Dyno Tune Misconceptions..

[Phoenyx]

All:

Ok, I'm starting yet another thread here because I'm a little surprised at the realworld results of my datalogging and tuning (Check out "Tuning the Phoenyx").

Correct me if I'm wrong, but it seems that if you tune on the dyno, the tuner will try to meet the correct a/f ratio for the temperature inside the dyno shop. Even with a fan blowing air into the radiator, this is not representative of the correct a/f ratio one would recieve outside with temperatures much lower.

So am I correct in that if you tune to say, and a/f ratio of 11.3:1 at 15psi throughout the rpm band and the inside temperature was about 75 degrees F. Then you go outside at night and punch it, your car isn't going to give enough fuel to keep that ratio, therefore the ratio increases!

Here's my logic. With high temperatures you are not getting as much air volume as you would if you were runnign the car outside at night where the temperatures are drastically cooler. More air volume would cause the car to run leaner since there is more volumetric air. If this is correct, wouldn't one want to tune to a much lower a/f ratio? Say, 10:1 or even lower? This would misrepresent the true power of the car, however, since you are almost guaranteed to get more power outside on the road.

Comments?

-M

 

[candela]

Well yes in a situation when its colder and the air is a bit more dense you will run leaner, but we all know leaner is meaner!

if you hae been on a dyno you would know that running 10:1 and even lower is quite rich and you prolly wouldnt make any power...unless you were in a deisel ofcourse. Anyhow most people usually do account for a loss/gain of temperatures but what can you do? You cannot predict temps/humidities nor do most people dyno there car for ever season of the year....thus is why (Ill use this as an example) someone may tune for something like 12:1 (which is pretty safe for our cars) on the dyno and then go out that night and be seeing closer to 12.5-13.0 (not likely to happen often).

Then in a nice shop there wont be much of a difference of outside temp and what the car is seeing. Fans, garage doors/bays, temp gauges to set a difference are all common in a good shop. And although its a little more tech thatn this and there is no way to tell real world conditions it is by far the most efficient way of tuning as your not driving around, running at the track...yoru just sitting there. I know lots of people drive around with widebands permanantly which is probably pretty cool. But still dynoing is really a great way to get a setup dialed in. Nothing is perfect!

 

[swedeGSX]

The MAF mesures intake air temp. and adjusts fuel to compensate for it. When you adjust with an AFC you are "lying" to the ECU about the amount of airflow the MAF sees and that way adds or subtracts fuel. But the rest of the functions are still active. The ECU takes in alot of different parameters not just airflow.

 

[candela]

 

[swedeGSX]

I'm I wrong? I know I'm not a expert but I think I got it down.

 

[JayHass]

Originally posted by candela

What? He's right.

Unless you are running a speed density set up without temp sensors (Even then they suck) the factory has put these factors into the ECU lookup table.

Hotwire MAF cars don't have this problem at all, and that's what makes them great. You can tune a MAF car in 10* dry weather and it will compensate on a 100* humid day and run the same. (From a tuning standpoint)

 

[candela]

What? He's right.

Yes he is right...I just didn't understand what he was trying to get at that pertained to the topic and until i reread it a few times I didnt get what you were pointing out. Sorry swede

 

[cazual]

Originally posted by JayHass


What? He's right.

Unless you are running a speed density set up without temp sensors (Even then they suck) the factory has put these factors into the ECU lookup table.

Hotwire MAF cars don't have this problem at all, and that's what makes them great. You can tune a MAF car in 10* dry weather and it will compensate on a 100* humid day and run the same. (From a tuning standpoint)

So technically if you are running the DSMtranslator you shouldn't have this problem. Correct. It uses a hotwire GM MAF.

 

[brads]

The DSM translator should work just like the OEM setup in that respect.

The problem is that this post is somewhat right, but for the wrong reasons. The volume of air coming in is identical, if its hot or cold, what changes is density. And as others have pointed out, the ECU accounts for the change in density. What the ECU doesnt know how to take into account, is things like the intercooler not getting enough airflow across it from being stationary. This results in a change of its efficiency, and a different manifold air temp.. A simple little fan in front of it, is NOT the same as 60mph air. It takes a pretty big fan to accurately simulate that. As for why his tuning was so far off on Vishnus dyno, which is an eddy current dyno from what I understand, I couldnt guess without knowing more about the test.

Brad

Originally posted by cazual

So technically if you are running the DSMtranslator you shouldn't have this problem. Correct. It uses a hotwire GM MAF.

 

[candela]

Well thats why I stated in my first post the denser is the word not VLOMUE as stated above. I have seen shops test intake air temps and then get as close to that as possible with any number of fans... this is what I meant to say in the first post about a good shop trying to duplicate this.

 

[Phoenyx]

Originally posted by brads
The DSM translator should work just like the OEM setup in that respect.

The problem is that this post is somewhat right, but for the wrong reasons. The volume of air coming in is identical, if its hot or cold, what changes is density. And as others have pointed out, the ECU accounts for the change in density. What the ECU doesnt know how to take into account, is things like the intercooler not getting enough airflow across it from being stationary. This results in a change of its efficiency, and a different manifold air temp.. A simple little fan in front of it, is NOT the same as 60mph air. It takes a pretty big fan to accurately simulate that. As for why his tuning was so far off on Vishnus dyno, which is an eddy current dyno from what I understand, I couldnt guess without knowing more about the test.

Brad

You are correct but so am I! Density of Air = Mass of Air / Volume of Air. The mass of air (or the amount of air) is directly related to the temperature of the air. The volume of air will increase as the mass of air increases, which in turn, would make the density of air decrease. Since volume stays the same (assuming a closed system like our turbo system with no leaks) from turbo outlet to intake manifold and into the combustion chamber, you are correct that it's not the volume that changes. However, since the temperature does change, this will change the density of air.

Since the ECU only reads outside ambient air (I'm assuming that the temp sensor is in the MAF) to measure density, it wouldn't be accurate in it's calculation of density in the intake manifold where the intercooler could be heat-soaked (as in a dyno room). Therefore, it would run a lot richer at the dyno than it would running outside with the wind blowing and properly cooling the intake temperature (as you stated)..

-M

 

[brads]

The mass of the air is directly related to the temperature, I will give you that. But the statement that the volume will increase as the mass increases, which in turn makes the density decrease, is crackhead logic, it just makes no sense whatsoever. Maybe if you explain it with respect to a temp that increases with volume, it might make sense, but thats a lot of assumption.

Also, you are making an erroneus assumption about the volume. The volume inside the intake tract along with the temp, isnt what the ECU is using to calculate fuel delivery as you seem to be implying. The ECU is measuring volume of flow AT the MAF, as well as barometric pressure, AT the MAF, and temperature AT the MAF. Knowing the volume, pressure, and temperature of the air at that one point, it is able to calculate the mass of air going into the system. Knowing the amount of air mass going into the system, the ECU doesnt have to calculate density at the intake manifold as you are suggesting it does. The ECU will calculate the mass of fuel required for the known mass of air. It does not get fooled by external temperatures like you suggest. If that were the case, then the cars would barely run. And it doesnt use the volume of the intake tract for calcs, if that were the case, anytime a FMIC, or new IC pipes were added, the ECU would go haywire, since the volume would have changed alot, and the ECU wouldnt know about it.

What the dyno room runs change, as I said, was the amount of cooling air passing over the intercooler, which results in it not working as well as it should.

Brad

 

[Phoenyx]

Originally posted by brads
The mass of the air is directly related to the temperature, I will give you that. But the statement that the volume will increase as the mass increases, which in turn makes the density decrease, is crackhead logic, it just makes no sense whatsoever. Maybe if you explain it with respect to a temp that increases with volume, it might make sense, but thats a lot of assumption.

When air gets heated up, molecules get excited and tend to push the surrounding non-excited molecules outwards. If the air isn't confined then the volume (the space it occupies) of that air increases.

Also, you are making an erroneus assumption about the volume. The volume inside the intake tract along with the temp, isnt what the ECU is using to calculate fuel delivery as you seem to be implying. The ECU is measuring volume of flow AT the MAF, as well as barometric pressure, AT the MAF, and temperature AT the MAF.

Correct. So if the temperature is being "perturbed" by the efficiency of the intercooler, you'll run leaner and timing will be more advanced which will lead to knock...

What the dyno room runs change, as I said, was the amount of cooling air passing over the intercooler, which results in it not working as well as it should.

Yep, and that could be significant as far as tuning goes. My SAFC settings were not just off by a couple of %, they were WAY off (-18% at 5000 to +8% at the same rpm).

-M

 

[brads]

But that is completely different from what you said before , which was the mass increases as the volume increases, therefore making the density decrease?

What you are saying below is finally correct. But you have to assume for this system, that since it is being drawn through the same sized MAF, that the volume taken in is going to be fairly constant. So as the air gets hotter, with the constant volume, its the density that goes down.

Originally posted by Phoenyx


When air gets heated up, molecules get excited and tend to push the surrounding non-excited molecules outwards. If the air isn't confined then the volume (the space it occupies) of that air increases.

This part you still arent getting. You have already measured the mass of air going into the system. Sure it gets compressed and heated by the turbo, but its still the same mass of air. The ECU knows the mass, and delivers the appropiate amount of fuel. If the intercooler doesnt work as well, and the temp is higher, then that would equal a lower density, and hence mass of air flowing into the intake. That mass of air flowing into the intake will equal the mass of air flowing out of the turbo. The mass of air exiting the turbo, will equal the mass of air entering the turbo. The mass of air entering the turbo, will equal the mass of air at the mass air flow sensor. SO, If you have lower air mass going into the intake manifold, you would have lower air mass being read at the MAF. The ECU accounts for it all, and will still give you the correct amount of fuel.

Originally posted by Phoenyx

Correct. So if the temperature is being "perturbed" by the efficiency of the intercooler, you'll run leaner and timing will be more advanced which will lead to knock...

They set it at -18% on the dyno? And you found +8% on the street? That is a really big difference, I would wonder if it wasnt more along the lines of some other problem with the car.

Originally posted by Phoenyx


Yep, and that could be significant as far as tuning goes. My SAFC settings were not just off by a couple of %, they were WAY off (-18% at 5000 to +8% at the same rpm).

 

[kpt4321]

The basic point that you're tying to make, that temperature and other atmospheric conditions effect the density of the air, is correct. However, it's damned basic, and doesn't really tell us much.

Your theory that the tuning difference between the dyno settings for your car, and the settings from you logger on the street is based solely on temperature of the air and such, is not feasible. The ECU will compensate for 99% of this change, and even if it didn't, it's not very likely that it would be a change of over 20% on the SAFC.

 

[Phoenyx]

Originally posted by kpt4321
The basic point that you're tying to make, that temperature and other atmospheric conditions effect the density of the air, is correct. However, it's damned basic, and doesn't really tell us much.

Your theory that the tuning difference between the dyno settings for your car, and the settings from you logger on the street is based solely on temperature of the air and such, is not feasible. The ECU will compensate for 99% of this change, and even if it didn't, it's not very likely that it would be a change of over 20% on the SAFC.

You are correct! But you aren't really stating a solution to the misalignment of the SAFC. The basic facts are:

1) The SAFC *was* indeed tuned too low. Not a single person with the same mods I have should be putting negative percent in the upper rpms unless they have bigger injectors. This does indeed indicate a boost leak.

-however-

2) When I tuned the car on the street, I had to raise the SAFC because I was running too lean! Once the settings were now in the positive range, the car is running fine. This does not indicate a boost leak!

Like I said before, I will get the pressure test, so lighten up on that. I'll post when its done.

-M

 

[brads]

This would be overly simple, but is it possible they put some race gas in during the dyno testing?
I'd be curious exactly what their dyno procedure is, ie. how many cool down minutes between runs, if they monitored head coolant or other temps between runs. I know some earlier DSM ECUs would increase the fuel delivery % a good bit if the engine started getting too warm. If they were tuning under that condition, they would take out a lot more fuel to get it dialed in right.

Brad

 

[Phoenyx]

Originally posted by brads
This would be overly simple, but is it possible they put some race gas in during the dyno testing?
I'd be curious exactly what their dyno procedure is, ie. how many cool down minutes between runs, if they monitored head coolant or other temps between runs. I know some earlier DSM ECUs would increase the fuel delivery % a good bit if the engine started getting too warm. If they were tuning under that condition, they would take out a lot more fuel to get it dialed in right.

Brad

I can tell you exactly what happened. There was no race gas put in my car. I was there and watched them.

They didnt' allow any cooldowns between runs. They just kept running the car and adjusting the fuel. It is possible that there was a boost leak somewhere, but the boost graph looks pretty good. It went to 16psi and hovered around 14psi. The wideband they used goes in the tailpipe. Would this cause an erroneous reading in a/f?

-M

 

[brads]

Good question about the wideband in the tailpipe, I guess it would depend on if it was the type properly designed to go there. I am more familiar with dyno runs on bikes(no cats), so I am not too familiar on what sensors they are using to get accurate AF readings after the cat. I would expect that Vishnu would get that right though. But if your exhaust system had leaks, that could throw it off some.

Back to back dyno runs sounds like it could have gotten the thing pretty darn hot. You didnt happen to log while they did the dyno runs did you? Or did you see what the coolant temp gauge was reading?

Have you contacted them to tell them how far off the results were? Politely of course But ask if maybe you can get them to do it again? This time, log the runs, and watch the coolant temps between runs. Maybe even put the settings that they had worked out, back in, and see how well that works first, when the car is cool.

Brad

Originally posted by Phoenyx


I can tell you exactly what happened. There was no race gas put in my car. I was there and watched them.

They didnt' allow any cooldowns between runs. They just kept running the car and adjusting the fuel. It is possible that there was a boost leak somewhere, but the boost graph looks pretty good. It went to 16psi and hovered around 14psi. The wideband they used goes in the tailpipe. Would this cause an erroneous reading in a/f?

-M

 

[Phoenyx]

Originally posted by brads
Good question about the wideband in the tailpipe, I guess it would depend on if it was the type properly designed to go there. I am more familiar with dyno runs on bikes(no cats), so I am not too familiar on what sensors they are using to get accurate AF readings after the cat. I would expect that Vishnu would get that right though. But if your exhaust system had leaks, that could throw it off some.

Back to back dyno runs sounds like it could have gotten the thing pretty darn hot. You didnt happen to log while they did the dyno runs did you? Or did you see what the coolant temp gauge was reading?

Have you contacted them to tell them how far off the results were? Politely of course But ask if maybe you can get them to do it again? This time, log the runs, and watch the coolant temps between runs. Maybe even put the settings that they had worked out, back in, and see how well that works first, when the car is cool.

Brad

Brad,

I appreciate all your help in this. I hope I am not coming off as a troll. My exhaust system is solid. All the pipes are welded together so...

I probably will contact them later about it, but I'm satisfied with my own tuning right now. The car feels good at high rpms and high gears, so I don't want to mess with it much. I didn't log while they were running it cause I didn't have a pocketlogger at that time.

As far as their settings - they just don't make since. Every setting should be in the positive range with stock injectors. Even PSI gives some sample settings on a stock car with just intake cone and they are in the positive. Lowest value at Hi-settings is 0%.

It just doesn't make sense..

-M

 

[autronicDSM]

Placement of wideband o2 sensor does matter. Usually Sensor/analyzer have to be calibrated for where it goes. I had Autronic A/F analyzer and it had 2 options: in header or in the tailpipe. After I tuned my car, few days later I went out and fine tuned air charge modifier table because weather chaged since the first time. When I hooked up my analyzer readings were way off until I realized it was set up for a sensor placed in a header. After changing setting I got the same reading as few days before. Also it matters too but after you install wideband either in tailpipe or header it goes through self calibration again to get the right reading just because it would read less oxygen in the header then tailpipe.

 

[brads]

No problem, I am curious as to the actual reason for their weird settings. -18% is way out of wack, unless there were serious boost leaks, and then it shouldnt change from dyno to street. If they didnt calibrate the meter for its placement like autronicDSM mentioned, that could explain it, although you would think those guys would be aware of that issue.

You will have a lot better likelihood of getting the thing dynoed again for free if you talk to them right away. I know if someone came back to me 6 months later, and told me the mods I made to his car didnt work, and he wants to get another dyno run, I'd be very suspicious about that story. If you do the redyno, maybe you can figure out why their settings were so far off, as well as find out how much power you are making with your self adjusted settings.

Brad

 

[crazyjoe]

Just a quick chemistry lesson here:

MASS of air does not change with temperature. That's like saying that 1 lb of water is lighter if you heat it up. Mass is only adjustable by either adding or removing from the molecules present.

The amount of VOLUME a given mass of air occupies does (or can)change with temperature, however, and I would assume that is what some were trying to say. The calculation mentioned, DENSITY = VOLUME/MASS is correct and shows more of what you were trying to say.

End chemistry lesson.

 

[GrocMax]

The engine is at a different load on an inertia dyno vs. on the road, thus the difference. We see, on 400-500 WHP cars, about an .4-.8 AFR leanout when at the track/on road with everything else remaining the same.

Even see this on S/D systems, not just MAS.

 

[NosLaser]

Originally posted by GrocMax
The engine is at a different load on an inertia dyno vs. on the road, thus the difference. We see, on 400-500 WHP cars, about an .4-.8 AFR leanout when at the track/on road with everything else remaining the same.

Even see this on S/D systems, not just MAS.

You're kidding? You mean LOAD actually makes a difference on an engine? You mean when I sit in my driveway, and free-rev my motor, it doesn't make the same horsepower as it does when I am doing a 3rd gear pull uphill? You mean my aluminum flywheel doesn't increase spoolup because it's lighter and provides less load? The horror!!!

Regards,