This document covers the following topics:

1. General discussion of Blend Door design.

2. Jeep Grand Cherokee dual control experiments.

3. Ford Explorer “Nail Fix” experiments.

4. Jeep Grand Cherokee Evaluation.

5. Jeep Grand Cherokee Re-circ door replacement process (VIDEO)

6. Ford Explorer 02+ alternative installation process

7. Ford Explorer 02+ Technical Problems and Solution

8. Jeep Grand Cherokee Blend Door fix (VIDEO)

9. Ford Taurus/Mercury Sable Diagnostic Process Video

10. Motor Madness — Information on how to diagnose actuator motor problems

11. Grand Cherokee Dual control Wiring diagrams

12. Motor Calibration Procedure for Jeep Grand Cherokee 99-04 (bottom of page)

13. Ford F150 Video

14. Ford Expedition/Lincoln Navigator Video

15. Dodge RAM 02-08 Video

16. Jeep Grand Cherokee Right Hand Drive Customer Installation set

17. Ford Explorer (96-02), Ranger, Mercury Mountaineer, Mazda installation video

18. Ford Windstar Video

History and Root Cause of Blend
Door Failures

On earlier automobiles there was generally a slide lever
on the HVAC control panel (remember?). This slide lever was connected by a cable
to the blend door and you had the control to slide the lever to full heat, full
AC, or anything in between. In the 90?s most automobile manufacturers moved
from manual blend door

control to mechanized computer control.

Generically, the HVAC system operates by providing air
flow that can be directed through the AC evaporator and the Heater Core. The
AC evaporator is cooled by the engine driven compressor and a flow of refrigerant,
and the Heater Core is heated by a flow of hot radiator fluid through the core.
The air flow is usually directed through the evaporator core and to the blend
door where the air can be directed down two different paths to the vent system.
One path goes directly to the vents, and the other path is directed through
the heater core.

For AC, the compressor is on and air is directed by the
blend door directly into the vent system. For heat, the compressor is off and
air is diverted through the heater core. These are the two extremes, and the
system can split the amount of air flowing down the two paths and control on/off
of the compressor to achieve any desired temperature between the two extremes.
When the blend door fails, control over which path the air flow takes is lost
and dependent on the design of the system and where the broken door falls, you
will lose heat, AC, or both.

When the manufacturers went to electronic control, different
variations of a stepper motor were deployed to provide fine control over the
position of the blend door. Some have variable resistance sensors built into
the motor, or time measurements to gauge how long to run the motor to get to
a desired position, or counting commutator clicks. Independent of the method
used for fine control, the methodology requires that the system “knows” the
extents of movement. The system has to measure where the door stops against
the two end points of movement. The computer does this by driving the motor
to a stall point and measuring a voltage surge or lack of change in the position
monitor input. The DC motors are geared to move slowly with a great deal of
force. The root cause of most blend door problems is the over-stress of this
calibration process, and the plastic doors break over time like bending a coat
hanger back and forth. The computer system generally goes through this calibration
routine every 20 times that the car is started. Time-to-failure is not a traditional
mileage, age, or driving habit function, but just how many times the car is
started. A pizza delivery guy that stops the car for each delivery will fail
much faster than someone commuting to work once a day.

Automobiles were not built with this failure mechanism
in mind and generally the process of replacing the blend door is involved and
expensive. The usual dealer fix is to remove the steering column, pull the dash
panel, evacuate and disconnect the AC system, drain and disconnect the radiator,
and then remove the HVAC plenum (heater box). Once the box is out, it’s fairly
simple to open it and replace the cheap plastic door, and then put the whole
thing back together hoping you don’t screw up something else. HeaterTreater
has developed processes for replacing the blend door without having to remove
the dash and heater box. Cuts are made into the box in non visible areas with
a Dremel tool, and the blend door is replaced with precision machined metal
components. The process results in an easy, cheap fix and replacement components
that are designed to outlast the automobile (much stronger than OEM plastic).

To date, thousands of our HeaterTreater repair kits have
been installed in multiple different models with 100% installation success and
correct operation. We are intent on maintaining superior customer service and
would appreciate your consideration of our products.

 

Jeep Grand Cherokee Dual Blend Door Control Analysis

Jeep Setup:

  • Dual zone control has two motors, one for the driver’s
    side blend door

    and one for the passenger side.
  • Driver’s side is easily accessible through the glove
    box, but passenger motor is located behind the heater box and is inaccessible
    without removing the heater box.
  • The motors both have a limit pin that constrains movement
    between two stops molded into the heater box. Force from the motor during
    calibration is absorbed by the limit pin.
  • The axle connector to the blend door is a 5/16? nylon
    rod with two grooves cut into it to mesh with the door. The length of the
    grooves (effective connection contact) is ?”
  • Connection contact is not symmetric and must be aligned
    for insertion into the door.

Evaluations:

  • First experiment was to attach a metal blend door
    to the axle by inserting the axle connector into a 5/16? ID tube with the
    door attached and cinching the door to the axle using a set screw. Limit pin
    was cut from the connector so that full force was applied to the axle and
    door.
  • Door functioned correctly and was able to withstand
    the force from driving the motor to a stall point. No lateral movement of
    the door was observed, but strain was evident at stall point.
  • Door was planned to go through 25 calibration cycles
    to evaluate reliability of the connection. Incremental slippage was observed
    and total failure occurred at 21 cycles. Picture shows groove worn in the
    axle and complete stripping of connection.
  • The nylon axle is not strong enough to avoid deformation
    from a set screw.
  • Second experiment was to machine a connector that
    would fit the axle pattern with the full 1/2? possible contact length, and
    weld the blend door to a tube with this connector on the end. Connector was
    a tight fit, but would slide on without excessive force required. Again, limit
    pin was cut.
  • First observation was a lateral movement of the door
    at the stall point. The nylon axle had a twisting action at the stall point
    and would force the door away from the motor, causing the door to scrape against
    the far side of the door tunnel. Experiment was done with the driver’s side
    door and no passenger side door in place. Passenger side was added to provide
    resistance to the driver’s side lateral movement.
  • o   

    With both doors in place lateral force was

    observed, but was constrained by the presence of the passenger door.
     
    There

    was still some deflection which reduced the margin between the door and

    edge of the box, but the door moved freely with no scraping.

  • System was taken through 25 calibration cycles with
    no failures, but lateral movement appeared to continue and door resting position
    moved 1/16? away from the motor.
  • Calibration cycling was continued due to small changes
    observed and the axle totally broke at cycle 31. Picture of the broken axle
    is shown. Note that the axle is hollow and the break point was into the hollow
    area just outside the ?” connection area. There didn’t appear to be a gradual
    failure, it just snapped. Temperature was in the 50?s that day, so the fail
    point may take longer in warm

    weather or conversely less time in cold climates.

Conclusions:

The limit pin on the axle is the first point of failure.
When the pin shears off, the force of stalling the motor is transferred to the
blend doors themselves. The second point of failure is the plastic blend door.
The plastic breaks where the connection is made to the axle and the doors fall
off resulting in a loss of temperature control.

A secure connection of a metal blend door to the plastic
axle using a set screw will not withstand long term stresses. The nylon material
is too soft for a secure connection and the set screw will strip the axle.

Using a machined connection to attach to the axle has
three major problems:

  1. The axle is capable of rotating a full 360 degrees,
    but the door has movement between the two stall points of 85 degrees. The
    axle is keyed and must be position within the 85 degree window for the machined
    connector to be inserted. On a broken door, the axle can stop in any position
    and the chances are 85/360 (23.6%) that the axle will fail in a position where
    the connector can be inserted without indexing the motor.

    On the driver’s side this is easy since the motor
    is accessible, but indexing the passenger motor is difficult. The passenger
    side door won’t fit until the axle is indexed with the 85 degrees of door
    movement.

  2. The nylon axle is flexible and will twist when force
    is applied. With a smooth fit connection, this action tends to force the door
    away from the motor. This lateral movement can be constrained against the
    partner door and into the motor connection on the opposite side. The system
    appears to be strong enough to sustain this force, but alignment of the doors
    and force against the opposing motor could be problematic

    over time.
  3. Complete failure of the nylon axle was observed after
    31 cycles. The GC runs the cal cycle every twenty times that the ignition
    switch is turned on. Complete failure would be expected after ~600 engine
    starts. Dependent on frequency of use, we would reasonably expect the fix
    to last 2-18 months. In the picture above, an intact GC axle is shown next
    to the same axle from a Ford F150. The Ford does not have a limit pin and
    the system is designed for the blend doors to absorb the force of the calibration
    (doesn’t work any better than the GC system…the door still breaks). The Ford
    axle is designed to withstand the force of the motor and is 7/16? diameter
    with 1? effective contact to the blend door socket and made of a hardened
    phenolic plastic, compared to a 5/16? diameter 1/2? nylon GC connector. The
    GC axle is designed to operate with force constrained by a limit pin, not
    the axle.

From the above evaluations done in 2006, we at HeaterTreater
elected not to at tempt to develop a solution using the nylon GC axle to withstand
the force of calibration. On the passenger side door, the motor is not accessible
without removal of the heater box and the limit pin would be known to be broken
if the blend door is broken.   With a broken limit pin on
the hidden motor, we do not believe that a reliable connection can be made to
the axle stub and if(when) the nylon axle breaks, the full system must be disassembled
and the back connector replaced.

In our opinion, the only viable fix for the GC is to
connect the blend doors together and use the back axle stub as a pivot point
for the axle. Once the back stub breaks, it cannot be used as a pivot point
and a single control solution is not possible. Our solution solves the blend
door problem and cannot create further damage that would result in a very expensive
repair. On our Ford products, we do rely on the strength of the axle connection
to withstand calibration because it is designed to withstand that force, and
we have gone through 100 stress cycles with no evidence of lateral movement
or breakage. The GC system was not designed to withstand the same forces and
a hollow nylon axle will break over time.

Ford “Nail” Blend Door Fix

This is a description from the HeaterTreater Engineers
on the Internet “Nail” fix for the Ford Blend door. The description will describe
the fix and outline pros and cons of the fix. HeaterTreater does not accept
any liability for the correctness of the enclosed information or the efficacy
of the fix.

The Ford Blend Door has two primary failure patterns
shown in Fig.1. The door typically fails with either the connector splitting
in half or totally breaking off at the top of the door. The Nail fix can only
be done if the failure is of the split connector type.

The first step is to remove the actuator motor and check
to see if you have a split or broken connector (or no problem at all). There
are two motor attachment schemes used on Ford. The older models used screws
mounting the motor directly to the plenum box.  Access the
motor by fully opening the glove box and looking for the motor on top of the
heater box in the upper left hand corner of the glove box opening. With the
screw attachment, you will see two 8mm head screws(note that a 5/16 socket will
also work) in the front of the motor. These are easy to remove. There is a third
screw that is directly behind the white connector socket that has to be removed
with a thin profile ratchet. Remove the three screws and pry up on the motor
to remove it.

The second attachment method used on later models is
a plastic housing that is screwed to the heater box with a plastic pinch attachment
that the motor fits on to. The motor is removed by prying up first on the front
of the motor and when the front connector releases, pry up on the back connector
directly under and behind the white electrical connector. The motor will just
pop out.

Refer to Fig.2 for the next step. Note that the photo
is take on a box that has been removed from an Explorer to get a clear view
of the axle. You can use a small mirror or just feel the hole in the top of
the box where the motor axle extends into the box. Fig.2 shows what the top
of the blend door looks like with the split connector fail mode. The “V” shaped
remainder of the axle will protrude slightly through the axle hole. If the axle
is completely broken, there will be no remnants of the door connector visible.

If the blend door connection looks like Fig. 2, the nail
fix can be attempted. For a completely broken axle, the blend door needs to
be replaced.

The motor connector(shown in Fig. 3) is a “dog house
door” shaped axle. Looking down at the connector, the right angle of the remaining
portion of the connector in Fig. 2 would align with the lower right hand corner
of the dog house.

The end of the axle protruding from the motor will look
like this:

The trick is to get a nail through the axle to contact
the two edges of the remaining half connector as shown below

The nail is placed through the axle, contacting the two
sides of the remainder of the connector. Rotation is achieved by constraining
the broken axle against the heater box opening and turning the motor axle. Note
that the nail is shown going through the axle from the lower left to upper right
of the dog house. This

forces the lower right corner into the connector as it was originally positioned
before the connector broke. The standard fix found on the internet positions
the nail as shown below.

Our best estimation is that this fix will kinda-sorta-almost-sometimes
work.  We have no way to conduct long term reliability assessments,
so cannot make any statements about the life or potential damage that could
be done by the nail fix. One concern is that the axle is not held securely in
the center of the blend door connector. This will allow the axle to have some
deflection from rotating about a constrained center point. The motor gears are
designed to provide rotational force to the axle. If the axle moves out of alignment,
this will produce a “twisting” force in the interface between the plastic gears.
This could cause stripped gears or a meshing problem that could cause an unconstrained
movement of the gears resulting in a “clicking” noise. In our best estimation,
the forces generated by the motor during the calibration cycle that caused the
original problem of a broken axle will most likely also cause this fix to fail
over time. We would consider this to be a temporary fix at best.

We at HeaterTreater have designed a bullet proof fix
for the blend door problem that entails replacing the plastic components with
metal parts designed to withstand the forces in the system. The repair is done
by cutting an opening in the bottom of the heater box, removing the broken door,
and replacing it with new hardware. The fix is simple to install and will provide
a long term solution that will outlast the vehicle. If you choose to attempt
the nail fix, please keep us in mind should the system require further intervention.

Grand Cherokee Heater Treater Fix

First, to understand the solution, you have to understand
the problem. When heater control went from the familiar slide lever with a cable
connect to electronic control, there was a need for a way to modulate door position
just as we used to do with the lever. What Chrysler came up with was a “poor
man’s” stepper motor. The system detects the endpoints and can control position
by running the motor for the percent of the cycle time between endpoints to
hit the right spot. As with all stepper motors, the trick is to find the endpoints
or extents of door movement.

This can be done with a mechanical switch, magnetic switch,
optical system, etc. The method chosen by Chrysler was to run the motor until
it hits a stall point and sense the voltage surge that indicates that the door
has reached the full extent of possible motion. This would have been just great
except for two problems. 1) The software guys never believe the hardware guys
and wrote the program to go through the

calibration routine of detecting the endpoints to initialize the system every
20 times the engine is started. 2) The components are plastic and cannot take
the repeated force of calibrating the system over time. The fail mechanism is
not a function of age, mileage, or driving habits, but just how many times the
car is started. If you run 10 errands a day, the failure will occur sooner than
someone that commutes 200 miles, but only starts the car twice.

The GC is designed with a limit pin molded into the axle
like a “Q” that is limited by two points molded into the heater box. 

This pin is undersized and is the first thing to break. 
Once the pin is broken, the calibration forces are transferred directly to the
blend doors(and re-circ door), and the doors themselves will break over time.
The failure is simply a wear-out mechanism and will occur on all GC’s given
enough time.

Once the doors break, there are two basic ways to address
the fix. 1) replace the doors with stronger components that can withstand the
forces OR 2) reprogram the computer to calibrate less frequently. Note that
this doesn’t actually solve the problem, just delay it out in time. If it took
40 yrs for the system to fail, no one would notice.

Chrysler chose a solution midway between the two basic
fixes. The factory fix is to replace the plastic axle with a new axle that has
a thicker limit pin.  This thicker pin requires that stop points be molded
into the heater box to allow the same extents of motion. The fix requires replacement
of both the doors and the plastic housing. The system will still break over
time, but the intention would be for it to last long enough that it will fall
off the Chrysler radar screen.  If you check some of the forums, you can
find multiple examples of GC owners that have replaced the doors more than once. 

We at HeaterTreater have addressed the problem by constructing
metal components that are engineered to withstand the forces for the life of
the GC. We have two blend door products.  One that can be installed
through the glove box with a simple cut into the housing, and a second(HeaterTreater
Pro) that replaces both axles on a Limited model and maintains dual control. 
The problem with the pro is that you have to completely disassemble the dash
and remove the heater box to get access to the passenger side actuator motor
that is buried behind the box, just like the factory fix.  This product
maintains dual control, but is significantly more difficult to install and we
have found that most customers elect not to go to that much trouble unless there
is also a need to replace the heater core or evaporator. The simple fix
is a compromise between cost and functionality. Easy fix/single control
vs difficult fix/dual control.  We have found that with a 5-10yr old automobile,
most are choosing the cheaper labor costs of the easy fix. 

One of the comments was that the goal should be to do
the “correct” fix.  I fully agree with that.  However, my definition
of “correct” is to do it once with components that will last for the life of
the GC.  Whether a customer chooses a low cost fix that gives up dual control,
or undertakes the more extensive process of removing the heater box, the HeaterTreater
is a one time permanent solution.  With all due respect, the HeaterTreater
is the “correct” fix.


Ford
Taurus/Mercury Sable Diagnostics Video

This video shows how to remove the center control panel to access the blend door motor and diagnose blend door problems.  On this particular automobile, the blend door was OK and the car was repaired by replacing the blend door motor.  However, the video will give a good overview of how to get to the motor, remove it and check for problems.