I would expect the ultimate to pickup raw knock voltages. The raw knock voltages by themselves aren't useful unless you can apply a filter. Here's the filter or voltage cutoff that AEM EMS uses to determine if it's real knock or just noise. At low RPMS, any significant raw knock voltages is intrepreted as real knock, but as you go higher up the RPMS the valvetrain noise gets worse so the cutoff is adjusted higher.
so, there isn't a working feature that monitors knock count?
You seem to be very familiar with the software... Is it possible that a filter can be written to provide a better interpretation of knock? Do you know if the support tool uses a driver to communicate with the system, or is it a direct communication? (probably is).
How exactly did you mount the sensor after you tapped the Y-pipe? Did you just epoxy it up or something else?
Well the threads come with a small amount of compound on them; although I used a little bit of 'the right stuff' (RTV) to make a complete sealed gasket.
...I've been obsessing about the temp and baro signal thing, and its interesting that the stock sensors are mounted in the stock airflow meter, even though a much more accurate temp reading would be sampled from the Y-Pipe or even the plenum. The architects must have assumed this temperature inaccuracy when they wrote the ECU firmware.
The best temp readings would come from mounting the GM IAT sensor in the Y-pipe or plenum.
I no longer believe that last statement. Our ECU programming operates from the assumption that the IAT is mounted in the intake as part of the stock airflow meter. It 'expects' the standard temperature at that point to be 77*F, and if the temps are above that, it decreases injector pulse width (and if temps are below, it increases IPW.) In my experience, intake air temps run 15-25 degrees (F) more than ambient temperature depending upon whether you are moving or not.
Temperatures in the Y-Pipe or plenum are much higher than 77*F. If the IAT sensor is mounted there, our ECU will constantly decrease injector pulse width to attempt to compensate. You can tune around it, but why?
The point is, the ECU programming was written with the assumption that the temp reading is taken from the intake, not the IC piping or plenum. To be consistent with that, it makes sense to mount the temp sensor in or near your intake.
Then, using a resistor to provide a fixed barometric pressure reading is problematic IMHO. A fixed 760 mm Hg equals barometric pressure of 29.9, or 1 atmosphere. Barometric pressure is a statement of air density, and our ECU needs that to accurately calculate mass airflow. If the ECU is told baro pressure is always 29.9 when in fact it is higher or lower than that, it will make no adjustments, and you'll be running leaner or richer than you should. Worse than that, your tune will drift depending upon the weather and your altitude. So if a low pressure weather front comes in (thunderstorms, tornadoes, etc.), and actual barometric pressure is now less than 29.9, you'll be running rich because the ECU doesn't realize that barometric pressure has fallen. Same thing if you drive up into the mountains. Or conversely, if a high pressure weather front comes in (beautiful sunny summer day perfect for drag racing or auto-crossing), you'll be running lean because ECU doesn't realize the pressure is higher, and can't adjust.
Bottom line: I now believe the right place to mount a temp sensor is in or near the intake, and I believe it is critical to have a barometric pressure sensor to allow the ECU to fine tune the IPW according to existing conditions.
... using a resistor to provide a fixed barometric pressure reading is problematic IMHO.
When you convert to speed density, the pressure sensor is already compensating for barometric pressure. Therefore you should keep the barometric pressure sensor input fixed, as you don't want to compensate for it twice.
so, there isn't a working feature that monitors knock count?
You seem to be very familiar with the software... Is it possible that a filter can be written to provide a better interpretation of knock? Do you know if the support tool uses a driver to communicate with the system, or is it a direct communication? (probably is).
The raw knock voltages aren't meaningful without knowing the threshold for real knock, that is why I included the diagram. It would be nice if greddy improved the software and applied a user definable filter to the raw knock voltages and determined if you had knock, and how much.
Also it wouldn't be too hard to write a seperate program that read the greddy datalog files and applied the knock filter. This would be really handy for 96-99 car owners.
When you convert to speed density, the pressure sensor is already compensating for barometric pressure. Therefore you should keep the barometric pressure sensor input fixed, as you don't want to compensate for it twice.
OK, that makes sense (doh!). And what's your opinion on the temp sensor? Intake or IC pipes?
Ellery, how difficult would it be to code a program like that? I'm a 96 and need all the help I can get.
Quote:
Originally Posted by BlackStealth
The raw knock voltages aren't meaningful without knowing the threshold for real knock, that is why I included the diagram. It would be nice if greddy improved the software and applied a user definable filter to the raw knock voltages and determined if you had knock, and how much.
Also it wouldn't be too hard to write a seperate program that read the greddy datalog files and applied the knock filter. This would be really handy for 96-99 car owners.
OK, that makes sense (doh!). And what's your opinion on the temp sensor? Intake or IC pipes?
Well I'm pulling intake temps from 118 degrees fahrenheit (at cruise) and upwards of 150- 165 degrees (WOT), from the Y-pipe. So if your speculation is correct, then I could try to mount the IAT in the intake pipes and log the IPW.
Go to www.stealth316.com and do a search for Jeff's turbo outlet temperature calculator where you can enter ambient and intake temps, boost pressure, etc., and it will calculate your plenum temp for you. See how accurate Jeff's calculators with what you are seeing.
The raw knock voltages aren't meaningful without knowing the threshold for real knock, that is why I included the diagram. It would be nice if greddy improved the software and applied a user definable filter to the raw knock voltages and determined if you had knock, and how much.
Also it wouldn't be too hard to write a seperate program that read the greddy datalog files and applied the knock filter. This would be really handy for 96-99 car owners.
Well, I'd imagen that it's a matter of coming up with an algorhymth to distinguish knock from the other noises in the engine. So it's probably not so much a threadshold as it is keeping track of the reading's delta. Also, reading the logs would be an easy way to do it, but I'd probably be more interested in getting a live feed of the knock. But yes, simply interprenting the logs correctly would be a great tuning tool.
Quote:
Originally Posted by Mitsugirl3kgt
Ellery, how difficult would it be to code a program like that? I'm a 96 and need all the help I can get.
How hard is it to write? Well, if it's a matter of reading the logs... it's easier... I'd just have to come up with the algorhymth that works best. For a live feed, I'm not sure as I don't own an emanage and I'm not familiar with the interfaces available.
Anyone want to send me their logs so I can take a look at the structure? (efamilia@gmail.com)
The AutoGuide.com network consists of the largest network of enthusiast-owned enthusiast-operated automotive communities.
AutoGuide.com provides the latest car reviews, auto show coverage, new car prices, and automotive news. The AutoGuide network operates more than 100 automotive forums where our users consult peers for shopping information and advice, and share opinions as a community.
Re: -= Ultimate Emanage Thread =-
