Project Analysis Paralysis

I'm in ur Honda, swappin' ur motorz
User avatar
Rocwandrer
Posts: 7152
Joined: Mon May 15, 2006 4:04 pm

Re: Project Analysis Paralysis

Postby Rocwandrer » Fri Aug 09, 2013 3:20 pm

saabman wrote:I am still confused, it doesnt spoof the ECU about air?? Typically piggy backs scale the air value down on the way to the ECU and scale the fuel up on the way out of the ECU. Is this what yours is doing?

Chris


We talked through this and other stuff on the phone (thanks, Chris) but the reason I don't do sensor scaling is that scaling the air value changes both fueling and ignition timing at the same time, and the resulting output (at least with different sized injectors) is unlikely to be able to be correct on both parameters at the same map point for sensor signal mod.

Instead, I apply a percentage duty cycle trim to injector outputs from the stock ECU, and an offset to the spark signal output that is positive or negative measured in degrees.

User avatar
MR Chuck
Posts: 552
Joined: Thu Apr 15, 2010 10:31 am

Re: Project Analysis Paralysis

Postby MR Chuck » Fri Aug 09, 2013 4:18 pm

Rocwandrer wrote:I'd have to weld in a bung or lose eyes on AFR by pulling my wideband to put in an actual narrow band sensor. I have the bung, but don't want to take the time to install it if I can avoid it. By idle changes and soot and so on, I have no reason to suspect a bad reading from my wideband (not conclusive, I know).


My suggestion is more directed at seeing if a valid signal from a narrowband 02 to the ECU makes a change in the right direction to the fuel trims in case you find that you dont have a short and no other means of really disabling it and forcing open loop

Id hook up the stock 02 and let it idle to see what happens to fuel trims. No experience with this...but can you temporarily put the aftermarket WB in the tailpipe for a test like this? Chris, dont you have something of the sort?


Rocwandrer wrote:...I have the option to modify inputs to the stock ecu. All of those maps are zeroed out or turned off entirely. I plan to repin so I can use that approach on the TPS signal only.


So you have the ability to force open loop with only ECU changes?


Rocwandrer wrote:We talked through this and other stuff on the phone (thanks, Chris) but the reason I don't do sensor scaling is that scaling the air value changes both fueling and ignition timing at the same time, and the resulting output (at least with different sized injectors) is unlikely to be able to be correct on both parameters at the same map point for sensor signal mod.

Instead, I apply a percentage duty cycle trim to injector outputs from the stock ECU, and an offset to the spark signal output that is positive or negative measured in degrees.


Youre right that altering the airflow signal changes load and moves you left or right in the maps. And you cant have more fuel and more timing without directly changing the maps. But maybe you dont want to. :wink:

If you do...Can you control the fuel one way and the timing the other?
Charlie #12
2004 Infiniti G35 ESP-T

User avatar
Rocwandrer
Posts: 7152
Joined: Mon May 15, 2006 4:04 pm

Re: Project Analysis Paralysis

Postby Rocwandrer » Fri Aug 09, 2013 4:49 pm

I can add or subtract 20 degrees of timing at the same time as adding or subtracting 400% duty cycle on the injector signal. I can trim timing and fuel for individual cylinders as a function of RPM, etc etc etc.

A good thumb rule would be that you can create any fuel or timing relationship you want, as long as it is based on being a function of the output from the stock computer.

Within the limitations of having to make my fuel adjustments as a percentage of either stock computer duration or duty cycle (and all fuel affecting maps have to be the same, duty cycle or duration based), I can call for any fueling I want using pretty much any parameter vs rpm to generate a scalable interpolated and extrapolated 16x16 map.

Within the limitations similar to above I can change timing as a percentage of stock timing (why would anyone ever want this?) or as an offset value applied to stock timing in degrees.

On top of this, because there are multiple maps for timing and fuel, I can make a multidimensional map for each. So for a single example of many, I can modify fuel based on MAF voltage and MAP pressure in separate cumulative 16x16 maps and add to that 1x16 modifications for water temp, air temp, etc. I can then modify fueling based on a threshold TPS signal and the IACV duty cycle, etc. Similarly with ignition timing.

I think the point of the EMU was supposed to be that it becomes any fraction of a standalone that you need, but the tuning required to work is limited by the degree to which you have deviated from stock, instead of having a massive cost/learning curve hurtle just to install it on a stock vehicle like a stand alone. This is an awesome goal, and they missed the mark pretty badly in my view, but I think on the other side of the learning curve it will be a workable solution.

User avatar
Rocwandrer
Posts: 7152
Joined: Mon May 15, 2006 4:04 pm

Re: Project Analysis Paralysis

Postby Rocwandrer » Fri Aug 09, 2013 5:03 pm

MR Chuck wrote:My suggestion is more directed at seeing if a valid signal from a narrowband 02 to the ECU makes a change in the right direction to the fuel trims in case you find that you dont have a short and no other means of really disabling it and forcing open loop

Id hook up the stock 02 and let it idle to see what happens to fuel trims. No experience with this...but can you temporarily put the aftermarket WB in the tailpipe for a test like this? Chris, dont you have something of the sort?


Confirmed working. If I change the piggyback fuel trims, the stock computer compensates to seek about 14.3-15.2 on the wideband near idle, and constant oscillation much tighter to 14.7 at higher air flow rates.

User avatar
saabman
CMC Weather Chair
Posts: 6186
Joined: Tue Apr 11, 2006 7:50 am
Location: Standish, ME
Contact:

Re: Project Analysis Paralysis

Postby saabman » Fri Aug 09, 2013 5:06 pm

Now that sounds like progress.
Chris

1971 Saab Sonett, #97
CMC 2002-2011 A/X Chair

User avatar
Rocwandrer
Posts: 7152
Joined: Mon May 15, 2006 4:04 pm

Re: Project Analysis Paralysis

Postby Rocwandrer » Sat Aug 10, 2013 4:49 pm

If I disconnect the O2 from stock ecu and ground the signal wire, I don't get fuel trims. If I leave it disconnected and electrically isolated, I eventually, after no trim at all for a while, get rapid positive trims of 25%.

So it looks like I'm leaving the stock ecu connected to O2 and using trim reduction to build the closed loop part of the map...

User avatar
Rocwandrer
Posts: 7152
Joined: Mon May 15, 2006 4:04 pm

Re: Project Analysis Paralysis

Postby Rocwandrer » Sat Aug 10, 2013 7:30 pm

Pins removed from spare boomslang harness, TPS spoof wiring, knock datalog wiring, IACV duty cycle tap wiring, and re-pining for water temp instead of AIT in progress. Data logging computer will charge while off, but not power the computer from my inverter... has about a 1 hour battery life....

User avatar
Rocwandrer
Posts: 7152
Joined: Mon May 15, 2006 4:04 pm

Re: Project Analysis Paralysis

Postby Rocwandrer » Sat Aug 10, 2013 11:47 pm

Opened the boxes that came today - intercooler pump, hose, and heat exchanger. Also some much needed tools and some cold side insulation.

Reinstalled the valve cover rain shield

Tee'd in for a boost signal for the dyno

Rewired TPS from tap to intercept

Wired IACV to the karman vortex signal input so I can hopefully raise the hot idle and stop the valve from going nuts when the ac shuts off

Changed from knock to water temp on ch12

Re-pinned IAT as knock on ch13

Added output pins for fan relays and EBC while I had it apart

Found some loose wires that were stripped at the end, but not isolated and taped them up

Opened up the case and fixed/double checked all jumpers on the EMU including 12 and 13 from above

Autumn says she can hear me coming 4 blocks away, and even at idle it is too loud to start at 10:30 at night, so I find out in the morning if I made any mistakes.

User avatar
MR Chuck
Posts: 552
Joined: Thu Apr 15, 2010 10:31 am

Re: Project Analysis Paralysis

Postby MR Chuck » Sun Aug 11, 2013 5:28 am

More progress. :D

Rocwandrer wrote:So it looks like I'm leaving the stock ecu connected to O2 and using trim reduction to build the closed loop part of the map...


:thumbsup:

Im sure there is more going on in the code than previously thought. 100% open loop is best, but if you got to chase fuel trims to make the ECU function properly...so be it.
Charlie #12
2004 Infiniti G35 ESP-T

User avatar
Rocwandrer
Posts: 7152
Joined: Mon May 15, 2006 4:04 pm

Re: Project Analysis Paralysis

Postby Rocwandrer » Sun Aug 11, 2013 10:17 am

Warning lights came last week and Autumn conveniently put them in the truck of her car and forgot about them, so now everything I ordered to put in for Loring is here.

User avatar
Rocwandrer
Posts: 7152
Joined: Mon May 15, 2006 4:04 pm

Re: Project Analysis Paralysis

Postby Rocwandrer » Sun Aug 11, 2013 12:55 pm

The reason I don't need anti-engine stall is because I have no bov, and my maf is very close to the throttle body (Patrick might be interested).

User avatar
Rocwandrer
Posts: 7152
Joined: Mon May 15, 2006 4:04 pm

Re: Project Analysis Paralysis

Postby Rocwandrer » Sun Aug 11, 2013 11:40 pm

Intercooler pump has 3/4 npt threads, even thought the product info (including the box it came in) says half. I'm hoping Lowes will have the fittings I need. Also, the intercooler fan wasn't in the box. Amazon says I'll have one Tuesday.

I'm tired and itchy, but the intercooler is somewhat better insulated, the coolant return line is MUCH BETTER insulated, and a bunch of wiring is done.

Oh, and I have tires for loring now.

User avatar
Rocwandrer
Posts: 7152
Joined: Mon May 15, 2006 4:04 pm

Re: Project Analysis Paralysis

Postby Rocwandrer » Tue Aug 13, 2013 12:50 am

Holy crap what a day :!:

Thanks Andrew for the sweet deal on tires :!:

Thanks Patrick for setting me up with a reputable brand replacement for my questionable map sensor :!:

Thanks Chris for all your time today :!:

bottom line up front: Thanks to the person(s) who will end up letting me drive their car(s) when this doesn't work out in time for Loring....

list:

call borg Warner and ask them why my 600 mile old turbo is eating itself

check sparkplugs for fouling

compression test

swap MAP sensors

source 3/4 npt x 1" fittings, 1" heater hose

do more intake insulating

find a way to block hot air from getting to filter (buy some aluminum sheet to replace the aluminum sheet I already own that I can't find)

execute plan to get cold air to filter (block side vent air to engine bay, reroute to fender well.)

mount pump

mount heat exchanger and fan

wire pump and fan with relays and ECU switches

insulate water lines in engine bay

find someone with a truck to transport my jug of 93 octane for me for loring

find someone to drive while I road tune

Figure out what the heck is going on with the piggyback wiring to the IACV.

Figure out how to create a dummy water temp signal, so I can enable autotune in the EMU without screwing up the water temp signal to the stock ecu.

label critical wiring (labels already printed)

get front tires rebalanced when OPR wears off (remount fronts on the rear, remove OPR by force?)

Find alternate car(s) to actually drive at Loring.

User avatar
pdool09
Posts: 623
Joined: Mon Apr 04, 2011 8:16 am
Location: Maine
Contact:

Re: Project Analysis Paralysis

Postby pdool09 » Tue Aug 13, 2013 7:55 pm

Hope it's helpful
Patrick D
#173 2005 SMF Scion Tc
#113 2013 SM Scion FR-S
Sold 2013 VW Golf GTI
Sold 1992 Acura Integra

User avatar
Rocwandrer
Posts: 7152
Joined: Mon May 15, 2006 4:04 pm

Re: Project Analysis Paralysis

Postby Rocwandrer » Tue Aug 13, 2013 8:30 pm

Oil line can't go under the axle, not touch, and not run uphill in order to get back to the pan bung. Oil level in the pan is above the bottom of the oil line entrance without any accelerating forces. With the existing (extended to get the oil line away from the turbo) fitting, the line can't go above the axle without going up after going down and without kinking.

And I tried to lift the line by cutting it shorter (touches the driveshaft now), so now the car is out of commission.


Return to “Project and Technical Forum”

Who is online

Users browsing this forum: No registered users and 1 guest