Acceleration enrichment

Official FreeEMS vanilla firmware development, the heart and soul of the system!
Post Reply
tpsretard
QFP80 - Contributor
Posts: 99
Joined: Thu Mar 19, 2009 3:05 am

Acceleration enrichment

Post by tpsretard »

Based on my hell with the megasquirts AE/EAE implementations and out of the thread here, viewtopic.php?f=6&t=929&start=12

I decided to open a little discussion on it. I think every ecu i have used has its own setup on how to do it. Some work better than others. But quite simply the megasquirt's is the hardest to setup. It is not intuitive at all. In fact it is down right strange.

What is it
The accel (acceleration) pump is used to add a bit of extra fuel the second the throttle is depressed, the harder, or rather faster you press it, the more fuel u will need to add, this is to stop the engine from missing from a sudden lean condition and the air rushes into the engine. Some will need more some will need less.

Tables you need
  • Delta TPS = The initial value of fuel from the rate of tps or map change.
  • Accel vs RPM = A scaler for the accel pump size based on rpm, you always need more at low engine speeds
  • Coolant temp = A correction to the accel amount based on coolant temp
Settings you need
  • Trigger = The delta rate that the accel pump engages. this will stop miss triggers
  • Max size = The max amount of change that the accel pump will change the PW by.
  • Decay time = How fast the pump will decay. this should be done as a percentage decrease based on the last value.
There is another setting that i can think of, but i do not know if it should be a setting or a table also something that is a big problem on some engines is the initial split second injection time or sensitivity if you will. the higher the value the higher the initial pump size. This is why EAE needs the normal AE to sharpen the edge of the accel pump on the megasquirt.
MotoFab
1N4001 - Signed up
Posts: 307
Joined: Thu May 29, 2008 1:23 am
Location: Long Beach CA

Re: Acceleration enrichment

Post by MotoFab »

Those AE/EAE schemes do not maintain the desired AFR during transient throttle. No accurate AE adjustment is possible because there are more variables. In addition, several of the sensed variables must be continuously high speed sampled, and streamed into a set of accumulator registers. One set of accumulators per cylinder. A single sample will not work.

Then, for each fuel calc, when each injector PW is about to start, the transient fuel algorithm is performed on the set of accumulators for that cylinder.

The process isn't really very complicated, once you understand what's going on. It can help to think about running the algorithm on a single cylinder motor. Then as more cylinders are added, interleave/overlap an additional process for each cylinder.

Here is a excellent primer. Of the many I've read, this one is the easiest to understand.
Attachments
US Patent 4,388,906 - Toyota - Fuel injected engine control device and method performing wall-adhered fuel accounting - Sugiyama, et al. - June 21, 1983.pdf
(4.12 MiB) Downloaded 1032 times
tpsretard
QFP80 - Contributor
Posts: 99
Joined: Thu Mar 19, 2009 3:05 am

Re: Acceleration enrichment

Post by tpsretard »

good reading.

That was only the tables for accel enrichment, but you will need for deccel enleanment also.

was working on a pectel SQ6 this morning. they have 14 options/tables for transient corrections alone..

Wonderful ecu to work with, the settings do what they suggest. This is what we need to achieve, even if there is an adaptive algorithm that diyefi uses, it should be simple to setup, where a positive increase is just that. not something that is based on the shape of a curve.
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Acceleration enrichment

Post by Fred »

tpsretard wrote:was working on a pectel SQ6 this morning. they have 14 options/tables for transient corrections alone..

Wonderful ecu to work with, the settings do what they suggest. This is what we need to achieve, even if there is an adaptive algorithm that diyefi uses, it should be simple to setup, where a positive increase is just that. not something that is based on the shape of a curve.
Screenshot(s)?
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
MotoFab
1N4001 - Signed up
Posts: 307
Joined: Thu May 29, 2008 1:23 am
Location: Long Beach CA

Re: Acceleration enrichment

Post by MotoFab »

tpsretard wrote:good reading.

That was only the tables for accel enrichment, but you will need for deccel enleanment also.
It's hard to get what's going on in the first couple of reads. It's a normalized factor.

Depending on the input data streamed into the accumulator registers during the interval between injections, the formula produces either a greater than 1 or less than 1 factor to apply to the result of the current injector pulsewidth calculation.

The process of transient enrichment depends on continuous sampling during the interval between injections for each cylinder. It will not work otherwise. As a minimum example, the 'throttle transient' adjustment to the current pulsewidth is based on the manifold conditions during the last 720° crankshaft degrees, for that cylinder.

Any throttle transient method based on the rate-of-change of the inputs isn't realistic. The ROC factor is there to try and compensate for the lack of streamed sampling, data storage, math, and timing. The reason it kind of works is due to the large range of combustible fuel/air ratios. That, and most of time the throttle isn't moving, so the regular fuel calculation works fine.

Dumping more fuel in when the throttle is advanced is what a carburetor does. Current aftermarket/diy designs can sort of be thought of as a 'smart' accelerator pump. A processor can do accurate, genuine, transient throttle, but not all processors are capable.
tpsretard
QFP80 - Contributor
Posts: 99
Joined: Thu Mar 19, 2009 3:05 am

Re: Acceleration enrichment

Post by tpsretard »

fred just finished a stage rally last night, when i unpack everything i will take some screen shots of the pectel and gems if you want, i think i can even toss in from motec too if you would like.
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Acceleration enrichment

Post by Fred »

The more input the merrier. There is a thread for general screen shots around, but AE ones can live here :-)
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
Post Reply