AEM F/IC writeup
- Thread starter Ricktalife
- Start date
OK, if it's in closed loop up until 4250 (could be slightly off since there is no way to get a calibration perfect on this thing), then that makes a little bit of sense, but then what is the first spike in fuel delivery?
You can tell the car is thirsty before those spikes, and as soon as it gets that fuel it takes off. I can't find any good dynoplots of our car, but I remember there is two obvious spikes in power.
I want those spikes gone.
I think it's time to start playing with the O2 signal, if I can override the closed loop pulses, then I can get the car the fuel it needs without the engine taking it away.
You can tell the car is thirsty before those spikes, and as soon as it gets that fuel it takes off. I can't find any good dynoplots of our car, but I remember there is two obvious spikes in power.
I want those spikes gone.
I think it's time to start playing with the O2 signal, if I can override the closed loop pulses, then I can get the car the fuel it needs without the engine taking it away.
Ricktalife
Member
- :
- 2012 Subaru WRX
Well today I hooked up the FIC to my wideband and tucked the ECU back underneath the kickplate. All went well except for the fact that the wires leading from the FIC are too short... so much for mounting in the glove box. I guess I'll use some 3m double sided tape or maybe velcro and find a good spot to hide it while still having access to the usb port. I downloaded some log information from the FIC to my netbook so I'll see what it looks like. I forget if I made any pulls lol.
- :
- 03.5 YLW MSP
Will this work? The blue lines where before a boost leak test, new plugs, and ECU reset. I also had a issue with my internal WG getting stuck open at that time. I don't recall what gear the pulls were in.
http://i223.photobucket.com/albums/dd231/StelathWyvern/msp_dyno.jpg
Ricktalife
Member
- :
- 2012 Subaru WRX
Ok how the **** do you open log files with AEMlog? I downloaded the s*** from the FIC software but it created a text document and AEMlog won't read it.
Yeah it's retarded. AEM's own software doesn't open it's own data log format, which is simply tab delimited text. You have to click on the file menu, then import, then point to the txt file, and then select the floating point options before continuing. It will then ask you to save the file in stf format, the native format for AEM Log.
Here's the whole video on it.
http://www.youtube.com/user/AEMpowerTV#p/u/17/Y5QrNM6lKHQ
Thanks for the dynoplot, Stealth. I don't see the big spikes I thought we had though, you look fairly smooth.
I don't know if spoofing the O2 will work, haven't gotten a chance to try it yet. Even if I do, I doubt it is possible to tune out those spikes in fuel completely. The are instantaneous, the injectors go from 8 to 24, which is 300%.
Ok, just got done watching the O2 narrowband video. They just do the offset. Although they also use a scan tool to measure fuel trim and add fuel so the PCM doesn't eventually alter the long trim fuel maps. I cheaped out and bought the $60, code reader instead of the $120 scan tool. I wonder if Harbor Freight will let me trade up.
http://www.youtube.com/watch?v=Z_IyLEJFEB4
Another thing I learned from watching the videos is you can change the values of the X/Y break points. I could make the break points equal sides of the fuel spikes to smooth them out.
Here's the whole video on it.
http://www.youtube.com/user/AEMpowerTV#p/u/17/Y5QrNM6lKHQ
Thanks for the dynoplot, Stealth. I don't see the big spikes I thought we had though, you look fairly smooth.
I don't know if spoofing the O2 will work, haven't gotten a chance to try it yet. Even if I do, I doubt it is possible to tune out those spikes in fuel completely. The are instantaneous, the injectors go from 8 to 24, which is 300%.
Ok, just got done watching the O2 narrowband video. They just do the offset. Although they also use a scan tool to measure fuel trim and add fuel so the PCM doesn't eventually alter the long trim fuel maps. I cheaped out and bought the $60, code reader instead of the $120 scan tool. I wonder if Harbor Freight will let me trade up.
http://www.youtube.com/watch?v=Z_IyLEJFEB4
Another thing I learned from watching the videos is you can change the values of the X/Y break points. I could make the break points equal sides of the fuel spikes to smooth them out.
Last edited:
Ricktalife
Member
- :
- 2012 Subaru WRX
Yea sorry I already knew that you could change the breakpoints. I would have told you but I thought you knew. As far as the AEMlog goes, I got to the floating point thing but had no idea what to do from there.
No way you could have known I didn't know that. My fault for not noticing it in the manual or watching the videos first.
I did find some stuff about the acceleration enrichment in the latest manual.
I am really interested to see your fuel logs and see if you have the same spikes as me. Since it is closed loop it's kind of random whether the PCM is rich or lean when it switches over to open loop mode. If it is at the rich end of its curve, we get a nice smooth progression into super rich open loop mode, although more often than not, it is at its lean peak and you get that hesitation as the car waits for more fuel.
I am going to play with the O2 today and see if I can't get the car to run richer in the lower rpms. I think this could cure the 4k rpm hesitation. The 2700 rpm spike is another story since that should be in closed loop on both sides of the spike. I am not sure how I am going to get rid of that.
I did find some stuff about the acceleration enrichment in the latest manual.
I am really interested to see your fuel logs and see if you have the same spikes as me. Since it is closed loop it's kind of random whether the PCM is rich or lean when it switches over to open loop mode. If it is at the rich end of its curve, we get a nice smooth progression into super rich open loop mode, although more often than not, it is at its lean peak and you get that hesitation as the car waits for more fuel.
I am going to play with the O2 today and see if I can't get the car to run richer in the lower rpms. I think this could cure the 4k rpm hesitation. The 2700 rpm spike is another story since that should be in closed loop on both sides of the spike. I am not sure how I am going to get rid of that.
Ricktalife
Member
- :
- 2012 Subaru WRX
Well I actually just did a quick datalog from my house back to work after my break, and the graph makes no sense lol. I have to check the settings for the AUX input (aka wideband) , but to me it seems like it's taking the MAF reading and displaying it backwards. I looked at the plot points when I gave it a decent amount of throttle, and the A/F reading was 14-15... wtf? I was watching my wideband anytime I hit the throttle hard, and it was jumping to 10.0 every time. I even had some nice bucking at one point, so it's not just hesitation.
Ricktalife
Member
- :
- 2012 Subaru WRX
Now that I think of it, I wonder if leaving the MAF wire disconnected is messing s*** up? I may have to bag that method and just log the injector duty cycles. But I don't think that would work as well. Damnit, I need time to look at this tonight.
Ricktalife
Member
- :
- 2012 Subaru WRX
Ok apparently the aux gauge settings are for something completely unrelated to an A/F gauge, because they are completely backwards lol. I thought maybe I was wrong at first, but I did some reading and verified that at idle the MAF draws about 1 volt (roughly) and at WOT it can draw anywhere up to 5 volts. So basically it means that when the MAF is close to 0 volts, there is little to no fuel being added aka leaning, and when the MAF is higher than 1-2 volts (again roughly) it is adding fuel aka enriching. So by that logic, one can assume that 0 volts would show 20.0 on an wideband and 5 volts would show 10.0 or maybe even lower. Here is what my settings were (yea I failed again)
edit: Oh yay I just found out by watching the AEM youtube videos that I need to connect the 13a pin (User Switch) to a ground. I just took my good old time yesterday putting the ECU back together and tucking it away all clean and s***. I checked the instruction manual and it doesn't say anything about connecting that wire to a ******* ground. All it says is "N/A"... how the **** am I supposed to decode that as being connected to ground? I'm wondering if I could just get away with putting a terminal on that wire and leading it to a bolt on the frame. Not because I'm lazy, I just don't want to ever **** with the ECU again in fear that my harness may be damaged beyond repair.
edit edit: Scratch that last rant. I only need that wire for internal logging of the FIC. I can still datalog normally with my PC connected. Bullet dodged.
edit: Oh yay I just found out by watching the AEM youtube videos that I need to connect the 13a pin (User Switch) to a ground. I just took my good old time yesterday putting the ECU back together and tucking it away all clean and s***. I checked the instruction manual and it doesn't say anything about connecting that wire to a ******* ground. All it says is "N/A"... how the **** am I supposed to decode that as being connected to ground? I'm wondering if I could just get away with putting a terminal on that wire and leading it to a bolt on the frame. Not because I'm lazy, I just don't want to ever **** with the ECU again in fear that my harness may be damaged beyond repair.
edit edit: Scratch that last rant. I only need that wire for internal logging of the FIC. I can still datalog normally with my PC connected. Bullet dodged.
Last edited:
justanothermp5
Banned
- :
- Rust Bucket
lol nice, at least you figured it out, and as far as where to mount the unit...I have my SSAFC in the space under the cup holder,
works awesome, can reach my laptop by just pulling the cup holder out
works awesome, can reach my laptop by just pulling the cup holder out
Ricktalife, do you buck with only the stock ECU in? Leaving the MAF wire completely untapped going straight into the PCM should be completely stock.
Not sure where you are going with the MAF and wideband voltages findings. With just the ignition on our MAF is 1.5v, then around 2.0 at idle, and hits 4.0 around 3k rpms at WOT. I peak at 4.6.
But this is pretty much unrelated to AFR. For instance, I am up to 16:1 AFR in closed loop with my MAF as high as 4.4v.
The default settings for the Aux in should match up pretty well with most widebands. I have an LC1 which actually goes from 7.35 to 22.39 so I had to adjust mine a little bit, but 10 at 0v and 20 at 5v is pretty typical.
If you ever get running, the O2 sensor modification doesn't work with the 1k resistor. The offset doesn't seem to do anything, and the voltage just seems to kill the signal completely. Oddly, I didn't get a cell with my O2 voltage steady at 0.018 volts. Maybe a higher resistance will work, but I am getting pretty fed up. Nothing I've tried to reduce the AFR in closed loop works, because the PCM just takes it out.
I guess I'll have to put in some more resistance.
Not sure where you are going with the MAF and wideband voltages findings. With just the ignition on our MAF is 1.5v, then around 2.0 at idle, and hits 4.0 around 3k rpms at WOT. I peak at 4.6.
But this is pretty much unrelated to AFR. For instance, I am up to 16:1 AFR in closed loop with my MAF as high as 4.4v.
The default settings for the Aux in should match up pretty well with most widebands. I have an LC1 which actually goes from 7.35 to 22.39 so I had to adjust mine a little bit, but 10 at 0v and 20 at 5v is pretty typical.
If you ever get running, the O2 sensor modification doesn't work with the 1k resistor. The offset doesn't seem to do anything, and the voltage just seems to kill the signal completely. Oddly, I didn't get a cell with my O2 voltage steady at 0.018 volts. Maybe a higher resistance will work, but I am getting pretty fed up. Nothing I've tried to reduce the AFR in closed loop works, because the PCM just takes it out.
I guess I'll have to put in some more resistance.
Ricktalife
Member
- :
- 2012 Subaru WRX
Yea my MAF wire is untapped going straight to the PCM. I had bucking with the bypass connected as well, which would mean it occurred with nothing altering the stock ECU signal.
I was simply using the voltages for the purpose of an example, but I am completely confused on the concept. I have been reading several places that idle (you said is 2 volts) would be reading stoich on the wideband, correct? Then at WOT the voltage jumps to 4 volts (roughly) and when we are in WOT we will be dumping more fuel into the mixture, which would indicate a reading normally in the rich range for a boosted car. So why is the AUX gauge setup to read completely opposite of that?
I was simply using the voltages for the purpose of an example, but I am completely confused on the concept. I have been reading several places that idle (you said is 2 volts) would be reading stoich on the wideband, correct? Then at WOT the voltage jumps to 4 volts (roughly) and when we are in WOT we will be dumping more fuel into the mixture, which would indicate a reading normally in the rich range for a boosted car. So why is the AUX gauge setup to read completely opposite of that?
Ricktalife
Member
- :
- 2012 Subaru WRX
Okay well I just logged a few minutes of data and after staring at the graph wondering wtf was going on... I realized the throttle % never went above 0. I remember calibrating it when I created the map, but the engine was off when i did it. I assume it has to be running when we calibrate it? Until I fix that, I can't make any sense of what's going on but I do know if I give it more than 50% throttle it will buck for sure.
Though what confuses me is, the 'gauges display' shows the tps going above 0% when I blip the throttle.
Though what confuses me is, the 'gauges display' shows the tps going above 0% when I blip the throttle.
Last edited:
Outlawstar98
Member
- :
- 2003 MazdaSpeed Protege / 2013 BRZ
I added this page to the "quicklinks" here:
http://www.mazdas247.com/forum/show...-More!-READ!&p=5513732&viewfull=1#post5513732
http://www.mazdas247.com/forum/show...-More!-READ!&p=5513732&viewfull=1#post5513732
Ricktalife
Member
- :
- 2012 Subaru WRX
Awesome, thank you sir.
