 63 throughout the table.
Next start by entering 1.10 in b1651 and b1652. This is a good starting point.
B1844 and B1845 set to zero throughout the whole tables and save(please keep a copy of your current tune to keep these values we need them later)
Run standard timing corrections and keep spark to around 15 degrees for testing at idle.
Start the engine(warm up) and log timing and airflow, remember these work together. Please use STFT's aswell as it will trim fuel as idle vacuum improves.
If timing is being corrected to low ie around 5 degrees. We know that max idle air is to high, it needs to be decreased. Also turn on a/c and watch the timing it will increase a few degrees. If the car is an auto put in in "D" and watch the timing figures. If timing is to high increase max idle air by 0.04 at a time. change by 0.02 when close.
Step 3.
Adjust max idle air up or down to get desired idle and timing. In a manual I would go for around 15 deg idle a/c off and 18-20 a/c on. Auto pretty much the same but in D maybe let it creep into the 20s.
Now we have that right time to get min idle air correct. Min idle air is actually desired airflow. Also it is overdone in factory trim and works well for a stock cam but it needs to be simplified to make the car run smoother and prevent hunting. Once its correct it can be massaged.
Populate the integral and proportional tables I mentioned earlier. In the past I spent much time on these when I realised that its not overly important. To get simple effective results min idle air and max idle air and timing are most important. All that is required is to reduce these values by say 50%. I go into integral and reduce by 50% same as proportional. Even 20% is enough in both. In fact plenty as long as the 4mm bypass hole has been drilled its fine. Ensure these are populated and save the tune.
If min idle air is to high the idle becomes "maxxed out" as I call it meaning that it is set higher then max idle air. eg 10 g/s could equal 1.10 max idle area this isnt always the case but in general its close.
Set the whole table to 14g/s. Yes the whole min idle air table. Start the engine and while logging etctp watch what it says. Please note this is the main value you check while adjusting max idle area and min idle air.
Once the engine is running take note of the TP%. Rev the engine let it settle to idle and see how long it takes for the TP to start trimming. If min idle air is to high it will take 10-15 seconds before TP reduces from the max idle area position and starts to trim. If its to low and min idle air needs to increase. It will trim immediately if it is low. We increase until it begins to max out then we keep those values. Again, when there is a delay in trimming by 10 or more seconds we know it is maxxed out. Experiment with 14 s a starter and likely you will need to reduce some calibrations are different, Generally 8g/s is where it ends up.
Now to finish off the min idle air. Again, this table is overdone. When infact all gears including N and D can be all the same. The table should minorly taper upwards towards higher rpm. Remember as soon as the values are above max idle area ie(what we checked for before) this is the value that the figures are maxxed out and overlooked. So anything over say 12 is a waste of time as its not even used. Not even the GM engineers sat down to work this one out.
So set upto 1000rpm your final value of min idle air that you worked out. ie could be 8g/s. from 1000-1200 go 9 1600 up 10 and 2000 up 12. Again leave all the gears and P N etc the same value. This is ALL that is required here. Keep it simple like this to avoid hunting or overcorrection.
Step 4.
We have established correct desired airflow by using our max idle area to find it. Now we have it right we can slightly increase max idle area. This allows the pcm to learn and trim within a fine limit and allows perfection of idle in all conditions cold and hot start, winter and summer, Also full lock, with a/c on in gear and nice throttle feel with no surging or overshoot.
I have found over time by tuning a variety of e38 combos from blown to 240s cams on tight lsa to baby cams that an extra 4% add on is sufficient.
So if we first found max idle area for correct idle was 1.10 , our final figure will be 1.14.
This allows more more air at cold start, The factory pcm adjustment will increase min idle air at cold ect by a multiplier. But we need to ensure that we have enough min idle air at cold start to get a nice idle cold. Cold start enrichment here is very important but that's another topic.
Add 0.04 to the figure, If it is not enough add 0.02 more. But generally 0.04 is enough with good enriched afrs cold off 13:1.
This will get your idle correct. Small changes will be required depending on combo and pcm cal. But overall there all similiar outcomes. Max idle area and min idle area always vary from car to car hence why its so important to do a custom tune on each vehicle.
If you are having troubles idle tuning, this will fix the issues. So many tuners believe min idle air solves everything but it doesn't and those cars tuned like that show it because they surge or hunt slightly which is annoying. Some e38s will get the idle right straight up without these procedures but overtime with increased cam overlap mostly the idle has a LTIT effect that learns upwards which is a bad thing. The e38 has outstanding control. Why not use it and get it idling better then that small block? My 240-242-111 now idles like a 220s cam in a LS1 so give these a try and you maybe want to try that bigger cam. Please feel free to pm me on any of this stuff.
Last edited by hymey; October 18th, 2010 at 12:43 AM.