Jump to content

Fault Detection System


Recommended Posts


Hi, per our Formula SAE rules when using an electronic throttle we must be able to shut off power to the fuel system and throttle in the event that a TPS/APPS/BrakePressureSensor is disconnected. However, the Link ECU has a 2second delay before generating the fault which is far to slow for our rules (needs to be 100msec). From playing around with the ECU it seems like when a sensor is disconnected the voltage the ECU reads slowly drops. I was wondering if some explanation can be provided as to how the fault detection works on a link ECU for the sensors.

Link to comment
Share on other sites

1 hour ago, UVic Formula Racing said:

However, the Link ECU has a 2second delay before generating the fault which is far to slow for our rules (needs to be 100msec).

Im not sure how that works with any ecu, there is a mechanical limit to how fast the throttle blade can be moved due to the inductance, motor power, inertia of all the gears and spring etc.  A typical electronic throttle will take more than 100ms to reach target during fast/large pedal movements.  This means the TP doesnt match target for more than 100ms.  If that statement above is true you would kill the engine everytime you stabbed the throttle.   

 

1 hour ago, UVic Formula Racing said:

From playing around with the ECU it seems like when a sensor is disconnected the voltage the ECU reads slowly drops.

This is a normal electronic phenomenon known as a floating input, when you have a micro pin not connected to vcc or ground then it "floats" somewhere in between.  There are various techniques to work around this but they all reduce the accuracy of the input.  The stronger you pull it up or down the larger the effect on accuracy.  The analog inputs have a weak pull-down to ground so when disconnected it will drop to 0V. 

 

1 hour ago, UVic Formula Racing said:

I was wondering if some explanation can be provided as to how the fault detection works on a link ECU for the sensors.

If the voltage received on the analog input goes above or below the set Error High/Error Low, the the input is considered faulty so the measured value will be ignored and overwritten with the set "Error Value". At the same time a fault code will be generated and the CE light will be activated.  

Link to comment
Share on other sites

  • 2 weeks later...
On 5/27/2022 at 3:16 PM, Adamw said:

Im not sure how that works with any ecu, there is a mechanical limit to how fast the throttle blade can be moved due to the inductance, motor power, inertia of all the gears and spring etc.  A typical electronic throttle will take more than 100ms to reach target during fast/large pedal movements.  This means the TP doesnt match target for more than 100ms.  If that statement above is true you would kill the engine everytime you stabbed the throttle.  

We do have a rule that a 10% deviation between e-throttle target vs TPS that exists for more then one second must be detected and power to the e-throttle be shut off, this is something we have solved by sending the two pieces of data over CAN to our PDU which does the mathematical check and then removes e-throttle power. The 100ms rule is for in the event that a sensor signal open or short circuits (the simplest test being unplugging the sensor) the fault must be detected within 100ms and the e-throttle power shut off. The original post I was inquiring about the ECU's 2 second delay which I now understand from your reply is due to it having a weak pull down.

On 5/27/2022 at 3:16 PM, Adamw said:

This is a normal electronic phenomenon known as a floating input, when you have a micro pin not connected to vcc or ground then it "floats" somewhere in between.  There are various techniques to work around this but they all reduce the accuracy of the input.  The stronger you pull it up or down the larger the effect on accuracy.  The analog inputs have a weak pull-down to ground so when disconnected it will drop to 0V.

This I understand, I had not realized the Link ECU was using a weak pull-down and that's why it was so slow when the sensor was disconnected, I thought maybe some other detection system was being used because of the slow speed. Is the pull-down hardwired into the ECU? I know several other manufactures allow for the resistor to be software select-able.

Through some testing we have realized by adding an additional resistor in a pull-down configuration we are able to effectively modify the 2-second detection speed which solves our issues with the slow detection currently built into the ECU for sensor open/short circuits.

 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

×
×
  • Create New...