The Tuning Talk Thread - Forced Induction EFI & JZ Tuning.

[OnAll-FOUR]

Just found this astract on IEEE explore

Self-learning FNN (SLFNN) with optimal on-line tuning for water injection control in a turbo charged automobile .
1461308 abstract .

.
Chi-Hsu Wang Jung-Sheng Wen
Dept. of Electr. & Control Eng., Nat. Chiao Tung Univ., Hsinchu, Taiwan

This paper appears in: Networking, Sensing and Control, 2005. Proceedings. 2005 IEEE
Issue Date: 19-22 March 2005
On page(s): 878 - 882
Print ISBN: 0-7803-8812-7
INSPEC Accession Number: 8504865
Digital Object Identifier: 10.1109/ICNSC.2005.1461308
Date of Current Version: 05 July 2005

Abstract
This paper proposes a new architecture of self-learning fuzzy-neural-network (SLFNN) for water injection control in a turbo-charged automobile. The major advantage of SLFNN is that no off-line training is needed for initialization. The SLFNN will initialize itself with a random set of initial weighting factors (normally zeros) and a specifically designed on-line optimal training algorithm is invoked immediately after the engine of the automobile is turn on. The on-line optimal training can guarantee that the weighting factors will be directed toward a maximum-error-reduced direction. Although this SLFNN can also be used as a controller for fuel injection, we adopt the SLFNN as the water injection controller to reduce the knocking effects of a turbo-charged engine and therefore the emission is cleaner with less petrol consumption. Real implementation has been performed in a Saab NG 900 (1994 -1998) automobile with excellent results.

[punkture]

hello, to start off i have nil efi tuning experience.. i am contemplating a MS3 for my gt4 (st185) im good with electronics and most mechanical things, but was wondering if anyone has ever used one before? and if local dyno tuners could tune it etc? thanks

[mr1jz]

Are you talking megasquirt?

[punkture]

Are you talking megasquirt?

yeah mate =)

[mr1jz]

What about this.

http://www.vi-pec.com/page_files/V44.html

Far superior ecu. I have dealt with both and the Vipec reigned supreme.

[punkture]

yeah i was contemplating a plug in one for the 3sgte, but was wondering what kind of $$ theyre worth

[mr1jz]

They are going to be more then the megasquirt but at the end of the day, you get what you pay for.
Just jump on the Vi-pec wesite and check for your local dealer. From memory there was only one is SA.

[JP]

the v44 plug-in will set you back around the $1600-1700 mark + tuning. I'd also budget for a wideband sensor and controller.

[thechuckster]

msefi ECU - $300-600 depending on how much you want to build
wideband - $150-300 depending on quality/features

When you search the MSextra forums for '3SGTE' you get a lot of hits suggesting that a lot of 3S folks are participating but haven't got around to putting up 'it runs' threads. However, most tend to put up details of their engines and settings (e.g. MSQ) files) which will give you an idea of timing and fuel maps.

a few build threads i quickly found:
http://www.msruns.com/viewtopic.php?f=89&t=19709
http://www.alltrac.net/phpBB2/viewtopic.php?t=31249

The http://www.msextra.com/ forums are the place to join up and start researching.

[OnAll-FOUR]

Does anybody use their intake air temps to remove a certain amount of timing once it exceeds a set threshold for excessive temps?

I have heard of setups that do use this and it is part of the default wolf v500 configuration, I was wondering if anybody here has experiences with it. Maybe also some suggested settings?

For instance if you know that while tuning the ignition and maps you know that temps never exceeded a certain value, would it make sense to activate this feature around this level because you know this is outside the range the maps have been tuned for?

Or is it just adding complication and confusion?

[thechuckster]

you'd probably set that up after you've done some dyno time to get peak power from your timing maps (in conjunction with some IAT logging).

I think this thread has some previous discussion on IAT compensation?

[OnAll-FOUR]

yeah a little but I think that was more in terms of a general trim for the different densities at different temperatures for a given pressure?

[1JZ-Rolla]

There are two aspects to this, I assume you are approaching this in relation to the most critical operating aspect that you'd wish to avoid, being detonation?

Firstly, in its simplest form, with increasing IAT you reduce tolerance to detonation, with the corresponding increase in combustion charge temperature right? However the inherant effect on fuel mixtures (if not corrected in IAT vs injector duty maps) as temperature rises is to become more rich, since the density of the incoming air decreases. The effect of this richer mixture is both that it will cool the intake charge somewhat, and the cylinder walls & head, masking the expected detonation to a degree, with no change to ignition advance.

Secondly and I assume this is your main point Ben, In a setup where intelligent knock sensing devices (and associated programmed ignition retard) is NOT available, determining the appropriate point at which the IAT is considered too hot for the given ignition advance is probably best commented on by a proficient and experienced tuner, so I won't throw any rudimentary figures from my tune up here. I would however consider it appropriate to use a progressive ignition retard vs IAT compensation past this given point in the interest of preserving your engine (dare I say "safe" Todd ). Generally IAT is not going to come back down on its own if the same operating conditions are maintained (such as extended WOT in circuit laps), so this is probably the point at which you should consider having a break in any case

[thechuckster]

the more i think about this (from the 'reduce tolerance to detonation' position), i'd suggest another strategy:
use a high IAT trigger to change to a different set of maps (fuel & spark) that are aimed at avoiding engine damage. Effectively a temporary limp mode. This aligns with 1JZ-Rolla's suggestions about having a break when thigns get too hot

Otherwise, minor fluctuations in IAT are going to be dealt with by your fuelling equation (most Speed Density fuelling incorporates some kind of IAT compensation as well as being driven by RPM, MAT, VE & CLT). Your timing maps are going to incorporate timing retard in the boost cells anyway and a heat-soaked motor is probably going to show CLT increases which will in turn affect the fuelling equation.


fwiw:
the MS2Extra has this adjustment (MAT Based Timing Retard):

The inlet air temperature has a massive effect on how much advance you can safely run (especially on a boosted engine), if the temperature increases too much, due to sitting in traffic, long/hard runs in boost, etc, the amount of ignition advance can be retarded proportionately with the temp increase to help prevent detonation. The MS2-Extra code has a MAT Based Timing Retard table. This works in much the same way as the Cold Ignition Advance but it RETARDS the ignition from the base map to keep away from detonation. So in the example below, if the engine is running at 20deg on the base map, then at 93C the engine would be retarded to 18deg BTDC

(it also has a CLT-driven Cold Ignition Advance compensation tool for cranking the timing up when the motor is cold).

(from http://www.msextra.com/doc/ms2extra/...on.htm#moreign)

[OnAll-FOUR]

Thank you guys, That is exactly what I was trying to grapple with .