Change Default Behavior After Following Error - Stop Rapid After Following Error

When a joint has a following error, it seems the default behavior is to move opposite the direction of the motion which caused the following error at, or near, maximum speed. If this ever happens near the limit of the machine, then the joint will hit the limit switch and I'll have to turn off the machine and manually jog the joint off the limit switch. Is there any way to disable this rapid after a following error? Ideally, after a following error I'd like the machine to just turn off output to the joint. It seems like this should be the default behavior. 

A following error should disable the drives. This is a hardware setup issue,
not anything specific to LinuxCNC.

It makes sense that this wouldn't be the default behavior, but I am unsure of what sort of configuration problem, on my end, would cause this. 

I have the enable pin on my 3 servo drives tied to an opto-isolated relay, connected to a general purpose output pin on my 7i77. This output is connected to a fairly long chain of components, but is basically driven by the pid.xxx.enable, or pid.xxx.output for the spindle.  I had trouble getting the 7i77's built in servo drive enables to trigger the opto-isolated relay - I think because of a isolated ground plane, but I can't remember now. 

Should I add an OR component to add joint.x.amp-enable-out to be another source that can trigger the output to my servo drive enable pin? I assumed that the pid.pos.enable pin for the spindle would be triggered at the same time as the amp-enable-out. Should I OR all of the amp-enable-outs, or do they all get triggered at the same time?

Attached is a copy of my HAL. The servo drive enable pins are tied to hm2_5i25.0.7i77.0.0.output-03 on line 330.
 

EDIT: I was looking at an old backup version of my HAL file. The servo drive enable pins are actually connected to spindle.0.on and pid.s.enable. I'm not sure that this is much different than stated above, but worth mentioning I thought.

I added an OR component of spindle.0.on, pid.s.enable and joint1.amp-enable-out to be the switch sources for the servo drive enable pins. I'll see if that fixes the problem. 

I was totally wrong with the previous post. I was looking at the logic for the spindle drive enable, which is totally seperate from the servo drive enable. I was also wrong about using analogena, it is wired up to control the drive axis enable pins. I think at some point in time I was having trouble making analogena work, but it was due to a misunderstading in how the two pins should have been connected.

As you can see, hm2_5i25.0.7i77.0.1.analogena is tied to pid.x.enable, so the servo drives should turn off as soon as a following error occurs. 

Any idea on what the problem could be?

ENA0+ is connected to VFIELD
ENA0- is connected to the trigger pin of an opto-isolated relay
The opto-isolated relay connects GND to INHIBIT-IN on my servo drives (AMC 30A8) when enabled.

There are status LEDs on the servo drives and since day one, I've seen that the drives are inhibited when they should be.

The error seems to be unique to my Z axis and now that I think about it, it may be related to drive tuning. All the axes have the same model servo, but I have a much higher P value for the Z axis since it has almost no mass.

The .ini params for the servo are: