The FC Datalogit software has the recalc base feature which adds the fuel correction to the base fuel map and then reset all values of the correction map to 1.000.

But now have a look at the o2 feedback control and what the manual says about the fuel correction and base maps. The base map should be tuned to achieve a stoich mixture and then the fuel correction map should be tuned to achieve the desired A:F ratio.

Then the PowerFC has a setting for the o2 feedback control, that is the coefficient where it goes into open loop. The PFC looks into the current cell of the fuel correction map and then compares the value with the value from the o2 feedback setting, if it's above, the PFC goes into open loop, below that value the PFC is in closed loop.

Now if you have used "recalc base" you tell the PFC to achieve a stoich mixture over the whole map.

Does anyone think the o2 feedback control works different than described above?

 

I suspected the O2 control worked kind of like you are saying. I've tried to keep my Fuel Correction Map to reflect my target AFR. I even made a suggestion in the PowerFC forum to add an AFR button with a Stoich string gadget to change all values to AFR based on the Stoich value entered and allow editing the AFR. Maybe support Lambda too with a cycle gadget. I thought it fit better with the philosophy of how the PowerFC was designed to work rather than the recalc function which can cause problems but most tuners seem to like. I thought keeping the Fuel Correction Map accurate would allow the O2 control to work better and might make it easier with a self tuning function later also. I tried your FC Tune software v 0.50 and it looks like you tried to implement something like I'm talking about above with the use of the Show target lambda button but the button doesn't appear to do anything. Do you have a newer version of FC Tune? Just a minimal effort at http://www.fctune.com/ with a link to download the most recent FC Tune would be nice. You have spent hundreds of hours on the FC Tune software and not 5 minutes on the web site .

 

Recalc base makes it easier to tune the base map, but FC Datalogit doesn't disable the o2 feedback control, and that shows me that they haven't understood how the fuel correction map / o2 feedback control works.


In one of the next versions you can switch between coefficient and target lambda / AFR. The fuel correction map is then used for auto tuning. The base map can be switched between pulse width and duty cycle.

If someone wants to make a website for FC Tune...

 

Or... they don't care. I think several of the people that use the recalc base are aware of at least some of the problems it causes. They turn off the O2 control and work around the other problems. I think some of the other problems are not being able to lower the base map values enough in some cases and the uploaded data getting corrupted sometimes. They could tell you the problems better than me since I haven't had to deal with them. The problems could be due to software bugs also. It would be nice if you implemented the recalc base in FC Tune even if it's not a good idea. Some tuners wouldn't ever use FC Tune without it. Just make sure the undo works or a confirm requester appears for the rest of us in case we accidentally hit it.

Sweet! I'm liking the FC Tune more and more all the time. For auto tuning will a wide band O2 be read and the base map will have some user set percentage of the fuel taken out until the AFR matches what is set in the fuel correction map?

Can the base fuel map be specified as a duty cycle? Wouldn't some of the injector settings change this duty cycle like the lag time correction. People who have larger primary injectors have to make this value negative to get a lean enough idle and then retune the rest of the base fuel map. No one understands why and noone has found a way around it. Maybe you could shed some light.

 

I can implement recalc base, no problem. But i think a message should pop up that notifies about the downside of this feature. Also o2 feedback should be automatically turned of.


For auto tuning a wideband will be read and then the base map adjusted till the value from the correction map is reached.


Lag time etc. does not affect the duty cycle. Atleast not the on time you are interested when you tune the base map. The total on time is pulse width + lag time, but the lag time should not be thrown into the calculation.


The lag time IS NOT NEGATIVE! The lag time is "Lag time vs. battery voltage" + Lag time correction. Now if you set the lag time correction to a negative value, the lag time is still positive.

 

I agree.


I think we are on the same page. The adjustments should be made very slowly for a Wankel engine because they are not tolerant of lean conditions. Most auto tuning software is unusable on a Wankel because they are so fragile.


I think I understand how lag time works now but I still don't understand how the whole injection cycling works. The "lag time" is the adjusted time it takes the injectors to turn on but there is no feedback to tell when they did? Then the injector is open and injects fuel for the amount of time equal to the adjusted fuel map value ("injection time")? Then it turns off for a while and starts over?

"lag time"->"injection time"->"dead time" repeat

Reducing the lag time value should cause less fuel to be injected because the injector is not injecting fuel at the begining of it's "injection time" because it's still physically lagging. This is one way to reduce fuel at idle but not the best. Reducing the fuel in the fuel maps at idle with larger primary injectors doesn't take out any more fuel with the injector settings like Apex asks for. The idle magic must be harcoded into the PowerFC or are there other settings dealing with idle that the Datalogit doesn't allow editing. What will the settings in the advanced menu of FC Tune do?

 

Right.


Right.


The PFC has a "minimum injection time" setting, but as far as i know it is set to 0.000 so every time should be valid.


Idle ignition timing: You can set the base timing for idle from where the idle ignition control advances or retards.

Fuel cut recovery ignition: Retards the ignition for fuel cut recovery, also has a crank angle setting

Fuel cut recovery enrichment: Enriches the mixture for fuel cut recovery

Idle RPM vs. Water temperature: Corrects the idle RPM for cold start or when the engine is getting hot.

Fuel cut: Sets a load point for fuel cut and fuel cut recovery

Idle D: Unknown, MAP + Unknown setting

Idle Air: Unknown

 

Chiming in late on o2 feedback control. Many find it unnecessary since we can tune very lean (16-17 AFR) under steady state cruise to get decent mpg. Also allows the narrow band to be removed.

 

It's never too late . True, better fuel mileage can be achieved with O2 control off. What percentage of self tuners do you think have tuned their cars that well? A pretty fair number just tune for performance and would be better off keeping O2 control. If it were made easy enough to edit the base map while comparing the fuel correction map and WBO2 AFR then maybe people would tune this way. It would allow people the option of turning O2 feedback on or off, avoid problems with the recalc function, allow for future auto tuning, and make other people's maps easier to look at and edit as the target AFR's could be matched. Options are good in my book. FC Tune already has an advantage over the Datalogit because the base map window, fuel correction map window and a WBO2 history window could all be opened at the same time on the same screen. The color contoured values are very nice to watch for input mistakes. FC Tune has the potential to revolutionize how we tune while transparently directing us in the direction the PowerFC creater intended. It requires an open mind from us the tuners and a lot of work smoothing out the software interface. I'm not saying tuners that already have their PowerFC tuned using recalc should go out and retune but the tuners just starting tuning or tuning for someone elses car should look at doing things another way because there are advantages.

 

Hmm, I wonder why taking out the fuel doesn't change the AFR at idle then. If you figure out why the RX-7 community would like to know. How did you figure out all the PowerFC info? Did you disassemble the code?


Interesting stuff. Some may even be useful. Will this do anything on an RX-7? I have an idea what idle air may mean. The RX-7 and other modern cars often have what is called an (ISC) idle speed control. This is an electronically controlled (solenoid) valve that allows air to bypass the throttle plates to control idle as well as on deceleration to prevent afterburn (backfires and roughness). I believe it has a duty cycle (on a certain % of time) associated with it as I remember reading in the RX-7 Workshop Manual about a diagnostic computer that can be hooked up to read the ISC duty cycle as the throttle plates (opening) should be adjusted if the duty cycle is too high or low. I always thought it would be nice if the ISC duty cycle could be displayed with the Datalogit but it's probably not possible. Maybe the idle air setting is a base duty cycle for the ISC? I wouldn't think it would be necessary though as this value is learned. Anyway, I hope this helps.

 

I've just started looking at this recalc function and I have one question, is it possible (if one had the time) to convert back to the point before the recalc was clicked? My map was tuned using the recalc function so everything is showing 1's now. I was wondering if I went through each cell one by one and back calculated and then turned the o2 control back on if that would help. I'm not too worried about gas milage but it would be nice to get a little better milage on long trips... I'm just learning all this tuning stuff as I go so please be gentle...

 

There is an obvious way around this. Make a DAT copy to save the fuel adjustment values for the cells affect by the O2 control.

On your original dat, 1.00 out the saved O2 cells, do a recalc, save, go get the saved cells and put them back in. This way you only recalc for non O2 control cells.

You could posible also do a re-init to get the orignal values back unless you DAT was modified by some shop.

 

Unless I misunderstood, you currently have all 1's in your inj map, you have your car tuned the way you want it to be and don't want to go back to stock PFC maps, but you want to get your inj map back to being similar to the way the PFC came so O2 control will work, correct?

If that's the case, one way of doing this relatively quickly would be to create a spreadsheet. Copy and paste the base map, and copy and paste an original base map. Divide the cells in your corrected base map by the cells in the original base map, and it will result in a tuned correction map. In other words,
(Tuned Inj Map) / (Orig Base Map) = (Pre-recalc Inj Map)
Just in case it's too early in the morning and I'm telling you wrong, as a double-check, make sure that your new (Inj Map) * (New Base Map) = (Old Base Map) assuming you had all 1's in your old correction map.

Hopefully that makes sense......

 

That's what I did a couple of days ago at work but I ran into a small problem. I can only plug in a value of 1.496 in the Inj Map and I needed to put in a value of 1.6xx. Of course that was in the boost section so it probably wouldn't matter if I left that part all 1's and just adjusted my vacuum portion back...

I really just should quit while I'm ahead before I really screw something up...lol

 

A correction coefficient of 1.6 would mean an air fuel ratio of 9:1. I have never seen a car running that rich as long as you are not on methanol.

 

Glad you caught that! I had forgotten to mention that the correction map only allows +/-50% so if it's more than that you have to move over to the base map to do the corrections.

But I would say that's definitely possible to have a correction factor that big given that depending on your injectors and all that, sometimes the main injector settings are significantly overriding the map, such as the main injector percentage, lag, etc.

 

Maybe I've missed a point? As I understand, the O2 Feedback control switch is to enable closed-loop at all. The value you can enter there is the target AFR closed-loop will aim for. The PowerFC decides between closed- and open-loop depending on actual rpm and throttle position. At least this is what I see on the Evo 4G63 engine. E.g. cruising <4.000 is closed-loop. Everything above is open-loop...

Am I wrong?

Best regards,
Erik

 

From what I understand it works differently on the RX7. If the value in the correction map is less than the O2 feedback setting, then the car attempts closed loop, otherwise it is forced into open loop mode. That's where people run into problems when they do a recalc, since that causes the ECU to try to go into closed loop mode all the time. Also, I don't think you can set what AFR it aims for. It always tries for 14.7, which is all a narrowband is good for detecting anyway.

 

I disagree here, from my understandings the PFC aims for the value set in the fuel correction map. If you have done a recalc base that's actually 14.7:1.

Maybe someone could post a voltage/ratio chart of the stock o2 sensor?

 

The voltage from a stock narrow band O2 sensor varies a lot except at stoich. The RX-7 is no exception and some like me are running a standard bosch sensor. The wankel does produce more heat and varies more because of the heat. Here is a very general AFR chart for a narrow band as requested. Just remember the actual voltages will vary a ton with temperature especially away from stoich...

Volts A/F
0-.1 17:1
.1-.2 16:1
.2-.3 15.5:1
.3-.4 15:1
.4-.5 14.7:1
.5-.6 14.6:1
.6-.7 14.5:1
.7-.8 14.2:1
.8-.9 13.2:1
.9-1 12.5:1

The PFC may aim for whatever AFR is set in the fuel correction map but it's going to be inept at getting any closer than stoich using the narrow band O2.

 

Dig up an old thread. :P
From reading this thread my understanding is....

Using the datalogit.......
In area's of the map that you DONT want to change the base fuel map (i.e moderate/heavy load) you set the values to 1 prior to hitting the recalc base function. THEN after the recalc has completed you copy your old wot fuel correction values back in to the respective cells. Those respective cells would not have been adjusted via the recalc base function as they were 1 for the recalc base.

Does that sound correct?

Is it possible to have FC Tune prompt you to enter a load and rpm maximum value, this would then do the recalc only upto load point 10 and rpm say 4000rpm. Leaving all the other heavy load values alone.

AND OR

Have FC Tune use fuel correction trim values and apply it to the base map to suit.
i.e tab 'Settings 5' on the datalogit. This should leave your fuel correction map as it was with the old injectors but with the trim values applied to the base map.

This would make a really handy feature when fitting up larger injectors to the PFC.

-----------

FCTune's AFR target self tune...
Is it possible to have FCTUNE prompt you to enter a value for load/rpm that will have fctune only modifying afr's between the entered values.

This would be handy in the aspect of setting up different areas of the map for different afr's. i.e light load areas (say load points 0-10) with a lean 16-17:1 then setup moderate load areas (say load points 11-13 - 0-2psi boost) with a leanish 13.5:14:1 afr. And possibly wot area's from load points 14+ with an 11.5-12:1 afr.

As mentioned prior it may not be such a good thing for the rotors but for the old piston motors it would be nice.