RomRaider Logo

RomRaider

Open Source ECU Tools
 FAQ •  Register •  Login 

RomRaider

Documentation

Community

Developers

It is currently Fri Dec 15, 2017 12:10 am

All times are UTC - 5 hours [ DST ]





Post new topic Reply to topic  [ 7 posts ] 
Author Message
 Post subject: Injector flow compensation per battery volts?
PostPosted: Wed Nov 15, 2017 4:30 pm 
Offline
Experienced

Joined: Fri Sep 02, 2016 4:42 am
Posts: 250
Location: Riga, Latvia
Been working on injector latency and there's an interesting trend I have noticed
Attachment:
battv_ipw_afr_error.png
battv_ipw_afr_error.png [ 79.85 KiB | Viewed 136 times ]

It looks like the afr error is on lean side above 14V and on rich side below. What could be the possible explanation?

Also my latency is 0 at 16.5V and I still can't rid of the richer->leaner slope above 14V. Decapped light blues.

The colder the engine the higher the volts.
Under normal operating temperature it sits around 13.5-13.9V.


Top
 Profile  
 
 Post subject: Re: Injector flow compensation per battery volts?
PostPosted: Wed Nov 15, 2017 5:58 pm 
Offline
Senior Member

Joined: Thu Aug 03, 2006 10:40 am
Posts: 1901
trez0r wrote:
Been working on injector latency and there's an interesting trend I have noticed
Attachment:
battv_ipw_afr_error.png

It looks like the afr error is on lean side above 14V and on rich side below. What could be the possible explanation?

Also my latency is 0 at 16.5V and I still can't rid of the richer->leaner slope above 14V. Decapped light blues.

The colder the engine the higher the volts.
Under normal operating temperature it sits around 13.5-13.9V.


too many variables to extract meaningful trends from this kind of test.


Top
 Profile  
 
 Post subject: Re: Injector flow compensation per battery volts?
PostPosted: Wed Nov 15, 2017 6:23 pm 
Offline
Experienced

Joined: Fri Sep 02, 2016 4:42 am
Posts: 250
Location: Riga, Latvia
Care to list them so I can narrow it down?


Top
 Profile  
 
 Post subject: Re: Injector flow compensation per battery volts?
PostPosted: Thu Nov 16, 2017 10:01 am 
Offline
Senior Member

Joined: Thu Aug 03, 2006 10:40 am
Posts: 1901
rpm iat ect ipw as a start.

if you're trying to tune a compensation, you need to make every effort to ensure that it is the only variable in the test.

it can be tricky in the case of supply voltage, as you have both the battery and the alternator to contend with. i would only collect data at a single load cell with a stable (ie fully warmed up) engine/underhood environment. to vary supply voltage i would disconnect battery and alternator and use a benchtop regulated power supply (or something more ad hoc like a battery charger and power rheostat).

i think we can both agree that no injector has zero latency so the fact that afr error isn't corrected by entering a nonsensical value should be a red flag that you're either a) basing adjustments off irrelevant data or b) adjusting the wrong compensation.


Top
 Profile  
 
 Post subject: Re: Injector flow compensation per battery volts?
PostPosted: Thu Nov 16, 2017 2:56 pm 
Offline
Experienced

Joined: Fri Sep 02, 2016 4:42 am
Posts: 250
Location: Riga, Latvia
Yes, and I think we can also agree that it will never see those 16.5V under normal operating conditions. 1.14@14V and 0@16.5V interpolates to 0.94@14.5V which is the highest volts I've seen by logging.
I'm thinking if it might happen that there are several combinations of latencies + flow scale that result in more or less balanced AFR error across ipws. Say maybe somewhere between the latencies I run now and stock values there is another set that works. I never tried going lower than 1ms@14V on decapped, maybe I should. I incrementally adjusted them over the course of several weeks until I ended up at 1.5ms@11.5V, 1.14ms@14V and 0@16.5V - last one could easily be much higher as it does not have a profound impact at volts just slightly above 14V. This all works out to picture in OP which is actually fine, but could be better per mentioned observations. I should add that the table is generated from filtered data where CL/OL=8, throttle pos > 0 and cell hit count is > 10 and AFR_ERROR here being LTFT+STFT. One could argue, that this should be done on steady throttle too, but I think given enough data, it should normalize out.

Not quite sure I understand why ECT / IAT or other comps would impact anything as long as the target against which the error is measured includes those comps.
I think you said it yourself - there are several different ways how to achieve the same ipw - different RPM/MAP etc, but the point is to achieve nearly the same AFR error across all IPWs.


Top
 Profile  
 
 Post subject: Re: Injector flow compensation per battery volts?
PostPosted: Thu Nov 16, 2017 4:12 pm 
Offline
Senior Member

Joined: Thu Aug 03, 2006 10:40 am
Posts: 1901
any discussions of fuel error should not involve using fuel trims imo. they are by necessity prone to data scatter as they are part of a feedback system with inherent time delay. a long time ago i decided that all fuel error calculations should be done open loop, for some of the reasons that are discussed here:

viewtopic.php?f=36&t=10235

Quote:
I'm thinking if it might happen that there are several combinations of latencies + flow scale that result in more or less balanced AFR error across ipws. Say maybe somewhere between the latencies I run now and stock values there is another set that works


yes, they may "work" but i think we can agree that only one is an accurate representation of what's actually happening at the injector.

a "workable but inaccurate" scale/latency may be fine under "normal" conditions but what about that drive where you toss an alternator belt and batt voltage slowly but surely sags down to 10v? i bet the car won't make it home! whereas a "workable and accurate" scale/latency will keep the car rolling down the road and into your driveway.

most of the time consuming stuff in tuning isn't typically common. warm up, for example, is a right pain in the ass because you have to catch it while it happens, then you have to wait a long time for the block to be cold again. same thing with extreme ambient temps... here in new england it can be >100f in summer, and <0f in winter. or how about going WOT until 6k, then lifting completely, before mashing again, all in the span of a few log lines? suddenly you're knocking when and where you never were before. you don't see hidden compromises in your maps until you start hitting the uncommon conditions.

i mentioned IAT and ECT those since they can be directly logged, but you can be sure that if ECT and IAT are changing then fuel temperature is too, which is NOT compensated for.

bottom line: to measure the effect of varying battery voltage you must only vary battery voltage--not load, not rpm, not temperature, not throttle, etc.

fwiw i think your latencies are too high, at least compared to my last known good rom with stock dark blues and +20psi fuel pressure (which should make them open even slower).


Top
 Profile  
 
 Post subject: Re: Injector flow compensation per battery volts?
PostPosted: Thu Nov 16, 2017 5:38 pm 
Offline
Experienced

Joined: Fri Sep 02, 2016 4:42 am
Posts: 250
Location: Riga, Latvia
Thank you ride5000 for discussion.
When setting the latencies, I had removed the alternator wire to properly test the low volts area and it went really well (at the end of the test it was below 11.5V). I'm no electrician but I don't think I will ever encounter a failure scenario when there is significant overvoltage though.
I've scoured the internet and the general consensus is that decapped light blues feel "at home" somewhere around 1ms and 1.20ms @ 14V. I might some day try going significantly lower than that just to see if it yields anything workable.
Also I take from you that I probably should ignore the error on cold engine (where the volts typically exceed 14.2V).

I read your thread about forcing the open loop to scale the maf (same applies for latency testing) a while ago and I've been contemplating that. I'll probably do so soon too. I'm about to replace the clutch in next few days and while at it I'll helicoil the stripped stud I have on my turbo to eliminate the slight possibility of my wideband being a bit off on low exhaust flow.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 7 posts ] 

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: Bing [Bot], Google [Bot] and 6 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