Max,

My bung is close to the turbo and I run a RB dual tip catback downstream of a mid muffler.

I always run the WB in the catback to max the life of the O2 sensor. The only time I get inconsistencies is when idling in high wind conditions, the tailpipe sensor will go full lean having me troubleshooting like crazy until I realize it's just the wind blowing air into the exhaust tip! Once under way there's enough exhaust flow to negate the wind effect.

 

Max

If backpressure is a problem, looks like I should have it also. I am running a Magnaflow muffler in the, SR midpipe, exiting through an SR Dual Tip Exhaust.

Since the AFRs look pretty good, the only problem I see in running the sensor in the OEM position is the possibility that the EGTs exceed the recommended 850-900 C. operational limits suggested by the manufacturer. I have never had an EGT probe in this car so I really don't have a clue what temps I am pushing at the sensor. I am road course tracking the car this weekend and then it is back to the dyno with some friends. I may just check for consistency after exposing the sensor to a weekend of high temps.

 

Uhm for my car at part throttle, it is running mostly in the 12s right now and it runs ok but stinks and sucks up tons of gas. I am going to lean it out to around 14.5 in the NON BOOST areas I think... and blend it in around P8,9,10 where it goes into boost to get to my boost target AFRs which will be upper 11s for 0-5 PSI range. As I go higher in the boost, the AFR I will shoot for will go down to low-mid 11s.

Right now I'm running as rich as 9.8 in many high boost, high rpm (over 7500) cells and it runs like crap when it hits those 9.x cells!

Brian

 

Thanks for the info, Brian. I found the same thing; everything over 6400 rpm and in boost is in the 9s and most of my part-throttle settings are in the 12-13 range. Although with only intake, IC, and DP, I expected to be rich...

Does the PFC automatically go into closed-loop at a given AF ratio, say over 14? I would guess that if so, I should see average AF ratios for those cells right around 14.7. However, checking over the map, the only cells I see that lean are at idle. My autometer af gauge indicates that I'm in closed loop occasionally, so I'd expect to see a few cells at stoich.

During closed loop operation, should my WB be reading stoich and what is the deviation from stoich closed loop allows? I'd feel better about the accuracy of the WB if I saw more measurements corresponding to closed loop operation. BTW, I have a separate bung for the WB, so the stock O2 is still operating.

 

I don't know anything about closed loop because my car runs with O2 feedback disabled all the time. I haven't had an O2 sensor in my car besides the wideband one for like a year

 

A question for all the DIY Wideband + Datalogit types listening here....

Let's say you collect 50 datapoints for a given cell... and the min is 10.50 the max is 14.0 and the average is 11.0.

Looking at the average, you'd say you could lean it out to around 11.5 or so to pick up power... You notice that the datapoints are obviously more clustered around 11.0 and not as many are up at 14 ... but the max is already at 14... So would you lean it out more or not?

Regards,
Brian

 

Brian

The question you are really asking is why are you seeing the 14's. There may be at least 50 legitimate answers.

One of the most likely reasons for the 14's is that they represent the "tip in" curve. "Tip in" is the early part of the fuel consumption curve which occurs just after the TPS volts max out and represents the delay between asking for and getting the additional fuel to the ports.

Look at your chart function and see if the higher AFRs follow very closely to the TPS+ going max. If that is the case, it becomes more likely that the high AFRs are transient, delay dependent, and less dangerous (although you should try to get rid of them also with other techniques). If you see AFRs around 14 a second or two after WOT begins then you have another issue altogether and you should address that as a lean condition in need of richening up.

 

wargasm,

When I datalog a run I never get that many datapoints in a cell. I typically datalog one run at a time and get 3-8 data points per cell. I look at the max and average, should be within 0.1-0.2 AFR or I look at it closer before making changes.

The 14's are likely happening during decel ... 9's may be due to misfire. I had difficulty tuning for awhile as the cells would show a low 10 min value even after pulling out alot of fuel and seeing higher avg AFR.

Best advice is to question everything, if you get a large variation in a cell then don't use the data.

 

Excellent thoughts from both of you

I will study the relationship between AFR and tip-in conditions and see if that's where the higher numbers are coming from.

Perhaps I am logging for too long... some of my cells have several hundred points in them. I figured that more was better. Perhaps in this case, more can make too much clutter.

Brian

 

I checked my log files... I found that in the cells just before the one cell that is 14.xx, the PIMV jumps up from 1.x to 4.x indicating that the leaness was due to throttle tip-in.

So... the next question... what can you do to help with throttle tip-in issues? My double throttle system is removed if that is of interest. Should I put it back? Is there a parameter in the PFC to tinker with to try to eliminate the condition? Or maybe I should just ignore it?

Brian

 

Unless you are getting a lean stumble what's the big deal? Tip-in can occur almost anywhere, how are you going to tune your map to account for tip-in and not run rich as hell during cruise? If you want to tune out the lean condition then go to the accel maps and get to work or try adjusting the inj overlap settings.

 

As twokrx7 says tip in can occur almost anywhere on the maps but would logically be the most problematic at higher RPMs and Boost. To tell the absolute truth, I have the classic tip-in pattern and haven't done anything with it yet. The duration of the tip-in period is negligble and has no noticeable effect on acceleration or other engine parameters.

I have been thinking about playing with the "Accelerate Injector Time" settings in the Settings 2 Tab of the datalogit software, but haven't gotten around to it yet.

Has anyone else tried to kill off the tip-in and were you successful?

 

Good thoughts... that was my conclusion as well. You can't do much about tip-in because the car doesn't "know" you're about to jab on the gas. The whole problem stems from the fact that the car only knows you hit the gas by the TPS and/or MAP sensors and they both have a tiny bit of lag in them.

The car runs great, so I guess I won't worry about the tip-in maximums and I'll just tune for my 11.x value.

Brian

 

guys i am going to get the tech edge unit.... but what one do iget ... i dont need anything fancy and will the 260AU$ one do what i need for the pfc datalogit combo

 

Look at this for sale:

https://www.rx7club.com/forum/showth...hreadid=172413

Pretty good deal and it is the same setup I have. The 1.0 is the base unit and will be replaced by version 1.1 soon. If you buy new the cost is about $175 for the unit and cable and $140 for the sensor. There is also a AFR readout, but I opted not to get that.

 

i can get the sensor for 100 from napa

 

Time for me to interject.

Given the perfect excuse for spending money (tax refund) I came this close to getting some SSR GT1s and tires today, but I didn't get a quick response from Rishie . So, I think money would be better spent upon a water injection system and a wideband.

What would the congnoscenti here recommend.

1) I already have a Power FC
2) I have ordered, and am waiting for Rotary Performance to get the DataLogit in stock
3) I want a wideband, and I have been looking at the Tech Edge stuff.
4) I don't want to mount it, I would rather have something I can stick in the tailpipe when needed. Yes, I understand there will be an RPM delay in what I see but
a) the sensor will live longer
b) I will have a package that can be used on another vehicle. I don't see myself tuning for money, but I have a bunch of modding DSM friends, and I want a mobile, complete package.

So I guess I need logging capability for myself and a display. That way I can log for my FD, and operate remotely for other autos.

So would that be their (TechEdge's) solution:
NTK L1H1 + Cable + WB Unit(1.5) + TE5301 display?

On their site, they say provide your own L1H1, unless they have it in stock. Then they mention that they should have some Bosch LSU4 compatability soon...

I would prefer to pay a little extra for a working package, then possibly dorking things up assembling it.

Opinions?

 

I have the 1.0 NTK version with a 5301 and datalogit. I have never been sorry. However, knowing these guys at TE like I do, I would not hesitate to be one of the first to own the v2.0 when it becomes available. The lack of availability (and high $$$$ currently being charged) of the NTK sensor makes me wish I had an LSU4 option. So unless you need this setup before June (expected production date is May) I would look into preordering the v2.0.

BTW the v2.0 has a load of features that my v1.0 lacks, but with the datalogit I don't miss the additional features except the ability to trim the free air calibration.