EGO correction and it's importance/downsides/gotchas

Official FreeEMS vanilla firmware development, the heart and soul of the system!
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

EGO correction and it's importance/downsides/gotchas

Post by Fred »

austinbob wrote:The change in volumetric efficency is handled by the afr sensor. The afr sensor is used to correct the volumetric efficencies.
Ahhh, AFR corrections will be strictly optional. They are no use when you rip through first gear to 25psi anyway. This comes back to the tune your engine and tune it right, sure humidity will affect things, but insignificantly for 90% of users as the air in a given region usually stays within a given range of damp/dry. Wideband sensors can't *just* be trusted either, you need complex safety guards if you want to trust them. Yes, I know you can limit the effectiveness of the correction etc, but it's not a final answer to everything.

"EGO" correction is primarily useful for cruise and idle light load regions. Many OEM systems that have AFMs and oxygen sensors on them totally ignore those past a certain load point anyway.

Fred.
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!
austinbob
DIP8 - Involved
Posts: 29
Joined: Fri Jun 06, 2008 3:46 pm

Re: Fuel calcs

Post by austinbob »

AFR corrections are what gives the ems the auto-tune ability. With afr corrections the engine tune will get better and better with time. The afr corrections is what causes the fuel map to evolve with time and adapt to changes in engine components such as new exhaust or new heads or new camshafts. The afr corrections are very important.
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Fuel calcs

Post by Fred »

austinbob wrote:The afr corrections are very important.
On OEM computers it is because it has to go a million miles without a retune.

On a custom project where you *want* to fiddle anyway, not necessarily, esp without the safe guards that OEMs do use to ensure that sensor failure is reported and the sensors are ignored when they are bad. We will have AFR correction, but it is not high on the list of things to implement. It should be thought through now though regardless such that the calculations are still good when we add it later.

Fred.
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!
austinbob
DIP8 - Involved
Posts: 29
Joined: Fri Jun 06, 2008 3:46 pm

Re: Fuel calcs

Post by austinbob »

On motorcycles it is important fred because we are always changing our exhausts,heads, cams etc.

As I said before, the o2 sensors are used to update the volumetric efficenies. If you want accurate volumetric efficencies you need the o2 sensors.
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Fuel calcs

Post by Fred »

austinbob wrote:On motorcycles it is important fred because we are always changing our exhausts,heads, cams etc.

As I said before, the o2 sensors are used to update the volumetric efficenies. If you want accurate volumetric efficencies you need the o2 sensors.
I personally think it is extremely unreasonable to change major components of an engine and expect it to work well when using MAP to estimate airflow.

The processor has a limited number of erase/write cycles on the flash memory and you are just asking for trouble to be constantly writing to flash while driving/riding. You are also asking for trouble by allowing fundamental things to be changed permanently without your knowledge IMO. You must have a fair bit of spare cash to be changing them so often that you don't have time to retune the engine to make best advantage of them. Seems a shame to let a computer and some code *try* to optimise a tune based on what is effectively an inaccurate unreliable approximation to AFR esp when the readings are not usable without extensive processing during dynamic situations and with any decent amount of power all WOT regions are highly dynamic. Are you being lazy? or is an EMS a means to an end for you? or do you have some solid reason to not want to tune your setup when you make major changes? What's the story?

Those experienced tuners that I know steer clear of AFR corrections for very good reasons.

Fred.
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!
austinbob
DIP8 - Involved
Posts: 29
Joined: Fri Jun 06, 2008 3:46 pm

Re: Fuel calcs

Post by austinbob »

On a motorcycle changing exhaust, heads and cams is done quite frequently.

Also you are not using just map to estimate air flow, you are using the o2 sensors to MEASURE the air flow. From the 02 sensor measurement you can correct the volumetric efficencies.

Please forgive me for being so blunt here, but you are absolutely wrong in your opinion of the value of 02 sensors. They are not inaccurate and they are not unreliable. As I said before you need to think of engine control as a statistical process. I have two wideband 02 sensors on my bike right now and they work extremely well. My bike has the best tune that it has ever had. And I can change cam shafts or heads or exhaust when I want.

This is the ecm that I have on my bike. I previously had a speed-density style ecm on my bike.

lhttp://www.daytona-twintec.com/TCFI.html

The alpha-n system described above is superior to the speed-density. The only reason I mention this is to hope to convince you that you have a wrong idea about the value of 02 sensors.

Regarding constanly writing to flash memory I have not considered that. What I do know is the fuel table is not constantly written. It is only written when the user downloads the eeprom to the pc and and applies the corrections. Then the maps are uploaded to the ems. I am not sure how the data is stored in the microcontroller when riding. It seems to me that data logging requires writing to memory but if you are correct , it must not be written to eeprom.

So how would data logging work?
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Fuel calcs

Post by Fred »

There is no need to convince me that they are valuable :-)

I wouldn't attempt to tune an engine without one myself.

What you will have a very hard time convincing me of is that I should allow my computer to adjust the tune that I have set in it based on sensor readings when wires could fail, sensor could be damaged, sensor controller could fail, etc etc etc etc on a long term basis when these things really happen. This I simply do not buy.

I want to explicitly set a tune for a given hardware setup and then enjoy the vehicle KNOWING that my tune is NOT being messed with by a well meaning piece of code that might not be getting it quite right.

I'm not saying that it can't be done.
I'm not saying that it isn't useful.

I'm just saying that it can be dangerous, and it is most certainly a secondary thing for cruise/idle areas only. There is a lot more to it than you might guess if you want to use it at WOT and trust it there.

Data-logging should be done using an external card like SD/MMC/usb stick etc and not integrated into the CPU. A flash chip could be added on board with it's own external usb interface etc, but something external seems to me to be much much more useful overall.

I'm participating in derailing Gearheads thread, so I might split this into another topic now.

Save your reply for a few minutes till you see the new thread pop up.

Fred.
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!
austinbob
DIP8 - Involved
Posts: 29
Joined: Fri Jun 06, 2008 3:46 pm

Re: EGO correction and it's importance/downsides/gotchas

Post by austinbob »

What you will have a very hard time convincing me of is that I should allow my computer to adjust the tune that I have set in it based on sensor readings when wires could fail, sensor could be damaged, sensor controller could fail, etc etc etc etc on a long term basis when these things really happen. This I simply do not buy.
The fuel map that is generated is permanant. The sensors could fail and the map would be as good as of the last time it was applied to eeprom memory. YOu can go closed loop or open loop. In closed loop you are autotuning. In open loop you are running on the map that your auto tune produced. I could take the o2 sensors off my bike RIGHT NOW, and my bike would run just as good. That is because I created the fuel map using the feedback from the sensors. Over time the the map becomes statistically correct and the sensors are making very small corrections. At that point you could turn off the feedback.
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: EGO correction and it's importance/downsides/gotchas

Post by Fred »

austinbob wrote:YOu can go closed loop or open loop. In closed loop you are autotuning. In open loop you are running on the map that your auto tune produced.
Open loop is running on a map of some sort that is fixed. Closed loop usually means running on the map + corrections in real time, NOT adjusting the tune while doing that.
I could take the o2 sensors off my bike RIGHT NOW, and my bike would run just as good.
Will this be the case even if the cpu thinks they are connected? Not without significant time spent on code that could be spent on something more useful.
That is because I created the fuel map using the feedback from the sensors.
Exactly, YOU create that map using that feedback from the sensors with manual processes or good statistical analysis of the data at a later date. YOU, not it. Allowing it to change it is asking for trouble IMO and many others too.
Over time the the map becomes statistically correct and the sensors are making very small corrections.
This is the process known as tuning. I understand that tuning a bike or jetski etc is a pain in the traditional sense and can easily see your desire to let the ECU do the work while you ride, however this should still be a one time process that you define the time period of, not a continuing thing out into the future like OEMs do it. If your exhaust develops a leak upstream from the O2 your ECU will be "tuning" (read UNTUNING) the tables the whole time while it gets worse and worse. This is BAD. Any autotuning needs to be under controlled circumstances, not at it's own discretion (or lack there of). The easiest AND BEST way to do this is with a connected external tool. A palm computer is NOT too big to carry on the bike during the tuning period. It's not too big to always carry. Given that it's open source, when it becomes actually useful you will be absolutely free to modify it to suit your specific needs and distribute your modified copy with the licenses. I can't and won't stop you adding code for what the MS lads call "AMC" but it is WAY down my list below almost everything else. EGO correction is further up, but still lowish on the list.

It's good to have this discussion though. I know at least two other members want the computer to do it for them too. But I also know that more members don't want the computer meddling with their tune without their knowledge at all. No one with a good knowledge of the issues will want the ECU adjusting the tables in the WOT range. At least, no one that is making good power per litre where subtle things can cause melted things.

Fred.
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!
austinbob
DIP8 - Involved
Posts: 29
Joined: Fri Jun 06, 2008 3:46 pm

Re: EGO correction and it's importance/downsides/gotchas

Post by austinbob »

The programming for this is not significant AT ALL.

I think you misunderstood some of my words in my prevous post.

I dont want to argue about this anymore. Hopefully the ems will be designed in such a way that evolving it into different directions will be easy.
Post Reply