Ignition Timing on DI Engines

Discussion of tuning specific to MAZDASPEED6/MAZDASPEED Atenza/Mazda 6 MPS vehicles
mituc
VersaTuner guru
Posts: 1324
Joined: December 17th, 2011, 2:47 pm
Location: Iasi/Romania

Re: Ignition Timing on DI Engines

Post by mituc »

Dolfius wrote: June 28th, 2020, 9:52 pm What loads are you seeing in your 3 MPS @mituc? You're probably running some sort of aux fuelling there but @ 32psi you should see well over 2.5 load in your logs.
I target and see 2.6-2.65 but that's on the MAF cal that was trimmed down for meth a few good percent (12-15%). So the real load is actually 12-15% higher.
2008 Cosmic Blue Mazda 3MPS
Built engine + WMI + GTX3071 gen2, ~509BHP @34PSI
2008 Icy Blue Mazda CX7
Stock engine and exhaust (YES!!), JBR3" + GTX2867 gen2 + Autotech HPFP, self-tuned to 330-ish BHP
Dolfius
Posts: 64
Joined: May 31st, 2017, 2:54 am

Re: Ignition Timing on DI Engines

Post by Dolfius »

mituc wrote: June 28th, 2020, 9:28 pm
Dolfius wrote: June 28th, 2020, 11:48 am viewtopic.php?t=17

My friend came across this. I'm surprised I missed this topic.
The last entry before mine was in 2010. 10 years later and no update (solution)?
I tend to think it's tune related and I think I figured this out while I was tuning my CX7. I need to make more tests though.
Please let us know know what your findings are.
Lolmooses
Posts: 43
Joined: June 16th, 2020, 1:53 pm

Re: Ignition Timing on DI Engines

Post by Lolmooses »

Really strange that some have the issue and others don't. Would it be useful to share both of your tunes and compare to see if anything stands out?
2006 MS6: CS HPFP internals, stage 2 CS intake, CP-E TMIC, CS catback with racepipe, GFB hybrid BOV, CS EBCS
mfinlay04
Posts: 14
Joined: February 13th, 2015, 3:27 am

Re: Ignition Timing on DI Engines

Post by mfinlay04 »

Bringing this back from the dead. But here is a link to rfinkle2's base ignition timing for our cars. I have always started with this and made adjustments based on logs.

https://web.archive.org/web/20191115104 ... ax-137280/

post 24 starts the good stuff :woot

What most people don't understand is that all ignition tables should be set to the same values. Knock and No-knock tables is a bit deceiving as the ECU will pull timing regardless what these tables say if it detects knock.
mituc
VersaTuner guru
Posts: 1324
Joined: December 17th, 2011, 2:47 pm
Location: Iasi/Romania

Re: Ignition Timing on DI Engines

Post by mituc »

One other thing to mention is that at least across Europe with these new E10 fuels the engines seem to accept and like quite a bit more timing up to certain load values in each RPM point. I recently has the surprise to see that on a big turbo setup of one of my friends who is using 95RON gas (91OCT in the US) I could use slightly more timing than I was on my engine with exactly the same turbo (GTX3071R gen1, I sold mine to him) with 100/101RON fuel E5 back in 2013-2014.

I can't speak for the other regions where the fuels need to have 8-10% ethanol but here is how it works around here.
2008 Cosmic Blue Mazda 3MPS
Built engine + WMI + GTX3071 gen2, ~509BHP @34PSI
2008 Icy Blue Mazda CX7
Stock engine and exhaust (YES!!), JBR3" + GTX2867 gen2 + Autotech HPFP, self-tuned to 330-ish BHP
User avatar
ahayden04
Posts: 5
Joined: May 24th, 2022, 4:31 am
Location: South Carolina

Re: Ignition Timing on DI Engines

Post by ahayden04 »

Lolmooses wrote: June 28th, 2020, 1:21 am You don't seem to be having that load cap issue I've read about a few times here where they'll shoot for a load higher than 2.2 or so and they have to overshoot to hit it while fighting the AFR's going lean.
I just had this problem myself, and I was able to solve it. Operating Load Limit 1/2 (and I believe Load Limit Force Spark Retard, but cannot confirm without additional testing) are to blame. I set the table to the following:
0.803 0.856 2.5 5 5 5 5 5 5 5 5 5 5 5.
This allows the ECU to make calculations at higher loads. The table name/description is deceiving. ATR makes it clear that these values are calculation limits, not actual load caps. Furthermore, they seem to operate at a percentage. A cap at 2.5 load will cause calculated load to peak at approximately 2.2 on logs. Hopefully this is helpful to others going forward.
mituc
VersaTuner guru
Posts: 1324
Joined: December 17th, 2011, 2:47 pm
Location: Iasi/Romania

Re: Ignition Timing on DI Engines

Post by mituc »

AFRs going lean past some load/air flow is usually an indication of incomplete burn. There can be several reasons for that:
- bad spark plugs;
- improper spark plugs;
- worn out spark plugs;
- improperly gapped spark plugs;
- weak/work coils;
- weak battery (past a certain throttle input the alternator is disconnected, so if the battery is weak it's bad);
- bad or clogged injectors with damaged spray pattern, or low flow, assuming the fuel rail flow/pressure is in spec.

The load limit force spark retard table dictates when the ECU should start using the Overload tables instead of the open loop tables and has nothing to do with the AFRs going lean past some load, unless of course the overload tables have really lean target AFRs and the limits in the force spark retard table are exceeded.
2008 Cosmic Blue Mazda 3MPS
Built engine + WMI + GTX3071 gen2, ~509BHP @34PSI
2008 Icy Blue Mazda CX7
Stock engine and exhaust (YES!!), JBR3" + GTX2867 gen2 + Autotech HPFP, self-tuned to 330-ish BHP
User avatar
ahayden04
Posts: 5
Joined: May 24th, 2022, 4:31 am
Location: South Carolina

Re: Ignition Timing on DI Engines

Post by ahayden04 »

mituc wrote: August 11th, 2022, 3:34 am The load limit force spark retard table dictates when the ECU should start using the Overload tables instead of the open loop tables and has nothing to do with the AFRs going lean past some load, unless of course the overload tables have really lean target AFRs and the limits in the force spark retard table are exceeded.
After further testing, I agree. However, the lean AFRs were 100% due to the Operating Load Limit tables being set too low. The load flatlines at a predetermined value, and the AFRs continue to lean out due to the incorrect fuel calculation. In fact, the program ATR has a completely different description for these tables.
Post Reply