Well, in order to log any data with the eManage you need a laptop and I didn't think to bring mine with me. Although in response to your question, the eManage doesn't log A/F, the Wideband does that which also hooks to a laptop.

I talked to Steve (the guy running the dyno) about the possiblilty of a difference in the reading from the tailpipe (how they were reading A/F) vs. a sensor pre-cat. He said that there was absolutely no difference in the reading.

One other guy there, Phatmitzu also had the same wideband setup that I have, only he also had the digital gauge. I was curious to see what his realtime gauge was reading vs. what the dyno printout said, but I didn't get the chance to see it. he too had extremely low numbers.


THX723,

I can't remember the last time I reset the ECU, but I know that it's been a few months at least. Most likely it was when I was installing the eManage. I'm a little unsure about the MAF sensor seeing the air correctly as kwkslvr and KRZY1 mentioned. it would seem that if the ECU "sees" and adjusts over time to the mods we add, then what it the point of adding anything. Case in point, both bone-stock Z's dyno'd in the 242 to 245 hp range whereas cars with moderate to extreme mods (not including nos) only dyno'd in the 238 to 243 range.

It almost makes me want to rip out all the stuff I've installed and get re-dyno'd just to test that theory.

-Chris

 

I just realized I was not clear enough in that statement. You're probably thinking of the eManage. I meant that in respond to the ECU itself. 'Long Term Fuel Trim' is a mechanism for all OBD-II compliant ECU.

 

So in order to "see" gains from mods would one need to reset his/her ECU on a weekly basis? That way it doesn't get the chance to adjust itself and bring the hp down?

-Chris

 

chris, could you elaborate on this statement? how could there be no difference between the reading before the catalytic and after? could this have affected the readings at all? i don't know too much about the rxns. w/in the converter, but could this have had something to do with the erratic readings?

i would also tend to agree w/the velocity of the air entering the vehicle, but am sure that the MAF sensor would have accomodated for this. but THX, could the density of the air entering the vehicle have changed enough to where the MAF could not adjust itself that low, hence richening the mixture? this is somewhat confusing, and if i am digging myself deeper into a hole i can't get out of, someone please stop me!

n1cK!

 

N!ck!,
I'll answer your question first, because that's an easy one.

MAF is actually a very accurate device for deriving air mass (thus the name Mas$ Air Flow). You'll have noticed that I said 'mass' not simply volume or velocity. The later two does not actually reveal how many molecules of air exists, because as you pointed out the density of air changes with respect to temperature. This is exactly why all MAFs have a temperature sensing device also built-in. The ECU has this in mind to calculate the acutal mass of air. For a perspective, most multi million dollar wind tunnels also use the same technology to measure air flow ... because it is very accurate.

Oh and by the way, there is a definite difference reading AF from the tail pipe. The characteristic (curve) would not have changed greatly, but it will almost always read approx. 0.7-1.5 pts. higher (leaner) than it should. Mostly due to the oxygen gradient that exists between the outside air and the tail pipe section and facilitated by some back washing that is especially bad for those w. huge tail pipes.

 

n1ck,

Well, I thought the same as you that if there is a sensor reading the A/F before the cat and then a tailpipe sensor reading the A/F after the cat, there would be a difference in the readings. According to Steve, there is a difference, but it is so small that it barely registers and in most cases is the same for both sensor locations. Even after hearing that, I find it hard to believe... but I guess he would know.

When I did all my tuning and logged my A/F doing WOT runs, I logged a steady 12.5 to 12.7 A/F across the board up till redline where it went down to 12.1. The Graph from Powertrain Dynamics shows my A/F in the low 11's and even dipped into the high 10's near redline .

I am beginning to think that it was the difference in air volume that was causing my rich condition. Since the eManage works by adding an additional duty cycle to the injectors based on the number you enter into the cells of the fuel map, no matter what the base ECU map has for fuel input at any given cell, the eManage is going to add fuel based on the cell location. If, as THX said the ECU will adjust for the reduced air input on the dyno (fan only) and lower the amount of fuel to compensate for the air volume, that still will not effect the emanage. Let's look at this example:

Let's say that in actual driving conditions under WOT at Point X on the fuel map the ECU is injecting fuel at a 35% duty cycle. At the same point X on the GReddy eManage map it is adding an additional 20% duty cycle to the injector. So between both the ECU and eManage the injector duty cycle at point X is 55% (hypothetically).

At WOT on the dyno, at the same point X on the ECU fuel map there is less air being pulled into the engine so let's say that the ECU adjusts for this and reduces the duty cycle of the injector to 15%. The eManage is still adding the same 20% duty cycle to the injector at point X. The duty cycle of the injector is now 35%, which is less than the amount if you were actually driving WOT on a road, but since the ECU is only expecting 15% duty cycle and the eManage is providing an additional 20% on top of that, it could explain the rich condition on the dyno vs. the ideal A/F on the road. I'm not sure if that is entirely accurate, but that's how I understand the system to work. I know that scenario is kinda basic, but it could explain this discrepency in the A/F numbers. I would love to go test that theory by logging my A/F on a few WOT runs, but until this rain stops, I'll have to wait. When I am able to do some runs, if my wideband shows my A/F is still in the 12.5 range then either my sensor is defective and not reading correctly, or there really IS a significant difference in the A/F ratio under normal driving conditions and on the dyno.

-Chris

 

Chris,
Here’s my response to your concerns …



What a bummer! Always always take advantage of a wideband sampling of AF that’s pre-cat. There will have been no mistake about what your AF curve actually looked like.

Wow … if you were actually getting 11s and even 10s from the tail pipe, then you are running EXTREMELY pig rich, as tail pipe sampling almost always read 0.7-1.5 pts higher.



That is unfortunately not always correct. If the sniffer probe is the correct design and long enough than it will be very close, but of all the dyno shops I’ve seen in the past … I’ve yet been satisfied by the ones I’ve seen.

Steve is right however regarding the pre-cat and post-cat reading only very slightly. The bulk of the change in AF has to do with the gradient that exists at the tail end of the exhaust.



[QUOTE]One other guy there, Phatmitzu also had the same wideband setup that I have, only he also had the digital gauge. I was curious to see what his realtime gauge was reading vs. what the dyno printout said, but I didn't get the chance to see it. he too had extremely low numbers.[QUOTE]

Again, you missed out on a great opportunity to get to the bottom of this. That would have been an wonderful way to normalize the AF probe they were using … assuming Mark’s (Phatmitzu) AF equipment is properly calibrated.

This reminded me of the following …



Is your wideband properly calibrated to the free air? … before you started to tune w. eManage?!?!? … and please, don’t trust whatever factory claims that they pre-calibrated the unit or the fact that it need not EVER be recalibrated. It’s good to re-cal at least a few times a year … especially for those who have theirs operating under rich environment (mostly FI guys).



Well, it could only have benefited you. Although, in light of your extremely PIG PIG rich AF … that may have done more harm than anything else during the dyno.

For a full explanation of how and why ECU reset and also some goods on how our ECU works … read HERE.

I don’t feel like typing it all out again.



[QUOTE]I'm a little unsure about the MAF sensor seeing the air correctly as kwkslvr and KRZY1 mentioned. it would seem that if the ECU "sees" and adjusts over time to the mods we add, then what it the point of adding anything. Case in point, both bone-stock Z's dyno'd in the 242 to 245 hp range whereas cars with moderate to extreme mods (not including nos) only dyno'd in the 238 to 243 range.[QUOTE]

There isn’t any ‘mod defeating’ algorithm in the ECU. Believe me on this! There are however, two fuel trim correction factors that are always working and changing; one for short-term and the other long-term. This exists for, but not limited to, every OBD-II complaint engine management system. This is the part that a typical ‘piggy-back’ style fuel management correction unit (i.e. eMange, SAFC) can’t quite cope with; that is the ECU is a dynamic thing. Although with enough work … you can get to a point where it’s reasonable, but never ever perfect.



I feel for you … but welcome to modding 101. :P

If you do certain things correctly, it’s actually not that bad of a deal. I’ve got things down to a system and I can reproduce my dyno/AF plots with great accuracy any day any time.



If you haven’t read the ‘THREAD’ I linked above … please do.

Resetting ECU is beneficial in many ways and a REQUIRED procedure prior to Dyno sessions in my book. It is the surest way to be certain you are ways basing your power measurement from the same ignition timing map … a great source of output variance. I’m almost certain those that didn’t do so hot at the dyno day were probably operating off of their lower map.



Hope that helps,

 

Guys... keep in mind air density is variable based on temperature, humidity, and elevation.

Logging A/F at higher elevations and then again at sea level will show some discrepancy.

 

Clint,

Here are some dyno graphs. The first is of my base dyno 1 year ago vs. the best run from this Saturday. All are SAE corrected.

The second graph is of both above runs plus the dyno run I did in June with a cracked plenum. I actually got better numbers on that run than on the one from Saturday!! Everyone who was there on that day was complaining because they said the numbers were about 10-15hp low. Looking at all 3 runs, pretty much the only thing that really changed was my A/F. REALLY confusing. Anyway... here's the graphs.

-Chris

Base (blue) vs. Current on 1-8-05 (red)


Base (blue) vs. June run (red) vs. Current (green)

 

Chris,
Thanks for posting the plot. I'll review them tonight.

I was there with you at the last dyno in June. You may recalled the video I made.

But again, unless you've performed ECU resets before any of those previously established baseline ... they are just not a good point for reference. The cracked plenum didn't help either. But, we can safely say you are most definitely running too rich for this newest Dyno run.

 

Oh YEah!! I remember now.

Yeah, I definitely need to lean out my map. I know that I didn't reset the ECU before any of my dyno runs.

I know there has been a lot of debate about resetting the ECU, but just for clarification, you mean removing the negative battery cable for like 14 hours or something like that, right? I know I have also heard that just removing it and touching it to the positive, then back on the negative worked too, but who knows?

-Chris

 

There's no need to disconnect for 14 or so hours ... that is ludicrous ... LOL! The negative disconnect then touching the positive terminal for a few seconds will do the trick.

When you get a chance, recalibrate the ECU by performing the throttle pedal learning, throttle plate closing learning and finally the Idle air volume learning procedures. Be sure the engine coolant temp is at opertating point before performing them.

 

Well, that's what I wanted to hear. I was hoping it wasn't the 14 hour thing. I've tried to do those "learning" things, but I could never seem to get the timing correct so I never really saw that it was or wasn't working. I'll give it a shot once the rain lets up.

-Chris

 

I hear you on that ... it took me hours to do the first time I tried. It never worked. Finally, I got my self a watch and looked at the second hand ... and bang ... worked! It's all in the timing. Also, you must be certain your car's up to operating temperature. Go drive around for a bit then do it.

 

clint, so which is more accurate? the graph from datalogging in real environment, or dyno graphs? wouldn't it make sense to tune for the real world environment and not the dyno since i don't "drive" a dyno to work? and would chris be right about the eManage? i mean, it makes sense that it would add another percentage no matter what the ecu sees. i could see that being a problem.

i guess you should have several different maps depending on the condition then. say, a map for dyno day, a map for nasty humid weather, hot weather, cold, so on and so forth.

n1cK!

 

Road data is always more accurate. I'm not particularly fond of the DynoJet for that matter. I'd much prefer the 'load-based' DynaPack that can accurately simulate road loads and more. However, for WOT runs on the DynoJet ... it shouldn't have been that far off as seen in Chris' case. Something is definitely out of tune w. Chris's setup.

That is a down fall of all Piggy-Back unit. If the map isn't tweaked right, modern ECUs will tend to be thrown into a twist w. funny reaction after a period. You can get to a point where it's close and good enough, but never perfect. It's also important to not modify portions where close-loop operations exists ... or at the very least do minimal changed in those area.

 

What do you mean?

 

Close-Loop portions of the ECU is tricky business. That is the section where the ECU is most adaptive. If you use the eManage to tweak with it, it can respond in ways you didn't intend.

 

Well, I have modified the numbers in every cell of the injector map. I based my original map on the GReddy TT map, then added and removed numbers from there.

-Chris

 

Clint is a freakin genious.

So basically you can never tune it perfect, but close because the ECU will always tweak it because of POS OBD2

So i would conclude(not sure) resetting the ECU prior to dynoing will cause or help you to get the highest reading on the dyno? Won't the car lose power later on anyways due to the normal ECU tweaking? How often should it be reset?(still don't really know)