EGO versus Lambda correction
#1
Elite Member
Thread Starter
iTrader: (15)
Join Date: Dec 2007
Location: San Antonio, Texas
Posts: 4,847
Total Cats: 27
EGO versus Lambda correction
This is more of a confirmation than a question, just want to make sure I have this right after reading (and rereading, and rereading) the megasquirt and megatune manuals, as well as some threads here like this one
https://www.miataturbo.net/forum/sho...ght=ego+lambda
So I was confused as to the existence of EGO and Lambda as they both seem to be closed loop mixture control based on O2 sensor readings (WB02 in my case). I did not understand how both could run simultaneously without conflicting. But I think I get it now...
EGO is the 'simple' closed loop control that richens or leans the mixture to meet the EGO switch point set in the 'Exhaust Gas Settings'. This would work for narrow band sensors since they essentially only read lean-OK-rich. But EGO can also be used with a WBO2.
Lambda based closed loop control allows one to have variable O2 targets based on the 8x8 AFR Targets table (function of MAP and RPM). This is more sophisticated of course since you can richen the target mixture up for when you are in boost for example. When you set MS to 'Use Target Tables - Always' then it ignores the EGO target and only uses the AFR Targets. I also see the option to allow EGO closed loop control to run instead of the AFR targets if the TPS is below a certain setpoint.
So, at any given instant in time, the MS is only using EGO or Lambda. If you exceed the open loop threshold in the Open Loop Mode settings then neither EGO or Lambda are controlling anything.
So do I have it right?
https://www.miataturbo.net/forum/sho...ght=ego+lambda
So I was confused as to the existence of EGO and Lambda as they both seem to be closed loop mixture control based on O2 sensor readings (WB02 in my case). I did not understand how both could run simultaneously without conflicting. But I think I get it now...
EGO is the 'simple' closed loop control that richens or leans the mixture to meet the EGO switch point set in the 'Exhaust Gas Settings'. This would work for narrow band sensors since they essentially only read lean-OK-rich. But EGO can also be used with a WBO2.
Lambda based closed loop control allows one to have variable O2 targets based on the 8x8 AFR Targets table (function of MAP and RPM). This is more sophisticated of course since you can richen the target mixture up for when you are in boost for example. When you set MS to 'Use Target Tables - Always' then it ignores the EGO target and only uses the AFR Targets. I also see the option to allow EGO closed loop control to run instead of the AFR targets if the TPS is below a certain setpoint.
So, at any given instant in time, the MS is only using EGO or Lambda. If you exceed the open loop threshold in the Open Loop Mode settings then neither EGO or Lambda are controlling anything.
So do I have it right?
#4
Boost Czar
iTrader: (62)
Join Date: May 2005
Location: Chantilly, VA
Posts: 79,501
Total Cats: 4,080
the swtich point is meraly the voltage you reach 14.7....
a narrowband would have a switch point of .5 (half of 1v)
a wideband typically has a switch point of 2.25 (half of 5v)
lambda is your O2 reading. EGO is what you wanna do with it.
turn EGO off and you still get a lambda reading, it's just that MS doesn't do anything with it. Turn EGO on and you can alter your on-fly-on fueling based on real-time lambda readings.
if you turn off the targets table, EGO doesn't work. it's simply the map you base the adjustments off of.
say you read 16:1 AFR while cruising, and your targets table say you should be at 15:1, so your EGO kicks in and makes adjustments based on your table and your actual lambda reading.
a narrowband would have a switch point of .5 (half of 1v)
a wideband typically has a switch point of 2.25 (half of 5v)
lambda is your O2 reading. EGO is what you wanna do with it.
turn EGO off and you still get a lambda reading, it's just that MS doesn't do anything with it. Turn EGO on and you can alter your on-fly-on fueling based on real-time lambda readings.
if you turn off the targets table, EGO doesn't work. it's simply the map you base the adjustments off of.
say you read 16:1 AFR while cruising, and your targets table say you should be at 15:1, so your EGO kicks in and makes adjustments based on your table and your actual lambda reading.
#9
EGO settings in the below menu determine where the switchpoint is and basically the amount of change and frequency of the adjustments made.
Lambda AFR Targets are used by EGO, the adjustments made are based upon these tables. The are enabled in the Lambda AFR Settings menu as shown.
This target table is what EGO uses to base it's adjustments off of. When running closed loop it will make use the step size and authority to adjust the VE table to try and hit the target value. This table is shown in volts, do not use it as it is very lean. AEM UEGO's use volts but the LC1s use AFR values instead.
All of this is ignored when in open loop.
This is how I understand all of this anyway.
(Do not copy or use these settings, these images are used just to show which menu i'm talking about. These settings will most definitely cause major issues and possible engine damage.)
Lambda AFR Targets are used by EGO, the adjustments made are based upon these tables. The are enabled in the Lambda AFR Settings menu as shown.
This target table is what EGO uses to base it's adjustments off of. When running closed loop it will make use the step size and authority to adjust the VE table to try and hit the target value. This table is shown in volts, do not use it as it is very lean. AEM UEGO's use volts but the LC1s use AFR values instead.
All of this is ignored when in open loop.
This is how I understand all of this anyway.
(Do not copy or use these settings, these images are used just to show which menu i'm talking about. These settings will most definitely cause major issues and possible engine damage.)
#10
Elite Member
Thread Starter
iTrader: (15)
Join Date: Dec 2007
Location: San Antonio, Texas
Posts: 4,847
Total Cats: 27
OK late lunch today so I can respond
Thanks for the screen shots Chad, fantastic. I was going to do this myself this evening but you beat me to it
OK I think that answers my question. In the 'Lambda Sensor Targets' window if you set the 'Use target tables' to 'always' it will ignore the 'EGO switch point' in the 'EGO Control' window, and use the set points from the 'AFR Targets' table instead. However it still uses other parts of the 'EGO Control' window, like 'Controller Step Size' and 'Controller Authority'.
From the MS extra software manual "If you are using a WideBand Lambda then the AFR Target Table is another option (an 8x8 table), if using the table then the EGO switch point is ignored."
Furthermore, if you use the 'Change EGO Limit above (KPa)' setting you can override the 'Controller Authority' value (in the 'EGO Control' window ) to the value entered in the 'Change EGO Limit'.
From the manual "You can now change the Ego Limit depending on the MAP KPa so the ECU can have a larger authority during cruising and a smaller authority when accelerating hard. If you don't want it to change simply put 255 in the KPa setpoint."
OK I think I have it straight?
BTW I know that this does not permanently alter the VE table, it just does on-the-fly tweaks to try to meet the AFR targets. It is more of a fine-tune than a general tune tool. I'll have all this disabled when I am doing the initial VE tuning (open loop). I just want to know how it works should I choose to enable it later on. If you were wondering I am not planning on using closed loop control via EGO to compensate for a poorly set up VE table.
I also understand that even when running open loop (EGO disabled) that MS will still measure and display the WBO2 readings. This is an important fact for datalogging.
This MS programmability is really cool. Once you understand all it can do, it really is pretty sweet.
Thanks for the screen shots Chad, fantastic. I was going to do this myself this evening but you beat me to it
OK I think that answers my question. In the 'Lambda Sensor Targets' window if you set the 'Use target tables' to 'always' it will ignore the 'EGO switch point' in the 'EGO Control' window, and use the set points from the 'AFR Targets' table instead. However it still uses other parts of the 'EGO Control' window, like 'Controller Step Size' and 'Controller Authority'.
From the MS extra software manual "If you are using a WideBand Lambda then the AFR Target Table is another option (an 8x8 table), if using the table then the EGO switch point is ignored."
Furthermore, if you use the 'Change EGO Limit above (KPa)' setting you can override the 'Controller Authority' value (in the 'EGO Control' window ) to the value entered in the 'Change EGO Limit'.
From the manual "You can now change the Ego Limit depending on the MAP KPa so the ECU can have a larger authority during cruising and a smaller authority when accelerating hard. If you don't want it to change simply put 255 in the KPa setpoint."
OK I think I have it straight?
BTW I know that this does not permanently alter the VE table, it just does on-the-fly tweaks to try to meet the AFR targets. It is more of a fine-tune than a general tune tool. I'll have all this disabled when I am doing the initial VE tuning (open loop). I just want to know how it works should I choose to enable it later on. If you were wondering I am not planning on using closed loop control via EGO to compensate for a poorly set up VE table.
I also understand that even when running open loop (EGO disabled) that MS will still measure and display the WBO2 readings. This is an important fact for datalogging.
This MS programmability is really cool. Once you understand all it can do, it really is pretty sweet.
Last edited by ZX-Tex; 03-07-2008 at 03:03 PM.
#11
Bumping this thread. Trying to figure out how to adjust lambda sensor targets and Ego control so I get more closed loop on the fly adjustment. Just not sure where to even start.
What i want to achieve is less leaning at crusing. I set my initial VE table to pretty much what i want but I get some surging and leaning when crusing.
What i want to achieve is less leaning at crusing. I set my initial VE table to pretty much what i want but I get some surging and leaning when crusing.
#12
Elite Member
iTrader: (1)
Join Date: Jun 2006
Location: Warrington/Birmingham
Posts: 2,642
Total Cats: 42
If it's surging when cruising it's too lean, or too advanced add a bit of fuel, or trim a bit of timing.
Using EGO corrections to fix this is bad form, they should be used for minor corrections in fuelling, not corrections of your VE table due to bad tuning.
Using EGO corrections to fix this is bad form, they should be used for minor corrections in fuelling, not corrections of your VE table due to bad tuning.
#16
Elite Member
iTrader: (1)
Join Date: Jun 2006
Location: Warrington/Birmingham
Posts: 2,642
Total Cats: 42
Increase resolution? No, it's hardware limited, (there's not enough memory for bigger tables)
If you're desperate for more bins, just move other more redundant ones around to suit.
I run 40deg at cruise
If you're desperate for more bins, just move other more redundant ones around to suit.
I run 40deg at cruise
Thread
Thread Starter
Forum
Replies
Last Post
bigmackloud
Miata parts for sale/trade
19
01-08-2021 11:24 AM
Zaphod
MEGAsquirt
47
10-26-2018 11:00 PM
Frank_and_Beans
Supercharger Discussion
13
09-12-2016 08:17 PM
Motorsport-Electronics
ECUs and Tuning
0
09-05-2015 08:02 AM