RomRaider

Open Source ECU Tools
 FAQ •  Register •  Login 

RomRaider

Documentation

Community

Developers

It is currently Fri Oct 24, 2014 4:24 am

All times are UTC - 5 hours [ DST ]





Post new topic Reply to topic  [ 25 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Injector Duty Cycle
PostPosted: Thu Jul 12, 2007 2:33 am 
Offline
Moderator

Joined: Wed Nov 22, 2006 10:23 pm
Posts: 2306
I've read stuff from people who are running over 100% IDC, and that makes me wonder... what exactly is 100% IDC? If 110% IDC is possible, then IDC is not really "duty cycle" in the usual sense. How can the injector be "on" 110% of the time?

I notice that the logger definition for IDC is derived from RPM and injector pulse width (IPW), like so:

IDC=(RPM*IPW)/1200

So, as your RPM increases, IDC goes up, even if IPW remains the same. Or, the IPW for a given duty cycle gets smaller:

At 1200 RPM, 100% IDC is 100ms IPW (that will never happen).
At 3600 RPM, 100% IDC is 33ms IPW (still seems unlikely).
At 7200 RPM, 100% IDC is 16.7ms IPW (that's conceivable).

Then the question is, why 1200 for the constant in that formula? At 7200 RPM, you get 120 revolutions per second, or 8.3ms per revolution, or 16.7ms per two-revolution cycle, which also happens to be the maximum IPW. Coincidence? I think not.

Then the question is, what's really happening with IDCs over 100%?

I still have no idea. I guess the injectors just never close after 100%, but that would mean these peoples' AFRs just get leaner as RPM increase beyond that point. But I don't see how the ECU could ever report > 100% IDC in the first place, so I must be overlooking something. Also, it bothers me that at moderate IDCs (say, over 50%) the injector is spraying on the back of a closed intake valve most of the time. Is that really the case?

What am I overlooking?


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jul 13, 2007 12:42 am 
Offline
Experienced
User avatar

Joined: Mon Sep 18, 2006 10:55 am
Posts: 229
bump


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jul 13, 2007 3:36 am 
Offline
Experienced

Joined: Mon Oct 23, 2006 2:57 am
Posts: 174
Location: Carol Stream, Illinois
basically you have the right idea, once IPW hits 100% the injectors just don't shut off. I think the over 100% is just the ECU's way of displaying that it no longer can add anymore fuel in order to reach the targets AFR's.

Personally I never want to see my car over 85% duty cycle, if it gets that high it's time for larger injectors.

_________________
Jeremy
2001 Impreza RS w/ 2003 USDM WRX Swap
VF-23, TXS Turboback, Hyperflow Top Mount, Modded Stock Injectors, Warlbro Fuel Pump


Top
 Profile  
 
 Post subject: Re: Injector Duty Cycle
PostPosted: Fri Jul 13, 2007 7:14 am 
Offline
RomRaider Donator
User avatar

Joined: Sun Apr 09, 2006 12:05 pm
Posts: 866
Location: Indianapolis, IN
NSFW wrote:
What am I overlooking?


Nothing. What you wrote is correct. Total available time for injection is RPM/1200. You cannot physically pass this upper limit. If IPW equals this value it means the injector is just full stuck open, spraying 360 degrees around the crank. There is nothing contradictory here.

The constant 1200 is just the result of using milliseconds as your unit of time, just like any other constant in physics equations (like gas constant, Planck's constant, etc.) You practically wrote the whole proof in your post.

I imagine the ECU might calculate injector pulse width and send that to the logger before checking to see if it will be over the available time for injection, and that's why you see numbers over 100%. It seems IDC is a purely calculated-by-the-logger value (checked Enguinity logger XML), not an SSM item itself, so that'd have to be the explanation if you see IDC > 100% in your logs.

I think it is actually optimal that the injectors fire upon a closed valve, just before the valve opens, but not when it is actually open. The intake valve can really only be open about ~1 out of the 4 strokes of the engine cycle, give or take, so keep that in mind when thinking about IDC and if the fuel injector must be spraying on an open or closed valve. I'm not 100% sure how the ECU times injection at medium duty, but at low, it fires on a closed valve just before the valve opens, and if it is high enough you necessarily must be firing on both open and closed times. You can find your intake valve's "duty cycle" by dividing the cam duration by 720 degrees (4 strokes).


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jul 13, 2007 11:10 am 
Offline
Experienced

Joined: Wed Jul 26, 2006 3:19 pm
Posts: 650
Location: Connecticut, USA
The attached document provides an overview of what the ECU is trying to do.


Attachments:
ECM.pdf [140.74 KiB]
Downloaded 230 times
Top
 Profile  
 
 Post subject:
PostPosted: Fri Jul 13, 2007 11:40 am 
Offline
Administrator
User avatar

Joined: Wed Mar 29, 2006 10:38 pm
Posts: 5340
IPW as logged (SSM) includes latency. IDC is not an SSM parameter and is calculated from the IPW. Maybe this explains why the IDC are over 100%. Should latency be included when calculating IDCs?


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jul 13, 2007 11:54 am 
Offline
RomRaider Donator
User avatar

Joined: Sun Apr 09, 2006 12:05 pm
Posts: 866
Location: Indianapolis, IN
I'd assume so, because you want to know when you're about to max your injectors.

Going from 97% duty cycle to 100% could cause AFR to jump because of the dead time being eliminated if the injectors are just being held open. Part of the reason to not run that high! :) When dead time is close to the total low signal time of the injector cycle I assume it is going to start acting unstable. The ECU may be including dead time, but if the injector doesn't even have time to fully close after release, it might take less dead time at the start of the next injection period.

I'm not 100% positive the car will actually hold the injectors open solid. It's possible it will briefly close them at the end of every intake cycle. Does anyone know for sure? The ECU does some really weird s*** with the injector pulse time before it somehow gets sent off to the one-shot controller. Since it is a one-shot controller, it may require a brief downtime. Maybe not. Again, I'm not completely sure. I'd guess it will hold them solid if total (with dead time) injector pulse is greater than available pulse time.


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jul 13, 2007 1:39 pm 
Offline
Newbie

Joined: Sun Mar 05, 2006 7:06 pm
Posts: 6
In the past, there was talk on NASIOC about latency issues with datalogging injector duty cycle. If the RPM data is retrieved first, followed later by the injector pulsewidth, it's possible that the displayed IDC is going to be ever-so-slightly artificially high. Remember, we are talking about milliseconds here, and most of us are getting data snapshots every ~250ms... which is relatively slow.

In short, if you can increase IDC, even beyond a datalogged 100%, and the car runs richer, then you *ARE* injecting more fuel. Still, I wouldn't recommend regularly running higher than 85% without keeping a VERY close eye on things. ;)


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jul 13, 2007 1:51 pm 
Offline
Moderator

Joined: Wed Nov 22, 2006 10:23 pm
Posts: 2306
Quote:
I imagine the ECU might calculate injector pulse width and send that to the logger before checking to see if it will be over the available time for injection, and that's why you see numbers over 100%.


That would make sense.... so semantically it's really how much IDC the ECU wants, rather than how much IDC it actually uses. In that light it seems even more likely that people with IDCs over 100% are leaning out.


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jul 13, 2007 5:55 pm 
Offline
Administrator
User avatar

Joined: Wed Oct 25, 2006 12:32 am
Posts: 3040
To summarize what Freon said in post #4.

At 100% IDC, each combustion cycle takes 2 revolutions, so

IPW_max = 2 rev x (1/RPM) x (60000 ms/min)
IPW_max = 120,000 / RPM

The injector pulse width, IPW, is the time the injector is open in units of ms and includes latency. IDC is in units of % so:

IDC = (IPW/IPW_max) x 100

IDC = IPW x RPM / 1200


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jul 13, 2007 6:17 pm 
Offline
Newbie

Joined: Thu May 03, 2007 9:22 pm
Posts: 41
Location: 07 STi Limited #482
x99percent wrote:
Still, I wouldn't recommend regularly running higher than 85% without keeping a VERY close eye on things. ;)


That's funny considering the stock STi can hit 95% and beyond. Not aiming that at you, but Subaru. I agree.


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jul 13, 2007 8:40 pm 
Offline
Experienced

Joined: Wed Jul 26, 2006 3:19 pm
Posts: 650
Location: Connecticut, USA
Freon wrote:
Going from 97% duty cycle to 100% could cause AFR to jump because of the dead time being eliminated if the injectors are just being held open. Part of the reason to not run that high! :) When dead time is close to the total low signal time of the injector cycle I assume it is going to start acting unstable. The ECU may be including dead time, but if the injector doesn't even have time to fully close after release, it might take less dead time at the start of the next injection period.
Actually, going from 97% duty cycle to 100% will cause the AFR to fall (i.e. become richer than desired) because the interval between injections becomes less than dead time and the injector doesn't fully close.

Here's an interesting thread on this topic, where EcuTeK and Prodrive were attempting to tout the new ability to flash the US STi. Unfortunately, the links to dyno plots and a log are now dead.
http://forums.nasioc.com/forums/showthread.php?t=553419


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jul 13, 2007 9:37 pm 
Offline
Experienced
User avatar

Joined: Mon Sep 18, 2006 10:55 am
Posts: 229
wow, what a great thread !!

Thanks Jon!


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jul 13, 2007 10:20 pm 
Offline
RomRaider Donator
User avatar

Joined: Sun Apr 09, 2006 12:05 pm
Posts: 866
Location: Indianapolis, IN
Jon [in CT] wrote:
Actually, going from 97% duty cycle to 100% will cause the AFR to fall (i.e. become richer than desired) because the interval between injections becomes less than dead time and the injector doesn't fully close.

Here's an interesting thread on this topic, where EcuTeK and Prodrive were attempting to tout the new ability to flash the US STi. Unfortunately, the links to dyno plots and a log are now dead.
http://forums.nasioc.com/forums/showthread.php?t=553419


Sorry, I didn't mean to imply with "jump" that it would go in the lean direction. I agree it would go in the rich direction. It would drift rich right (possibly suddenly, or by a pretty decent margin of AFR) when you approach and reach 100% if the injector no longer has a physical dead time. I guess that thread confirms the injectors will actually stay open and not be force to go through an off cycle every intake stroke.

I guess you could take (RPM/1200) - Deadtime as a practical limit on injector pulse width for a given RPM before you start to get unpredictable or inconsistent results. Especially since you won't get exactly the same airflow, load, and pulse width on the stock ECU every full throttle run.


Top
 Profile  
 
 Post subject:
PostPosted: Sun Jul 15, 2007 5:59 am 
Offline
Experienced

Joined: Fri Apr 06, 2007 6:09 am
Posts: 116
Location: France
Well, I do not 100% agree. In the past, before flashing tools exists, I developped an electronic device fireing the injectors in place of the ECU (because of the CEL and safe mode generated by a MAF voltage read above 4,7 Volts that I can't change without a remap). I saw that with an IDC of up to 130%, increasing the IDC consistently increase the AFR (richer).
Since we have the flashing tools, I've tuned a lot of WRX and STi's with IDC above 100% (110 to 120%), and I can guaranty that going from 100% to 120% consistenty richen the AFR. The limit seems to be somewhere between 125 to 130% :wink:
Don't ask me why :lol: I do agree that 100% would indicates constantly wide open injectors, but it's not until 130% is reach.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 25 posts ]  Go to page 1, 2  Next

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 4 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
Style based on FI Subsilver by phpBBservice.nl