Page 5 of 9
Posted: Thu Dec 15, 2005 7:12 pm
by Legacy777
Found my answer. In addition to the *.raw files there are *.001, *.002, etc files. These files have the actual raw data logged, and can be imported into excel.
Posted: Sat Dec 17, 2005 6:31 am
by Legacy777
I'm reposting my other post in here, since it has some info about some things I'm dealing with right now.
I am working on a plan to flow test both MAF sensors, and datalog the output so I have a graph/data of airflow vs. MAF voltage. With that data, I should be able to get the differences between the two, input the voltage changes in the perfect power map and have a pretty decent map without doing any tuning. There will probably need to be some slightly tweaking, but it's a starting point. Then when I add larger injectors, I will have to play with the map again.
I'm looking into this because already, at stock boost levels I'm hitting like 4.9 V on the MAF sensor.
One other thing is I've been talking with a guy on the PP forums about using the MAF sensor input as the deflection instead of the TPS signal. I'm still researching that though.
Posted: Sat Dec 17, 2005 6:36 am
by Legacy777
I've tried inputting values for the Temp & AMP sensors, and it doesn't appear to be working. I'll have to look into it more closely tomorrow or something.
I did some data logging this evening, and ran into something very weird. I was in 2nd gear, low rpms, at full throttle, and accelerating where the car just seemed to lose all power and do one big buck. I don't know what happened. I was data logging, however I didn't notice anything out of the ordinary.
Here are all the files I've data logged. (right click & save as if you want to play with them in excel)
http://www.main.experiencetherave.com:8 ... r/SMT6.004
http://www.main.experiencetherave.com:8 ... r/SMT6.005
http://www.main.experiencetherave.com:8 ... r/SMT6.006
http://www.main.experiencetherave.com:8 ... r/SMT6.007
http://www.main.experiencetherave.com:8 ... r/SMT6.008
006 is the one where I had the loss of power.
Here's an excel spreadsheet with data & graphs. The first one is all over the place. The second two I've trimmed down to just show a pull to full throttle.
http://www.main.experiencetherave.com:8 ... gging1.xls
Posted: Sat Dec 17, 2005 9:36 pm
by bmxpunk
Josh, cranking out the posts, hahahaha
SO in complete fairness, I heard back from Sean at rallitek, It took awhile but it has eased a bit of my tension in this situation. Thank you doug for having him look at that . Ireally appreciate it.
Apparently my injectors are maxing out?!?!?!?
Josh pm me your email if you wanna check out my maps and data logs.
Looks like you have a lean spot or two but looks pretty good most of the way around. I am assuming the long lean spot is deceleration? ( log #8) at first it definately scared me.
Posted: Sun Dec 18, 2005 5:09 am
by Legacy777
Yeah, there's a lean spot when I'm off the accelerator.
Also, I've noticed that there are times where there's no signal coming in on the O2 sensor. So that may be the spot towards the end you're seeing. More then likely the ECU's just ignoring it, and not sending a signal....I don't know.
I went to the dyno today, one thing Dan, the tuner mentioned was that I'm running a little lean until about 4000 rpm or so. More then like it's because the smaller turbo spools so fast. He also mentioned I'm a little rich at the top end.
http://bbs.legacycentral.org/viewtopic.php?t=24511
I want to work on the MAF sensor issues right now. When and if I get things figured out, I'll probably have him tune/tweak things so the AFR is a little better. We'll see.
I was datalogging during the dyno runs, however I haven't had time to look at the data yet.
I'll pm ya with my email addy.
Posted: Sun Dec 18, 2005 5:12 am
by Legacy777
Posted: Mon Dec 19, 2005 2:37 am
by Legacy777
I added to my earlier post, but to calibrate the temp & AMP sensor, you have to divide the voltages you are going to enter by 0.019mV.
I got the engine temp working, however the scale is still off.
The MAP sensor is working but the scale is completely wrong. I'm not sure if I have to enter the values in bar or what.
Posted: Mon Dec 19, 2005 2:41 am
by Legacy777
I've been discussing with another guy on the perfect power forums about using the MAF sensor as the analog deflection instead of the TPS.
He says that for turbo cars, it's a more accurate representation of load, and increases gradually instead of all the way like if you used the TPS signal.
Here's the thread. You can read the details.
http://www.perfectpower.com/forum/forum ... opicId=500
I'm going to be switching my deflection to the incoming MAF signal. Probably won't get around to it until after Christmas though.
Posted: Mon Dec 19, 2005 4:16 pm
by bmxpunk
Legacy777 wrote:I've been discussing with another guy on the perfect power forums about using the MAF sensor as the analog deflection instead of the TPS.
He says that for turbo cars, it's a more accurate representation of load, and increases gradually instead of all the way like if you used the TPS signal.
Here's the thread. You can read the details.
http://www.perfectpower.com/forum/forum ... opicId=500
I'm going to be switching my deflection to the incoming MAF signal. Probably won't get around to it until after Christmas though.
that is the exact reason sean told me to use an aftermarket MAP sensor for deflection.
Posted: Sat Dec 24, 2005 7:41 pm
by bmxpunk
Josh, can the select monitor tell you when it switches from closed to open loop?
I completely redid my map and it is far more drivable and much better gas milage. But my bucking at 3500 is still there and it feels harder.
What I am thinking: two theories I am researching:
1. What is the injector impulse width on the 22t, or 22e (for doug) I am prettty sure my friend got 1.8ms on his select monitor on our 1.8 ecus. more on this after i get some answers.
2. the switch between loops is around 3500rpm and 50% throttle on obd1 cars . Can anyone confirm this.
Posted: Mon Dec 26, 2005 12:32 am
by Legacy777
I don't believe there is anything displaying when it switches from closed loop to open loop.
The only thing I can think of is watching the O2 sensor voltage. It will no longer oscilate between high and low voltage. You should be able to see this on the perfect power, assuming you hooked up the O2 sensor wire.
On the select monitor there is a blinking O2 monitor light, that I believe goes steady....which is similar to the voltage no longer oscilating. This would be the only thing I can think of to differentiate when you have gone from closed loop to open loop.
Are you asking about the injector pulse width at idle?
Posted: Mon Dec 26, 2005 5:30 am
by douglas vincent
Switching from closed to open loop........Hmmmm, interesting.
I hate this bucking. I almost want to get rid of the Perfect Power, but I know it does help keep the det down, so thats the only reason I am keeping it for now.........
Posted: Tue Dec 27, 2005 3:02 am
by douglas vincent
OK, did some grounding today. I ran 12 guage Romex from the battery to the manifold, and to both strut towers.
Did it work?
I think it did! But with hesitation....since i have only driven it a little but everytime I would punch it, no matter what gear or rpm, it was buck free.
But.......
When I first took it out for the drive to lunch, I was getting "some" of the buck/hesitation. But when we came back, it was 100% gone. So it also may have been a computer learning thing to.....
Happy for now.
Posted: Tue Dec 27, 2005 4:16 am
by magicmike
I hate to ask but I searched for about a half an hour earlier today. please dont shit on me if this has been answered but how much do these things cost? If I see that this hesitation problem is figured out I think I'll need one of these to get to my 20lb goal. Thank you
Posted: Tue Dec 27, 2005 4:50 am
by douglas vincent
Posted: Tue Dec 27, 2005 5:47 am
by magicmike
ahh looks like they are having a holiday special. $499 with a halman boost controller. thanks for the link
Posted: Wed Dec 28, 2005 4:08 am
by douglas vincent
Still some hestation/buck, but almost gone. And it seems better, much better, once you drive around for awhile, like once it warms up (not just warmed up, but some boost time warmed up) then it virtually dissappears.
But... its still there. I am going to attached the SMT6 by the heat sink tab to steel ground next.
Posted: Wed Dec 28, 2005 4:21 am
by Legacy777
Your grounding from the ECU to the battery/chassis may not be as good as it could be too.....not just the engine.
Posted: Wed Dec 28, 2005 9:56 pm
by bmxpunk
ok I just got a line on some 440 injectors to see if that will fix some of my porblems.
Will these be enough when I upgrade to a 16G?
I am going to try new grounding like doug did this friday.
Posted: Wed Dec 28, 2005 11:59 pm
by bmxpunk
new theory. so my bucking problems seem to be air-fuel related.
I am wondering and thinking that the 02 sensor grounds out in the exhaust system. well my exhasut is coated so could I be having grounding issues?
Posted: Thu Dec 29, 2005 3:57 am
by Legacy777
If you don't have a 4-wire O2 sensor, then yes it grounds through the exhaust. If it's coated, the grounding will probably not be so good. Try C-clamping a ground wire to the body of the O2 sensor and see if you notice a difference. I'd probably reset the ECU to get rid of any "bad" learning from the bucking.
Posted: Thu Dec 29, 2005 3:58 am
by Legacy777
As for the 440's. I don't know....they'll probably be on the border line.
I think I'm probably going to try and find some 550's (or whatever the sti's are) to be on the safe side.
Posted: Thu Dec 29, 2005 7:07 am
by douglas vincent
Still getting some issues, but nowhere near what I had though.
Havent had the time (or weather) to get in there and do some more grounding.
I am going to run a heavy line from the ECU ground straight to the battery AND T it to the block and chassy.
Posted: Fri Dec 30, 2005 3:29 am
by Legacy777
I finally got the damn MAP sensor calibrated. I wrote up a quick overview on the rallitek forums.
I've had my issues with calibrating the MAP/AMP sensor values in the PP. I finally got it calibrated. Here's a quick synopsis.
The factory manuals have a voltage vs pressure curve as well as numbers, so I got the numbers I needed to calibrate the AMP values in the perfect power.
In the application notes it mentions where each step is 19mV, so I've divided the voltage I want by 0.019 V to get the correct number.
Where I ran into the biggest issues was in terms of the units. I only got the units of bar in the absolute scale to work correctly. I don't know if that's the only thing that works, but that's what worked for me.
Anyway...
My MAP sensor reads 2.6V at 0 mmHg & 4.02216V at 568.864 mmHg.
That works out to 0 bar atmospheric and 0.7584 bar atmospheric.
To get bar in absolute terms, add 1 bar to each value.
So that would be 1 bar abs and 1.7584 bar abs.
These numbers are for the stock 91-94 turbo legacy pressure sensor. The curve is linear, so if you want to calibrate the sensor at a different point, you can use those numbers to interpolate the voltage.
Here's what I entered in the dos version of the perfect power.
AL 0 1 136.84
AH 15 1.7584 211.69
So hopefully this will help save someone a little of the frustration I went through.
Posted: Sat Dec 31, 2005 6:40 pm
by bmxpunk
SO I went in and added some ground wires manifold/coilpack to the battery, both strut towers together and to the battery, and I used a hose clamp to place one on the 02 sensor to the strut tower.
I alos changed the diff fluid and the engine oil.
I installed new spark plugs- bkr7es-11 (one range colder gapped to .038. and reset the facotry ecu.
The car is a hell of a lot smoother but I dont have enough gas in the tank to try WOT to see if it bucks still, though doesnt feel like it is going to.
I did get to lower some of the injector duty cyle at lower deflections on the PP.
My A/F ratios read much more consistantly now that I grounded it better. Idle is about 10000 times smoother even wit h the cam lop. I also noticed driving around A/F ratios stayed at 14.7 almost the whole time.
The car still runs a little lean up top but I have new injecotrs coming (440's used safely on and early sti with a 16G @ 16-18psi and 80%duty cycles) and I am going to due a RRFPR at the same time to be on the safe side.
Any Reccomendations on a RRFPR? What ratio should I look for.