Mesa 7i39 + 5i20 + Pittman ELCOM ST BLDC

Question for the experts:

I setup the following combination of components:

Pittman ELCOM ST BLDC Motor - uses 34 pattern for hall effect sensor commutation. 4 poles.
24VDC power supply
USDigital Encoder model E5S-1024-315-IG - no index, no compliments for A or B signal.
two Mesa 7i39-LV - about 2 years old, but never got to work before recent LinuxCNC update. These are on connectors P2 and P3 of the 5i20.
Mesa 5i20 card - works well; I also have a 7i33 that I've used to test other motors and amplifier combination. The 7i33 is not wire into this setup.
using LinuxCNC 2.5
Used PNCconf, and with some tweeking, I have movement from the above setup.

Problem is there is no feedback from the encoder in LinuxCNC, and I'm sure I'm missing the needed HAL components for this. When running the Axis demo g-code, seems to run away, but does reverse directions. When I turn motor shaft, no feedback; display never changes x axis values.

I only have one axis, the X-Axis, wired for now. Motor is not coupled with anything.

Could someone who has worked with both Hall Sensors and Encoder Feedback post a HAL and INI file that they have working? I'm glad I have movement, and really anxious to tie the x-axis into LinuxCNC, before I wire up the remaining motors.

I've attached my HAL and INI files for this setup.

Thanks!
Karl

Board has no jumper for encoder selection; it has just A, B and Index connections, along with power and ground.

First I would verify encoder operation by itself by looking at encoder 0 postion with a halmeter when you rotate the motor shaft. If this does not work, verify the 7I39/encoder connections, and the 7I39/5i20 connections (check dmesg for which 5I20 pins and connectors the encoders are on)

Thank you Peter for the insight.

I am embarrassed to say that the wiring was correct, but the lack of any signal was purely a rookie mistake.
I literally have a hundred of these motors in my scrap bin - all next to new. I happen to grab one for this testing that I had removed the encoder's optical transmissive module; this happened about a year ago when I was troubleshooting an encoder problem. The electrical connector snapped and latched into the housing. I never bothered to check "under the cover", and when I finally removed it to replace what I thought was a blown module, I was shocked to see it missing. Funny that I should meet it half way with a signal.

I just need to get the rotation of the motor feedback to match the proper direction in LinuxCNC - throws a joint 0 error when commanding a positive motion and feeds back a negative. Anyone have a quick solution?

I works now! I ended up using the setp bldc.0.rev true and also assigned a negative number to the scale.

Studied the following link:

www.linuxcnc.org/docview/2.5/html/man/man9/bldc.9.html

Thanks all! I will post my final set up with pictures. Stay tuned.