Retrofitting Broken 2002 Fanuc O-MD based mill reusing servo drivers

The computer in my Fanuc based CNC mill died, so I am wishing to update the mill to use Linux CNC. The mill has an A06B-6102-H211#H520 Fanuc spindle ampliier driver module and a 3 axis A06B-6080-H304 servo amplifier module which both seem to be still in good working order.  Both have inputs that are labeled PWM inputs.  I think from researching the manuals and what I could find on the internet that the servo amplifier boards simply take in a 3 phrase like low voltage and low current PWM inputs and they amplify the voltage and current that is provided to the spindle and 3 access drive motors. 

I found that Mesa CNC makes a SPINx1 driver board that could likely geneerate the PWM input signa that the spindle ampliier driver module and a 3 axis input require to get the mill back up and running. The manuals for the Fanuc modules provide a PIN OUT of the PWM input and it lists which pins are for the 12 PWM channels.  The PWM channels are labeled PWMA to PWMF. The interface cable also has 12 COMA to COMF connections too.

The manual for the input connector for the mill's Fanuc Amplifier Modules has 3 signals labeled IRn, ISn and IMCONn which seem to have 3 return signals or grounds which are labeled GDRn, GDSn and GNDn.  I beleive the three Ixxx and Gxxx signals are provided so that the current being supplied to the servos can be meaured which would support detecting if the servos stall so that the PWM module being supplied to the Servos can be turned off quickly and so an alarm can be triggered through Linux CNC too.  In that sense IRn and needs to be connected through a sensing resistor to GDRn and ISn needs to be connected to GDSn through a sensing resistor too.  I think that MCONn needs to be connected through a sensing resistor GNDn.  However the GNDn likely needs to be to the ground of the driver board that provides the PWB inputs.

The installation manual also said relative to that interface  that  To protect the signals from external noise, assign the cable’s central pairs to each pair of current feedback signal and ground signal (i.e., IRn and GDRn, and ISn and GDSn). Otherwise, external noise may result in uneven feed or abnormal sound. For for the benefit of any trying to use my method later of replacing the PC, the caution note in the manual said the PWM signals outputs should be well shielded too. 

I am thinking about using a SPINx1A PWM which is made by Mesa to generate the PWM input  signals for the 3 axes's and the spindle servo. I will also likely purchase a Mesa board wtih ethernet that is compatible with the SPINx1A PWM cards but I have not gotten that far yet.

Q1) I suspect that the servo amplifier inputs need a low voltage such as 10v perhaps.  If anybody can confirm the voltage input, that would be helpful?
Q2) I was unsure as to how to replicate the overcurrent detection and protection circuit that would insure that the power to the servos are turned off in cases like when an over travel happens which causes larger than normal current to happen which can destroy the servos if not building the protection circuit.  Any confirmation of how I am planning to connect the overcurrent would also be welcome.  
3) I would also like to confirm if anyone knows if he PWM signal that is input into the servo amplifiers are basically intended to be 3 phase signals such as what the SPINx1A PWM boards ouput. In that sense I think the resulting drive signals to the servo and spindle are intended to match the input PWM.

For the benefit of any trying to retrograde Fonuc based equipment, the benefit of the approach described here is that the original servos and position sensors are being kept which are expensive. I will post later if the upgrade is successful.  However for the mill I am working on, the feedback of the servo is done in analog by what Fonuc calls a servo amplifier.  All the high voltage and currents are left stock and as long as the over current circuits work as I suspect, it should be simple to build an alarm like circuit to turn the Mesa or other PWB board outputs off in an over current event, doing that and using CNC Linux as the user interface should get my mill back up and running with minimal muss and fuss and minimal cost too.





 

I am working on a similar project , robot six axis, and would like to drive the fanuc servo drives even if I have to bypass the the front and just run the transistors directly, any way any advice on which boards to run would be helpful.

Thanks
   

I'm also retrofitting a fanuc 0MC from that time. I chose to just replace everything by ethercat and delta equipment.
It is a bit more expensive. But much easier to maintain.