forum.cncdrive.com

I have your DG4S – 16035 drives on a controls update for my Shizoka mill. The Fanuc servos have differential encoders (A, B,-A & -B).

Does it make any sense to use your differential line driver that only requires A and B channel inputs or use all four encoder inputs, (A, B,-A & -B).

The reason I'm asking is these Fanuc encoders are getting hard to find and are expensive.

If I only used the A & B channels and for some reason (short or overvoltage) they were damaged, I would still have the other two channels to fall back on. Whereas I would not have a fallback position if I used all four channels and one or two were damaged.

Also If it was ok to use your Differential line drivers using only A and B encoder signals, could I get better performance by not using the Differential line drivers and using all four encoder channels (A, B,-A & -B)?

Thanks,
Chuck

Theoritically it should work.
However differential line drivers have current driven outputs, so it is not guaranteed that the diff.line driver in the encoder will output TTL voltage levels (0/5Volts).
For example the 26LS31 diff.line transmitter guarantees a 2.5V output voltage (defined at 20mA output) and the 26LS32 diff.line receiver guarantees a 2.0V high input voltage level detection, so this combination will work, but depending on what line transmitter does your encoder has it may not work, if the output voltage is lower than 2.0Volts, because in our diff.line driver modules there is 26LS31 transmitter IC.
So, it is likely to work, but not 100% guaranteed, you should measure the voltage levels on the outputs of the encoder to make sure.
And if the high level output voltage of the encoder is lower than 2.0Volts then you will have to find a paired line received anyways or somehow convert the voltage level.

Thank you for the quick response.

So if I understand correctly, you would like me to test my Fanuc encoder to see if the output voltage is greater than 2V.
This test would measure the voltage is on A,B,A-and B- with 5 V applied to the encoder's main 5 V input terminals.

The initial testing on the encoders verified that all four channels gave the proper traces on the oscilloscope, but I did not verify the actual voltage levels. I assumed that since 5V was the input voltage I was also getting 5V output.

I will reconnect the encoder to the scope and read the voltages and get back to you.



Not being electrically inclined, I really appreciate your expertise. I began this controls update using Gecko drives and after several instances of electronic related smoke, Arturo from CNC4PC recommended your drives. He is also recommending the AMT102V encoders but since I'm already wired up with the Fanuc encoders I would like to use them if possible.

Thanks again,
Chuck

I measured the voltage on one of the encoder signals at 3.5V with 5.1V Applied to the encoder. This was from the oscilloscope upon rotating the encoder shaft slowly with my fingers.

So based on this 3.5V it sounds like I can use only two of the Fanuc encoder channels?

Please confirm I'm testing this properly and can proceed.

Thank you,
Chuck

Chuck767 wrote:I measured the voltage on one of the encoder signals at 3.5V with 5.1V Applied to the encoder. This was from the oscilloscope upon rotating the encoder shaft slowly with my fingers.

So based on this 3.5V it sounds like I can use only two of the Fanuc encoder channels?

Please confirm I'm testing this properly and can proceed.

Thank you,
Chuck

Please confirm that I can proceed using only two of the four channels on my Fanuc differential encoder.

Also, with using only two of the four channels from the encoder, am I losing any accuracy?

Thanks,
Chuck

Yes, the mentioned voltage range should work.
I think you thought about resolution, accuracy is an incorrect therm here.
You not loosing resolution, because the inverse channel not carrying further resolution information.
The signals A to _A (and B to _B) have the inverse logic level to eachother, so I think it is easy to see that the second signal carrying no extra resolution information, because it switches on the exact same point as the first one, just with inverse sign.

The idea behind using differential lines for an encoder channel is:

1.) To have the ability to detect if the lines are broken. If there are 2 signals which always have to be in opposite logic level to the other then it is detectable if the output is broken, when the 2 signals are not on the opposite logic level to the other. (Broken wire, broken line driver.)

2.) Because the signals have the opposite logic level the current flow in the wires are also opposite and so if the wires are in a twisted pair then the 2 wires are kind of neutral when viewing from the environment, so the differential wire pair is more immune to external magnetic fields, electrical interferences.