"Violent"... When doing a first gear downshift and when you get down to 3000 RPM, you grab a big fist full of throttle. Do you prefer "sudden"?

Andy

 

How do you downshift from first gear? Anyway, if you are going from a long decel to a WOT accel condition you aren't in closed loop. The ECM isn't using O2 feedback. It is running on the VE table and if the VE table and enrichment table is set correctly. This isn't a problem and not something that needs to be addressed. Acceleration from 3000 is acceleration from 3000 as seen by the ECM. So, if you can steady state this when mapping VE's. Your VE's are set. That leaves the enrichment table. Let's keep in mind. The closer the state of tune is. The less enrichment we need. So, we can tune this table by holding a steady state and accelerate from this point.


Something else that needs to be kept in mind. An engine is going to pump so much air at a given rpm. Where the throttle blade is positioned isn't going to have a huge effect on exhaust flow if 100kpa is being seen. (open throttle) This means that if more exhaust is flowing through the exhaust system. RPM has changed. RPM doesn't change sudden. It isn't going to go from 3000 to 4000 in-between firing events. It is going to accelerate from 3000 to 4000 rpm as it goes through all the rpm points in-between. Gear ratio or leverage is going to change the acceleration but it will never rev as fast as when it is in neutral. This holds true for rapid temp change as well.


I think something that needs to be understood is having a calibration set up for closed loop operation does not mean that the bike is in closed loop 100% of the time. Even if you are in the closed loop area 100% of the time. This is even true for target tune. There are times when the ECM is going to do nothing with the feedback system. It is going to be running on the VE's and the adaptive table. If adaptive is set up and on. If not. Just VE's. This is why it is important to have the VE's mapped correctly in the first place. Even with target tune.


How do we work around and trouble shoot all of this? Well, the first step in trouble shooting is to know how it is suppose to work in the first place. Only then can one compare what it is doing to what it is suppose to be doing.


Good luck

 

Good question ...is the Delphi running PID, PI or simple on/off?

I might have thought that a full PID loop would not just give up and revert to a table. Transient changes would be catered for by the Derivative (or rate of change) part of the feedback loop and would be tuned by the Acel/Decl tables ...no??

Or is Delphi just running PI or on/off ...as we see with the controller only reading high/low from the O2 sensor??

 

 

I don't think giving up is the correct way to look at it. The ECM's programming will direct it to use non-closed loop methods when ever the set closed loop parameters are not met. There's more than the O2 sensor data to consider. One of the reasons why some of the calibrations have and use Tables for PE A/F, PE Spark, ect along with Accel/Decell tables. So one way to look at it in my mind is that even when it seems to go Open loop operation, it is still in Closed Loop mode as the other mode is just part of the operation. That's the way I look at it, kinda a whole system approach.

 

When you are collecting closed loop data while tuning with vision. Running AT-Basic. Ever notice that a message pops up? O2 0 or something? Also look though your data. Ever been in a closed loop area holding steady state and the CLI is 100. Either short term is perfect or it is not paying attention to it. This can be seen in all the tuning products in different ways. SE and TTS have blocks that are yellow or un highlighted that turn green when it is paying attention to feed back.


One place with the vision where it ignores CLI is right around 47kpa and with TTS it is 55kpa. This does depend on base calibration.

 

Kind of what I thought. Put simply closed loop is monitoring and reacting, open loop is relying on pre populated tables, so if the system is running closed loop, I wouldn't expect it to give up at all ...I think its just the state of the system and using appropriate sensors and tables to decide what to do. I wondered if hrdtails original words were maybe a bit confusing - I think that if the system is in closed loop, it IS in closed loop, even if it has to switch strategy but that is where PID programming comes in ...unless the system falls outside of the limits, when DTCs get set

 

I was wondering about CLI recently, was why I was hoping Steve would pick up on your suggested next line of conversation.

Rather than talk about my situation I'd like to hear what Steve says first. However, the only time mine drops out to 100/100 is in open loop areas of my AFR map. Steady state in an open loop area is still open loop. Closed loop I always see the CLI chattering away.

Curious to hear why it would be ignoring 47/55 kpa ??

 

There are things in the system that do what they do, I don't always agree with the way it was done but we have to learn and understand them to be able to work with it. That said the ECM has many inputs that it test to see if it is to operate in Closed Loop Fuel control. Any of those inputs drops out of range and the system now goes into Open loop control. Now that is the basic operation, what one needs to understand is that the system uses the Long term (Adaptive) corrections even when operating in Open Loop. So how can you tell which mode it's really in? The adaptive system is a set of memory locations that are defined in the calibration of the engine. How and where each of those memory locations get used, again is setup in the calibration, and NO they are not all the same. So if you do not know wheer they are or how they are setup and if they are active or not, just looking at the raw data isn't going to tell you much now is it? TTS has the necessary information in it's data set that we had to write into the Code side of the ECM to be able to get it. Sure we could have not used anything and tried to look at what was there and then make assumptions but why, when we could just rewrite it to do what we needed.

Just because you see a value in the data does not mean it is active, correct or being used. The adaptive area is what stores in the ECM, the learned fuel corrections. The rest of it tells you what, when and where to use it.

 

I appreciate the other inputs like temperature and MAP that define the closed loop range, we can see some of those in the various tables, but are the adaptive values not just part of the fuel equations, and not necessarily anything to do with open vs closed loop as such??

One thing that jumps out a wee bit in what you say there is ...we don't know where everything is in the calibration ...so we wrote our own ...that might worry me under normal circumstances.

Do we presume other manufacturers are doing the same thing?