Odd CA670 behavior

I've installed a CA670, but when arming with the lock button, the door locks are not always activated. After sending an disarm/unlock from the remote, the next lock/arm signal only arms the alarm, and does not actuate the locks. Sending another lock/arm from the remote locks the doors. This behavior is repeatable. This must be some "feature" that can be selected, but I cannot find it documented in the install or user manual. I would like it to work normally, i.e. one press of lock on remote causes the doors to lock and the system to arm. Anyone have a clue on this?

Taht's a new one on me.  What is the year, make, model of the vehicle it's installed in?Cause I'm So white and nerdy...

First Class Certi-fried installer

Vehicle is a new body F-150. The CA670 door output triggers the body control module (double diode isolated) to activate door actuators; door solenoids used for CA670 arm/disarm to retain factory remote use. It's definately wierd. I'll throw a light or meter on it tomorrow to make sure it is the CA670 that is not sending the signal, rather than the BCM ignoring it. The BCM hasn't had time to go into sleep mode.

This is definately a CA670 behavior issue. Removing the CA670 door lock wire and monitoring it with a voltmeter, it pulses to ground only on the second and subsequent push of the lock button on the remote. Does anyone know of a "master reboot" like sequence for the CA670 so I can start from a known state?

Disconnecting the green (positive on lock) wire caused a change in the lock behavior. The lock now signaled correctly. The original (weird) behavior returned when unlocking by remote, then signaling an unlock with the door switch (i.e. on the next lock signal from the remote, the CA670 would not signal for the doors to lock.) The signal was absent from both the blue wire (negative) and green wire (positive.) This probably indicates that there is an issue with signal timing on the arm/disarm wires (hooked to the door lock solenoids,) since with the green disconnected, the lock solenoids no longer pulsed from the CA670 and the behavior went away. Manually pulsing the door unlock between the remote unlock signal and the next remote lock signal caused the behavior to return. Doors are set for 1 second unlock pulse.

I explained it backwards above. I don't have it connected backwards, i.e.: the doors lock on arm (the second time) and unlock on disarm.   Also, the diodes are not the issue, the lock pulse (positive blue, negative green) simply is not generated by the CA670 on the first arm signal. If it was, the diodes are set up correctly to pass the signal to the BCM to lock the door (like it does on the second arm signal.) Disconnecting the blue with black stripe MANUAL DISARM signal from the positive on unlock door lock solenoid signal causes the system to work correctly from the supplied remote (except, of course, the whole point of this wire is to make it work with the factory remote door lock system, so now unlocking the F-150 with the factory remote does not disarm the CA670.) I've restarted the system by disconnecting the battery. I've restarted the system with ignition on then putting the fuses in. The CA670 is very confused about the manual disarm signal. It is hooked up like this: CA670             F150               F150-BCM Green ---|<|--- Pink/Yel ----|<|--- Pink / YELLOW Blue ---|<|--- Pink/LtGn ---|<|--- Pink/Light Green LtBlue/Red ----- Pink/Black        (Manual Arm) LtBlue/Black --- RED / Orange        (Manual Disarm)