Thermocouplers | FerrariChat

Thermocouplers

Discussion in '348/355' started by Roth, Apr 13, 2018.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, Skimlinks, and others.

  1. Roth

    Roth Formula Junior

    Apr 1, 2016
    433
    Pepsi Generation
    Gentlemen, here we are again at the virtual round table. It seems most things, the 355 are vague. Facts mix with contradicting opinions and circulated informations on the web make it very difficult to pin down the cause of a problem and come with a solution. What is on my mind today is the 355 thermocouples malfunction, the SD Light. I hope to get some facts.


    A thermocouples is a simple device. It’s consists of two dissimilar metals(conductors) bonded together at one end. When heat is applied to this junction, the opposite ends of the metals generate electricity. The amount of electricity generated is proportional to temperature applied. It can be used to measure temperature of equipments. This simple fact makes thermocouples useful in wide variety of applications where high fluctuating temperature sensing is crucial. One particular example is the Ferrari 355 exhaust. My questions for everyone,


    1. What is the temperature threshold of the exhaust, the point where temperature is too high and trigger a SD Light to come on?

    1. Does the ECU sense temperature threshold by the output of the voltage generated by heat measured ?

    1. If the thermocouples are disconnected, does the ECU throws a SD Light when temperature is below the threshold while engine running?

    If we can answer these questions factually, it is plausible and cheap to rid the SD Light forever.




    A lot of folks post positive experiences with after market thermocouples. I hope to understand something. The ECU throws a SD Light when exhaust reaches above a temperature set point. If the temperature reading is below that set point, the SD Light never come on. With that in mind how do we know those after market thermocouples work? We can’t just assume they work because the SD Light doesn’t come on. For argument, the thermocouples can be a piece of metal that doesn’t do anything and because it doesn’t generate a voltage, the ECU doesn’t know the exhaust temperature and never give out a SD Light. The exhaust temp. could be 1000 degrees. The ECU doesn’t know. This can be a dangerous situation. More so than a Nuisance SD Light.
     
  2. fatbillybob

    fatbillybob Two Time F1 World Champ
    Consultant Owner

    Aug 10, 2002
    26,294
    socal
    Stoogepalooza may 19th. Its a long discussion with simple answers many before you have beat.
     
  3. taz355

    taz355 F1 Veteran
    Owner Silver Subscribed

    Feb 18, 2008
    5,942
    Indio Ca/ Alberta
    Full Name:
    Grant
    Bob had a good write up on this i think. It explained everything
     
  4. allandwf

    allandwf Formula Junior

    Aug 27, 2011
    267
    Scotland UK
    Thermocouples, something I know about :) It is a simple device, if the old one is replaced with the same type, J K E etc. (The number relates to the useable range and what dissimilar metals are used,) then manufacturer should not be an issue as the two dissimilar metals used will be the same. Remember though if they are a TC and not a RTD then they are polarity conscious and cannot be connected just anyway round, where a resistance type device can be.
     
  5. m.stojanovic

    m.stojanovic F1 Rookie
    Silver Subscribed

    Dec 22, 2011
    2,743
    Malaysia - KL
    Full Name:
    Miroljub Stojanovic
    The voltage produced by the thermocouples is amplified 100 times by the thermocouple controller ("control station") and this amplified voltage is read by the engine ECU. I have a graph showing the SDL trigger points but I cannot remember whether it is for 348 or 355 (should be the same?).

    Image Unavailable, Please Login
     
    Qavion likes this.
  6. Roth

    Roth Formula Junior

    Apr 1, 2016
    433
    Pepsi Generation
    #6 Roth, Apr 14, 2018
    Last edited: Apr 14, 2018
    Eureka,good find Stojanovic. These numbers are crucial for trouble shooting or make one. Please post the other graph when you find it. I suppose the 1st alarm is the flashing SD and the second is solid.?

    Interesting the signal is amplified 100 folds. The only thing separates real numbers and fractions is a decimal.
     
  7. m.stojanovic

    m.stojanovic F1 Rookie
    Silver Subscribed

    Dec 22, 2011
    2,743
    Malaysia - KL
    Full Name:
    Miroljub Stojanovic
    Yes, 1st flashing, second solid. One calrification:

    On the 348 (2.7 at least), the thermocouple controller (the black one that always fails) does not send any SDL voltage to the engine ECU but it itself lights the SDL, flashing and solid, and it only sends a bank shutdown signal (a simple switch on) to the engine ECU.

    On the 355 (green thermocouple controller) the controller sends the amplified voltage to the engine ECU which, based on that voltage, controls the SDL.

    I only have the diagram that I attached earlier. I believe the SDL points are the same for both systems, whether using the black or the green thermocouple controllers.
     
  8. Roth

    Roth Formula Junior

    Apr 1, 2016
    433
    Pepsi Generation
    [QUOTE="m.stojanovic,ECU.

    On the 355 (green thermocouple controller) the controller sends the amplified voltage to the engine ECU which, based on that voltage, controls the SDL..[/QUOTE]

    Based on your assessment, the 355 ECU is passive. It only send out a SD signal when temperature goes above a set point. This contradicts my experience. Sometimes I get a SD Light within 5 minutes of driving on city road and it goes off eventually. Im curious how long the SD blinks every time it gets a signal. It’s plausible normal operating exhaust temp. and higher temp. set point is such a narrow gap, the couplers occasionally register a high temp. reading. This can happen when coming to a stop. Lack of air moving thru the engine bay raises couplers temp. briefly enough for the ECU to trigger a SD. If this is the case, it can be “fixed” by manipulating the voltage output from the couplers .Im speculating of course. Gonna have to pick one up and play with it. It would be a shocker to learn the ECU multiplier is the problem. Anyhow, I’m hoping someone can answer this. To all with an after market couplers, how do you know they work? The absence of SD Light doesn’t mean they do.
     
  9. taz355

    taz355 F1 Veteran
    Owner Silver Subscribed

    Feb 18, 2008
    5,942
    Indio Ca/ Alberta
    Full Name:
    Grant
    Based on your assessment, the 355 ECU is passive. It only send out a SD signal when temperature goes above a set point. This contradicts my experience. Sometimes I get a SD Light within 5 minutes of driving on city road and it goes off eventually. Im curious how long the SD blinks every time it gets a signal. It’s plausible normal operating exhaust temp. and higher temp. set point is such a narrow gap, the couplers occasionally register a high temp. reading. This can happen when coming to a stop. Lack of air moving thru the engine bay raises couplers temp. briefly enough for the ECU to trigger a SD. If this is the case, it can be “fixed” by manipulating the voltage output from the couplers .Im speculating of course. Gonna have to pick one up and play with it. It would be a shocker to learn the ECU multiplier is the problem. Anyhow, I’m hoping someone can answer this. To all with an after market couplers, how do you know they work? The absence of SD Light doesn’t mean they do.[/QUOTE]

    This is all explained in the wsm. Kinda long explanation but once triggered it depens for how long and then also temp as well as how many times. This is just from memory so may not be exact but also if it is triggered a second time it takes longer to go off.
     
  10. m.stojanovic

    m.stojanovic F1 Rookie
    Silver Subscribed

    Dec 22, 2011
    2,743
    Malaysia - KL
    Full Name:
    Miroljub Stojanovic
    The green thermocouple controller is a simple analogue linear amplifier. Its output voltage (100 times the thermocouple voltage) will vary instantaneously (I have bench tested it), following the variations of the thermocouple voltage. I don't know how exactly the engine ECU interprets the received voltage. It may have a delay circuit which will keep the SDL on (flashing or continuous) for a period of time after the incoming voltage threshold has been exceeded, even though that voltage has only briefly exceeded and is back below the threshold.

    One could simulate a permanent "good" voltage (which otherwise goes from pin C of the controller to pin 44 of the engine ECU) but there is another pin, pin D, also connected to the engine ECU, to pin 11. I have marked it "Free" for my bench test, which was for the purpose of adapting the green controller for my 348, but on the 355 this pin has some function. Possibly just to signal to the engine ECU that the controller is present (connected) so, if one disconnects the controller and simulates voltage of its pin C (to ECU pin 44), the ECU may still recognise non-presence of the controller (nothing coming from the controller's pin D) and shut down the engine.

    Image Unavailable, Please Login
     
  11. Qavion

    Qavion F1 World Champ
    Silver Subscribed

    Feb 20, 2015
    11,484
    Sydney
    Full Name:
    Ian Riddell
    #11 Qavion, Apr 16, 2018
    Last edited: Apr 16, 2018
    Did you see this thread, Miroljub?

    https://www.ferrarichat.com/forum/threads/simple-2-7-slow-down-ecu-replacement-only-for-no-cats.538719/

    "wbt" has this pin disconnected and his F355 2.7 car seems to be suffering no ill effects. The 5.2 car has no wire going to pin D at all.

    (EDIT: Caveat. Pin D may not have any effect on the 2.7 with "wbt"'s simulator fitted because the other output is ok).
     
  12. m.stojanovic

    m.stojanovic F1 Rookie
    Silver Subscribed

    Dec 22, 2011
    2,743
    Malaysia - KL
    Full Name:
    Miroljub Stojanovic
    Yes, I also remember that someone found out that there is nothing coming out of the controller's pin D. One thing I forgot to mention in my previous post: if the thermocouple is disconnected from the green controller, the controller will immediately send the maximum voltage (~4.9 V) to the engine ECU and cause bank shutdown (I have verified this during my bench test).
     
    Qavion likes this.
  13. drbob101

    drbob101 F1 Rookie
    Owner

    Nov 26, 2012
    3,814
    Tinton Falls, NJ
    Full Name:
    Bob Ferraris
    Based on your assessment, the 355 ECU is passive. It only send out a SD signal when temperature goes above a set point. This contradicts my experience. Sometimes I get a SD Light within 5 minutes of driving on city road and it goes off eventually. Im curious how long the SD blinks every time it gets a signal. It’s plausible normal operating exhaust temp. and higher temp. set point is such a narrow gap, the couplers occasionally register a high temp. reading. This can happen when coming to a stop. Lack of air moving thru the engine bay raises couplers temp. briefly enough for the ECU to trigger a SD. If this is the case, it can be “fixed” by manipulating the voltage output from the couplers .Im speculating of course. Gonna have to pick one up and play with it. It would be a shocker to learn the ECU multiplier is the problem. Anyhow, I’m hoping someone can answer this. To all with an after market couplers, how do you know they work? The absence of SD Light doesn’t mean they do.[/QUOTE]

    There are so many threads on this subject so I'll just share a couple tidbits specific to your post.

    We don't know what exactly what the Motronics looks for or what triggers out of norm sdl. We know what triggers them in "normal", meaning all components working optimally and that is shown in the graph posted above. Brand new TCs, brand new SDECUs, everything connected properly, that's how it works. Flashing at 3.5-3.6V and bank shut down at 3.9-4.0V.

    Reality is that transient spikes in the voltages as a result of intermittent conductivity issues will generate sdl and bank shut downs. I have observed this real time and posted videos of it while driving around with a DVOM connected to my SDECU ouputs. I my case it was an intermittent short in the TC lead,

    I have also posted videos of sdl being generated as a result of after market amplifiers installed that start cold at .2V as opposed to OEM SCECUs that start at .5V. These sdl go away after the voltage rises on engine warming.

    If you are dealing with intermittent sdl that you have fully determined ARE NOT the result of overheated cats you are most likely dealing with faulty components of some sort. The culprits are the TC, usually the leads and/or SDECUs. The black potted ones have been known to be more faulty than the green potted ones but the green ones go south as well.

    By getting known working/new components and swapping things around you can usually troubleshoot and fix the issue. In my case I tapped the sdecu output and drove around with a DVOM and saw the problem real time and fixed it. If you have a 5.2 car, sometimes the CEL code that you MAY get will tell you if it is left right or center.
     
  14. Roth

    Roth Formula Junior

    Apr 1, 2016
    433
    Pepsi Generation
    #14 Roth, Apr 17, 2018
    Last edited: Apr 17, 2018


    Gentlemen, thank you for the wealth of information. While the control method is unnecessarily complex, the components are basic as the alphabets. A thermocouples is only slightly more than a piece of metal. It transfers heat variations into electrical signals. I can not get over the idea of shelling hundreds of dollar to replace a device simple as this without understanding why. Maybe there’s more than my limited logic suggests. As I see it, these things shouldn’t cost more than $50 a pop to stamp out. Better yet fix the one that is failing cheaper. Of course one can’t just make or fix something without a thorough understanding of its operating parameters. Fortunately, everyone out there in the 348/355 community knows something. Just need to put the Letters together. Thats the purpose of this thread. So far there’s A,B and C but missing everything from there to Z. The answer to question #3 in my first post will get it to W. And until the experienced of the 348/355 come out and help arrange the Letters, the X,Y and Z remain missing. Looks like I may have to use two month worth of my “happy meal” lunch money on a coupler to play with. I’ll starve temporary but the knowledge I gain will be forever.

    Dr.Bobb, Im told you and others did a write up on the subject. If possible can you verified question #3 in my first post with known good couplers? This will get me to Letter W. I appreciate it. Thanks for understanding.



    Sent from my iPhone using FerrariChat.com mobile app
     
  15. m.stojanovic

    m.stojanovic F1 Rookie
    Silver Subscribed

    Dec 22, 2011
    2,743
    Malaysia - KL
    Full Name:
    Miroljub Stojanovic
    I understand you would like to have the answer to your question:

    "If the thermocouples are disconnected, does the ECU throws a SD Light when temperature is below the threshold while engine running?"

    As I informed in my post above (#12), if a thermocouple is disconnected (from a green controller), the controller will immediately send a high voltage (~4.9 - 5.0 V) to the engine ECU which will then not only light the SDL but also shut down the bank as the ECU will understand this received high voltage as if the exhaust is very hot (even though it is actually not). The engine ECU is only looking at, and acting upon, the voltage received from the thermocouple controller which it compares with its look-up table; it does not know directly how hot is the exhaust.
     
  16. G. Pepper

    G. Pepper Three Time F1 World Champ
    Rossa Subscribed

    Mar 15, 2012
    33,723
    Texas/Colorado
    Full Name:
    George Pepper
    Sub. This is a 360 issue also.
     
  17. drbob101

    drbob101 F1 Rookie
    Owner

    Nov 26, 2012
    3,814
    Tinton Falls, NJ
    Full Name:
    Bob Ferraris
     
  18. Roth

    Roth Formula Junior

    Apr 1, 2016
    433
    Pepsi Generation


    My head hurts trying to piece it together. Either we’re not diagnosing it correctly or the guy who designed this control wasn’t up to task. Maybe both.

    All we have is two components; a sender and a receiver. The sender(TC) sends out voltage based on it’s temperature at the point of contact. The receiver(ECU Controller) determines if the temp. Is normal or too hot by doing nothing or trigger a SD Light. If temperature goes even higher, the ECU Controller shuts down the cylinders corresponding to the heat. This is the simplest logic. You are saying otherwise. According to your statement, a bad wire, loose connection or anything happen upstream that doesn’t register as “normal” voltage, the ECU Controller will throw a SD and shut those cylinders. Electrically, it doesn’t not make sense. Not even remotely logical. This implies the ECU Controller constantly actively monitors voltages it receives. Anything that fall below or above it’s preset “normalcy”, will send a SD and or shut down the affected bank. That means a loose TU wire will cause a SD and shut down a bank. That means loose a connection will will cause a SD and shut down a bank. That means at engine start up, there will be a SD and possibly a bank shut down because the exhaust is cold thus no voltage. To override it there must be a delay control circuit added to the ECU Controller. More over, anything that interruptions the communication between the TU and the ECU Controller even for a millionaire second will cause a SD. Possibly a bank shut down.

    This is my theory how it works or should. The ECU Controller has small relays designed to act at a specific voltage that correspond to temperature it deems too hot by sending aSD Light. There maybe a second relay to shut a bank down at a higher voltage. Once a SD or shut down is triggered, a delay circuit keeps those relays from resetting until a time period has lapsed. If high heat persists, the relays remain close.

    So again, how does anyone know their aftermarket TU works? It’s so easy to fool the ECU Controller. It will never send out a SD. I think I’ll pick one up(green) and get to the bottom of it.



    Sent from my iPhone using FerrariChat.com mobile app
     
  19. Roth

    Roth Formula Junior

    Apr 1, 2016
    433
    Pepsi Generation



    Sent from my iPhone using FerrariChat.com mobile app
     
  20. m.stojanovic

    m.stojanovic F1 Rookie
    Silver Subscribed

    Dec 22, 2011
    2,743
    Malaysia - KL
    Full Name:
    Miroljub Stojanovic
    Don't worry, we will get to the bottom of it. This is the flow of the electrical signals representing the exhaust gas temperature:

    1. The thermocuple ("TC") senses the exhaust temperature and produces millivolts, proportional to the temperature; the TC is connected to, and sends the millivolts to, the TC controller ("TCC"), a small black box sealed with green resin, mounted in the engine compartment.

    2. The TCC receives the millivolts from the TC, amplifies them 100 times (millivolts become volts, proportional to the exhaust temperature) and the TCC sends the volts to the engine ECU. The TCC has another built in feature which is that, if there is less than 5 mV (or zero mV, like on start-up) coming from the TC, the TCC will send fixed 0.5 V (5 mV x 100) to the engine ECU; when the signal from the TC exceeds 5 mV, the TCC will start sending 100 times the actual millivolts received from the TC. The fixed 0.5 V sent by the TCC to the engine ECU, right from the ignition "on", is what prevents the ECU from doing anything stupid.

    3. The engine ECU continuously receives the volts from the TCC, checks how high they are (against its own table of exhaust temperature volts), and acts accordingly - switches the blinking or fixed SDL or shuts down the offending bank.
     
  21. Qavion

    Qavion F1 World Champ
    Silver Subscribed

    Feb 20, 2015
    11,484
    Sydney
    Full Name:
    Ian Riddell
    I'm trying to understand how you can put a diode and a resistor in series on the plug going to the Motronic ECU harness and get the light to go out. Doesn't this suggest that the ECU is sending a voltage to the TCC/TCU? Is the TCU supplying a voltage or a resistance to the ECU?
     
  22. emac

    emac Formula Junior
    Silver Subscribed

    Sep 14, 2014
    851
    upstate SC
    Full Name:
    ernest
    I replaced my black Pots with green ones and TC with ones from DrBob. That fixed my SDL and I never thought about it again. Ignorance is bliss!
     
    taz355 likes this.
  23. m.stojanovic

    m.stojanovic F1 Rookie
    Silver Subscribed

    Dec 22, 2011
    2,743
    Malaysia - KL
    Full Name:
    Miroljub Stojanovic
    The engine ECU (Pin 44) has an internal permanent very low current 5 V on this pin (bias voltage). The voltage from the TCC to this ECU pin has stronger current and, when applied, completely overrides the bias voltage so the ECU reads the TCC voltage and not its internal bias voltage. This ECU feature is there to cause bank shut down if the TCC is disconnected. The circuit with a resistor and a Zener diode, connected between the ECU's pin 44 and ground (with the TCC disconnected), will bring the ECU's bias voltage down to, say, 2.5 V (if a 2.5 V Zener diode is used) and this will simulate a "good" exhaust temperature voltage to the ECU. However, the ECU expects to see a fixed 0.5 V on its pin 44 on engine start-up (which is not present with the Zener diode but 2.5 V instead) so it is not clear whether this solution can fully "full" the ECU.

    It has been reported that some ECU-s do not light the SDL or shut down the bank if the TCC is disconnected meaning that such ECU-s do not have the 5 V bias voltage on the pin 44. Possibly, there were different ECU versions (or different programming) for different markets so some tolerate TCC disconnection and some don't.
     
    Qavion likes this.
  24. m.stojanovic

    m.stojanovic F1 Rookie
    Silver Subscribed

    Dec 22, 2011
    2,743
    Malaysia - KL
    Full Name:
    Miroljub Stojanovic
    Sorry, I meant "fully fool the ECU"
     
  25. drbob101

    drbob101 F1 Rookie
    Owner

    Nov 26, 2012
    3,814
    Tinton Falls, NJ
    Full Name:
    Bob Ferraris
    So as not to confuse acronyms, TCU or TECU is used by most Fcar techs in referring to the Transmission Control Unit, not the thermocouple control unit. There is no right or wrong on this but FBB coined the SDECU (Slow down ECU) acronym and it seems to me to be most appropriate. Just my two cents.

    I have noted the following and I have no proof of it but just an empirical observation. The SDECU was used in 348s and earlier models as well. This was before the 2.7 or 5.2 motronics with more sophisticated central engine control. At some point in my quest to understand how this system works I had a conversation with Dave Helms who many on here know of. At this time I had installed aftermarket voltage amplifiers in my car and other than the beginning voltage of .2V that triggered sdl at cold start up, they worked very well and just fine. These are very simple devices with minimal components on a small ( 1" x 1") board. Having dissected a few SDECUs, the board in them is chock full of stuff and is large. I cannot read the components on that board and give any clue as to what they are supposed to do but in comparing a simple voltage amp to what is in a Ferrari SDECU, it is obvious that there is a big difference.

    Anyway, I asked Dave what the heck all that gadgetry in there does when all we need is a simple amp. He said he didn't know for sure but he thought that a lot of it was left over from 348s when the actual control and shutdown of cylinders and trigger of sdl was handled by the SDECUs. There was no Motronics to do that. This make sense as there are more outputs on the 348 SDECUs connector wise although the physical size and case are identical.

    I think that when 2.7 and later 5.2 came out, Ferrari just adapted the 348 SDECUs for use in 355s using just the voltage amp. capabilities of them and bypassing the actual control outputs that were in 348 versions. This allowed them to use the same physical case and vendor for them with out a lot of fuss. What does this all mean? Not much other than possibly explaining the fragility of these things when the simple devices I have used are far cheaper and work just fine. I am sure that Gothspeeds versions are most likely simple like the off the shelf units I have used. Encasing them in a box that mounts like OEM and is protected with potting compound raises them up to a better level than the things I have used for sure.

    The outputs of a correctly working SDECU are easily measured by tapping the signal wire and running long leads into the cockpit and watching the voltages real time. I have done this many times and it is what lead me to have a complete understanding of how it works and whats wrong when it doesn't work. Roth asked how do you know if the SDECU is working. This is how you know.
     
    Qavion likes this.

Share This Page