Welcome to the North American Subaru Impreza Owners Club Saturday December 20, 2014
Home Forums WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Here you can view your subscribed threads, work with private messages and edit your profile and preferences Home Registration is free! Visit the NASIOC Store NASIOC Rules Search Find other members Frequently Asked Questions Calendar Archive NASIOC Upgrade Garage Logout
Go Back   NASIOC > NASIOC Technical > Factory 2.0L Turbo Powertrain (EJ Series Factory 2.0L Turbo)

Welcome to NASIOC - The world's largest online community for Subaru enthusiasts!
Welcome to the NASIOC.com Subaru forum.

You are currently viewing our forum as a guest, which gives you limited access to view most discussions and access our other features. By joining our community, free of charge, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is free, fast and simple, so please join our community today!

If you have any problems with the registration process or your account login, please contact us.
* Registered users of the site do not see these ads.
Reply
 
Thread Tools Display Modes
Old 07-18-2007, 04:50 PM   #1
hurricane123
Scooby Newbie
 
Member#: 121116
Join Date: Jul 2006
Chapter/Region: W. Canada
Location: Calgary
Vehicle:
2004 WRX
v7 w/VF30

Default Aftermarket EGT Connected To ECU

Has anyone tried this?

It should be possible to "repeat" the signal from a k-type thermocouple mounted in the header into the the stock EGT sensor input. It is possible the scaling might need to change but it shouldn't since I believe all k-types put out the same voltage/temperature ratio ~41uV/*C

Can anyone confirm that the stock EGT sensor is a k-type thermocouple? I threw mine out when I gutted my uppipe so I can't even test it.

I also notice that the scaling in the ECU is in volts, although the k-type outputs in mV ... is there something in the ECU to amplify the signal?

This came up because I'd like to be able to log my EGT's in Enginuity ... I've learned from past experience that trying to watch a gauge while doing a WOT run is very inaccurate. I also want to retain the in-cabin gauge for daily driving, and don't want to have to mount another EGT probe.

Jeff
* Registered users of the site do not see these ads.
hurricane123 is offline   Reply With Quote
Old 07-18-2007, 05:24 PM   #2
Mulder
Trust no one
Moderator
 
Member#: 11170
Join Date: Oct 2001
Chapter/Region: Tri-State
Location: NYC
Vehicle:
02 WRX
15 WRX/05 GTO

Default

Won't work, this has been discussed many times.
The factory probe is a negative-coefficient thermistor not a thermocouple. The ECU won't read a different probe.
Mulder is offline   Reply With Quote
Old 07-18-2007, 06:02 PM   #3
hurricane123
Scooby Newbie
 
Member#: 121116
Join Date: Jul 2006
Chapter/Region: W. Canada
Location: Calgary
Vehicle:
2004 WRX
v7 w/VF30

Default

Hmmmm ..... ok, thanks for the info Mulder. I'll look into another method then.

Jeff
hurricane123 is offline   Reply With Quote
Old 07-19-2007, 12:42 PM   #4
Turn in Concepts
NASIOC Vendor
 
Member#: 93646
Join Date: Aug 2005
Chapter/Region: MWSOC
Vehicle:
Many Track Records
Let us help you go fast!

Default

You can read EGT off the factory probe, but it is very unresponsive and quite a bit off on the actual reading. IMO, not a good option.

Tony
Turn in Concepts is offline   Reply With Quote
Old 07-19-2007, 02:35 PM   #5
hurricane123
Scooby Newbie
 
Member#: 121116
Join Date: Jul 2006
Chapter/Region: W. Canada
Location: Calgary
Vehicle:
2004 WRX
v7 w/VF30

Default

Quote:
Originally Posted by Turn in Concepts View Post
You can read EGT off the factory probe, but it is very unresponsive and quite a bit off on the actual reading. IMO, not a good option.

Tony
Thus why I wanted to be able to bring the signal from the k-type thermocouple I mounted in the header into the ECU for logging purposes.

There's more than one way to skin a cat, I'll keep thinking about it.

Jeff
hurricane123 is offline   Reply With Quote
Old 07-24-2007, 10:45 PM   #6
hurricane123
Scooby Newbie
 
Member#: 121116
Join Date: Jul 2006
Chapter/Region: W. Canada
Location: Calgary
Vehicle:
2004 WRX
v7 w/VF30

Default

Has anybody tried this:

http://cgi.ebay.com/ebaymotors/ws/eB...m=320133245563

Looks like exactly what I need.

Any comments?

Jeff
hurricane123 is offline   Reply With Quote
Old 08-06-2007, 02:28 AM   #7
hurricane123
Scooby Newbie
 
Member#: 121116
Join Date: Jul 2006
Chapter/Region: W. Canada
Location: Calgary
Vehicle:
2004 WRX
v7 w/VF30

Default

Bump for any input.

Jeff
hurricane123 is offline   Reply With Quote
Old 08-06-2007, 01:50 PM   #8
Robertob
Scooby Newbie
 
Member#: 147254
Join Date: Apr 2007
Chapter/Region: SCIC
Location: Los Angeles - Atwater Village
Vehicle:
2005 9-2x
Silver

Default

Try it and let us know

I've been thinking of the same thing (I think I responded on enginuity too).

I'm sure it will work - just use the vref for power to the thermocouple amplifier, and send the output to the ECU like the stock NTC. Rescale the table in the ROM to match the output and you're good to go. Everything will work the way the ECU expects.

In fact, you could even use the closed loop EGT stuff as a way to force lean spool, I'd think - very cool.
Robertob is offline   Reply With Quote
Old 08-06-2007, 02:44 PM   #9
Turn in Concepts
NASIOC Vendor
 
Member#: 93646
Join Date: Aug 2005
Chapter/Region: MWSOC
Vehicle:
Many Track Records
Let us help you go fast!

Default

Quote:
Originally Posted by hurricane123 View Post
Has anybody tried this:

http://cgi.ebay.com/ebaymotors/ws/eB...m=320133245563

Looks like exactly what I need.

Any comments?

Jeff
I spoke at length with Omega Engineering about a year ago when I was thinking the same thing, but for a different purpose. I was trying to find a way to adapt a K type thermocouple to a Thermister. After they tried a few things to see if they could create an adapter and after I got a EE friend of mine involved, there was no solution found to adapt. I worked with a company out of England who was able to provide an adapter, but I believe (as it was being used with a data logger) software was required to get it to convert the signal. I thought of extending the wires and relocating the stock sensor to the manifold, but due to the strange compression fitting that holds it in place and the fragility of the sensor I decided against it.

In the end, there is apparently no easy way to make this happen. If you find out let me know.

Tony
Turn in Concepts is offline   Reply With Quote
Old 08-06-2007, 02:50 PM   #10
Jon [in CT]
*** Banned ***
 
Member#: 2992
Join Date: Nov 2000
Chapter/Region: NESIC
Location: Connecticut, USA
Vehicle:
02 WRX Sedan
Silver

Default

Quote:
Originally Posted by Robertob View Post
I'm sure it will work - just use the vref for power to the thermocouple amplifier, and send the output to the ECU like the stock NTC.
You can't read the output of the thermocouple amplifier via the stock EGT sensor connection.
Jon [in CT] is offline   Reply With Quote
Old 08-06-2007, 10:36 PM   #11
Robertob
Scooby Newbie
 
Member#: 147254
Join Date: Apr 2007
Chapter/Region: SCIC
Location: Los Angeles - Atwater Village
Vehicle:
2005 9-2x
Silver

Default

Quote:
Originally Posted by Jon [in CT] View Post
You can't read the output of the thermocouple amplifier via the stock EGT sensor connection.

Why not? The stock EGT is a NTC thermistor, right? So the ECU "expects" a voltage between 0 and Vref (5volts), higher V for higher temps.

So why wouldn't an amplified EGT sensor value of between 0 and 5 volts work?
Robertob is offline   Reply With Quote
Old 08-07-2007, 01:34 AM   #12
hurricane123
Scooby Newbie
 
Member#: 121116
Join Date: Jul 2006
Chapter/Region: W. Canada
Location: Calgary
Vehicle:
2004 WRX
v7 w/VF30

Default

I ran this scenario past my company's resident hardware R&D lead (we build hardware to acquire/record/display various sensor data on oil rigs) who had no reason to believe that this wouldn't work.

As Robertob said the ECU is just looking for a 0-5v signal. Since we can scale the EGT input now with Enginuity it shouldn't present any problems.

I have the hardware coming anyway so I will let you all know the results.

Jeff

We are also coming up with an inexpensive "smart" IC sprayer control that could be adapted for many other things as well. *stay tuned for that thread*
hurricane123 is offline   Reply With Quote
Old 08-07-2007, 02:27 PM   #13
Robertob
Scooby Newbie
 
Member#: 147254
Join Date: Apr 2007
Chapter/Region: SCIC
Location: Los Angeles - Atwater Village
Vehicle:
2005 9-2x
Silver

Default

Quote:
Originally Posted by hurricane123 View Post
I ran this scenario past my company's resident hardware R&D lead (we build hardware to acquire/record/display various sensor data on oil rigs) who had no reason to believe that this wouldn't work.

As Robertob said the ECU is just looking for a 0-5v signal. Since we can scale the EGT input now with Enginuity it shouldn't present any problems.

I have the hardware coming anyway so I will let you all know the results.

Jeff

We are also coming up with an inexpensive "smart" IC sprayer control that could be adapted for many other things as well. *stay tuned for that thread*
Let us know how it goes (and about the IC sprayer) - there isn't any good reason it won't work. I'd do it but I have too many other projects on my plate right now.
Robertob is offline   Reply With Quote
Old 08-08-2007, 01:20 AM   #14
hurricane123
Scooby Newbie
 
Member#: 121116
Join Date: Jul 2006
Chapter/Region: W. Canada
Location: Calgary
Vehicle:
2004 WRX
v7 w/VF30

Default

Will do. Still waiting for the part, it always takes forever to get across the border.

Jeff
hurricane123 is offline   Reply With Quote
Old 08-08-2007, 12:31 PM   #15
Unabomber
Big Ron
Moderator
 
Member#: 18062
Join Date: Apr 2002
Chapter/Region: MAIC
Location: I can save you a ton of cash
Vehicle:
on car parts so PM
me b4 j00 buy

Default

No offense, but even if you found a fix that cost $1 and took 6 seconds to install......

EGT readings from the uppipe are about as useful as dog crap. You can not use them for any useful tuning functions and all they will let you do is PERHAPS see some trends, but even then, it's skewed all to hell due to the mixing of 4 EGTs.

I mean, knock yourself out and all, but there's no purpose unless someone can enlighten me to something new.
Unabomber is offline   Reply With Quote
Old 08-08-2007, 02:10 PM   #16
hurricane123
Scooby Newbie
 
Member#: 121116
Join Date: Jul 2006
Chapter/Region: W. Canada
Location: Calgary
Vehicle:
2004 WRX
v7 w/VF30

Default

Quote:
Originally Posted by Unabomber View Post
No offense, but even if you found a fix that cost $1 and took 6 seconds to install......

EGT readings from the uppipe are about as useful as dog crap. You can not use them for any useful tuning functions and all they will let you do is PERHAPS see some trends, but even then, it's skewed all to hell due to the mixing of 4 EGTs.

I mean, knock yourself out and all, but there's no purpose unless someone can enlighten me to something new.
Quote:
Originally Posted by hurricane123 View Post
Has anyone tried this?

It should be possible to "repeat" the signal from a k-type thermocouple mounted in the header into the the stock EGT sensor input.

Jeff
In my first post I say that I want to bring in the EGT from the header location (where everyone mounts aftermarket EGT thermocouples) not the uppipe. If I wanted the uppipe EGT I would just use the stock sensor.

Jeff
hurricane123 is offline   Reply With Quote
Old 08-25-2007, 11:58 PM   #17
Ezmendriz
Scooby Newbie
 
Member#: 10747
Join Date: Oct 2001
Location: Puerto Rico
Vehicle:
2002 WRX
Black

Thumbs up Please let us know

I has been looking for something like this for a while... Let us know haow it turns out.


Ezmendriz
Ezmendriz is offline   Reply With Quote
Old 08-26-2007, 12:06 PM   #18
hurricane123
Scooby Newbie
 
Member#: 121116
Join Date: Jul 2006
Chapter/Region: W. Canada
Location: Calgary
Vehicle:
2004 WRX
v7 w/VF30

Default

Well .... so far just splitting the output of one thermocouple to two inputs (gauge/translator) has failed. When I get some time I will try a direct thermocouple->translator->ECU setup. Will update with the results.
Jeff

Last edited by hurricane123; 08-26-2007 at 01:35 PM.
hurricane123 is offline   Reply With Quote
Old 08-27-2007, 10:35 AM   #19
ride5000
Scooby Guru
 
Member#: 32792
Join Date: Feb 2003
Chapter/Region: NESIC
Location: lincoln, ri
Vehicle:
2003 GGA MBP
12.9 / 105+

Default

Quote:
Originally Posted by Robertob View Post
Why not? The stock EGT is a NTC thermistor, right? So the ECU "expects" a voltage between 0 and Vref (5volts), higher V for higher temps.

So why wouldn't an amplified EGT sensor value of between 0 and 5 volts work?
dude, it takes two seconds to google "thermistor."

your understanding of their theory of operation needs work.

what you SHOULD be focusing your energies on is identifying an additional 0-5v analog input to the ecu that is not critical for operation. this has already been done for a different purpose, that of logging wideband uego controller analog translated outputs.
ride5000 is offline   Reply With Quote
Old 08-27-2007, 06:11 PM   #20
LyveWRX
Scooby Specialist
 
Member#: 19057
Join Date: May 2002
Chapter/Region: MWSOC
Location: The Lost Valley, Detroit, MI
Vehicle:
2002 WRX
& Red Ford & '72 CB350

Default

How about just logging it with a LabJack and ECUTek software, thats what I do, I have four thermocouples and a pressure sensor, along with four unused analog 0-10v inputs......
LyveWRX is offline   Reply With Quote
Old 08-27-2007, 07:29 PM   #21
hurricane123
Scooby Newbie
 
Member#: 121116
Join Date: Jul 2006
Chapter/Region: W. Canada
Location: Calgary
Vehicle:
2004 WRX
v7 w/VF30

Default

Quote:
Originally Posted by ride5000 View Post
dude, it takes two seconds to google "thermistor."

your understanding of their theory of operation needs work.

what you SHOULD be focusing your energies on is identifying an additional 0-5v analog input to the ecu that is not critical for operation. this has already been done for a different purpose, that of logging wideband uego controller analog translated outputs.
The translator takes the k-type thermocouples output and "translates" it to a 0-5v output. That is what the stock EGT sensor input is. Why would I go looking for a different 0-5v input?

Also using the stock EGT input would allow me to utilize some of the failsafe features in the ECU ... ie: if EGT>1650* then put the car into failsafe mode, thus cutting boost and going to the failsafe fueling map.

Jeff
hurricane123 is offline   Reply With Quote
Old 08-27-2007, 10:50 PM   #22
hurricane123
Scooby Newbie
 
Member#: 121116
Join Date: Jul 2006
Chapter/Region: W. Canada
Location: Calgary
Vehicle:
2004 WRX
v7 w/VF30

Default

Quote:
Originally Posted by LyveWRX View Post
How about just logging it with a LabJack and ECUTek software, thats what I do, I have four thermocouples and a pressure sensor, along with four unused analog 0-10v inputs......
That is another option, but I want to be able to log it along with all the other parameters (timing, WBO2, manifold pressure, MAF, etc) in one place using Enginuity. As well as the added insurance mentioned above.

Jeff
hurricane123 is offline   Reply With Quote
Old 08-28-2007, 07:41 AM   #23
ride5000
Scooby Guru
 
Member#: 32792
Join Date: Feb 2003
Chapter/Region: NESIC
Location: lincoln, ri
Vehicle:
2003 GGA MBP
12.9 / 105+

Default

Quote:
Originally Posted by hurricane123 View Post
The translator takes the k-type thermocouples output and "translates" it to a 0-5v output. That is what the stock EGT sensor input is. Why would I go looking for a different 0-5v input?
NO IT IS NOT.

why don't you pop your head out of your ass and get with the program?

if you can't go to google, type in "thermistor," hit the "i feel lucky" button, read and comprehend what you see on the page in front of you, then have fun ruining your car because you're obviously waaaaaaaaaaaaaaaaaaay too stupid to be successfully modding it.
ride5000 is offline   Reply With Quote
Old 08-28-2007, 10:45 AM   #24
Worxman02
Scooby Specialist
 
Member#: 66563
Join Date: Jul 2004
Chapter/Region: Tri-State
Location: Jersey
Vehicle:
2002 Impreza WRX
WR Blue Pearl

Default

Quote:
Originally Posted by ride5000 View Post
NO IT IS NOT.

why don't you pop your head out of your ass and get with the program?

if you can't go to google, type in "thermistor," hit the "i feel lucky" button, read and comprehend what you see on the page in front of you, then have fun ruining your car because you're obviously waaaaaaaaaaaaaaaaaaay too stupid to be successfully modding it.
Why don't you try offering a helpful post instead of just calling him an idiot?

I looked up both thermistor and thermocouple and by glancing over the wikipedia articles, it seems that a thermistor measures temp at a linear rate while a thermocouple measures it at a non-linear rate defined by a polynomial.

So all he would have to do is have a controller that used the equation to convert the thermocouple data into a linear output. Or am I missing something too?
Worxman02 is offline   Reply With Quote
Old 08-28-2007, 11:34 AM   #25
ride5000
Scooby Guru
 
Member#: 32792
Join Date: Feb 2003
Chapter/Region: NESIC
Location: lincoln, ri
Vehicle:
2003 GGA MBP
12.9 / 105+

Default

yes, you're missing something too.

a thermISTOR presents a varying RESISTANCE wrt temperature. they may have negative or positive temperature coefficient characteristics. positive means resistance goes up proportionally with temp, and negative means resistance goes down inversely proportionally with temp. basically every conductor in the world exhibits varying resistivity wrt temperature. those elements specifically used for measuring temperature have very carefully chosen characteristics to improve linearity. in order to "read" a temperature from a thermistor you have to read the resistance. a resistance DOES NOT GENERATE A VOLTAGE--it is a passive element that MUST have current flow through it in order to determine the value of resistance.

a thermocouple relies on the seebeck effect which basically states that a temperature gradient along ANY conductor will generate an electrical potential across it... aka VOLTAGE. under normal circumstances this potential is minuscule and is partially offset by attempts to measure it. by using two dissimilar metals the electrical potential can be increased which improves resolution. this is typically called a thermocouple since it is made of two dissimilar metallic conductors COUPLED together. it is important to note that the thermocouple does not "measure" the absolute temperature at the junction itself since there is no temperature gradient there. it actually measures the gradient along the lengths of the connection leads. this is why you must use the proper wire to extend thermocouple leads. you can literally take a thermocouple and measure the voltage across its leads with a garden-variety multimeter. however this voltage must be compensated for in order to arrive at a more accurate temperature. in the old days it used to be performed by placing the "local" ends of the thermocouple wires in an ice bath which will self-maintain at 32*F... a known temperature. then the difference in temps across the lengths of the TCs could be calculated. these days this task is typically performed by a thermocouple interface chip which carries out the "ice bath" compensations by electronically measuring its own internal temperature and using that as a reference.

the bottom line:

thermistors present a changing RESISTANCE wrt temperature and must have a metering circuit which MEASURES RESISTANCE by using ohm's law, which requires imposing a VOLTAGE and a CURRENT through that unknown resistance to arrive at a value.

thermocouples present a changing VOLTAGE SOURCE wrt temperature and must have a metering circuit which MEASURES VOLTAGE. although not strictly necessary, thermocouple "adapters" perform a linearization of the voltage--temperature characteristic and correct for the local junction temperature. typical adapters will output a certain VOLTAGE per DEGREE which can then be logged with some device that MEASURES VOLTAGE.

the OEM ECU is expecting a PURE RESISTANCE between pins B16 and B19 and will use a varying voltage and current to determine it. it will not be useful AT ALL in measuring a voltage. go take a DMM, set it to measure resistance, and put the leads across a battery. tell me if the numbers make any sense whatsoever.

what is necessary--as i've said before--is to use an ecu input that is expecting an analog voltage between 0 and 5vdc. this has already been done by some folks who have used the fuel level sensor input, for example, to directly log a wideband uego controller's 0-5vdc analog output.

is that helpful enough for you?
ride5000 is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Unmarried my V1 AP now no connection to ECU WTF Cobb!! spoolinsti05 AccessPort 23 08-03-2009 03:27 PM
Logger will not connect to ECU....? OMG_WTF Open Source Reflashes 2 07-20-2008 03:36 AM
problems connecting to ecu with green connectors and jumper connected claudefrog Open Source Reflashes 3 05-15-2008 12:58 PM
WTB: Wire That Connects to ECU strikers Private 'Wanted' Classifieds 2 02-20-2008 12:17 AM
Automatic transmission connected to ECU? lordzappo Newbies & FAQs 1 07-18-2005 02:59 AM


All times are GMT -4. The time now is 11:14 AM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2014 Axivo Inc.
Copyright ©1999 - 2014, North American Subaru Impreza Owners Club, Inc.