A recent experience has clarified some aspects of the two methods for me and I thought it worth sharing .

I often hear that a dyno tuned car ends up being richer on the street and I have also experienced a street tuned car running leaner on the dyno .
It has been suggested on here that this is related to the way the AFRs are measured ie the sensor position is very different .

I believe this to be only part of the story and the main reason has to do with the acceleration on the drums vs real life acceleration .
I have noticed that if all the gears are run off the same fuel map - AFRs get progressively richer as you row through the gears .
I believe this has to do with the time taken to accelerate in each gear and subsequently the amount of fuel needed is less the slower you accelerate.

Dynos are set up for a wide range of cars and power levels - it is near impossible for them to match real life acceleration to your exact car when setting the ramp rate on the dyno . From the few examples iv'e seen it would seem they are set to accelerate faster than our cars would do on the street - hence the reason the dyno tuned car ends up running richer on the street.

 

From Greg Banish Designing and tuning high performance fuel injection systems.
Read Tuning Only On The Street and Tuning Only On The Dyno

 

interesting.
Would airflow have anything to do with it?

 

In principle, the MAF alone should account for everything, including changes in temperature and pressure, as those don't affect the % of available oxygen. However, clearly it doesn't work perfectly, as most all MAF systems also measure inlet air temp, and the RX8 also measures barometric pressure.

Really, all the sensors are pretty flawed in terms of performing their putative tasks, which I think is why they often seem to be performing redundant measurements.

Hopefully they are designed so they can't all be wrong enough at the same time to allow the total system to exceed acceptable parameters.

 

Legal vs Illegal!

 

Might you think it actually has something to do with the delta tables.

RPM delta has a powerful effect on resultant AFR and timing. YOu have to understand the way the fuel tables interact with the Ve table and the delta tables.

 

Actually the velocity vs time delta i.e. acceleration, which is directly tied to rpm by gearing (ignoring wheel and drivetrain slippage)

 

My point was that there are actually tables that control the AFRs separately over time and RPM rise.

Besides understanding how they work, it also pays to use a load-bearing dyno that has a controllable rise time.

I use a Dyanopack and I can set the rise time to match reality for each gear.

 

A dyno will never match the road for actual tuning as it will never load the car the way the road does. Also you can't hold the car at 6000 rpms constant on the highway. I guess you could but there are huge risks. I prefer the track or street over the dyno if I have to choose one over the other. I have seen it time and time again were a car tuned on the dyno isn't quite right on the street or the track, but thats just me and my results.

 

Both are operator error.

Of course, the skill set needed to properly tune a car on the dyno for maximum drivability is well beyond that of most tuners.

The best option in that situation is a combination since, just as you suggested that the "dyno will never match the road", nor will the road match the dyno since there are many scenarios that are very difficult (if not impossible) to repeatedly replicate while driving the car.

 

The VE tables interact with the fuel tables by addition of 1-byte ints, is that what you mean? :P

Are there delta tables for fueling that aren't exposed in ATR? I've been street tuning doing runs up a steep hill on occasion to get greater resolution, now I feel foolish.

 

Think about how the Ve table modifies the fuel tables.

It isn't a static process.

 

OK - thanks for the comments .

FWIW , Here is a Brettspeed tip for those street tuners who want to take their car onto the dyno to test power output:

Tune in 3rd gear and tune AFRs into the low 11s high 10s .
Use this same map for 4th 5th and 6th . The car will run a tad richer in those gears on the street but without any noticeable power penalty. However on the dyno it will run a tad leaner but still in the safe range.


Unless MM wants to chime in and explain how to use these mystical delta rpm tables to prevent this variance from happening .

 

It's impossible to have a proper tune w/ the dyno only. The load is constantly varying on the street/track. You will never simulate this on the dyno.

My favorite is to the the tune on the street and then fine tune it on the dyno.

This is of course if I am doing on from scratch. Makes it alot easier when you have already had a set up several times and can just upload and then clean up the map.