Re-Circ bypass: 1st attempt results

[jblc]

Some of you may have seen in my various threads, that I've been trying to find a way to keep recirculation from turning off every 5 minutes. (Smoke, smog, etc reasons).
Since documentation says keeping recirc On is contingent on humidity readings, my strategy was to trick the truck into thinking the humidity sensor is giving a "low humidity" reading, so it won't need to turn off recirc.

For background, most recent thread is here. Thanks @airline tech for discussions and helpful documentation

(1) I made a custom harness to go between the humidity sensor and vehicle. I can toggle between a manual humidity input to the vehicle, and the stock humidity sensor output.
I can also adjust the manual humidity i want the truck to read.
I verified that it works, and also calibrated how outputs track with humidity (eg 2.7 V is very humid, 1.5 V is dry)

I tested during a drive, with 1.5, 1.0, and 0.5 V manual override. The output meter built into the custom harness showed that the stock sensor was reporting 2.9 V (although the truck wasn't seeing that), so very humid in the cab.
Windows were fogging up from moisture on this rainy day, with me running very warm air from the HVAC to promote condensation.

(2) Interestingly, no matter how low I manually set the humidity output, the truck STILL turns off recirc every 5 minutes.
This is contrary to the documentation that says the switch is contingent on humidity readings.
I attached the two most pertinent docs that @airline tech posted in that thread; see the PDF markings that they added.

Does anyone have thoughts? Some specific questions:
- Is there more than one humidity sensor?
- is the documentation inaccurate, or is it correct but not complete? (something left out)
- any other ideas on why this might not be working? It should have worked, based on "Climate Control Auto.pdf".

Sponsored

 

[airline tech]

I am attaching another Diagram - It is highlighted with notes (wish I could post color vs, easier to see)
I am also attaching Connector Pinout with notes

After reading your post, yes Pin 4 is Humidity but Pin 2 is Temp, possibly try tapping into both with your setup, to force Low Humidity and Low Temp but wonder if forcing low temp would cause temp control issues (constant heat) - IDK at this point it's worth trying

Note: Due to Pins 3 & 5 - tying into so many sensors is exactly why my co-worker & I decided the easiest way was to focus on the motor (driving it closed)
But it appears you may have a good starting point now, i made notes on the attached for you to see and ref to what wires you tapped into - Note the name of the module (Temp & Humidity Sensor) it is reading both, if you read the decryption yes it says humidity sensor controls the door but also temp and then the sun load sensor
It is a very busy system
and keep in mind that wiring diagrams are sometimes misleading with the nomenclature on things
Like, you pointed out (humidity, its shortened for humidity fan Grd) - that's all it can be as the way I view it, that fan is always running - to draw air in across the sensor - (to get accurate readings)

 

[jblc]

Thanks for you your thoughts

BTW, i still have one more overall idea that may not trigger any FCIM errors, if this sensor- or blend-door-based approach doesn't work: tap into the recirc button. Have a manually selectable mode that automatically resets to recirc the instant it turns off. If the recirc LED turns off, electrically close the circuit of the button immediately, thereby (almost) fully keeping the vehicle in recirc.

(1) Thanks; agreed on possibly temp being a factor too -- i already started changing the harness to include a manual override of that (see (4) below)
Since each vent has its own temp sensor, perhaps the combined temp/humidity is only used for state change (eg recirc, change mixing to be cooler or warmer, etc). Hopefully it won't cause any errors, but I agree with you that it seems a complex system that might not control temp well if tampered with. We'll see shortly

(2)

Due to Pins 3 & 5 - tying into so many sensors is exactly why my co-worker & I decided the easiest way was to focus on the motor (driving it closed)

That could still be a good way to go, if this doesn't pan out. One worry is that the technical docs seem to indicate that FCIM will throw an error (after an accumulated error count) if the door position doesn't match the command, so that's why I went this route ultimately
I'm hoping to avoid mismatched states in the vehicles computers, and changing sensor inputs are one way to avoid that.

(3) Agreed on sunload. I saw the sunload text as well, but for now am assuming that will be a second-order effect (just a speed optimization, via sunload feedforward into the HVAC control loop), compared with internal temp and humidity.
I also was wondering whether the camera can detect fog, and perhaps open the circuit from that. Or perhaps that's sealed against the windshield and won't get foggy -- I don't know which?

(4)

I am attaching another Diagram - It is highlighted with notes (wish I could post color vs, easier to see)
I am also attaching Connector Pinout with notes

Thanks I can clarify your question that you annotated on the doc, where you're wondering whether temp output is combined or separate, and why humidity sensor not shown. The humidity sensor is shown, but it's shown as a box with three wires; the humidity sensor has pwr/gnd/out (it's just not labeled). The temp sensor output is on pin 2, but it's not an output, as much as an input to measure the temp.
Earlier on I did tests on the temp sensor as well, and it's a standard variable resistance thermistor. So, it's just the resistance between temp output, and gnd.
My method to bypass it is to add a digital potentiometer to the harness, so that i can stably and easily control the high resistance (100s of kOhm to ~1 MOhm)

This is certainly a complex system. Hopefully something good will come out of this, for myself and for others who are interested in fixing this...
A Ford FW update would be easiest, of course

 

[airline tech]

So, Just Focusing on the Temp/Humidity Sensor

Pin 2 = Actual Sensor Temp Reading (Imputing to FCIM) - so unaltered voltage
Pin 4 = Actual Sensor Humidity Reading (Imputing to FCIM) - so unaltered voltage

Pin 3 = Full 5 volts power to one side of the Temp/Humidity sensor and tying into (One side) of each of the Actuators and Temp sensors (In the system)

Pin 5 = Varying (5 Volt Return) - Voltage depending on resistance (Temp & Humidity) within the sensor
Higher Voltage = Higher Temp/Humidity
Lower Voltage = Lower Temp/Humidity
This return voltage ties into the other side of each of the actuators and temp sensors
Thus (In-Turn) is the driving (ref) voltage for the door positions to open or close
Example: Push and Resist - 5 Volts pushing on one side and depending on how much resistance on the other (determines door position)

Then the Feedback signal from each door actuator, signals the door is in the commanded position and cuts power (each door actuator has its separate feedback signal)

Is this your findings?

 

[jblc]

So, Just Focusing on the Temp/Humidity Sensor

Pin 2 = Actual Sensor Temp Reading (Imputing to FCIM) - so unaltered voltage
Pin 4 = Actual Sensor Humidity Reading (Imputing to FCIM) - so unaltered voltage

Pin 3 = Full 5 volts power to one side of the Temp/Humidity sensor and tying into (One side) of each of the Actuators and Temp sensors (In the system)

Pin 5 = Varying (5 Volt Return) - Voltage depending on resistance (Temp & Humidity) within the sensor
Higher Voltage = Higher Temp/Humidity
Lower Voltage = Lower Temp/Humidity

Mostly, but not quite:
pin 1: sensor fan motor gnd
pin 2: temp sens resistance input (It's connected to pin 5/Gnd)
- Higher resistance = lower temp
- Lower resistance = higher temp
pin 3: +5V
pin 4: humidity reading
Higher Voltage = Higher Humidity
Lower Voltage = Lower Humidity
pin 5: Gnd (and thermistor resistance reference)
pin 6: sensor fan motor voltage

Then the Feedback signal from each door actuator, signals the door is in the commanded position and cuts power (each door actuator has its separate feedback signal)

Partially yes. The command to the blend door is a feedback command, so it cannot move with a power cut. All positions are an active powered position, relative to a reference command trajectory. It's moved through a geared actuator, so passive motion (eg with a spring) isn't possible.
Errors relative to the reference command trajectory lead to a FCIM accumulator register increase, and after a certain threshold (I think 600 of these events), an overall CEL error code.
So bypassing the motor will cause a discrepancy between the actual position, and commanded position.
Related to that, I think bypassing the position sensor (as a way around the above) might not work, because it still won't match there reference trajectory.

 

[jblc]

Followup:
1) Manually changing the newly-added temp input, and the humidity input, did nothing.
This was in the dark, so no sunload (ie not a factor per documentation).
I tried a range of temp/humidity inputs to FCIM; no change to the 5 minute timer.

2) I'm starting to think their owner's manual and technical documentation is false or intentionally misleading.
What if sensors aren't what stop recirculation, and they just set a hard-coded 5 minute timer?
This is (of course) opposite of what Ford claims in the manual.
Someone might not have done their job, and just said that it works with sensors -- or maybe they intended it to work, but don't care that recirc keeps stopping every 5 minutes.

At this point, there's only one test left: open all windows while running HVAC, and see if it still switches off recirc.
If it still does, that means the computer is NOT actually using sensors, since the point of turning off recirc is to pull in outside air -- but the inside would already have the same properties as outside air.

Sponsored