| Author |
 Topic  |
|
|
enetec
Un(deleted)
Virgin Islands (United Kingdom)
4611 Posts |
 Posted - 17 May 2005 : 18:48:46
|
Di solito, fino a che numero di giri e/o a che pressione dell'acceleratore le ECU in generale (o la ECU MKIII in particolare) lavorano in closed loop e quando abbandonano questo tipo di funzionamento?  |
TESSERA SOCIO TITOLARE 2013 N°120
No sign. No comment. |
|
|
marco.sart
Pischello della Domenica
49 Posts |
Posted - 20 May 2005 : 17:22:31
|
Il controllo motore cerca di mantenere il controllo in closed loop in ogni circostanza in cui sia possibile. Ciò significa al minimo (a motore caldo, perchè in condizioni di avviamento a freddo entrano in gioco le strategie di light off di sonda lambda e cat) e in funzionamento stabilizzato (carico+giri costante).
In realtà, durante i transitori, la "lentezza" dei vari sensori non consente di ottenere una retroazione in tempo reale sul motore, per cui si fa riferimento, per calcolare la quantità di benzina da iniettare, alle tabelle nella ROM della centralina. Un'altra condizione particolare è quella di WOT, in cui si lavora in open loop utilizzando una miscela ricca, al fine di scongiurare la detonazione, ottenere il massimo delle prestazioni e diminuire la temperatura in camera.
Il discorso che ho fatto è molto semplificato, perchè in realtà intervengono molti altri parametri nella determinazione del controllo, uno fra tutti la temperatura del refrigerante.
Infine, due sono i fattori determinanti:
-la centralina deve giudicare la sonda lambda operativa (il segnale in tensione non deve rimanere troppo a lungo nella zona di transizione)
-non si è in cut off.
Ciao, Marco |
TESSERA SOCIO TITOLARE N°175
Pagina Personale
|
 |
|
|
enetec
Un(deleted)
Virgin Islands (United Kingdom)
4611 Posts |
Posted - 20 May 2005 : 18:14:53
|
Oh... FINALMENTE qualcuno interviene con una risposta tecnica a questo topic... siccome però qui le risposte "latitavano", nel frattempo avevo aperto analogo topic su Spyderchat e dopo qualche scambio di idee siamo (o meglio, sono...) arrivato a queste conclusioni, che mi piacerebbe tu analizzassi e commentassi...
Premessa: è importante capire come mai chiedevo ciò. Io ho la mia macchina "fortemente elaborata" - vedi sign - e, fra le altre cose, ho una centralina aggiuntiva che interviene nel modificare la carburazione e l'accensione con sue mappe.
Da qui il mio dubbio su come e quanto il sistema di closed loop potesse intervenire nel "modificare le modifiche" se capisci cosa intendo... 
Ti posto parte del post su Spyderchat direttamente in inglese (non vorrai mica che lo traduca, vero?  ), se hai qualche dubbio su qualche punto cmq sono a disposizione...
quote:
"LittleRocket"
I don't know enough about your system to adequately answer that question.
They were only two examples to let us argue a bit... let's go with "general" discussion about closed loop...
quote:
"LittleRocket"
Does the e-manage intercept the o2 sensor voltage and condition the signal before it goes back to the ecu to prevent correction? What size injectors are you running? Fuel pressure? o2 sensor? Bosot levels? And about 10 other questions I need answered. :lol:
Respectively: no way, 420 cc, stock, all installed as stock, is it important at low-medium level TP?
Only changes e-Manage can do on inputs signal to stock ECU are MAF values correction... BUT since O2 signals are still stock, ECU "should" be able to make its corrections.
BUT if it would be this way, even stock maps shouldn't be so... "useful" until ECU can work with closed loop only... :?
So, what I think is that closed loop corrections can only change AFR of a (very?) limited amount respect of what is fixed by maps...
This could explain my issues too: at low RPMS and low TPS usually stock ECU make some closed loop corrections on my (modified by e-Manage) maps BUT, when engine is hot, it probably can't compensate so much as needed and so I get this low (12.x) AFR... possible? :?:
quote:
"wts"
I have a different scenario in that mine goes rich when coming to a stop and RPM <1000. For instance, I'm cruising in closed loop and AFR is stoich , as I come to a stop, depress clutch, shift out of gear the AFR goes way rich for a second with zero throttle input. Engine doesnt stumble/stutter or anything and the AFR quickly goes back to 14-16 as I sit at an idle.
I think it is a quite normal behavior... even when you shift a gear you should see about the same (some moments of 12.x AFR).
If your temporary high AFR comes together with a bit higher idle, instead you probably have only to reduce a bit the fuel in one or two cell just over "idle cell" at 0% TP in your map... :wink:
quote:
"blackwidow"
Is this with a wideband and controller or the stock o2?
I'm not sure if you know this, but stock o2 sensors can be up to 2 points off when not at stoich. This might have something to do with your readings.
I would think the only problem with the e-manage is partial throttle. I believe it has been discussed before that the stock ecu will try to compensate for whatever the emanage does as long as it is in closed loop mode. Which would cancel out the partial throttle control.
Now then i believe there is a boost controlled device that tricks the ecu into open loop mode. How it does this i forget. Try to search for one i know i remember seeing it. If this thing works the way it should, you could possibly, almost, maybe get the emanage into standalone territory.
Well... to be honest I don't want to "disclose" a lot how my system is tuned since it hasn't been (well, until now... :wink:) my work BUT a work done by someone else (Jay Hass) and I don't think he could like a lot his work "explained for free" if you well understand what I mean...
Anyway, only for understanding purposes...
Stock ECU has (I guess...) two maps: injection and ignition.
They are based on some input values (I think RPM and MAF values) and modified by some other, like TP, (first two only...?) O2 sensors, water temp, knock sensor and so on... (am I forgetting something?)
e-Manage and Profec in my system make some changes over it using three maps: Ignition variations, Additional injection and MAF values modification. First two works by RPM and Profec MAP inputs, 3rd by RPM and TP!
So, this way we have: stock ECU still works by MAF and with original O2 sensors (BUT with ignition and injection values post-changed by e-Manage AND MAF values pre-changed by e-Manage too...)
e-Manage EVER works by RPM and Profec MAP sensor (never by MAF in my system!) plus TP.
So no "switch" between MAF and MAP but both working together!!
(with MAF values probably corrected by e-manage when they aren't "so sharp" anymore...)
Systems without a MAP have their e-manage maps based on MAF too...
This if I don't go wrong and without disclosing any "customized details".
(these are only standard Greddy e-manage and Profec ways to run...)
So, since what I've understand, if you make some "minor changes" to original maps by e-Manage you could have them compensated by stock ECU "closed loop" BUT if you make some "main changes", stock ECU probably can't compensate so much and you HAVE your changes EVEN if you are still in closed loop.
So, let's make me some examples:
Closed loop situation
Standard output value (of fuel or what you want): 5
Wanted output value (as above): 8
e-Manage map changing value - output obtained value
-1 - 5
0 (as stock) - 5
1 - 5
2 - 5
3 - 6
4 - 7
5 - 8
6 - 9
Open loop situation
Standard output value (of fuel or what you want): 5
Wanted output value (as above): 8
e-Manage map changing value - output obtained value
-1 - 4
0 (as stock) - 5
1 - 6
2 - 7
3 - 8
4 - 9
5 - 10
6 - 11
What do you think about my hypothesis?
DISCLAIMER: I'm sorry for my English, but when discussions begin to become more complex probability that I insert some mistakes arise too!!
|
TESSERA SOCIO TITOLARE 2013 N°120
No sign. No comment. |
|
|
|
marco.sart
Pischello della Domenica
49 Posts |
Posted - 20 May 2005 : 19:02:32
|
Capito il problema..
Bisogna considerare che la maggior parte dei controlli è di tipo proporzionale-integrale. L'uso della lambda, in grado di trasmettere un segnale ON-OFF, comporta l'innesco di un comportamento a ciclo limite, ovvero una continua oscillazione tra valori ricchi e valori poveri nel titolo.
La formula utilizzata dalla centralina per stabilire, in closed loop, la quantità di benzina da iniettare, è:
Benzina_corretta= Benzina_tabella_ROM * (0.75+KO2)
dove KO2 è una grandezza derivata direttamente dal segnale della lambda. Come vedi, quindi, la "libertà" di controllo del closed loop non è assolutamente trascurabile, arrivando fino alla correzione del 25% del valore imposto dalle tabelle!!
Mi potresti dire esattamente quali valori leggi (ricco-magro) e quando (condizioni di carico, riscaldamento)? |
TESSERA SOCIO TITOLARE N°175
Pagina Personale
|
|
|
| |
Topic |
|
|
|
Closed loop
