Page 2 of 3 FirstFirst 123 LastLast
Results 11 to 20 of 25

Thread: EGT K type Thermocouple Probe

  1. #11
    Lifetime Member
    Join Date
    Jan 2007
    Posts
    298

    Default

    Mine are T21's from Temprel. You can order them in a huge variety of formats. They look exactly like the ones on the front page but mine have yellow bodies.

    http://www.temprel.com/

    I can't tell you who has the best thermocouples out on the market though. I have only ever owned these. Seem to work ok. This kind of sensor definetly has more latency than a WB. Perhaps choosing the shortest probe lenght you can will help with the fact the sensor has to change temp.


    I can tell you this. The thermocouple won't be a good indication of the exact temp but it will give you the jist of where it is within maybe 5-10C.
    Last edited by 2002_z28_six_speed; December 29th, 2007 at 01:10 PM.

  2. #12
    Lifetime Member
    Join Date
    Jan 2007
    Posts
    298

    Default

    Hey RHS! I had a second thought. You couldn't log this but you might be interested in getting a laser thermometer. Those are highly accurate as far as reading surface temps. Best of all they are relatively cheap and useful for MANY things.

  3. #13
    Lifetime Member killerbee's Avatar
    Join Date
    Jan 2006
    Posts
    1,079

    Default

    do you mean infrared?

    Notoriously innacurate, though they have great repeateable innaccuracy.

    They are highly prone to absorbtivity and emissivity error, the texture and color of the object component is the difference. 2 objects, one shiny aluminum, and the other black rubber, might have the same surface temp, but can show 100 degrees different
    Michael, Systems Engineer 04.5 D-max LLY, Phoenix, Arizona Email
    Custom Tuning Services
    DURAMAX/CUMMINS ECM's TCM's for sale, all years, 20% wholesale discount available
    EFILive's first VGT "Software Wastegate Tune"

  4. #14
    Senior Member
    Join Date
    Dec 2007
    Posts
    134

    Default

    http://www.omega.com has a great technical on TC's and a few helpers for purchasing what you need for a given application. (gas, liquid, etc.)

    I could go into exhausting detail about shielding material, how the tip is presented, special limits of error, table look-up vs. sensitivity curves, bla, bla, bla, HOWEVER most of the that all becomes irrelevant when you consider how the raw voltage signal is acquired, processed and presented to us via EFILive hardware.

    EGT probes:
    http://forum.efilive.com/showthread....&highlight=egt

    It looks as if the way EFILIVE handles TC's could be the primary limiting factor in accuracy and sensor choice.

    Maybe one of the guys could chime in on an official detail as to how TCs are processed to clear things up?

    I'd like to know;
    1. Do they use a NIST table Look-up or an approximate sensitivity of 41 µV/°C method?
    2. What is the resolution of the analog inputs? (Bits)
    3. What is the sample rate, filter and sampling method?

    Once we know this then I can suggest specific TC's as inexpensive as $20 a piece and an accuracy of 1°C/sec.

    --
    -Adam Chant
    Tune it or lose it!
    2007 Saturn SKY 2.4L Automatic - E67
    EFILive Tuned

  5. #15
    Senior Member
    Join Date
    Dec 2007
    Posts
    134

    Default

    Quote Originally Posted by killerbee
    do you mean infrared?

    Notoriously innacurate, though they have great repeateable innaccuracy.

    They are highly prone to absorbtivity and emissivity error, the texture and color of the object component is the difference. 2 objects, one shiny aluminum, and the other black rubber, might have the same surface temp, but can show 100 degrees different
    I have to agree with this 100%. Infrared TC's are generally useless and basically an example of technology not improving accuracy or performance.
    They do have a place in controlled environments where physical contact is impossible. Then and only then they can be very reliable and accurate.

    I deal with TC's daily in my 9to5 with engines and their supporting systems and I have done a few reports on them.
    -Adam Chant
    Tune it or lose it!
    2007 Saturn SKY 2.4L Automatic - E67
    EFILive Tuned

  6. #16
    Lifetime Member
    Join Date
    Jan 2007
    Posts
    298

    Default

    Quote Originally Posted by killerbee
    do you mean infrared?

    Notoriously innacurate, though they have great repeateable innaccuracy.

    They are highly prone to absorbtivity and emissivity error, the texture and color of the object component is the difference. 2 objects, one shiny aluminum, and the other black rubber, might have the same surface temp, but can show 100 degrees different
    I don't think that is what I have. I have a handheld gun with a class 2 laser. It says it is a laser thermometer and the light is definetly visable.

    The laser thermometer and factory IAT sensor were pretty close. My TC ended up -+ about 10 C.


    Quote Originally Posted by scdyne
    http://www.omega.com has a great technical on TC's and a few helpers for purchasing what you need for a given application. (gas, liquid, etc.)

    I could go into exhausting detail about shielding material, how the tip is presented, special limits of error, table look-up vs. sensitivity curves, bla, bla, bla, HOWEVER most of the that all becomes irrelevant when you consider how the raw voltage signal is acquired, processed and presented to us via EFILive hardware.

    EGT probes:
    http://forum.efilive.com/showthread....&highlight=egt

    It looks as if the way EFILIVE handles TC's could be the primary limiting factor in accuracy and sensor choice.

    Maybe one of the guys could chime in on an official detail as to how TCs are processed to clear things up?

    I'd like to know;
    1. Do they use a NIST table Look-up or an approximate sensitivity of 41 µV/°C method?
    2. What is the resolution of the analog inputs? (Bits)
    3. What is the sample rate, filter and sampling method?

    Once we know this then I can suggest specific TC's as inexpensive as $20 a piece and an accuracy of 1°C/sec.

    --

    Sounds like from your experiences+knowledge it might be better to get EGTs connected to an aftermarket controller and then have the controller feed the FSV2 a 0-5V signal? That could reduce error if an answer can not be obtained.

  7. #17
    Lifetime Member
    Join Date
    Jul 2005
    Posts
    702

    Default

    scdyne, you sound like you know your stuff, so just please tell us what's the best thermocouple(without completely breaking the piggy bank) for measuring airtemp in the intake tract. i've looked at hundreds of thermocouples, with different parameters and i couldnt find one that would be a drop-in solution, just much faster and more precise. people use the omega 44005 probe (http://www.omega.com/ppt/pptsc.asp?r...NTS&Nav=temd11) but there's gotta be a faster one out there.

  8. #18
    Senior Member
    Join Date
    Dec 2007
    Posts
    134

    Default

    Update:
    http://forum.efilive.com/showthread.php?t=4593
    I'm guilty of not using the search button. Doh!

    Counts = 850
    Sample Rate = 10 Hz
    Filter = ?? maybe a 5 point average or 10 Hz butterworth ??

    2002_z28_six_speed,
    Sorry the LED laser you see is nothing more than a pointer for the IR laser that is actually reading the temperature.


    Sounds like from your experiences+knowledge it might be better to get EGTs connected to an aftermarket controller and then have the controller feed the FSV2 a 0-5V signal? That could reduce error if an answer can not be obtained.

    solution:Use a higher resolution acquisition system with 16bits of resolution or greater. Also one that uses thermal compensation at each input and a dynamic adjustable Butterworth filter. And for good measure a TC calibrator to verify your system. All said and done you'll spend a thousand dollars on hardware, software and calibrator for an increase in sensitivity from 4°C @ 10Hz to 1°C @ 100 Hz.

    Serious high end mobile acquisition hardware that is often used by GM and BMW (for example) is HBM MGCPlus and/or National Instruments PXI chassis. Each of these systems run in the $50,000+ range when you add the necessary cards and software and provide the ultimate in resolution. (0.025°C @ 9600Hz+)

    We use HBM and NI hardware on a regular basis in our testing, but unless it's a critical safety of test temperature it's measured with a $100 Fluke TC reader or a scanner that scans 64 channels a second.

    I think my point is that while the FSV2 offers a smaller resolution it's accurate enough for most anything we would want to measure on a vehicle with the right TC.

    For IAT, CHT (open Air temps) I would pick exposed element TC assemblies http://www.omega.com/ppt/pptsc.asp?ref=JTIN
    or just simple TC wire
    http://www.omega.com/ppt/pptsc.asp?ref=5TC

    For liquid based temps I would pick an ungrounded element
    http://www.omega.com/ppt/pptsc.asp?ref=JTIN
    or even the NPT threaded units (like the ones listed in other threads)
    http://www.omega.com/ppt/pptsc.asp?ref=TC-NPT
    -Adam Chant
    Tune it or lose it!
    2007 Saturn SKY 2.4L Automatic - E67
    EFILive Tuned

  9. #19
    EFILive Developer Site Admin Blacky's Avatar
    Join Date
    Mar 2003
    Posts
    9,490

    Default

    FlashScan's thermocouples are sampled at 5Hz.
    They are controlled by two MAX6675 K-Type Thromocouple controller chips.
    http://www.maxim-ic.com/getds.cfm?qv_pk=3149
    12 bit resolution, 0.25degC accuracy. 0-1024 degC range.
    Temperature updates are sent from the MAX6675's to FlashScan via an SPI link every 200ms.

    Regards
    Paul
    Before asking for help, please read this.

  10. #20
    Lifetime Member
    Join Date
    Jan 2008
    Posts
    1,169

    Default

    Would these http://www.omega.com/Temperature/pdf/TC-NPT.pdf be suitable for EGT probes?

Page 2 of 3 FirstFirst 123 LastLast

Similar Threads

  1. EGT Thermocouple Question
    By chevyspy in forum Duramax 06 LLY / 06+ LBZ & LMM
    Replies: 1
    Last Post: August 17th, 2007, 09:33 AM
  2. thermocouple not working when in manifold
    By kyle03d in forum Diesel
    Replies: 9
    Last Post: August 14th, 2007, 12:44 AM
  3. Thermocouple input thread
    By Drake in forum General (Petrol, Gas, Ethanol)
    Replies: 14
    Last Post: September 21st, 2005, 03:22 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •