The complete automotive resource for buyers, sellers, and owners like you.
Car, Truck and SUV Forums at Roadfly
+ Bentley Forums
+ BMW Forums
+ Cadillac Forums
+ Chevrolet Forums
+ Ferrari Forums
+ Jaguar Forums
+ Lamborghini Forums
+ Lotus Forums
+ Mercedes-Benz Forums
+ Maserati Forums
+ MINI Forums
+ Porsche Forums
+ General Discussion
+ Marketplace Forums
Re: Actually before I was going to do it (more) (archive)

[ Follow Ups ] [ Z3 Message Board ] [ Msg. Board FAQ ]

Posted by Shawn Fogg on February 26, 2001 at 16:34:03:

In Reply to: Actually before I was going to do it (more) posted by John D. on February 26, 2001 at 15:01:45:


"Any suggestions? The way we suggest with trucks it to put it to the floor and pull the rpms down in increments of about 250 rpm. Or the other way around pull it down to about 1000 rpm and bring it up in 250 rpm increments. I have written the software so that you can create a pattern and then hit the start button and it automates everything."

Lots of suggestions... you may regret asking for them. ;)

The first thing you must keep in mind is that the process of Sharking/Unsharking the DME resets the cars long term fuel trim... adaptation. If you just drive onto the dyno... test it... Shark it... then test again you are comparing Apples to Oranges. Stock is in an adapted state, Sharked is in an unadapted state. This doesn't give you anything meaningful.

You could Shark... test... unshark... test and compare unadapted to unadapted but that isn't much more meaningful as that isn't how it will be on the street. There are lots of other concerns I'll detail below.

So, since the car is driven on the street everyday you should test it in its adapted state. What does this mean. First make sure the car is fully warmed up to get it into closed loop quickly after you do the next step. Now you reverse the software (go from stock to sharked or vice versa). This resets adaptation. Then you need to take the car for a specific drive. Say 20 minutes long... mostly cruising as that is when the car adapts. Then bring it to the dyno and test it if you have to turn off the engine while you are strapping it into the dyno let it idle for 3-5 minutes before the first run. That should get you out of the startup routines when you are testing. Then you reverse the software again, go on the same drive (try to drive the same way) and then dyno it once more again waiting if you had to turn off the engine. This way you are at least testing the two states with a similiar amount of adaptation occuring.

However, there is a bunch more you need to watch out for on the dyno itself.... namely heat and knocking. Hopefully your guys will have LOTS of cooling fans at your dyno. The more the better you simulate real world driving conditions. Not for airflow into your engine but airflow over the engine for cooling. Heat plays a big part in the dyno tests. Probably more on your car then it would on the Diesels you normally test. Couple of reasons for this.

A) You car uses an air temperature sensor to trim ignition timing. That sensor is located in your intake manifold. On a dyno (or even when you stop your car at the side of the road) the temperature that sensor sees is effected by the temperature of the intake manifold. The hotter it gets the more timing the car will remove in an attempt to avoid knocking. If you don't have enough airflow over the engine the manifold will get hotter then it normally would on the street and you can loose some timing from this. One of the things the Shark Injector does is advance timing as shown in my earlier post. If you loose some timing from heat soak that will influence the results.

B) Along the same deal extra heat + advanced ignition timing (from the Injector) make it more likely that the car will knock. If the DME detects knock (before you can probably even hear it) it will retard the timing to get rid of knock. Most likely it will take out more timing then the Injector added to get rid of the knock. Again lose the timing and you will loose the power. Loads of airflow will help keep the heat out and high octane fuel will help resist knock.

Since this is a load dyno (and you are very familiar with the software on the dyno) you may be able to look for tell tales if your knock sensors are kicking in. At each RPM point you load it up at watch for the torque to flutter more then normal. What would be occuring is the DME will retard the timing to get rid of the knock then it will start bringing the timing back in. If it starts knocking again it will take the timing back out. You may see a pattern of the torque rising and falling if it is knocking. If it knocks too much I think it will dial out some timing and keep it out for some amount of time. The stock software will be less likely to knock because its timing isn't as advanced.

Checking for knock would be *much* easier with OBDII software. If you have a computer available there I could lend you Vehicle Explorer and you could just watch your ignition timing directly while you dyno it. It will also let you watch things like your adaptation and such.

Not having an ignition timing makes things a lot easier for you. ;)

My brain is a little bit of mush right now but if I forgot anything hopefully Dpeete will jump in and square me away.


Follow Ups:

[ Follow Ups ] [ Z3 Message Board ] [ Msg. Board FAQ ]
Questions, comments, or problems, please visit the Roadfly help desk. Logo © 1997 - 2018 Jump Internet Inc. All rights reserved.