Simulating an Evinrude TRIM Sensor

Information about Evinrude I-Command, ICON Pro, and ICON Touch Color Displays
jimh
Posts: 5101
Joined: Fri Oct 09, 2015 12:25 pm
Location: Michigan, Lower Peninsula
Contact:

Simulating an Evinrude TRIM Sensor

Postby jimh » Thu Mar 22, 2018 9:30 am

Simulating an Evinrude TRIM Sensor

--or--

How to not go crazy trying to perform the CAL SENDER procedure

According to some reference material, the resistive sensor on an Evinrude outboard engine that is used for the TRIM position ranges from 10-Ohm (fully up position) to 88-Ohms (fully down position). I wanted to create a variable resistor for bench testing that would produce this same range of resistance. Using a 100-Ohm linear-taper potentiometer and some standard value resistors, I was able to make a sender with a resistance range of 10.6-Ohm to 83-Ohm.

EvinrudeSenderEquivalent.jpg
Using a 100-Ohm potentiometer and two fixed resistors, an equivalent of the Evinrude TRIM sensor is created. The 11-Ohm resistor is actually made by paralleling two 22-Ohm resistors.
EvinrudeSenderEquivalent.jpg (4.75 KiB) Viewed 553 times


I connected my bench-test TRIM sensor to my bench-test ICON Pro RPM gauge using the marked "ANALOG TRIM P1-7" and to the 12-Volt negative circuit (the 0-Voltage reference). Then I performed the CAL SENDER procedure.

The CAL SENDER procedure is found in the menu path:

Normal operation ---> EDIT MENU ---> DATA SOURCES ---> TRIM ---> CAL SENDER

The procedure prompts to "HOLD MODE >1 SEC WHEN TRIM IS FULL DOWN." This calibrates the DOWN position and informs the operator "SETTING SAVED."

Next the procedure prompts "HOLD MODE >1 SEC WHEN TRIM IS FULL UP." This calibrates the UP position and informs the operator "SETTING SAVED."

The TRIM sender is then calibrated.

This sounds very straightforward, but accomplishing this nearly drove me crazy.

After performing this procedure several times, the result was always the same: the instrument showed the TRIM position was "TRIM 0%" no matter where I set the potentiometer. My first thought was I had performed the calibration procedure backwards, that is, I set the calibration for UP when the potentiometer was in at maximum resistance, and for DOWN when the potentiometer was at minimum resistance. I re-did the calibration procedure several times, with no change in the outcome. I was becoming very frustrated. I had gone to rather elaborate lengths to create a simulated TRIM sender from a few standard parts, and the resistance range of my bench-test trim sender was within the limits of the actual sender. But the ICON Pro RPM gauge would not be calibrated to this faux TRIM sender. What was wrong? I could not see anything unusual about the resistive input, nor anything unusual about the calibration procedure, yet the TRIM readings on the gauge were stuck at "TRIM 0%.

I took a break for an hour, and concentrated on some other task. Then I returned to the bench and tried the calibration procedure again--with the same result: "TRIM 0%" as a constant reading on the instrumentation.

As I was just about to go insane, a thought occurred to me: better check to see exactly what analogue input has been designated as the TRIM input. To check this, the menu path is

Normal operation ---> EDIT MENU ---> DATA SOURCES ---> ANALOG ---> #2 P1-7

This menu option allows the physical wire input to be configured for a particular category of data input. The choices are

  1. N/C (No Connection)
  2. FUEL
  3. BALLAST
  4. BAIT WELL
  5. OIL LEVEL
  6. FRESH WATER
  7. GRAY WATER
  8. BLACK WATER
  9. TRIM

I discovered that the P1-7 input had been configured for "N/C", so the instrument was not even looking at this signal for a resistance that was proportional to tank level. No wonder the CAL SENDER function was not working. I changed the P1-7 input DATA SOURCE setting to be "TRIM."

As expected this change solved all the problems. Once again I performed the CAL SENDER procedure, and finally the desired results were obtained.

jimh
Posts: 5101
Joined: Fri Oct 09, 2015 12:25 pm
Location: Michigan, Lower Peninsula
Contact:

Re: Simulating an Evinrude TRIM Sensor

Postby jimh » Thu Mar 22, 2018 9:40 am

I should also confess: the above account is the short version of the story. Here is a more comprehensive account.

I began the bench testing with just the 100-Ohm potentiometer connected to (what I though was) the TRIM analogue input. When the outcome was not successful, my first thought was that my simple substitution of a resistive input in the range 0 to 100-Ohm was not going to work, and I must get closer to the OEM sender's range of 10 to 88-Ohm range. After a few calculations and some searching in my low-wattage resistor junk drawer, I found I had some 22-Ohm and 330-Ohm resistors on hand, and I eventually came up with the more complicated arrangement shown above in the schematic diagram. Of course, I did not get to that configuration on the first iteration. I started with just a 22-Ohm series resistance to prevent the sender resistance from going all the way to 0-Ohms. When that did not work, I went through a few other configurations, eventually getting to the version shown above (and not quite the "final" version).

I did this because I thought that perhaps the ICON Pro RPM gauge was being very fussy about what resistance values it would tolerate for the calibration. I wanted to build an equivalent sender that would remain within the stated range, 10 to 88-Ohms, at all settings.

As I described above, the resistance range was NOT the problem: the real cause was much simpler--the physical input was not properly configured. Of course, this leaves open the question: would the CAL SENDER have worked with just the 100-Ohm potentiometer connected? I am afraid I am not going to test that possibility right now. I have too much time invested in creating the more complex simulation.

In my simulation, the potentiometer has a linear taper. I notice that the TRIM readings have more granularity at the "down" end of the range than at the "up" end. Near the fully-up end of the potentiometer, the slightest movement jumps the display from 95% to 100% TRIM. At the fully-down end and in the midrange, the % TRIM readings change very nicely and very slowly with rotation of the potentiometer control.

jimh
Posts: 5101
Joined: Fri Oct 09, 2015 12:25 pm
Location: Michigan, Lower Peninsula
Contact:

Re: Simulating an Evinrude TRIM Sensor

Postby jimh » Fri Mar 23, 2018 2:22 pm

The final version of my trim sensor equivalent changed the 330-Ohm resistor to a 470-Ohm resistor. The final circuit looks like this:

schematic470.jpg
Evinrude TRIM sensor equivalent circuit.
schematic470.jpg (7.5 KiB) Viewed 529 times


The equivalent circuit consists of two resistors in parallel. On the right side the 470-Ohm resistor is a constant. On the left side the resistance is either:

  1. 0-Ohm + 11-Ohms = 11-Ohms
  2. 100-Ohms + 11-Ohms = 111-Ohms

The parallel resistance is equal

    Rtotal = (R1R1) / (R1+R1)

For the case when the potentiometer is at 100-Ohms, the parallel combination is then

    R = (111×470) / (111+470)  = 90-Ohms

For the case when the potentiometer is at 0-Ohms, the parallel combination is then

    R = (11×470) / (11+470)  = 10.7-Ohms

The earlier version with a 330-Ohm resistor gave values of 83 and 10.6-Ohms, a ratio of 7.8:1 change. The OEM sensor, 10 to 88-Ohms, is a ratio of 8.8:1 change. The 470-Ohm version with 90 and 10.7-Ohms is a ratio of 8.4:1 change, and closer to the OEM range of change.

resistors.jpg
Three resistors and a potentiometer make an equivalent of the Evinrude trim sensor.
resistors.jpg (23.2 KiB) Viewed 509 times