Idle Spark Analysis

[Gelf VXR]

I posted over on the efilive forum, its not very active these days so I thought id post here for some insight.

Its an LS2 with stock LS3 heads and mild cam 216/229, .6100"/.6105"

I got the idea to do idle spark analysis from this guy, http://www.carprogrammer.com/Z28/PCM/Ignition/idle...

So the plan was to log MAP kPa, TB%, airflow g/sec and fuel consumption while varying idle spark up from 10* until I saw minimum MAP kPa. I set all minimum idle air to zero, idle adaptive spark to zero, fuel trims enabled to maintain stoich AFR, set idle 800rpm and I used DVT to adjust timing. The only thing controlling idle @ 800rpm is the electronic throttle.

I was surprised at the results, I could observe stable idle at 50* idle spark timing and MAP was still decreasing, in the other analysis above, MAP bottomed out around 35* then started increasing.

Most notable is the reduction in fuel from 3L down to 2L per hour

Interestingly, there is a table in the LS1, B5919 optimal timing "This table is used as a reference for what would be best timing, which is then used for various torque calculations"

Which would indicate 50* at 0.08gcyl @ 800rpm and 38* at 0.16gcyl @ 800rpm, inline with my results.

Edited by Gelf VXR on Thursday 15th January 04:31

[KMud]

Interesting stuff - Was the idle noticeably different between these different spark advances? Have you tried driving around with such a large idle spark advance, and if so, how was the transition? Have you kept an eye on what the actual spark advance (rather than commanded) is doing?

[M11 MFP]

My warm idle consumption was 2.46 l/hr with an LS1 cam it's now 2.54 l/hr with an LS6 cam. Unsurprisingly little different, given neither are anywhere close to overlap.

2.0 l/hr at stable idle with a larger cam is quite an impressive feat. Does it affect take up on sudden throttle opening, or encourage near
stalling whilst stopping at a junction with the idle that advanced and lean?

Not having your knowledge I'm loath to mess with it, despite having EFI Live. Running a descreened MAF has roughened up the idle a little I reckon, I think the purpose of the screen it to straighten and smooth the airflow, rather than protect the heated wires as some seem to suggest.

[KMud]

M11 MFP said:

My warm idle consumption was 2.46 l/hr with an LS1 cam it's now 2.54 l/hr with an LS6 cam. Unsurprisingly little different, given neither are anywhere close to overlap.

2.0 l/hr at stable idle with a larger cam is quite an impressive feat. Does it affect take up on sudden throttle opening, or encourage near
stalling whilst stopping at a junction with the idle that advanced and lean?

Not having your knowledge I'm loath to mess with it, despite having EFI Live. Running a descreened MAF has roughened up the idle a little I reckon, I think the purpose of the screen it to straighten and smooth the airflow, rather than protect the heated wires as some seem to suggest.

I think your suspicions on the MAF screen are correct.

[Gelf VXR]

KMud said:

Interesting stuff - Was the idle noticeably different between these different spark advances? Have you tried driving around with such a large idle spark advance, and if so, how was the transition? Have you kept an eye on what the actual spark advance (rather than commanded) is doing?

I didn't drive it, no insurance at the time, previously i was at 25* which was the same in the HI oct table, so there were no transition.

At 13* stock value the rpms are unstable >+-100, it definitely needs the adaptive idle spark, at around 40* it idled nicely +-50max. Ringram has just noted on efi forum, and i agree with him,

"The only consideration I can see for not going too close to lowest MAP is for idle control with AC, power steering load, alternator load etc. If you are too close to MBT and LBT at idle you have no room to take on load."

I think there's room for improvement thou

This data is for actual spark, i was commanding the spark in the bidirectional controls, and observing on the scanner, they are the same.

[Gelf VXR]

KMud said:

I think your suspicions on the MAF screen are correct.

+1

[KMud]

Gelf VXR said:

I didn't drive it, no insurance at the time, previously i was at 25* which was the same in the HI oct table, so there were no transition.

At 13* stock value the rpms are unstable >+-100, it definitely needs the adaptive idle spark, at around 40* it idled nicely +-50max. Ringram has just noted on efi forum, and i agree with him,

"The only consideration I can see for not going too close to lowest MAP is for idle control with AC, power steering load, alternator load etc. If you are too close to MBT and LBT at idle you have no room to take on load."

I think there's room for improvement thou

This data is for actual spark, i was commanding the spark in the bidirectional controls, and observing on the scanner, they are the same.

I have(/had) an occasional issue where reversing then coming off the throttle (with clutch dipped) it would stall. I think I've cured this with additional spark. I do think I'm using more fuel than I need to on idle, though conventional wisdom says I'm doing well with respect to cam size and idle speed. I'll get my intake on to 'finalise' my engine, then may join the investigation.

[SturdyHSV]

Gelf VXR said:

KMud said:

I think your suspicions on the MAF screen are correct.

+1

+2

I am very sure that that is the case. The screen is there to provide a clean, consistent and predictable flow of air over the sensor so that it can give a reliable reading. The air filter should do the job of filtering stuff out of the air (not trying to sound like a smartarse there, apologies if it came across like that )

Also, Gelf, have you actually found commanding the idle rpm makes any difference? I can manually control mine down to all sorts of figures, but whatever I command it rests at about 750/800 (I also have the much ignored E40)

[KMud]

SturdyHSV said:

+2

I am very sure that that is the case. The screen is there to provide a clean, consistent and predictable flow of air over the sensor so that it can give a reliable reading. The air filter should do the job of filtering stuff out of the air (not trying to sound like a smartarse there, apologies if it came across like that )

Also, Gelf, have you actually found commanding the idle rpm makes any difference? I can manually control mine down to all sorts of figures, but whatever I command it rests at about 750/800 (I also have the much ignored E40)

If you relax the under-speed corrections, especially for the 50-100 rpm sort of ranges you will probably get it lower...whether that's a good idea

[stevieturbo]

Gelf VXR said:

I posted over on the efilive forum, its not very active these days so I thought id post here for some insight.

Its an LS2 with stock LS3 heads and mild cam 216/229, .6100"/.6105"

I got the idea to do idle spark analysis from this guy, http://www.carprogrammer.com/Z28/PCM/Ignition/idle...

So the plan was to log MAP kPa, TB%, airflow g/sec and fuel consumption while varying idle spark up from 10* until I saw minimum MAP kPa. I set all minimum idle air to zero, idle adaptive spark to zero, fuel trims enabled to maintain stoich AFR, set idle 800rpm and I used DVT to adjust timing. The only thing controlling idle @ 800rpm is the electronic throttle.

I was surprised at the results, I could observe stable idle at 50* idle spark timing and MAP was still decreasing, in the other analysis above, MAP bottomed out around 35* then started increasing.

Most notable is the reduction in fuel from 3L down to 2L per hour

Interestingly, there is a table in the LS1, B5919 optimal timing "This table is used as a reference for what would be best timing, which is then used for various torque calculations"

Which would indicate 50* at 0.08gcyl @ 800rpm and 38* at 0.16gcyl @ 800rpm, inline with my results.

Edited by Gelf VXR on Thursday 15th January 04:31

Trying to use vacuum as a measurement when the throttle blade is moving...sort of screws up any type of test you're trying to do as the movement of the blade will alter engine vacuum.

And yes, smooth airflow is vital to the operation of any MAF, hence straight tubes, screens and general placement of MAF units.

And I've never seen any OEM ever run such high timing at idle...so I'd be cautious that running such high timing could still be impacting the piston/cylinder even if there is no audible pinking.

Maybe run 30deg tops, but even that's high.

Plus as you've no way to tell how timing is affecting emissions ( unless you're using 4/5 gas analyser ) again, stay away from wild ideas.

[Gelf VXR]

SturdyHSV said:

+2

I am very sure that that is the case. The screen is there to provide a clean, consistent and predictable flow of air over the sensor so that it can give a reliable reading. The air filter should do the job of filtering stuff out of the air (not trying to sound like a smartarse there, apologies if it came across like that )

Also, Gelf, have you actually found commanding the idle rpm makes any difference? I can manually control mine down to all sorts of figures, but whatever I command it rests at about 750/800 (I also have the much ignored E40)

I can only command lower than 800rpm using the bi directional controls, EFILive doesn't have the minimum idle calibration to adjust, only the desired idle. It didn't like idling below 750

[Gelf VXR]

stevieturbo said:

Trying to use vacuum as a measurement when the throttle blade is moving...sort of screws up any type of test you're trying to do as the movement of the blade will alter engine vacuum.

And yes, smooth airflow is vital to the operation of any MAF, hence straight tubes, screens and general placement of MAF units.

And I've never seen any OEM ever run such high timing at idle...so I'd be cautious that running such high timing could still be impacting the piston/cylinder even if there is no audible pinking.

Maybe run 30deg tops, but even that's high.

Plus as you've no way to tell how timing is affecting emissions ( unless you're using 4/5 gas analyser ) again, stay away from wild ideas.

Not disagreeing, all good points, I was logging TB which was fairly stable, as timing advanced the TB% reduced with MAP, Its just an analysis, intersting all the same. The LS1 OEM optimum timing is where I was observing MBT.

I would only run 19* for the emmisions test which passed previously

The logs are posted on EFILive forum for those who have EFILive software

Edited by Gelf VXR on Thursday 15th January 12:59

[Gelf VXR]

This paper explains the principles of idle spark control nicely

http://people.eng.unimelb.edu.au/manziec/resources...

[KMud]

"Furthermore, reductions in vehicle NVH may result in [...] reduced cost in engine mounts."

Penny-pinching of the highest order!

[Gelf VXR]

After studying the available E40 idle calibrations, I conclude that the torque reserve needs to be at least 13* {B1530} below MBT with -11* over speed reduction {B5131}.

The calculated torque is based on throttle angle vs RPM {B2513}, there isn't an optimal timing calibration like the LS1.

The extra load from AC auxiliary is compensated by adding airflow to the minimum idle airflow calibration {B1829}, opening the throttle further, there is a delay timer between switching on and AC on to allow rpm to increase accordingly. There is no viewable calibration for power steering load.

{B1829} can be considered to be the same as a stop screw on a carburettor, the minimum TB angle.

An idle tune strategy might be if your have changed cam and heads (or improve fuel consumption over stock),

Determine idle MBT and airflow, log airflow closed loop and trims to correct the airflow, MAP, fuel IPW and or Litres per hour and electronic throttle position, in efilive you can create a custom calc pid to increase the resolution for the TB (8/10ths per 1%), while increasing idle spark * incrementally until minimum airflow/MAP/IPW/TB% / diminishing returns.

Create a table to display airflow per 0.5* timing advance

Decide on the optimal idle timing value at least 13* back from MBT, look at the airflow at -11* from that point, that should be the minimum idle airflow value to place in {B1829} so that the throttle can close to lower rpm.

Test these settings with AC on /off, turn steering wheel to full lock and check for stall.

If it stalls, greater torque reserve is required, decrease idle timing a decrement ally and repeat above.

[MyM8V8]

Gelf VXR said:

After studying the available E40 idle calibrations, I conclude that the torque reserve needs to be at least 13* {B1530} below MBT with -11* over speed reduction {B5131}.

The calculated torque is based on throttle angle vs RPM {B2513}, there isn't an optimal timing calibration like the LS1.

The extra load from AC auxiliary is compensated by adding airflow to the minimum idle airflow calibration {B1829}, opening the throttle further, there is a delay timer between switching on and AC on to allow rpm to increase accordingly. There is no viewable calibration for power steering load.

{B1829} can be considered to be the same as a stop screw on a carburettor, the minimum TB angle.

An idle tune strategy might be if your have changed cam and heads (or improve fuel consumption over stock),

Determine idle MBT and airflow, log airflow closed loop and trims to correct the airflow, MAP, fuel IPW and or Litres per hour and electronic throttle position, in efilive you can create a custom calc pid to increase the resolution for the TB (8/10ths per 1%), while increasing idle spark * incrementally until minimum airflow/MAP/IPW/TB% / diminishing returns.

Create a table to display airflow per 0.5* timing advance

Decide on the optimal idle timing value at least 13* back from MBT, look at the airflow at -11* from that point, that should be the minimum idle airflow value to place in {B1829} so that the throttle can close to lower rpm.

Test these settings with AC on /off, turn steering wheel to full lock and check for stall.

If it stalls, greater torque reserve is required, decrease idle timing a decrement ally and repeat above.

Excellent. Thanks.

[Gelf VXR]

MyM8V8 said:

Excellent. Thanks.

By the way, the three airflow pids,

SAE.MAF is the returned value from the MAF calibrations {B1098} & {B1099}

GM.DYNAIR_LS2 is the dynamic air mode flow

GM.AIRPERSEC tracks the higher value of either of the two above pids.

Edited by Gelf VXR on Thursday 22 January 22:41

[mikeyb1987]

Gelf VXR said:

After studying the available E40 idle calibrations, I conclude that the torque reserve needs to be at least 13* {B1530} below MBT with -11* over speed reduction {B5131}.

The calculated torque is based on throttle angle vs RPM {B2513}, there isn't an optimal timing calibration like the LS1.

The extra load from AC auxiliary is compensated by adding airflow to the minimum idle airflow calibration {B1829}, opening the throttle further, there is a delay timer between switching on and AC on to allow rpm to increase accordingly. There is no viewable calibration for power steering load.

{B1829} can be considered to be the same as a stop screw on a carburettor, the minimum TB angle.

An idle tune strategy might be if your have changed cam and heads (or improve fuel consumption over stock),

Determine idle MBT and airflow, log airflow closed loop and trims to correct the airflow, MAP, fuel IPW and or Litres per hour and electronic throttle position, in efilive you can create a custom calc pid to increase the resolution for the TB (8/10ths per 1%), while increasing idle spark * incrementally until minimum airflow/MAP/IPW/TB% / diminishing returns.

Create a table to display airflow per 0.5* timing advance

Decide on the optimal idle timing value at least 13* back from MBT, look at the airflow at -11* from that point, that should be the minimum idle airflow value to place in {B1829} so that the throttle can close to lower rpm.

Test these settings with AC on /off, turn steering wheel to full lock and check for stall.

If it stalls, greater torque reserve is required, decrease idle timing a decrement ally and repeat above.

Great information Gelf, thanks for sharing.

[Gelf VXR]

Oops, correction, minimum idle airflow value is looked up at idle spark +13*, not -11.

If more torque is required to maintain idle rpm, idle spark increases, the TB angle needs to decrease (more torque from less airflow), if minimum idle airflow is too high, TB can not close further so rpm will increase to equilibrium.

Edited by Gelf VXR on Thursday 22 January 03:05