jcorenman

All you gotta do is solder a cable to the EZK coding pins on coding plug, with a switch on the other end (three-position single-pole, double-throw, center-off). The problem, if I recall correctly, is that you wind up one pin short so you need to scrounge up another pin from a dead harness.

I'll find my drawing and maybe even take a picture

(Be wary of any sentence that begins with "all you gotta do...". That is usually a cover for weeks of intensive effort that the author neglected to mention).

hacker-pschorr

That's hardly unique to 928 engines. We can be thankful 928's have such strong connecting rods (and other components) or they might be our weakest link. Something has to go first.
In some cases the head gaskets and even cylinder walls wave good-buy before the bearings.
Ovalized cylinders don't get mentioned very much anymore, use to be all the rage in the 928 world.

As Smokey said a few decades ago, if you make a pile up all the motors killed by detonation, then make a pile of every motor that failed from anything else the detonation pile will be bigger by a huge factor.

/not an exact quote.....

There is only one way to properly tune for optimal ignition timing and that's on a load bearing dyno (chassis or engine).
If possible, with the knock sensors disabled (as far as I know, the S4/GT/GTS EZK doesn't have this capability)

dr bob

Jim,

Is there consideration for the response time for the oxy sensor? Early sensors had some nasty settling time issues so we couldn't use as much prroportional component in the control as we would have liked. Of course, we were also implementing PI control with differential amplifiers and RC networks. Back in the day...

fraggle

I analyze multivariate data for a living. Multiple inputs, outputs and modeling for optimization is my specialty.

Oh the joy I would have with optimizing this. I need a sharktuner.

justaguy

I have been a machinist for 30 yrs and can fix just about any thing mechinical. I have no clue about multivarate data. I have a whole new respect for all the computer nerds out there.
Thanks Jeff for taking the time to figure out the tuner
Sean

dr bob

I could use something similar to analyze gas turbine tuning. I have constant RPM's, no way to measure airflow accurately, but do have adjustable inlet vanes and fuel valves to vary total load. And combustor temp and exhaust gas temp monitoring. Then need to integrate the results into real-time adjustment signals to what are effectively fuel and very non-linear air inlet valves. Oh, the energy content of the fuel varies and the analysis often lags by a minute or more. First demo of the Sharkplotter at the DFW OCIC really piqued my interest.

jcorenman

And I am sure that was true, then. And it may still be true. But there weren't any knock sensors then, either, and not much chance of hearing audible knocking over an open race-car exhaust. It was a tricky business, tuning those engines

You've mentioned this before, but I am still not getting my head around it. Help me out here, please.

Fine-tuning ignition advance on a dyno to maximize torque (or horsepower, or kiborts... choose one) in each cell, yes I get that, absolutely. That is the objective after all, to maximize efficiency at each load/RPM point. And the optimum ignition timing will generally be close to the knock threshold, hopefully a few degrees shy but often a few degrees beyond (i.e. unreachable without detonation).

And I understand that a skilled tuner would not want the ECU fiddling with his ignition timing, while he is busy fiddling with timing to optimize performance. But what I don't get is why he would not want the information that the knock sensors are picking up something. That seems nuts, to me.

Without knock sensors (as separated from knock-retard), how are you going to know how close you are to the knock threshold? If you can hear it, then you are way past it. That's the part I don't understand.

And I readily agree that I am a doofus when it comes to tuning, I know almost nothing. But I do take comfort that the EZK is covering my six, first taking out advance as needed, and then telling me about it -- if the ST2 is hooked up. Because sometimes we've got to buy less-than-optimum gas, sometimes we haven't had our foot in it for a while and things get a bit carboned, sometimes it sucks some oil down the intake, things vary.

Thanks!

fraggle

I would be feeding the raw data into JMP and give it a working over. When a mathematical predictive model is done, you then allow the JMP to simulate it and choose the best settings for a given set of controllable input factors such as throttle position, rpm, octane, outside temp, etc.

Where excel and other graphing systems start to get weak is in this area. You just can't easily visualize the relationships between all the variables, especially if they aren't independent and non-leanear. A proper model will reveal the finer interactions, and the right supporting software will allow the transforms and advanced **** you need for non-linear responses.

If your data collection is adequate and your model robust, you should be able to predict that knock window without running your motor on the wrong side - or with sensors disconnected.

Bob you should look into a design of experiments method for characterizing your turbine. That's where I WILL go when I nut up and buy a sharktuner.

killav

I would like to see a stand alone "read only" software version of the sharktuner where one could view (but not edit) the bin files for any tuning session, any car. As it stands now, you must have the sharktuner hooked up to a computer in order to load everything up and have a look through the sharkplotter screens.

Using an 85/86 as an example (ST-1), I would like to view the bin files through the sharktuner program, but have to have the car to do this as it stands now. And its not as simple as hooking up the diagnostic ports on the early 32V.


How difficult would this be to create???

John Speake

It wouldn't be dificiult but is set that way for security reasons.

Speedtoys

I just wanna hook up an EGT, and figure out what Lean-Of-Peak is with it.

killav

Ahhh, I see. Thank you John.

IcemanG17

great update........................ Can't wait to get the Widow sharktuned on Friday..............

I wonder if the tuning will gain any HP to your 928 too....not like 288whp is weak to start with!

S4ordie

Have you heard of or use Spotfire?

Speedtoys

The Assometer is never a true tool, but it feels better.